Potential of <i>in vivo</i> stress reporter models to reduce animal use and provide mechanistic insights in toxicity studies

Francisco Iñesta Vaquera; Febe Ferro; Michael McMahon; Colin J. Henderson; C. Roland Wolf

doi:10.12688/f1000research.123077.1

Home Browse Potential of in vivo stress reporter models to reduce animal use and...

ALL Metrics

Views

Downloads

Get PDF

Get XML

Export

▬

✚

Brief Report

Potential of in vivo stress reporter models to reduce animal use and provide mechanistic insights in toxicity studies

[version 1; peer review: 2 approved]

Francisco Iñesta Vaquera¹, Febe Ferro¹, Michael McMahon¹, Colin J. Henderson¹, C. Roland Wolf¹

Francisco Iñesta Vaquera¹, Febe Ferro¹, [...] Michael McMahon¹, Colin J. Henderson¹, C. Roland Wolf¹

PUBLISHED 11 Oct 2022

Author details Author details

¹ Systems and Cellular Medicine, University of Dundee, Dundee, DD1 9SY, UK

Francisco Iñesta Vaquera
Roles: Conceptualization, Formal Analysis, Funding Acquisition, Investigation, Methodology, Project Administration, Resources, Supervision, Writing – Original Draft Preparation, Writing – Review & Editing

Febe Ferro
Roles: Writing – Original Draft Preparation, Writing – Review & Editing

Michael McMahon
Roles: Formal Analysis, Investigation, Methodology

Colin J. Henderson
Roles: Funding Acquisition, Project Administration, Supervision, Writing – Review & Editing

C. Roland Wolf
Roles: Conceptualization, Funding Acquisition, Project Administration, Resources, Supervision, Writing – Original Draft Preparation, Writing – Review & Editing

OPEN PEER REVIEW

REVIEWER STATUS

Abstract

Chemical risk assessment ensures protection from the toxic effects of drugs and manmade chemicals. To comply with regulatory guidance, studies in complex organisms are required, as well as mechanistic studies to establish the relevance of any toxicities observed to man. Although in vitro toxicity models are improving, in vivo studies remain central to this process. Such studies are invariably time-consuming and often involve large numbers of animals. New regulatory frameworks recommend the implementation of “smart” in vivo approaches to toxicity testing that can effectively assess safety for humans and comply with societal expectations for reduction in animal use. A major obstacle in reducing the animals required is the time-consuming and complexity of the pathological endpoints used as markers of toxicity. Such endpoints are prone to inter-animal variability, subjectivity and require harmonisation between testing sites. As a consequence, large numbers of animals per experimental group are required. To address this issue, we propose the implementation of sophisticated stress response reporter mice that we have developed. These reporter models provide early biomarkers of toxic potential in a highly reproducible manner at single-cell resolution, which can also be measured non-invasively and have been extensively validated in academic research as early biomarkers of stress responses for a wide range of chemicals at human-relevant exposures. In this report, we describe a new and previously generated models in our lab, provide the methodology required for their use and discuss how they have been used to inform on toxic risk. We propose our in vivo approach is more informative (refinement) and reduces the animal use (reduction) compared to traditional toxicity testing. These models could be incorporated into tiered toxicity testing and used in combination with in vitro assays to generate quantitative adverse outcome pathways and inform on toxic potential.

Keywords

Risk assessment, alternative in vivo methods, chemical toxicity, early biomarkers, in vivo toxicology, mechanistic toxicology, oxidative stress, DNA damage.

Corresponding author: C. Roland Wolf

Competing interests: No competing interests were disclosed.

Grant information: This work was supported by a Medical Research Council Project Grant (MR/R009848/1); SFC/GCRF Block Grant (SFC/AN/02/2018); Natural Environment Research Council Grant (NE/W002213/1).

Copyright: © 2022 Iñesta Vaquera F et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

How to cite: Iñesta Vaquera F, Ferro F, McMahon M et al. Potential of in vivo stress reporter models to reduce animal use and provide mechanistic insights in toxicity studies [version 1; peer review: 2 approved]. F1000Research 2022, 11:1164 (https://doi.org/10.12688/f1000research.123077.1) First published: 11 Oct 2022, 11:1164 (https://doi.org/10.12688/f1000research.123077.1) Latest published: 10 Aug 2023, 11:1164 (https://doi.org/10.12688/f1000research.123077.2)

Abbreviations

AhR: aryl hydrocarbon receptor

hCG: human chorionic gonadotropin

iAs: inorganic arsenic

NBF: neutral-buffered formalin

OCDE: Organisation for Economic Co-operation and Development

PBS: phosphate-buffered saline

PFA: para-formaldehyde

TCDD: Tetrachlorodibenzo-p-dioxin

2-MCPD: 2-Monochloropropane-1,3-diol

3-MCPD: 3-monochloropropane-1,2-diol

Introduction

Development of new manmade chemicals has underpinned an enormous rise in quality of life in areas such as pharmaceuticals, technology, or food production.¹ However, humans and the environment need to be protected from potential short/long-term toxic consequences of exposure to these agents and their derivatives. Regulatory procedures ensure appropriate risk assessments are conducted on novel agents before approval for commercial use. In the absence of human data, studies with multi-cellular experimental models are required to detect and estimate toxic potential. In vivo models, although with limitations, provide the closest estimation of recommended levels of safe exposure in a physiological context. The implementation of such tests must be appropriate, responsible, and humane, by reducing the numbers of animals in a study, refinement of measurements and implementation of less invasive endpoints.² This is achieved by continuous application of the 3Rs principles, particularly when vertebrate models such as rodents are being used and is reflected in OCDE guidelines updates. For example, in acute inhalation studies, death as an endpoint was substituted for clinical signs of toxicity such as hypoactivity, bodyweight loss, and irregular respiration.³ However, clinical signs often reflect major disturbances of homeostasis, which are very unlikely to occur at relevant human exposure levels and do not inform on the mechanisms involved. Additionally, these tests are prone to subjectivity, inter-animal variability, require harmonisation between testing sites and need large numbers of mice for prolonged periods. For example, the OCDE-451 guideline for carcinogenic analysis recommends studies lasting for two years and groups of up to 50 animals for each testing condition.⁴

New approaches are being proposed to reduce reliance on animal testing for regulatory or research purposes. Although progress has been made, most models still have significant limitations, including the inability to recapitulate the cellular complexity of physiological structures of mammals, or the biochemical and physiological differences between simple organisms and the pathways which define chemical toxicity. Despite promising efforts to develop organoids, a harmonized model has been difficult to establish due to issues such as marked cell variability, aberrant activation of signalling pathways and the lack of physiological complexity.⁵ For example, metabolic activation of some compounds can initiate toxicity in a different distal tissue. Also, invertebrates do not provide an alternative to higher organisms because of differences in the pathways that define the mode of action as well as the sensitivity of target cells to chemicals. For example, vertebrates are sensitive to the effects of non-genotoxic carcinogens such as TCDD. However, invertebrates are insensitive, consistent with vertebrates possessing an AhR that binds TCDD-like chemicals, whereas invertebrate AhR homologs do not.⁶ Consequently, the use of complex organisms such as rodents remains a recommended approach for toxicity testing.

To overcome these challenges, the introduction of new methodologies to conduct “smart” in vivo studies with the same level of risk confidence has been proposed.⁷ Consistent with this paradigm, in this communication we propose the incorporation of toxic stress reporter mice into such smart in vivo studies (Table 1). We describe how stress reporter mice can overcome some of the issues surrounding traditional in vivo toxicity studies, accelerate toxicity testing and aid in risk assessment while simultaneously reducing animal use. The rationale is that cells are protected from toxic stress through the induction of specific cytoprotective genes (Figure 1A). Activation of these reporters provides an early biomarker of toxic potential and provides mechanistic information with minimal inter-animal variability. Such models also provide a refined, highly sensitive method of relating dose to toxic effects, a prerequisite for risk assessment. At the molecular level, reporter mice generate easily measurable endpoints in a manner reflecting transcriptional activation of different endogenous genes (Figure 1B). For example, the p21 gene is rapidly activated in response to DNA damage. Therefore, p21 transcriptional activation provides excellent biomarkers of DNA toxic stress. These models carry multiple complementary reporters, including luciferase (non-invasive imaging) as well as β-galactosidase (post-mortem, fine-scale spatial mapping of stressed cells by histochemistry). Moreover, the Hmox1 model expresses hCG, a secreted marker that can be detected in blood. Molecular details for the generation of these lines are described elsewhere (Table 1). We also present a new stress reporter model, HMOX1 stress shadowing mice, where cells exposed to stress are permanently marked and their subsequent fate is tracked (Figure 1C).

Table 1. Stress reporter models generated in our lab.

References to the original publications and subsequent use in toxicologic studies are provided.

Reporter name	Gene	Transcriptional regulation	Toxicity mechanism
HOTT (HMOX1 triple transgenic); NRF2-KO-HOTT⁸^,¹¹^,¹⁴^,¹⁵	Hmox1; Nfe2l2-Hmox1	Nrf2, AP1	Oxidative stress; Inflammation
Hmox1_KI_Cre; ROSA26 multireporter	Hmox1	NRF2, AP1	Oxidative stress; Inflammation (stress shadowing)
p21¹¹^,¹⁷	Cdkn1a	P53	DNA damage
CYP1A1_KI_Cre; ROSA26 multireporter¹⁹^,²⁰	Cyp1a1	AhR	Xenobiotic metabolism (stress shadowing)
Dual specificity protein phosphatase 1	Dusp1	MAPK3/1	Inflammation and apoptosis
P1ROZ¹⁶	hGstp	NRF2, AP1	Oxidative stress, Inflammation
Aldo-keto reductase 1C1	hAkr1c1 promotor	NRF2	Oxidative stress
Interleukin-6	Il-6	NFkB/AP1	Inflammation
CYP2B6-LacZ¹⁸	hCYP2B6	CAR/PXR	Xenobiotic metabolism

Figure 1. A. Key stress cellular pathways activated by toxicants where reporters have been developed. B and C. Schematic representation of how direct stress reporters (A) and stress shadowing reporters (B) work at the molecular level.

Methodology

Animal husbandry

All animals used in our studies were bred and maintained at the University of Dundee. Mice were housed in open-top cages in temperature-controlled rooms at 21°C, with 45-65% relative humidity and 12 h/12 h light/dark cycle. Mice had ad libitum access to food and water. Animals were subjected to standard health and welfare monitoring (twice daily). Environmental enrichment was provided. All animal work was performed under appropriate ASPEL licencing and approved by the Welfare and Ethical use of Animals Committee of the University of Dundee. Experimental design was undertaken following the 3Rs principles and according to the ARRIVE guidelines.

Study design

Data was obtained using either male or female mice, unless specific sex sensitivity was detected. Mice were 10–18 weeks old, heterozygous for the reporter allele, littermates or age-matched (within three weeks). Animals were randomly assigned to treatment groups. The number of mice per group was determined using the G*power calculator. For bioluminescence and hCG expression, previous work confirmed normal distribution of data using a Kolmogorov-Smirnoff test with standard deviation of 0.35. Sample size required to detect a two-fold difference between groups was n=4, using a standard power of 80% and a type I error of 5%. To compare the expression levels between two different groups a t-test was run. Scientists were aware of the nature of the compounds been administered. Analysts can be blinded to sample identity, as was done in our study.

Tissue harvesting and processing for cryo-sectioning

Tissues were rapidly harvested post mortem and fixed for 2 h in 4% PFA (brain, small intestine, skin), 3 h in 10% NBF (liver) or for 18 h in Mirsky’s fixative (other tissues). Tissues were dehydrated for at least 24 h in 30% (w/v) sucrose/PBS buffer at 4°C. Organs were later embedded in Shandon M-1 Embedding Matrix (Thermo Fisher Scientific) in a dry ice-isopentane bath. Sectioning was performed on an OFT5000 cryostat (Bright Instrument Co.). Except for lung (14 μm) and brain (20 μm), all sections were cut at 10 μm thickness (all reagents from SIGMA Aldrich [Merck], unless otherwise indicated).

In situ β-galactosidase staining and histochemistry

Sections were rehydrated in PBS at room temperature (15 min) before incubation overnight at 37°C in X-gal staining solution: phosphate buffer solution (PBS) (pH 7.4) containing 2 mM MgCl₂, 0.01% (w/v) sodium deoxycholate, 0.02% (v/v) Igepal-CA630, 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide and 1mg/ml 5-bromo-4-chloro-3-indolyl β-D-galactopyranoside (Thermo Fisher Scientific). The next day, slides were washed in PBS, counterstained in Nuclear FastRed (Vector Laboratories) (4 min), washed twice in distilled water (2 min) and dehydrated through 70% and 95% ethanol (4.5 and 1 min respectively) before incubation in Histoclear (VWR) (3 min), air-dried and mounted in DPX mountant (all reagents from Sigma Aldrich, except otherwise indicated).

In vivo luciferase imaging and hCG measurement were performed as described before.⁸

Results

We exemplify the utility of our reporter models to address different toxicology scenarios using previously published studies. We describe for first time the HMOX1 “stress shadowing” model, that can be applied to the study of chemical exposure toxicity during development and effects in later life.

Case study 1. Chemicals in the environment

iAs) is present in drinking water and polluted air exposing millions of people.⁹ Epidemiological studies have linked iAs exposure to the development of numerous diseases including cognitive impairment, cardiovascular failure and cancer. One limitation in the study of iAs toxicity is that the concentrations required to identify potential toxic effects on mouse physiology exceed realistic environmental exposure levels.¹⁰ The Hmox1 reporter model provides a robust in vivo approach to detect iAs-associated toxicity in a cell- and tissue-specific manner, and, importantly from an NC3Rs point of view, prior to any overt toxic phenotype. In a series of dose- and time-response studies we identified tissue-specific HOTT reporter activation at environmentally relevant concentrations in studies that lasted a maximum of four weeks. Moreover, we demonstrated that oxidative stress is the major in vivo mechanism of iAs toxicity using pharmacological (antioxidant N-acetylcysteine) and in vivo genetic approaches (Figure 2A). This demonstrates how the models can also be used to rapidly test treatments of disease.¹¹ Similarly, we have used these reporter models to understand the toxicity of additional environmental chemicals, including cadmium (a nephro- and hepato-toxin found in cigarette smoke), “forever” chemicals and diesel emissions (Inesta-Vaquera, manuscripts in preparation). The information generated can be used in risk assessments and inform policy for health interventions.

Figure 2. Demonstration of the utility of stress reporter mice to define mechanisms of toxicity.

A. NRF2_Hmox1 reporter mice.¹¹ iAs 10 mg/kg, heart tissue 20×. B. Hmox1 reporter model.⁸ APAP (Acetaminophen), 300 mg/kg; NAC (N-acetylcysteine), 300 mg/kg; LacZ staining in liver tissue (10×). C. p21 reporter model, Cisplatin 10 mg/kg.¹⁷ LacZ staining in liver tissue (10×).

Case study 2. Chemicals in consumer products

2-MCPD and its isomer 3-MCPD are formed during the refinement of vegetable oils in processed foods. 3-MCPD has been classified as non-genotoxic, and possibly carcinogenic, to humans (category 2B; IARC, 2012). In a conventional two-year long feeding study in rats (group size=50), 3-MCPD-induced nephrotoxicity and testicular toxicity. Moreover, an increased incidence for Leydig cell tumours and kidney tubular adenomas was observed.¹² The mechanism of 3-MCPD toxicity was not determined, although studies suggested that induction of oxidative stress was involved.¹³ A shorter (28-day) study using the HOTT model (only five animals per group), reported nephrotoxicity, testicular and liver toxicity. Strikingly, these studies identified a previously unreported induction of oxidative stress in the brain.¹⁴^,¹⁵ Importantly, the effects of 2-MCPD, which has not been studied in carcinogenicity studies, were less pronounced, suggesting that the molecular toxicity mechanisms of 2-MCPD and 3-MCPD are potentially different.¹⁴^,¹⁵ A similar approach was used to demonstrate the capacity of the food preservative ethoxyquin to induce liver toxicity, using a combination of our hGSTP1 and NRF2-nulled reporter mice.¹⁶

Case study 3. Drug-induced toxicity

The safety of new therapeutics is paramount in the drug development process. Many drugs targeting the DNA damage response currently developed as anti-cancer therapies have the potential to increase toxicity to normal tissues. The use of reporter mice can accelerate the drug development process by determining which agents (and/or their combinations) possess a stronger toxic potential (Figure 2B and C). We have determined the capacity of several DNA-damaging agents (ionizing radiation, cisplatin, etoposide) to induce DNA damage in tissues using reporter mice.¹⁷ We also illustrated the power of this approach to evaluate the safety of new therapeutic regimens involving combination therapies. We demonstrated that a single exposure to olaparib, a PARP inhibitor licenced for the treatment of a number of cancers, caused DNA damage in mucosal cells lining the mouse large intestine, which was exacerbated in this organ, and the kidney by co-administration of ionizing radiation, suggesting that olaparib has carcinogenic potential.

Drug metabolism is a major step in drug development and pharmacology. Humanised models of drug disposition are described in an accompanying report (see Henderson et al.¹⁶). The reporter technology can also be applied to identify activators of transcription factors responsible for induction of hepatic drug metabolising enzymes. For example, the constitutive androstane receptor (CAR) activates genes involved in drug metabolism, lipid homeostasis, and cell proliferation. We have developed a humanised rodent model of hCAR/hPXR containing a CYP2B6-LacZ transgene that allows the screening of large numbers of compounds, and studied the capacity of phenobarbital to activate this pathway in vivo.¹⁸

Stress shadowing technology

We report a novel Hmox1 stress-shadowing model. Stress-shadowing is a fate-mapping approach where Cre is driven from environmentally responsive rather than lineage-restricted genes. We engineered a new ‘driver’ mouse strain in which Cre is expressed from the Hmox1 locus (Hmox1^Cre). We also generated a new reporter/tracer strain ROSA26^{LSL-LacZ-T2A-Luc} in which a floxed silencing element (Lox-STOP-Lox, LSL) prevents expression of a dual reporter element harbouring both LacZ and luciferase. By crossing these ‘driver’ and ‘tracer’ strains, we generated the HMOX1 stress-shadowing mouse. Stress-induced expression of Hmox1 lead to Cre expression and, consequently, excision of the floxed silencing element. This permanently switches on the reporter, providing an enduring molecular mark that can be read at any subsequent time point (Figure 1C). We present a study confirming a low expression of the reporter in tissues of untreated young mice (six weeks old). Interestingly, the reporter becomes activated at older ages in untreated animals, especially in long-lived cells such as cardiomyocytes, neurons, and smooth-muscle cells (Figure 3). We have confirmed that activation is dependent on Cre and there is no ‘leakiness’ of reporter expression in the ROSA26^{LSL-LacZ-T2A-Luc} line (data not shown, but available on request). Therefore, this line can track post-natally the fate of cells exposed during gestation to chemicals that induce Hmox1. This approach was successfully validated in another of our lines where Cre recombinase is driven from the endogenous Cyp1a1 locus, an AhR-responsive gene.¹⁹

Figure 3. Application of the Hmox1 stress shadowing reporter mice to track metabolic changes during ageing (triplicate mice; young = 9 weeks old; aged = 24 weeks old).

LacZ staining performed in indicated tissues.

Discussion

We propose the use of stress reporter models to reduce and refine the use of rodents in toxicity testing. We have demonstrated that reporter expression is a reproducible approach to quantify the potential toxic effects of a test substance, avoiding the variability of clinical scoring in conventional studies. Groups of only four animals are required to detect robust changes in luciferase/hCG expression differences between two treatments. Moreover, in our experience, groups of just three mice are sufficient to obtain reproducible results when measuring ß-galactosidase reporter expression as a qualitative endpoint (absence or presence of signal). This is achieved due to stable and specific reporter expression, which can also capture transient metabolic changes difficult to observe by other methods such as transcriptional profiling, as is the case with HO1.

The reporter approach offers a higher sensitivity in detecting homeostasis disturbance compared to traditional clinical signs of toxicity, paving the way to extensive study refinement. For example, test compound concentrations can be at relevant environmental/clinical exposure scenarios, below doses inducing overall toxicity. Similarly, study durations can be significantly reduced. These in vivo reporters can therefore be used as early toxicity biomarkers for quantitative risk assessment on test substances and provide valuable information on toxicity-associated mechanisms. Moreover, the ubiquitous distribution of these reporters ensures that relevant adverse effects in tissues not directly exposed to the test substance are not missed. Importantly, the assays required to measure reporter expression are inexpensive, fast, reliable and obviate the need for sophisticated laboratory equipment.

The application of reporter mice could provide critical information about potential toxicity in the drug discovery process to avoid unnecessary traditional toxicity studies, and/or initiate studies to establish whether such changes may be of clinical relevance. However, the interpretation of cellular responses observed using these reporters need careful considerations. For example, cisplatin can activate the DNA-damage reporter in liver, but it does not cause hepatotoxicity. The nature of the study will decide the next step in toxicity studies and will still depend on what murine parameters are present. We propose that the generation of stress reporter models on a humanized genetic background will help to overcome some of the potential caveats related to species differences in drug and foreign compound metabolism. For example, in drug toxicity, testing the incorporation of stress reporters into models humanized for metabolic enzymes (e.g. P450s) may further increase the predictive power of in vivo studies in relation to risk to man.

Data availability

Underlying data

figshare: F1000R-NC3R Publication - Potential of in vivo stress reporter models to reduce animal use and provide mechanistic insights in toxicity studies, https://doi.org/10.6084/m9.figshare.c.6049598.v2.²¹

Reporting guidelines

figshare: ARRIVE Essential 10 Checklist for “Potential of in vivo stress reporter models to reduce animal use and provide mechanistic insights in toxicity studies”, https://doi.org/10.6084/m9.figshare.20528352.v1.²²

Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).

Bibliography

1. Oeppen J, Vaupel JW: Demography. Broken limits to life expectancy. Science. 2002; 296(5570): 1029–1031. PubMed Abstract | Publisher Full Text
2. Fischer I, Milton C, Wallace H: Toxicity testing is evolving! Toxicol Res (Camb). 2020; 9(2): 67–80. Publisher Full Text
3. Sewell F, et al.: A global initiative to refine acute inhalation studies through the use of ‘evident toxicity’ as an endpoint: Towards adoption of the fixed concentration procedure. Regul. Toxicol. Pharmacol. 2015; 73(3): 770–779. Publisher Full Text
4. Felter SP, et al.: Hazard identification, classification, and risk assessment of carcinogens: too much or too little? - Report of an ECETOC workshop. Crit. Rev. Toxicol. 2020; 50(1): 72–95. PubMed Abstract | Publisher Full Text
5. Parish ST, et al.: An evaluation framework for new approach methodologies (NAMs) for human health safety assessment. Regul. Toxicol. Pharmacol. 2020; 112: 104592. Publisher Full Text
6. Ankley GT, et al.: Adverse outcome pathways: a conceptual framework to support ecotoxicology research and risk assessment. Environ. Toxicol. Chem. 2010; 29(3): 730–741. PubMed Abstract | Publisher Full Text
7. Ball N, et al.: A framework for chemical safety assessment incorporating new approach methodologies within REACH. Arch. Toxicol. 2022; 96(3): 743–766. PubMed Abstract | Publisher Full Text
8. McMahon M, et al.: Measuring in vivo responses to endogenous and exogenous oxidative stress using a novel haem oxygenase 1 reporter mouse. J. Physiol. 2018; 596(1): 105–127. PubMed Abstract | Publisher Full Text
9. Straif K, et al.: A review of human carcinogens--Part C: metals, arsenic, dusts, and fibres. Lancet Oncol. 2009; 10(5): 453–454. PubMed Abstract | Publisher Full Text
10. Tokar EJ, et al.: Cancer in experimental animals exposed to arsenic and arsenic compounds. Crit. Rev. Toxicol. 2010; 40(10): 912–927. PubMed Abstract | Publisher Full Text
11. Inesta-Vaquera F, et al.: Application of the in vivo oxidative stress reporter Hmox1 as mechanistic biomarker of arsenic toxicity. Environ. Pollut. 2021; 270: 116053. PubMed Abstract | Publisher Full Text
12. Cho WS, et al.: Carcinogenicity study of 3-monochloropropane-1,2-diol in Sprague-Dawley rats. Food Chem. Toxicol. 2008; 46(9): 3172–3177. PubMed Abstract | Publisher Full Text
13. Buhrke T, et al.: Oxidative inactivation of the endogenous antioxidant protein DJ-1 by the food contaminants 3-MCPD and 2-MCPD. Arch. Toxicol. 2018; 92(1): 289–299. PubMed Abstract | Publisher Full Text
14. Schultrich K, et al.: Effects of 2-MCPD on oxidative stress in different organs of male mice. Food Chem. Toxicol. 2020; 142: 111459. PubMed Abstract | Publisher Full Text
15. Schultrich K, et al.: Correlation between 3-MCPD-induced organ toxicity and oxidative stress response in male mice. Food Chem. Toxicol. 2020; 136: 110957. PubMed Abstract | Publisher Full Text
16. Henderson CJ, McLaren AW, Wolf CR: In vivo regulation of human glutathione transferase GSTP by chemopreventive agents. Cancer Res. 2014; 74(16): 4378–4387. PubMed Abstract | Publisher Full Text
17. McMahon M, et al.: Olaparib, Monotherapy or with Ionizing Radiation, Exacerbates DNA Damage in Normal Tissues: Insights from a New p21 Reporter Mouse. Mol. Cancer Res. 2016; 14(12): 1195–1203. PubMed Abstract | Publisher Full Text
18. McMahon M, et al.: Constitutive androstane receptor 1 is constitutively bound to chromatin and ‘primed’ for transactivation in hepatocytes. Mol. Pharmacol. 2019; 95(1): 97–105. PubMed Abstract | Publisher Full Text
19. Schiering C, et al.: Feedback control of AHR signalling regulates intestinal immunity. Nature. 2017; 542(7640): 242–245. PubMed Abstract | Publisher Full Text
20. Villa M, et al.: Aryl hydrocarbon receptor is required for optimal B-cell proliferation. EMBO J. 2017; 36(1): 116–128. PubMed Abstract | Publisher Full Text
21. Inesta F: F1000R-NC3R Publication - Potential of in vivo stress reporter models to reduce animal use and provide mechanistic insights in toxicity studies. figshare. [Dataset]. Collection.2022. Publisher Full Text
22. Inesta F: ARRIVE guidelines checklist - F1000 report. figshare. Dataset.2022. Publisher Full Text

Comments on this article Comments (0)

Version 2

VERSION 2 PUBLISHED 11 Oct 2022

Author details Author details

¹ Systems and Cellular Medicine, University of Dundee, Dundee, DD1 9SY, UK

Febe Ferro
Roles: Writing – Original Draft Preparation, Writing – Review & Editing

Michael McMahon
Roles: Formal Analysis, Investigation, Methodology

Colin J. Henderson
Roles: Funding Acquisition, Project Administration, Supervision, Writing – Review & Editing

C. Roland Wolf
Roles: Conceptualization, Funding Acquisition, Project Administration, Resources, Supervision, Writing – Original Draft Preparation, Writing – Review & Editing

Competing interests

No competing interests were disclosed.

Grant information

This work was supported by a Medical Research Council Project Grant (MR/R009848/1); SFC/GCRF Block Grant (SFC/AN/02/2018); Natural Environment Research Council Grant (NE/W002213/1).

Article Versions (2)

version 2

Revised

Published: 10 Aug 2023, 11:1164

https://doi.org/10.12688/f1000research.123077.2

version 1

Published: 11 Oct 2022, 11:1164

https://doi.org/10.12688/f1000research.123077.1

© 2022 Iñesta Vaquera F et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Download

Export To

metrics

	Views	Downloads
F1000Research	-	-
PubMed Central Data from PMC are received and updated monthly.	-	-

Citations

SEE MORE DETAILS

CITE

how to cite this article

Iñesta Vaquera F, Ferro F, McMahon M et al. Potential of in vivo stress reporter models to reduce animal use and provide mechanistic insights in toxicity studies [version 1; peer review: 2 approved] F1000Research 2022, 11:1164 (https://doi.org/10.12688/f1000research.123077.1)

NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article.

track

receive updates on this article

Track an article to receive email alerts on any updates to this article.

Open Peer Review

Current Reviewer Status: ?

Key to Reviewer Statuses VIEW HIDE

ApprovedThe paper is scientifically sound in its current form and only minor, if any, improvements are suggested

Approved with reservations A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.

Not approvedFundamental flaws in the paper seriously undermine the findings and conclusions

Version 1

VERSION 1

PUBLISHED 11 Oct 2022

Views

Reviewer Report 07 Jul 2023

Heather M. Wallace, School of Medicine, Medical Sciences and Nutrition, Institute of Medical Sciences, University of Aberdeen, Aberdeen, UK

Approved

https://doi.org/10.5256/f1000research.135145.r153304

This is an interesting and well written overview addressing the the use of in vivo reporter models and how these might provide mechanistic detail not available via other methods. The paper discusses toxic risk and the text would benefit from this being clearly defined. The use of "smart" approaches is good but is this being used in the well known sense of smart goals? The authors argue that n = 4 is sufficient for the in vivo studies. It would be good to provide the evidence supporting this statement. It is stated that risk assessment is qualified by biomarkers but this statement needs support and explanation.

A minor correction OCDE should be OECD through out.

Is the work clearly and accurately presented and does it cite the current literature?

Yes
Is the study design appropriate and is the work technically sound?

Yes
Are sufficient details of methods and analysis provided to allow replication by others?

Yes
If applicable, is the statistical analysis and its interpretation appropriate?

Not applicable
Are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions drawn adequately supported by the results?

Yes

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Toxicology and risk assessment particularly of contaminants

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

CITE

Report a concern

Author Response 29 Aug 2023

Francisco Inesta-Vaquera, Systems and Cellular Medicine, University of Dundee, Dundee, DD1 9SY, UK

29 Aug 2023

Author Response

1.- The paper discusses toxic risk and the text would benefit from this being clearly defined.

We now define “Toxic risk” as the “likelihood of a chemical causing an ... Continue reading 1.- The paper discusses toxic risk and the text would benefit from this being clearly defined.

We now define “Toxic risk” as the “likelihood of a chemical causing an adverse health effect”.

2.- The use of "smart" approaches is good but is this being used in the well-known sense of smart goals?

The term “smart approaches” has been defined as “tier 3 non-conventional in vivo toxicity testing systems” (7).

3.- The authors argue that n = 4 is sufficient for the in vivo studies. It would be good to provide the evidence supporting this statement.

As per above reviewer one’s comment 1.

4.- It is stated that risk assessment is qualified by biomarkers but this statement needs support and explanation.

A new sentence has been added to the manuscript: “These in vivo reporters can therefore be used as early biomarkers of toxicity for quantitative risk assessment by detecting adverse effects at low levels of exposure.”

5.- A minor correction OCDE should be OECD through out.

This has been corrected.
1.- The paper discusses toxic risk and the text would benefit from this being clearly defined.

We now define “Toxic risk” as the “likelihood of a chemical causing an adverse health effect”.

2.- The use of "smart" approaches is good but is this being used in the well-known sense of smart goals?

The term “smart approaches” has been defined as “tier 3 non-conventional in vivo toxicity testing systems” (7).

3.- The authors argue that n = 4 is sufficient for the in vivo studies. It would be good to provide the evidence supporting this statement.

As per above reviewer one’s comment 1.

4.- It is stated that risk assessment is qualified by biomarkers but this statement needs support and explanation.

A new sentence has been added to the manuscript: “These in vivo reporters can therefore be used as early biomarkers of toxicity for quantitative risk assessment by detecting adverse effects at low levels of exposure.”

5.- A minor correction OCDE should be OECD through out.

This has been corrected.
Competing Interests: No competing interests were disclosed. Close
Report a concern
Respond or Comment

COMMENTS ON THIS REPORT

Author Response 29 Aug 2023

Francisco Inesta-Vaquera, Systems and Cellular Medicine, University of Dundee, Dundee, DD1 9SY, UK

29 Aug 2023

Author Response

1.- The paper discusses toxic risk and the text would benefit from this being clearly defined.

We now define “Toxic risk” as the “likelihood of a chemical causing an ... Continue reading 1.- The paper discusses toxic risk and the text would benefit from this being clearly defined.

We now define “Toxic risk” as the “likelihood of a chemical causing an adverse health effect”.

2.- The use of "smart" approaches is good but is this being used in the well-known sense of smart goals?

The term “smart approaches” has been defined as “tier 3 non-conventional in vivo toxicity testing systems” (7).

3.- The authors argue that n = 4 is sufficient for the in vivo studies. It would be good to provide the evidence supporting this statement.

As per above reviewer one’s comment 1.

4.- It is stated that risk assessment is qualified by biomarkers but this statement needs support and explanation.

A new sentence has been added to the manuscript: “These in vivo reporters can therefore be used as early biomarkers of toxicity for quantitative risk assessment by detecting adverse effects at low levels of exposure.”

5.- A minor correction OCDE should be OECD through out.

This has been corrected.
1.- The paper discusses toxic risk and the text would benefit from this being clearly defined.

We now define “Toxic risk” as the “likelihood of a chemical causing an adverse health effect”.

2.- The use of "smart" approaches is good but is this being used in the well-known sense of smart goals?

The term “smart approaches” has been defined as “tier 3 non-conventional in vivo toxicity testing systems” (7).

3.- The authors argue that n = 4 is sufficient for the in vivo studies. It would be good to provide the evidence supporting this statement.

As per above reviewer one’s comment 1.

4.- It is stated that risk assessment is qualified by biomarkers but this statement needs support and explanation.

A new sentence has been added to the manuscript: “These in vivo reporters can therefore be used as early biomarkers of toxicity for quantitative risk assessment by detecting adverse effects at low levels of exposure.”

5.- A minor correction OCDE should be OECD through out.

This has been corrected.
Competing Interests: No competing interests were disclosed. Close
Report a concern

Views

Reviewer Report 06 Jun 2023

Mark Miller, University/BHF Centre for Cardiovascular Sciences, University of Edinburgh, Edinburgh, UK

Approved

https://doi.org/10.5256/f1000research.135145.r174523

This is a short review with original data outlining the use of the investigators’ mouse models with reporters of stress pathways. Data is shown demonstrating the presence of the reporter in histological sections from mice treated with various xenobiotics or aged animals for pathways that include oxidative stress and DNA damage, with prevention by anti-oxidants or genetic knockout. The authors conclude that these models could be used to detect activation of key stress pathways at more environmentally relevant exposures of stimulants without inducing overt toxicity, while offering the potential to reduce the length of exposure and number of animals per group. The paper is clearly written and the conclusions are reasonable for the examples shown in the paper.

Minor comments:

Please state the source of the SD=0.35 used for the power calculations. I accept that the biological variability of the reporter expression is very low, however, n=4/group is very low for statistical analysis and risk of type I and II errors. The authors should comment on this. Similarly, the authors may also note that the presence of the reporter will likely be one variable measured in any detailed study using these mice but inevitably other functional and pathological parameters will be measured as well – these will still require higher n-numbers. Nonetheless, experiments with n=4 with reporter expression alone would still have great value as a screening and dose-finding tool.
It would be useful to show some quantitative data as well as the qualitative images. In particular, this could be used to demonstrate the variability of the reporter parameter or that more environmentally-relevant doses of stressors can be used, i.e. a dose response curve to a stimulant required to induce the reporter in these mice compared to a typical parameter used assess toxicity in normal mice to demonstrate.
If space permits, I would like to see more data from the examples outlined in the narrative text. The authors mention that these models can be used for non-invasive imaging – this data in particular would be good to show.
Please add a little further discussion on the physiological relevance of these pathways. For example, playing devil’s advocate, a dose of xenobiotic/drug may stimulate the reporter, however, this could be functionally harmless and it is only from greater/longer exposure that other biological responses become impaired causing a detrimental effect on health. In which case the reporter may give a false impression of the dose of exposure that is harmful, and potentially cause a drug to fail screening despite being promising in terms of risk-benefit in moderated doses.
Is there any impact on breeding or known non-specific effect of the reporter compared to the wild-type mice?
Ensure that figure legends are fully explained, e.g. describe why APAP and cisplantin are used – many readers will not be familiar with the type of stress/toxicity caused by these agents.

Is the work clearly and accurately presented and does it cite the current literature?

Yes
Is the study design appropriate and is the work technically sound?

Yes
Are sufficient details of methods and analysis provided to allow replication by others?

Partly
If applicable, is the statistical analysis and its interpretation appropriate?

Partly
Are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions drawn adequately supported by the results?

Yes

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: in vivo rodent models, air pollution, cardiovascular disease

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

CITE

Report a concern

Author Response 29 Aug 2023

Francisco Inesta-Vaquera, Systems and Cellular Medicine, University of Dundee, Dundee, DD1 9SY, UK

29 Aug 2023

Author Response

1. Please state the source of the SD=0.35 used for the power calculations. I accept that the biological variability of the reporter expression is very low, however, n=4/group is very ... Continue reading 1. Please state the source of the SD=0.35 used for the power calculations. I accept that the biological variability of the reporter expression is very low, however, n=4/group is very low for statistical analysis and risk of type I and II errors. The authors should comment on this. Similarly, the authors may also note that the presence of the reporter will likely be one variable measured in any detailed study using these mice but inevitably other functional and pathological parameters will be measured as well – these will still require higher n-numbers. Nonetheless, experiments with n=4 with reporter expression alone would still have great value as a screening and dose-finding tool.

The SD=0.35 is based on previous experience on using reporter animals (references 8, 17). We have extensively characterized the basal and inducible reporter activation across sex, age and tissues of the mentioned reporter mice. A sentence has been added to clarify this point. “The changes in reporter activation are highly reproducible, and therefore we are confident 4 animals/group is sufficient to avoid Type I and II errors. We acknowledge that where additional end points are to be measured simultaneously, this may be required to use a higher number of mice.”

2. It would be useful to show some quantitative data as well as the qualitative images. In particular, this could be used to demonstrate the variability of the reporter parameter or that more environmentally-relevant doses of stressors can be used, i.e. a dose response curve to a stimulant required to induce the reporter in these mice compared to a typical parameter used assess toxicity in normal mice to demonstrate.

Table 1 provides multiple examples of where quantitative luciferase reporter activity (quantitative data) was measured after chemical exposure (references 8 and 17). For example, when mice were exposed to 30 mg/kg of haemin, a 10-fold induction of bioluminescence due to Hmox1 reporter luciferase activity was observed (ref. 8).

3. If space permits, I would like to see more data from the examples outlined in the narrative text. The authors mention that these models can be used for non-invasive imaging – this data in particular would be good to show.

Due to space limitations we have summarized data on non-invasive imaging in table 1, including references where the images are shown.

4. Please add a little further discussion on the physiological relevance of these pathways. For example, playing devil’s advocate, a dose of xenobiotic/drug may stimulate the reporter, however, this could be functionally harmless and it is only from greater/longer exposure that other biological responses become impaired causing a detrimental effect on health. In which case the reporter may give a false impression of the dose of exposure that is harmful, and potentially cause a drug to fail screening despite being promising in terms of risk-benefit in moderated doses.

We now include the following sentence in the discussion of the manuscript:
“Cellular responses to toxic agents activate adaptive responses that protect cells from cell death. However, when these pathways are overcome toxicity results. Therefore, the activation of these pathways does not necessarily correlate with toxicity, but are indicative of toxic insults”.

5. Is there any impact on breeding or known non-specific effect of the reporter compared to the wild-type mice?

No, both heterozygous and homozygous reporter mice are viable and fertile.

6. Ensure that figure legends are fully explained, e.g. describe why APAP and cisplantin are used – many readers will not be familiar with the type of stress/toxicity caused by these agents.

Figure legend 2 now reads:
“Figure 2. Demonstration of the utility of stress reporter mice to define mechanisms of toxicity. A. NRF2_Hmox1 reporter mice [11]. iAs, a naturally occurring metalloid, was dosed at 10mg/kg, p.o. and 24h heart tissue (20x) was processed for LacZ staining. B. Hmox1 reporter model [8]. Acetaminophen (APAP; in excess, such as an overdose, results in rapid depletion of glutathione within the liver), 300mg/kg p.o.; NAC (N-acetylcysteine), 300mg/kg; LacZ staining in liver tissue (10x). C. p21 reporter model was tested with the anti-cancer drug Cisplatin 10mg/kg, which causes damage to nuclear and mitochondrial DNA [17]. LacZ staining in liver tissue (10x).”
1. Please state the source of the SD=0.35 used for the power calculations. I accept that the biological variability of the reporter expression is very low, however, n=4/group is very low for statistical analysis and risk of type I and II errors. The authors should comment on this. Similarly, the authors may also note that the presence of the reporter will likely be one variable measured in any detailed study using these mice but inevitably other functional and pathological parameters will be measured as well – these will still require higher n-numbers. Nonetheless, experiments with n=4 with reporter expression alone would still have great value as a screening and dose-finding tool.

The SD=0.35 is based on previous experience on using reporter animals (references 8, 17). We have extensively characterized the basal and inducible reporter activation across sex, age and tissues of the mentioned reporter mice. A sentence has been added to clarify this point. “The changes in reporter activation are highly reproducible, and therefore we are confident 4 animals/group is sufficient to avoid Type I and II errors. We acknowledge that where additional end points are to be measured simultaneously, this may be required to use a higher number of mice.”

2. It would be useful to show some quantitative data as well as the qualitative images. In particular, this could be used to demonstrate the variability of the reporter parameter or that more environmentally-relevant doses of stressors can be used, i.e. a dose response curve to a stimulant required to induce the reporter in these mice compared to a typical parameter used assess toxicity in normal mice to demonstrate.

Table 1 provides multiple examples of where quantitative luciferase reporter activity (quantitative data) was measured after chemical exposure (references 8 and 17). For example, when mice were exposed to 30 mg/kg of haemin, a 10-fold induction of bioluminescence due to Hmox1 reporter luciferase activity was observed (ref. 8).

3. If space permits, I would like to see more data from the examples outlined in the narrative text. The authors mention that these models can be used for non-invasive imaging – this data in particular would be good to show.

Due to space limitations we have summarized data on non-invasive imaging in table 1, including references where the images are shown.

4. Please add a little further discussion on the physiological relevance of these pathways. For example, playing devil’s advocate, a dose of xenobiotic/drug may stimulate the reporter, however, this could be functionally harmless and it is only from greater/longer exposure that other biological responses become impaired causing a detrimental effect on health. In which case the reporter may give a false impression of the dose of exposure that is harmful, and potentially cause a drug to fail screening despite being promising in terms of risk-benefit in moderated doses.

We now include the following sentence in the discussion of the manuscript:
“Cellular responses to toxic agents activate adaptive responses that protect cells from cell death. However, when these pathways are overcome toxicity results. Therefore, the activation of these pathways does not necessarily correlate with toxicity, but are indicative of toxic insults”.

5. Is there any impact on breeding or known non-specific effect of the reporter compared to the wild-type mice?

No, both heterozygous and homozygous reporter mice are viable and fertile.

6. Ensure that figure legends are fully explained, e.g. describe why APAP and cisplantin are used – many readers will not be familiar with the type of stress/toxicity caused by these agents.

Figure legend 2 now reads:
“Figure 2. Demonstration of the utility of stress reporter mice to define mechanisms of toxicity. A. NRF2_Hmox1 reporter mice [11]. iAs, a naturally occurring metalloid, was dosed at 10mg/kg, p.o. and 24h heart tissue (20x) was processed for LacZ staining. B. Hmox1 reporter model [8]. Acetaminophen (APAP; in excess, such as an overdose, results in rapid depletion of glutathione within the liver), 300mg/kg p.o.; NAC (N-acetylcysteine), 300mg/kg; LacZ staining in liver tissue (10x). C. p21 reporter model was tested with the anti-cancer drug Cisplatin 10mg/kg, which causes damage to nuclear and mitochondrial DNA [17]. LacZ staining in liver tissue (10x).”
Competing Interests: No competing interests were disclosed. Close
Report a concern
Respond or Comment

COMMENTS ON THIS REPORT

Author Response 29 Aug 2023

Francisco Inesta-Vaquera, Systems and Cellular Medicine, University of Dundee, Dundee, DD1 9SY, UK

29 Aug 2023

Author Response

1. Please state the source of the SD=0.35 used for the power calculations. I accept that the biological variability of the reporter expression is very low, however, n=4/group is very ... Continue reading 1. Please state the source of the SD=0.35 used for the power calculations. I accept that the biological variability of the reporter expression is very low, however, n=4/group is very low for statistical analysis and risk of type I and II errors. The authors should comment on this. Similarly, the authors may also note that the presence of the reporter will likely be one variable measured in any detailed study using these mice but inevitably other functional and pathological parameters will be measured as well – these will still require higher n-numbers. Nonetheless, experiments with n=4 with reporter expression alone would still have great value as a screening and dose-finding tool.

The SD=0.35 is based on previous experience on using reporter animals (references 8, 17). We have extensively characterized the basal and inducible reporter activation across sex, age and tissues of the mentioned reporter mice. A sentence has been added to clarify this point. “The changes in reporter activation are highly reproducible, and therefore we are confident 4 animals/group is sufficient to avoid Type I and II errors. We acknowledge that where additional end points are to be measured simultaneously, this may be required to use a higher number of mice.”

2. It would be useful to show some quantitative data as well as the qualitative images. In particular, this could be used to demonstrate the variability of the reporter parameter or that more environmentally-relevant doses of stressors can be used, i.e. a dose response curve to a stimulant required to induce the reporter in these mice compared to a typical parameter used assess toxicity in normal mice to demonstrate.

Table 1 provides multiple examples of where quantitative luciferase reporter activity (quantitative data) was measured after chemical exposure (references 8 and 17). For example, when mice were exposed to 30 mg/kg of haemin, a 10-fold induction of bioluminescence due to Hmox1 reporter luciferase activity was observed (ref. 8).

3. If space permits, I would like to see more data from the examples outlined in the narrative text. The authors mention that these models can be used for non-invasive imaging – this data in particular would be good to show.

Due to space limitations we have summarized data on non-invasive imaging in table 1, including references where the images are shown.

4. Please add a little further discussion on the physiological relevance of these pathways. For example, playing devil’s advocate, a dose of xenobiotic/drug may stimulate the reporter, however, this could be functionally harmless and it is only from greater/longer exposure that other biological responses become impaired causing a detrimental effect on health. In which case the reporter may give a false impression of the dose of exposure that is harmful, and potentially cause a drug to fail screening despite being promising in terms of risk-benefit in moderated doses.

We now include the following sentence in the discussion of the manuscript:
“Cellular responses to toxic agents activate adaptive responses that protect cells from cell death. However, when these pathways are overcome toxicity results. Therefore, the activation of these pathways does not necessarily correlate with toxicity, but are indicative of toxic insults”.

5. Is there any impact on breeding or known non-specific effect of the reporter compared to the wild-type mice?

No, both heterozygous and homozygous reporter mice are viable and fertile.

6. Ensure that figure legends are fully explained, e.g. describe why APAP and cisplantin are used – many readers will not be familiar with the type of stress/toxicity caused by these agents.

Figure legend 2 now reads:
“Figure 2. Demonstration of the utility of stress reporter mice to define mechanisms of toxicity. A. NRF2_Hmox1 reporter mice [11]. iAs, a naturally occurring metalloid, was dosed at 10mg/kg, p.o. and 24h heart tissue (20x) was processed for LacZ staining. B. Hmox1 reporter model [8]. Acetaminophen (APAP; in excess, such as an overdose, results in rapid depletion of glutathione within the liver), 300mg/kg p.o.; NAC (N-acetylcysteine), 300mg/kg; LacZ staining in liver tissue (10x). C. p21 reporter model was tested with the anti-cancer drug Cisplatin 10mg/kg, which causes damage to nuclear and mitochondrial DNA [17]. LacZ staining in liver tissue (10x).”
1. Please state the source of the SD=0.35 used for the power calculations. I accept that the biological variability of the reporter expression is very low, however, n=4/group is very low for statistical analysis and risk of type I and II errors. The authors should comment on this. Similarly, the authors may also note that the presence of the reporter will likely be one variable measured in any detailed study using these mice but inevitably other functional and pathological parameters will be measured as well – these will still require higher n-numbers. Nonetheless, experiments with n=4 with reporter expression alone would still have great value as a screening and dose-finding tool.

The SD=0.35 is based on previous experience on using reporter animals (references 8, 17). We have extensively characterized the basal and inducible reporter activation across sex, age and tissues of the mentioned reporter mice. A sentence has been added to clarify this point. “The changes in reporter activation are highly reproducible, and therefore we are confident 4 animals/group is sufficient to avoid Type I and II errors. We acknowledge that where additional end points are to be measured simultaneously, this may be required to use a higher number of mice.”

2. It would be useful to show some quantitative data as well as the qualitative images. In particular, this could be used to demonstrate the variability of the reporter parameter or that more environmentally-relevant doses of stressors can be used, i.e. a dose response curve to a stimulant required to induce the reporter in these mice compared to a typical parameter used assess toxicity in normal mice to demonstrate.

Table 1 provides multiple examples of where quantitative luciferase reporter activity (quantitative data) was measured after chemical exposure (references 8 and 17). For example, when mice were exposed to 30 mg/kg of haemin, a 10-fold induction of bioluminescence due to Hmox1 reporter luciferase activity was observed (ref. 8).

3. If space permits, I would like to see more data from the examples outlined in the narrative text. The authors mention that these models can be used for non-invasive imaging – this data in particular would be good to show.

Due to space limitations we have summarized data on non-invasive imaging in table 1, including references where the images are shown.

4. Please add a little further discussion on the physiological relevance of these pathways. For example, playing devil’s advocate, a dose of xenobiotic/drug may stimulate the reporter, however, this could be functionally harmless and it is only from greater/longer exposure that other biological responses become impaired causing a detrimental effect on health. In which case the reporter may give a false impression of the dose of exposure that is harmful, and potentially cause a drug to fail screening despite being promising in terms of risk-benefit in moderated doses.

We now include the following sentence in the discussion of the manuscript:
“Cellular responses to toxic agents activate adaptive responses that protect cells from cell death. However, when these pathways are overcome toxicity results. Therefore, the activation of these pathways does not necessarily correlate with toxicity, but are indicative of toxic insults”.

5. Is there any impact on breeding or known non-specific effect of the reporter compared to the wild-type mice?

No, both heterozygous and homozygous reporter mice are viable and fertile.

6. Ensure that figure legends are fully explained, e.g. describe why APAP and cisplantin are used – many readers will not be familiar with the type of stress/toxicity caused by these agents.

Figure legend 2 now reads:
“Figure 2. Demonstration of the utility of stress reporter mice to define mechanisms of toxicity. A. NRF2_Hmox1 reporter mice [11]. iAs, a naturally occurring metalloid, was dosed at 10mg/kg, p.o. and 24h heart tissue (20x) was processed for LacZ staining. B. Hmox1 reporter model [8]. Acetaminophen (APAP; in excess, such as an overdose, results in rapid depletion of glutathione within the liver), 300mg/kg p.o.; NAC (N-acetylcysteine), 300mg/kg; LacZ staining in liver tissue (10x). C. p21 reporter model was tested with the anti-cancer drug Cisplatin 10mg/kg, which causes damage to nuclear and mitochondrial DNA [17]. LacZ staining in liver tissue (10x).”
Competing Interests: No competing interests were disclosed. Close
Report a concern

Comments on this article Comments (0)

Version 2

VERSION 2 PUBLISHED 11 Oct 2022

Open Peer Review

Reviewer Status

Reviewer Reports

	Invited Reviewers
	1	2
Version 2 (revision) 10 Aug 23	read
Version 1 11 Oct 22	read	read

Mark Miller, University of Edinburgh, Edinburgh, UK
Heather M. Wallace, University of Aberdeen, Aberdeen, UK

Comments on this article

All Comments(0)

Add a comment

Browse by related subjects

Back to all reports

Reviewer Report

6 Views

18 Aug 2023 | for Version 2

Mark Miller, University/BHF Centre for Cardiovascular Sciences, University of Edinburgh, Edinburgh, UK

6 Views Cite this report Responses(0)

Approved

I have no further comments.

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

in vivo rodent models, air pollution, cardiovascular disease

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Respond to this report

Responses (0)

Back to all reports

Reviewer Report

8 Views

07 Jul 2023 | for Version 1

Heather M. Wallace, School of Medicine, Medical Sciences and Nutrition, Institute of Medical Sciences, University of Aberdeen, Aberdeen, UK

8 Views Cite this report Responses(1)

Approved

Is the work clearly and accurately presented and does it cite the current literature?

Yes
Is the study design appropriate and is the work technically sound?

Yes
Are sufficient details of methods and analysis provided to allow replication by others?

Yes
If applicable, is the statistical analysis and its interpretation appropriate?

Not applicable
Are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions drawn adequately supported by the results?

Yes

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

Toxicology and risk assessment particularly of contaminants

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Respond to this report

Responses (1)

Author Response

29 Aug 2023

Francisco Inesta-Vaquera, Systems and Cellular Medicine, University of Dundee, Dundee, DD1 9SY, UK

1.- The paper discusses toxic risk and the text would benefit from this being clearly defined.

We now define “Toxic risk” as the “likelihood of a chemical causing an adverse health effect”.

2.- The use of "smart" approaches is good but is this being used in the well-known sense of smart goals?

The term “smart approaches” has been defined as “tier 3 non-conventional in vivo toxicity testing systems” (7).

3.- The authors argue that n = 4 is sufficient for the in vivo studies. It would be good to provide the evidence supporting this statement.

As per above reviewer one’s comment 1.

4.- It is stated that risk assessment is qualified by biomarkers but this statement needs support and explanation.

A new sentence has been added to the manuscript: “These in vivo reporters can therefore be used as early biomarkers of toxicity for quantitative risk assessment by detecting adverse effects at low levels of exposure.”

5.- A minor correction OCDE should be OECD through out.

This has been corrected.

View more View less

Competing Interests

No competing interests were disclosed.

Back to all reports

Reviewer Report

10 Views

06 Jun 2023 | for Version 1

Mark Miller, University/BHF Centre for Cardiovascular Sciences, University of Edinburgh, Edinburgh, UK

10 Views Cite this report Responses(1)

Approved

Please state the source of the SD=0.35 used for the power calculations. I accept that the biological variability of the reporter expression is very low, however, n=4/group is very low for statistical analysis and risk of type I and II errors. The authors should comment on this. Similarly, the authors may also note that the presence of the reporter will likely be one variable measured in any detailed study using these mice but inevitably other functional and pathological parameters will be measured as well – these will still require higher n-numbers. Nonetheless, experiments with n=4 with reporter expression alone would still have great value as a screening and dose-finding tool.
It would be useful to show some quantitative data as well as the qualitative images. In particular, this could be used to demonstrate the variability of the reporter parameter or that more environmentally-relevant doses of stressors can be used, i.e. a dose response curve to a stimulant required to induce the reporter in these mice compared to a typical parameter used assess toxicity in normal mice to demonstrate.
If space permits, I would like to see more data from the examples outlined in the narrative text. The authors mention that these models can be used for non-invasive imaging – this data in particular would be good to show.
Please add a little further discussion on the physiological relevance of these pathways. For example, playing devil’s advocate, a dose of xenobiotic/drug may stimulate the reporter, however, this could be functionally harmless and it is only from greater/longer exposure that other biological responses become impaired causing a detrimental effect on health. In which case the reporter may give a false impression of the dose of exposure that is harmful, and potentially cause a drug to fail screening despite being promising in terms of risk-benefit in moderated doses.
Is there any impact on breeding or known non-specific effect of the reporter compared to the wild-type mice?
Ensure that figure legends are fully explained, e.g. describe why APAP and cisplantin are used – many readers will not be familiar with the type of stress/toxicity caused by these agents.

Is the work clearly and accurately presented and does it cite the current literature?

Yes
Is the study design appropriate and is the work technically sound?

Yes
Are sufficient details of methods and analysis provided to allow replication by others?

Partly
If applicable, is the statistical analysis and its interpretation appropriate?

Partly
Are all the source data underlying the results available to ensure full reproducibility?

Yes
Are the conclusions drawn adequately supported by the results?

Yes

Competing Interests

No competing interests were disclosed.

Reviewer Expertise

in vivo rodent models, air pollution, cardiovascular disease

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Respond to this report

Responses (1)

Author Response

29 Aug 2023

Francisco Inesta-Vaquera, Systems and Cellular Medicine, University of Dundee, Dundee, DD1 9SY, UK

1. Please state the source of the SD=0.35 used for the power calculations. I accept that the biological variability of the reporter expression is very low, however, n=4/group is very low for statistical analysis and risk of type I and II errors. The authors should comment on this. Similarly, the authors may also note that the presence of the reporter will likely be one variable measured in any detailed study using these mice but inevitably other functional and pathological parameters will be measured as well – these will still require higher n-numbers. Nonetheless, experiments with n=4 with reporter expression alone would still have great value as a screening and dose-finding tool.

The SD=0.35 is based on previous experience on using reporter animals (references 8, 17). We have extensively characterized the basal and inducible reporter activation across sex, age and tissues of the mentioned reporter mice. A sentence has been added to clarify this point. “The changes in reporter activation are highly reproducible, and therefore we are confident 4 animals/group is sufficient to avoid Type I and II errors. We acknowledge that where additional end points are to be measured simultaneously, this may be required to use a higher number of mice.”

2. It would be useful to show some quantitative data as well as the qualitative images. In particular, this could be used to demonstrate the variability of the reporter parameter or that more environmentally-relevant doses of stressors can be used, i.e. a dose response curve to a stimulant required to induce the reporter in these mice compared to a typical parameter used assess toxicity in normal mice to demonstrate.

Table 1 provides multiple examples of where quantitative luciferase reporter activity (quantitative data) was measured after chemical exposure (references 8 and 17). For example, when mice were exposed to 30 mg/kg of haemin, a 10-fold induction of bioluminescence due to Hmox1 reporter luciferase activity was observed (ref. 8).

3. If space permits, I would like to see more data from the examples outlined in the narrative text. The authors mention that these models can be used for non-invasive imaging – this data in particular would be good to show.

Due to space limitations we have summarized data on non-invasive imaging in table 1, including references where the images are shown.

4. Please add a little further discussion on the physiological relevance of these pathways. For example, playing devil’s advocate, a dose of xenobiotic/drug may stimulate the reporter, however, this could be functionally harmless and it is only from greater/longer exposure that other biological responses become impaired causing a detrimental effect on health. In which case the reporter may give a false impression of the dose of exposure that is harmful, and potentially cause a drug to fail screening despite being promising in terms of risk-benefit in moderated doses.

We now include the following sentence in the discussion of the manuscript:
“Cellular responses to toxic agents activate adaptive responses that protect cells from cell death. However, when these pathways are overcome toxicity results. Therefore, the activation of these pathways does not necessarily correlate with toxicity, but are indicative of toxic insults”.

5. Is there any impact on breeding or known non-specific effect of the reporter compared to the wild-type mice?

No, both heterozygous and homozygous reporter mice are viable and fertile.

6. Ensure that figure legends are fully explained, e.g. describe why APAP and cisplantin are used – many readers will not be familiar with the type of stress/toxicity caused by these agents.

Figure legend 2 now reads:
“Figure 2. Demonstration of the utility of stress reporter mice to define mechanisms of toxicity. A. NRF2_Hmox1 reporter mice [11]. iAs, a naturally occurring metalloid, was dosed at 10mg/kg, p.o. and 24h heart tissue (20x) was processed for LacZ staining. B. Hmox1 reporter model [8]. Acetaminophen (APAP; in excess, such as an overdose, results in rapid depletion of glutathione within the liver), 300mg/kg p.o.; NAC (N-acetylcysteine), 300mg/kg; LacZ staining in liver tissue (10x). C. p21 reporter model was tested with the anti-cancer drug Cisplatin 10mg/kg, which causes damage to nuclear and mitochondrial DNA [17]. LacZ staining in liver tissue (10x).”

View more View less

Competing Interests

No competing interests were disclosed.

Alongside their report, reviewers assign a status to the article:

Approved - the paper is scientifically sound in its current form and only minor, if any, improvements are suggested

Approved with reservations - A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.

Not approved - fundamental flaws in the paper seriously undermine the findings and conclusions

[1] 1. Oeppen J, Vaupel JW: Demography. Broken limits to life expectancy. Science. 2002; 296(5570): 1029–1031. PubMed Abstract | Publisher Full Text

[2] 2. Fischer I, Milton C, Wallace H: Toxicity testing is evolving! Toxicol Res (Camb). 2020; 9(2): 67–80. Publisher Full Text

[3] 3. Sewell F, et al.: A global initiative to refine acute inhalation studies through the use of ‘evident toxicity’ as an endpoint: Towards adoption of the fixed concentration procedure. Regul. Toxicol. Pharmacol. 2015; 73(3): 770–779. Publisher Full Text

[4] 4. Felter SP, et al.: Hazard identification, classification, and risk assessment of carcinogens: too much or too little? - Report of an ECETOC workshop. Crit. Rev. Toxicol. 2020; 50(1): 72–95. PubMed Abstract | Publisher Full Text

[5] 5. Parish ST, et al.: An evaluation framework for new approach methodologies (NAMs) for human health safety assessment. Regul. Toxicol. Pharmacol. 2020; 112: 104592. Publisher Full Text

[6] 6. Ankley GT, et al.: Adverse outcome pathways: a conceptual framework to support ecotoxicology research and risk assessment. Environ. Toxicol. Chem. 2010; 29(3): 730–741. PubMed Abstract | Publisher Full Text

[7] 7. Ball N, et al.: A framework for chemical safety assessment incorporating new approach methodologies within REACH. Arch. Toxicol. 2022; 96(3): 743–766. PubMed Abstract | Publisher Full Text

[8] 8. McMahon M, et al.: Measuring in vivo responses to endogenous and exogenous oxidative stress using a novel haem oxygenase 1 reporter mouse. J. Physiol. 2018; 596(1): 105–127. PubMed Abstract | Publisher Full Text

[9] 9. Straif K, et al.: A review of human carcinogens--Part C: metals, arsenic, dusts, and fibres. Lancet Oncol. 2009; 10(5): 453–454. PubMed Abstract | Publisher Full Text

[10] 10. Tokar EJ, et al.: Cancer in experimental animals exposed to arsenic and arsenic compounds. Crit. Rev. Toxicol. 2010; 40(10): 912–927. PubMed Abstract | Publisher Full Text

[11] 11. Inesta-Vaquera F, et al.: Application of the in vivo oxidative stress reporter Hmox1 as mechanistic biomarker of arsenic toxicity. Environ. Pollut. 2021; 270: 116053. PubMed Abstract | Publisher Full Text

[12] 12. Cho WS, et al.: Carcinogenicity study of 3-monochloropropane-1,2-diol in Sprague-Dawley rats. Food Chem. Toxicol. 2008; 46(9): 3172–3177. PubMed Abstract | Publisher Full Text

[13] 13. Buhrke T, et al.: Oxidative inactivation of the endogenous antioxidant protein DJ-1 by the food contaminants 3-MCPD and 2-MCPD. Arch. Toxicol. 2018; 92(1): 289–299. PubMed Abstract | Publisher Full Text

[14] 14. Schultrich K, et al.: Effects of 2-MCPD on oxidative stress in different organs of male mice. Food Chem. Toxicol. 2020; 142: 111459. PubMed Abstract | Publisher Full Text

[15] 15. Schultrich K, et al.: Correlation between 3-MCPD-induced organ toxicity and oxidative stress response in male mice. Food Chem. Toxicol. 2020; 136: 110957. PubMed Abstract | Publisher Full Text

[16] 16. Henderson CJ, McLaren AW, Wolf CR: In vivo regulation of human glutathione transferase GSTP by chemopreventive agents. Cancer Res. 2014; 74(16): 4378–4387. PubMed Abstract | Publisher Full Text

[17] 17. McMahon M, et al.: Olaparib, Monotherapy or with Ionizing Radiation, Exacerbates DNA Damage in Normal Tissues: Insights from a New p21 Reporter Mouse. Mol. Cancer Res. 2016; 14(12): 1195–1203. PubMed Abstract | Publisher Full Text

[18] 18. McMahon M, et al.: Constitutive androstane receptor 1 is constitutively bound to chromatin and ‘primed’ for transactivation in hepatocytes. Mol. Pharmacol. 2019; 95(1): 97–105. PubMed Abstract | Publisher Full Text

[19] 19. Schiering C, et al.: Feedback control of AHR signalling regulates intestinal immunity. Nature. 2017; 542(7640): 242–245. PubMed Abstract | Publisher Full Text

[20] 20. Villa M, et al.: Aryl hydrocarbon receptor is required for optimal B-cell proliferation. EMBO J. 2017; 36(1): 116–128. PubMed Abstract | Publisher Full Text

[21] 21. Inesta F: F1000R-NC3R Publication - Potential of in vivo stress reporter models to reduce animal use and provide mechanistic insights in toxicity studies. figshare. [Dataset]. Collection.2022. Publisher Full Text

[22] 22. Inesta F: ARRIVE guidelines checklist - F1000 report. figshare. Dataset.2022. Publisher Full Text

Potential of in vivo stress reporter models to reduce animal use and provide mechanistic insights in toxicity studies

Abstract

Keywords

Abbreviations

Introduction

Table 1. Stress reporter models generated in our lab.

Figure 1. A. Key stress cellular pathways activated by toxicants where reporters have been developed. B and C. Schematic representation of how direct stress reporters (A) and stress shadowing reporters (B) work at the molecular level.

Methodology

Animal husbandry

Study design

Tissue harvesting and processing for cryo-sectioning

In situ β-galactosidase staining and histochemistry

Results

Case study 1. Chemicals in the environment

Figure 2. Demonstration of the utility of stress reporter mice to define mechanisms of toxicity.

Case study 2. Chemicals in consumer products

Case study 3. Drug-induced toxicity

Stress shadowing technology

Figure 3. Application of the Hmox1 stress shadowing reporter mice to track metabolic changes during ageing (triplicate mice; young = 9 weeks old; aged = 24 weeks old).

Discussion

Data availability

Underlying data

Reporting guidelines

Bibliography

Comments on this article Comments (0)

Open Peer Review

Comments on this article Comments (0)

Open Peer Review

Reviewer Status

Reviewer Reports

Comments on this article

Browse by related subjects

Competing Interests Policy

Stay Updated