1. Introduction
Congenital adrenal hyperplasia (CAH) is a disorder of adrenal steroidogenesis, classified as one of the most frequent recessive inherited disorders. It is primarily caused by 21-hydroxylase deficiency (21-OHD) resulting from
CYP21A2 mutations. There are two forms of CAH: classic CAH, which is severe and life-threatening, attributable to the partial and complete ablation of residual enzyme activity and resulting in a lack of cortisol, and non-classic CAH, which is mild and caused by
CYP21A2 mutations that retain almost 20–50% residual enzyme activity and sustained cortisol biosynthesis [
1]. While classic and non-classic CAH differ in age of onset, symptoms, and severity, all patients primarily manifest with hyperandrogenism [
2]. Homozygous and compound heterozygous point mutations are typical for 21-OHD. In recent years, over 100 mutations in the
CYP21A2 gene have been identified to be associated with either the simple virilizing or the severe salt-wasting phenotypes of the classic form, as well as the milder non-classic phenotypes [
3,
4]. An example of such a mutation is the p.R484Q (Arg484Gln), resulting from a DNA alteration at position c.1451G > A in the exon 10 of the
CYP21A2 gene [
5]. The missense mutation p. R484Q resides in the C-terminal end of the 21-hydroxylase (21-OH) protein. In vitro assays with the point mutation R484Q revealed that there is a low but measurable activity of 21-OH-R484Q for the substrate conversion of 17-hydroxyprogesterone (1.1%; SD 0.7%) and progesterone (3.8%; SD 1.9%), the two natural substrates of 21-OH [
6]. Previous research on
CYP21 mutations has indicated that in vitro activities between 1–14% correlate with the simple virilizing phenotype, thereby suggesting that the R484Q missense mutation is part of
CYP21 mutations that are present in the simple virilizing form of the disorder [
7,
8,
9].
Individuals diagnosed with CAH require continuous medical intervention to compensate for deficient hormones and to regulate elevated androgen levels. Previous studies have shown that complications have persisted due to exposure to high doses of glucocorticoids and the continued high level of adrenal androgens [
10]. Over the last few years, there has been substantial improvement in the development of new drugs and therapies for CAH. Concomitantly, the testing of new therapies poses a challenge as the existing models have limited efficacy and are quite restrictive in scope for the disorder [
11].
The first genetically modified mouse model for 21-hydroxylase deficiency was the C57BL/10SnSlc-H-2
aw18 knockout mouse strain that showed increased levels of progesterone, typical to CAH. These H-2
aw18 mice were, however, difficult to handle and showed early postnatal mortality without treatment [
12]. Survival was therefore ensured by rigorous dexamethasone substitution for both dams and pups [
13,
14]. National Center of Biotechnology Information (NCBI) blast analysis shows that there is a 73% amino acid and 78% nucleotide sequence homology between the human
CYP21A2 and mouse
Cyp21a1 genes. Therefore, to study human mutations and further delineate CAH, we developed the humanized mouse model. In our previous work, we have shown that the humanized C57BL/6NCrl-
Cyp21a1tg(CYP21A2)koe mice showed no differences in fertility, viability, and growth and, therefore, have the potential to be an excellent model to introduce mutations relevant to CAH [
11,
15].
In this study, we generated and characterized a novel humanized CAH mouse model in which we introduced the CAH-causing mutation p.R484Q into the humanized C57BL/6NCrl-Cyp21a1tg(CYP21A2)koe mice. The CYP21A2-R484Q mice show typical CAH characteristics with hyperplastic adrenals, low levels of corticosterone, and higher concentrations of progesterone. This mouse model is the first viable rodent model available to test treatment and therapies relevant to CAH and has the potential to facilitate the transition from basic research into clinical application.
3. Discussion
In this study, we generated and characterized a knock-in mouse model integrated with a CAH pathogenic variant, p.Arg484Gln, by the name R484Q in the humanized mouse line C57BL/6NCrl-Cyp21a1tg(CYP21A2)koe. CAH triggered by 21-OH deficiency is considered a fully penetrant genetic condition showing a strong, although not complete, genotype–phenotype relationship in which the less severely affected allele determines clinical symptoms.
Most of the patients with 21-OHD present as compound heterozygotes for various mutations. Patients exhibit either complete deletions or crucial gene conversions affecting the entire
CYP21A2 gene. Furthermore, the nine most frequently occurring point mutations are also present in pseudogene
CYP21A1P, sharing 98% similarity with the coding gene sequence [
6,
16]. The genetic similarity and proximity between pseudogene
CYP21A1P and functional gene
CYP21A2 facilitate various cross-overs accounting for over 90% of all documented mutations linked to 21-hydroxylase deficiency [
17]. A comparative description of the differences between the human CAH patients and the
CYP21A2-R484Q mouse model is provided in
Table 1.
As in humans, the murine genome also has both the functional gene
Cyp21a1 and a corresponding pseudogene
Cyp21a2-p. The murine gene is positioned at chromosome 17, near the mouse H-2 class III region of the major histocompatibility complex (MHC) locus, and encodes a protein composed of 487 amino acids. The murine and human genes show a genetic homology of more than 75%, whereas their respective protein products share a 72.5% similarity [
15]. Although CAH is one of the most common disorders of adrenal steroidogenesis, advances in developing a suitable mouse model to study new treatments and therapies have been quite limited. Our knock-in mouse model with the point mutation p.R484Q in h
CYP21A2 on a C57BL/6NCrl genetic background exhibits an impaired steroid metabolism in the mutant mice. We suspect that the genotypic frequencies of the mice slightly deviated from the expected Mendelian distribution probably due to the loss of pups triggered by the phenotypic conditions of the mutant mice. Gene expression analysis indicates that the h
CYP21A2 gene under the control of the mouse promoter can fully substitute for the expression of the m
Cyp21a1 gene. While cortisol is the major endogenous adrenal steroid in humans, it is corticosterone for mice, which has the highest blood plasma concentration in comparison to other steroids produced by the adrenal gland [
18]. 11-deoxycorticosterone and corticosterone were measurable in significantly lower concentrations in the blood plasma of both male and female mutant mice in comparison to their wildtype littermates, indicating 21-hydroxylase deficiency (21-OHD). Progesterone, the corticosteroid precursor, was significantly elevated in the plasma of mutant animals in comparison to their respective wildtypes. Its counterpart in humans, 17-hydroxyprogesterone (17-OHP), serves as a successor metabolite to progesterone; notably, elevated levels of 17-OHP are identified as a primary diagnostic indicator for CAH in human patients. The accumulation of precursor molecules in the blood of mutant mice implies an enzymatic impairment of the subsequent steps in adrenal steroidogenesis. In 21-OH deficiency, there is an accumulation of both progesterone and 17-OHP due to the impairment in the conversion process. The histological analysis of mouse tissues shows a marked hyperplasia of the adrenal cortex of the mutated animals, indicating poor corticosterone production. Inefficient cortisol production in human CAH patients triggers the production of elevated levels of ACTH, thereby causing the hyperplasia of adrenocortical steroid-producing cells and leading to the development of hyperplasia [
19]. Although we observed higher ACTH concentrations in the male mutant mice, the same could not be seen in the female mutants. This is in contrast to the marked differences in ACTH levels seen in 21-OH knock-out mice [
20]. In addition, we did not succeed in observing significantly elevated expression levels of POMC. We suspect this might be a methodological issue or be due to the nature of the mutation and the residual enzyme activity. However, based on the observation of adrenocortical hyperplasia seen in mutant mice, we deduce that ACTH must be presumably elevated in the female mice as well, analogous to human CAH patients.
Aldosterone, another major glucocorticoid, is fundamentally responsible for maintaining salt and water homoeostasis and blood pressure regulation [
21]. Unlike male mice, the female mutant mice have significantly higher concentrations of aldosterone than their corresponding wildtypes. Although aldosterone is the primary regulator of blood pressure and blood volume, blood pressure did not differ between the mutant mice and their wildtypes for both sexes. Therefore, sexual dimorphism was only observed in aldosterone levels. However, the underlying mechanism for this disparity remains uncertain.
We were also successful in determining steroid hormone biomarkers of 21-OHD in mouse urine. Applying the highly specific technique of GC-MS urinary steroid analysis, we identified a metabolite of progesterone with a saturated ring system and three additional OH-groups. The fragmentation pattern suggested one of these groups to be located at position 16. The other two OH-groups are probably in position 11 and position 6, the latter being a typical hydroxylation site in the mouse. This parameter enabled a strong differentiation between the mice that were affected and those that were unaffected, aligning with the observations regarding progesterone levels in the plasma. The other metabolite is tetrahydrogenated (ring-reduced) corticosterone, the main glucocorticoid in mice. Taken together, these findings reflect the availability of a non-invasive monitoring of future treatment approaches.
Cyp11a1, a cholesterol side-chain cleavage enzyme, is responsible for catalyzing the first reaction in the steroidogenic pathway, the conversion of cholesterol to pregnenolone. Cyp11a1 gene expression was significantly higher in female mutant mice in comparison to their wildtype littermates but not in male mice. In contrast, the expression of steroidogenic acute regulatory protein (StAR), responsible for regulating the transfer of cholesterol from the outer to the inner mitochondrial membrane, did not differ between mutant animals and their wildtype littermates. The gene expression levels of Cyp11b1 did not change between mutant and wildtype groups, which rather contradicts the reduced corticosterone levels in mutant animals. Interestingly, the expression of Cyp11b2, an enzyme responsible for producing aldosterone, was significantly higher in both male and female mutant animals in comparison to their wildtype littermates. This is inconsistent with our finding of elevated aldosterone levels in the female mutant mice. 3ß-hydroxysteroid dehydrogenase (Hsd3b2), an enzyme catalyzing the conversion of pregnenolone to progesterone, did not vary between mutant and wildtype animals in both sexes.
Under normal conditions, the follicular and luteal phases of the menstrual cycle play a crucial role in the release of progesterone. However, untreated CAH leads to continuous adrenal hypersecretion and provokes a wide range of contraceptive effects. Furthermore, persistently high progesterone levels might impair folliculogenesis, resulting in follicular regression subsequently impairing follicle implantation [
22,
23]. This could be one explanation for our observations in the female R484Q mutant mice, where we noticed fewer follicles and signs of follicular regression. In mice, the estrus cycle is staged in four phases, proestrus, estrus, metestrus, and diestrus, repeating every 4–5 days. The diestrus phase in mice is found to be homologous to the human late secretory phase and is associated with high progesterone levels [
24]. When assessed for the estrus cycle over seven days, the female mutant mice were found to be in a consistent diestrus phase and showed signs of follicular regression. Therefore, female mutant mice projected fertility disturbances and were sterile. This condition of fertility impairment is comparable to the profound impact of CAH (21-OHD) on the ability of patients to reproduce [
23]. In the future, we intend to delve deeper into the fertility of the mutant mice and draw analogies to human patients. The testes of the mutated mice did not show any histological differences and had no fertility disturbances, unlike the female mutant mice.
Some other notable animal models in the field include the C57BL/10SnSlc-H-2
aw18 mouse strain, which was difficult to keep alive without the constant supply of dexamethasone to both the dams and pups [
12,
25]. Another example includes a
StAR knockout model developed to understand congenital lipoid adrenal hyperplasia; a condition characterized by impaired steroidogenesis with typical lipid deposits in the steroidogenic tissues. The mouse strain was generated by disrupting the
StAR gene by deleting a part of exon 2 and all of exon 3. Corticosteroids and 0.9% sodium chloride were supplied to rescue the knockout mice during their lifetime [
11,
26]. A more recent model for CAH includes the
Cyp11b1 null mouse, which was generated by exchanging exons 3–7 of
Cyp11b1 with the cDNA encoding for the cyan fluorescent protein. The mice were viable but suffered from mineralocorticoid excess, hypertension, glucose intolerance, and female infertility [
27].
In recent years, mice carrying a functional human gene have gained popularity in the field of biomedical research. Model animals are inherently different from humans in various aspects, but humanized mice offer the opportunity to study the development and functionality of human genes in vivo. Despite the various limitations associated with humanized mice and a need for caution while interpreting the obtained results, there is reasonable proof that their use is instrumental in the delineation of biomedical research [
28]. In general, the generated knock-in mouse model can be considered a significant step in exploring the functional significance of the
CYP21A2 gene. The integration of the point mutation c.1451G>A p.Arg484Gln has led to the development of a viable mouse line which can be beneficial in testing new treatments and therapies associated with CAH. The mutation p.R484Q is associated with the simple virilizing form characterized by the 21-OH enzymatic activity ranging from 1% to 4% [
6,
8]. One of the limitations associated with this CAH mutant mouse is that they will not be able to reproduce the excessive secretion of androgens observed in CAH patients. Unlike humans, the mouse adrenal cortex cannot produce androgens (specifically DHEA) due to the absence of
Cyp17a1 expression in the adrenal, as it is restricted to the gonads only. Overcoming the lack of adrenal
Cyp17a1 expression will be one focus of our future research.
In summary, the established transgenic mouse model integrated with the point mutation p.R484Q will be beneficial as an in vivo testing system for developing treatments and therapies for CAH. The thorough characterization of the humanized CAH mouse model has demonstrated that the human 21-hydroxylase enzyme effectively replaces the function of the mouse 21-hydroxylase enzyme. Mutant homozygous mice exhibit elevated progesterone levels with reduced 11-deoxycorticosterone and corticosterone levels, measurable in serum and urine. The adrenocortex of both male and female mutant mice were markedly hyperplastic. The 21-OHD led to the production of elevated progesterone, thereby influencing the development of follicles and influencing estrous cycle (
Supplementary Figure S3).
In conclusion, humanized homozygous mutant CYP21A2 mice not only serve as a robust model for studying the 21-hydroxylase function and the metabolic aspects of CAH, but they also represent a valuable model for clinical applications such as the testing of novel therapies in the future.
4. Materials and Methods
4.1. Sex as a Biological Variable
In our study, we examined male and female animals, and sex-dimorphic effects are reported if any.
4.2. Experimental Animals
All animal experiments were performed in compliance with the approved standards of animal care as described in the Ethical Guidelines of the German Animal Welfare Act and Directive 2010763/EU, which protects all animals used for scientific purposes. C57BL/6NCrl mice were purchased from Charles River (Sulzfeld, Germany). CYP21A2-R484Q heterozygous mice were generated and bred on a C57BL/6NCrl genetic background. The breeding and maintenance of the mouse strain were accomplished at the mouse facility of the Centre of Regenerative Therapies Dresden (CRTD), Technische Universität Dresden, Germany. The mouse facility was maintained under specific pathogen-free conditions in individually ventilated cages on a 12 h day/night cycle. Standard mouse chow and autoclaved sterile water ad libitum were supplied to the mice.
4.3. Integration of the Point Mutation Arg484Gln (p.R484Q) into Humanized CYP21A2 Mice
Humanized
CYP21A2 mice were generated previously by replacing the 2620 bp
Cyp21a1 mouse gDNA sequence with its orthologous 2713 bp
CYP21A2 human gDNA sequence using CRISPR/Cas9-mediated gene targeting [
15].
CYP21A2-R484Q mice were generated using the CRISPR/Cas9 technique with a specific single-guide RNA and a homology-direct repair oligo in zygotes obtained by intercrossing wildtype females with humanized
CYP21A2 males. Guide RNAs close to the specific point mutation site in h
CYP21A2 gene were chosen by low off-target activity and assessed using
http://crispor.tefor.net (accessed on 2 April 2020). The guide RNA 5′-CAAGTGCGGCTGCAGCCCCG-3′ was obtained as crispr RNA (crRNA) (Integrated DNA Technologies, Inc. (IDT), Coralville, IA, USA). The homology-directed repair oligo 5′-TCATCCTCAAGATGCAGCCTTTCCAAGTGCGGCTCCAACTCAGGCATGGGCGCTCACAGCCCGGGCCAGAGCCAGTGACAGGATAGGA-3′ was designed to contain the desired point mutation p.R484Q in addition to several silent mutations to prevent the re-cutting of the repaired allele and to facilitate genotyping. The repair oligo was ordered as custom PAGE purified Ultramer
TM DNA oligos (IDT, Coralville, IA, USA). Electroporation was used to integrate the point mutation into the generated zygotes and was performed using a 1 mm cuvette (BioRad, 1652089) and a BioRad Gene pulser XCell electroporator (Bio-Rad Laboratories GmbH, Feldkirchen, Germany). To electroporate the zygotes, Cas9 ribonucleoprotein particles were assembled by combining 4 µM Alt-R
® S.p. HiFi Cas9 nuclease V3 (10 mg/mL) (IDT, Leuven, Belgium) with 4 µM of crRNA:tracrRNA duplex and 10 µM of single-strand oligodeoxynucleotide (ssODN) in 10 µL Opti-MEM medium (Gibco
TM 31985062; Thermo Fisher Scientific, Darmstadt, Germany). Around 4 h post-electroporation of the CRISPR/Cas9 mix into fertilized zygotes, the surviving embryos were transferred into pseudo-pregnant recipient female mice. Positive pups carrying the point mutation p.R484Q were screened using PCR with primer pairs 5′-GCTCCAACCTCAGGGCAT-3′ and 5′-CTTCCCTTGACAACCCTCTCC-3′, which specifically amplify modified alleles. Furthermore, Sanger Sequencing was used to verify positive candidates using primer pairs spanning the mutated site. Founder
CYP21A2-R484Q heterozygous mice were bred on a C57BL/6NCrl genetic background. Next, two heterozygous mice with the point mutation were intercrossed to generate the homozygous mice.
4.4. Genotyping
Genomic DNA was isolated from the ear biopsies of the three-week-old transgenic mice using the One Step Mouse Genotyping Kit (Vazyme Internationa LLC, San Diego, CA, USA) in accordance with the manufacturer’s protocol. Two PCR reactions were performed to differentiate between the wildtype and the CYP21A2 allele and to verify the presence of the R484Q mutation. For genotyping, two forward primers, 5′-GTGTCATCCTCAAGATGCAG-3‘ and 5‘-CCAAGACCAGGGTGAGCGT-3‘, and one reverse primer, 5′-CTCACACCCCAGTAGAGAAG-3′ (Eurofins, Ebersberg, Germany), resulted in a 165 bp fragment for the wildtype mice, a 222 bp for the homozygous mice, and a double band for heterozygotes containing fragments of both wildtype and homozygous. Further on, a mutation-specific PCR reaction was performed using the primers 5′-GCTCCAACCTCAGGGCAT-3′ and 3′- CTTCCCTTGACAACCCTCTCC-5′ (Eurofins, Ebersberg, Germany). This resulted in a 322 bp fragment for HOM and HET mice and no fragment for the wildtypes. To further validate the results, Sanger sequencing was performed using the primers 5′-CTGCCGTGAAAATGTGGTGG-3′ and 5′-CTTCCCTTGACAACCCTCTCC-3′.
4.5. Mouse Specimen Collection
Blood samples were collected from eight-week-old mice in heparin tubes from the retrobulbar venous plexus. The plasma recovered from the blood samples was stored at −80 °C and subjected to only one freeze-thaw cycle. Between 20–22 weeks, the mice were anaesthetized with a ketamine–xylazine solution and subjected to a final cardiac puncture to collect blood, after which they were sacrificed by cervical dislocation. The brain, pituitary, liver, kidneys, adrenal glands, ovaries, and testes were removed and were both macroscopically and microscopically investigated. For microscopic investigations, the organs were stored at 4 °C and later processed for histological staining. For gene expression analysis, the organs were snap-frozen in liquid nitrogen and preserved at −80 °C for analysis via quantitative PCR. Metabolic cages were used to collect urine for a 24-h period, during which food and water consumption as well as weight were monitored. The urine was further stored at −80 °C and analyzed for urine metabolites using gas chromatography-mass spectrometry (GC-MS).
4.6. Total RNA Extraction and Quantitative Reverse Transcription PCR
The adrenals, pituitaries, ovaries, and testes of 20-week-old mice preserved at −80 °C were used to extract total RNA using NucleoSpin RNA II kit (Macherey-Nagel, Düren, Germany) according to the manufacturer’s protocol. The purity of the RNA was assessed using the NanoDrop Spectrophotometer (ND-1000) (NanoDrop Technologies, Wilmington, DE, USA). One µg of total RNA was reversely transcribed into cDNA using the GoScript Reverse Transcription System (Promega, Mannheim, Germany) according to the manufacturer’s protocol. Specific gene amplification primers for
CYP21A2,
Cyp11b1,
Cyp11b2,
Cyp11a1,
StAR, Hsd3b2, and
Nr5A1 were designed using Primer Express 3.0 and sourced from Eurofins Genomics (Ebersberg, Germany). For the analysis of
Cyp21a1 and POMC expression, a TaqMan Gene Expression Assay (Thermo Fisher Scientific, Darmstadt, Germany) was purchased. Real-time PCR was conducted using the GoTaq
® Probe qPCR MasterMix (Promega, Mannheim, Germany) in accordance with the manufacturer’s guidelines. Each sample was analysed in triplicate using the Quantstudio 5 (Thermo Fisher Scientific, Darmstadt, Germany. The genes were normalized using the housekeeping gene β-Actin (Actb) as the reference gene. The delta-delta threshold cycle (ΔΔCt) method was utilized to calculate the changes in mRNA expression levels [
29].
4.7. Plasma Steroid and ACTH Measurements
Plasma Steroid concentrations were determined using liquid chromatography-tandem mass spectrometry (LC-MS/MS) [
30]. The ACTH concentration was measured using the ACTH (Mouse/Rat) ELISA kit (Abnova, Taipei, Taiwan). All measurements were carried out according to the manufacturer’s protocol.
4.8. 24-h Urine Collection and GC-MS Determination of Urinary Steroids
The urine of 20-week-old mice was collected using metabolic cages (Techniplast Deutschland GmbH, Hohenpeißenberg, Germany) over 24 h. The mice were given free access to standard mouse chow and autoclaved water during the entire period of urine collection. Post collection, the urine was kept frozen at −80 °C and later used for urine metabolite analysis using gas chromatography mass spectrometry GC-MS [
31].
4.9. Blood Pressure Measurements
Blood pressure was measured in eight and 20-week-old mice using a non-invasive tail-cuff method using the NIBP instrument connected to a PowerLab 4/16 system (ADInstruments, Oxford, UK). During the experiment, the animals were under the influence of light isoflurane (0.5%) and subsequently transferred to a heating plate fitted with a inhalation mask to ensure the constant inhalation of isoflurane during measurement [
32]. The heartrate and mean systolic, diastolic, and arterial blood pressure values were measured three consecutive times. Data were processed and analysed using the Lab Chart Program (ADInstruments, Oxford, UK).
4.10. Assessment of Estrous Stage
The estrous cycle phase was assessed from vaginal epithelial cell smears taken by vaginal flush inserting sterile water into the vaginal opening of the mouse. The method is non-invasive and reliable, in which the vaginal cells were flushed gently by introducing water into the vaginal orifice using a pipette [
24]. The process was repeated a couple of times to obtain a good number of cells for investigation. A small amount of cell suspension was placed on a glass slide. The air-dried slides were stained with 0.1% crystal violet and covered with a slip. The slides were examined for different cell types under a light microscope (
Supplementary Figure S4).
4.11. Histological Analysis
To obtain tissue sections of different organs, eight mice of wildtype and homozygous genotypes were sacrificed (two males and two females of each genotype). After dissection, the organs were fixed in 4% paraformaldehyde (PFA) overnight. After fixation, the organs were washed in 1× phosphate buffered saline (PBS) for 10 min, transferred into fresh PBS, and stored at 4 °C. The organs were then embedded in paraffin at the Histology Facility of BIOTEC Dresden, Germany. Four µm sections of paraffin-embedded mouse organs were placed on glass slides and stained with hematoxylin and eosin (H&E). The microscopic images were captured using Keyence BZ-X700 (Keyence Corporation of America, Itasca, IL, USA).
4.12. Immunohistochemical Analysis
Immunohistochemically, the mice adrenals were analyzed using 4 µM sections obtained from cryo-frozen tissues. Adrenal glands were fixed in 4% formaldehyde for 4 h and transferred to 30% sucrose at 4 °C overnight. The following day, the organs were embedded in Tissue Tek® O.C.T.TM Compound Medium (Sakura Finetek Germany GmbH, Umkirch, Germany ) and immediately frozen at −80 °C. Four µM cryosections were cut using Cryostar NX70 cryostat. Next, tissue sections were blocked with a mixture of 5% normal goat serum, 0.1% bovine serum albumin, and 0.3% Triton X in PBS for 1 h. Post blocking, sections were incubated overnight at 4 °C with primary antibody tyrosine hydroxylase (TH) (Cat# AB152, Merck Millipore, Burlington, MA, USA). Subsequently, the sections were washed with PBS and incubated with fluorescence-labelled anti-rabbit Cy3 secondary antibody. Fluorescence was examined microscopically using a Zeiss Axiovert 200M inverse light microscope (Carl Zeiss Microscopy GmbH, Oberkochen, Germany) and recorded using the AxioVision SE64 Rel. 49 software (Carl Zeiss Microscopy Deutschland GmbH, Oberkochen, Germany).
4.13. Data Analysis
Figures and tables were generated using GraphPad Prism version 9.3.2 for Windows (GraphPad Software, La Jolla, CA, USA), Microsoft Excel, PowerPoint, and ImageJ (NIH) (
https://imagej.net/ij/).
4.14. Statistics
Data are presented as box plots displaying the median, first and third quartile, interquartile range, minimal and maximal values, and outliers. Differences between the wildtype (WT) and homozygous mutant (HOM) mice were analyzed using GraphPad Prism version 9.3.2. Steroid hormone concentrations below detection limits (BDL) were substituted with a 0.5× limit of detection (LOD) for statistical analysis. Tests for normal distribution were performed on all data sets using the Shapiro-Wilk test. Statistical significance for normally distributed data was analyzed using an unpaired t-test. The Mann-Whitney test was used for data which were not normally distributed. Significance p > 0.05 ns; p < 0.05 *; p < 0.01 **; p < 0.001 ***; p < 0.0001 ****.