Breeding and preliminarily phenotyping of a congenic mouse model with alopecia areata

Mei-Er GU, Xiao-Ming SONG, Chun-Feng ZHU，Hong-Ping YIN, Gui-Jie LIU, Li-Ping YU,Wei-Wei YANG, Li-Feng NI, Yan-Li ZHANG, Bao-Jin WU*

?

Breeding and preliminarily phenotyping of a congenic mouse model with alopecia areata

Mei-Er GU, Xiao-Ming SONG, Chun-Feng ZHU，Hong-Ping YIN, Gui-Jie LIU, Li-Ping YU,Wei-Wei YANG, Li-Feng NI, Yan-Li ZHANG, Bao-Jin WU

Laboratory of Experimental Animal Science, Hangzhou Normal University, Hangzhou 310036, China

In the current study, the alopecia areata gene was introduced into the C57BL/6 (B6)mouse through repeated backcross-ing/intercrossing, and the allelichomozygosity of congenicAAmice (named B6.KM) was verified using microsatellites. The gross appearance, growth characteristics, pathological changes in skin, and major organs of B6.KMmice were observed. Counts and proportions of CD4and CD8T lymphocytes in peripheral blood were determined by flow cytometry. Results show that congenic B6.KMmice were obtained after 10 generations of backcrossing/intercrossing. B6.KMmice grew slower than B6 control mice and AA skin lesions were developed by four weeks of age. The number of hair follicles was reduced, but hair structures were normal. Loss of hair during disease progression was associated with CD4and CD8T lymphocytes infiltration peri- and intra- hair follicles. No pathological changes were found in other organs except for the skin. In the peripheral blood of B6.KMmice, the percentage of CD4T cells was lower and percentage of CD8T cells higher than in control mice. These findings indicate that B6.KMmice are characterized by a dysfunctional immune system, retarded development and T-cell infiltration mediated hair loss, making them a promising new animal model for human alopecia areata.

Alopecia areata; Congenic mouse;T-lymphocytes; CD4; CD8

Alopecia areata (AA) is a common disease in hum-ans, can affect any hair-bearing region of the body, and usually shows with acute non-inflammatory alopecia parv-i-maculata, but unfortunately its underlying mechan-isms are not fully understood but it is correlated with genetic background, emotional stress, endocrine dyscra-sia and autoimmunity. Hereditary susceptibility is one of the most important factors of AA, and approximately 25% patients have affected family members. Due to increasing stress from study and work, a growing number of people are affected by hair loss. In some cases, it results in psychological effects such as reduced self-esteem, and in extreme cases may even induce depre-ssion or suicide. It is therefore vital to investigate the underlying mechanisms of AA, identify the related genes and explore possible therap-eutic targets (Sun et al, 2008).

AA mouse models have played important roles in AA studies. For example, Sundberg et al (1994) found that 20% of C3H/HeJ mice spontaneously develop adult onset AA by 18months of age, and the progression of this complicated multi-genetic disorder is quite similar to the clinical symptoms of human AA. In addition to laboratory mice, AA-like hair loss has been noticed in inbred Dundee rats (McElwee et al, 1996). Recently, with availabilities of C3H/HeJ mice and Dundee rats, concerted attempts have been made to understand the genetics, pathogenesis and treatment of AA (Michie et al, 1991; McElwee et al, 2002; Sundberg et al, 1996); however, due to the low incidence (20%) of AA in C3H/HeJ mice, validation of a new AA animal model is necessary.

The mice used here (current name: Alopecia Areata Tongji,AA) are characterized by progressive hair loss and were initially discovered 10 years ago during the propagation of KM mice by the Laboratory Animal Center at Tongji University. During years of conservation breeding, attempts have previously been made to introduce the mutant genes into other inbred strains by repeated back-crossing/intercrossing, but the documn-tation related to this process are missing. To establish an inbred mice strain with a purified genetic background, we introduced thegene into C57BL/6 (B6) mice by repeated backcro-ssing/intercrossing. The development and pathogenesis of this congenic inbred AA mouse strain (B6.KM) was then evaluated. The validation of this B6.KMmicemodelwill likely facilitate future AA mutant gene location, identification and related invest-igations into the biological characteristics of this condition.

MATERIALS AND METHODS

Experimental Animals

AAmice were acquired from Hangzhong Normal University four years ago. CleanAAmice and control B6 mice were provided by the Laboratory of Experim-ental Animal Science, Hangzhong Normal University (Laboratory Animal Production and Use License: SCXK [Zhejiang] 2011-0048; SYXK [Zhejiang] 2011-0157). Experimental animals were housed in protect-ive animal rooms with a 12/12 h light cycle, with temperature at 23±2 °C and humidity at 55±5%. Animals were fed Co60 irradiated food. Caging and bedding materials were changed freque-ntly and sterilized using heat.

Experimental Equipment and Reagents

Reagents used in polymerase chain reaction (PCR) were from Sangon Biotech (Shanghai, China). Monoc-lonal antibodies (CD4(ab25475), CD8(ab25478)) and Rat IgG (horse radish peroxidase) (ab6734) were produ-cts of Abcam (Cambridge, UK). CD3-PE/Cy7, CD4-PE, CD8-FITC and parallel control antibodies were from BD Bioscience (San Diego, CA). Experimental equipment and manufacturers were: PCR machine (BIOS, 1000), electrophoresis apparatus (Tanon, EPS300), gel imaging system (Tomon, 2500), paraffin embedding machine (MicRom, AP280), rotary microtome (MicRom, HM3-35E), pathological tissue floating and drying apparatus (Huali Electronics, Changzhou, China), flow cytometry (BD, FACSCalibur, Becton Dickinson, San Jose, CA, USA)), and upright camera microscope (Olympus, BX51).

Breeding of congenic B6.KMmice

F1mutant gene carriers without the AA pheno-type were hybrids of AA phenotype mice and B6 mice. F2 mice with either the AA or normal phenotype were bred through F1 mice intercrossing. F3 mice were obtained by backcrossing AA phenotype F2 with B6 mice. Then, F4 mice were bred through F3 mice intercr-ossing. After several generations of repeated intercro-ssing, AA phenotype F10 mice were established for strain conservation.

Homozygosity evaluation of congenic B6.KMmice

Genomic DNA extraction from the tip of the tails (0.5 cm) of F10 congenic B6.KMmice was purified by protease K digestion followed by phenol: chloroform extraction. Thirty-nine mouse microsatellites established by our laboratory (Wu et al, 2003) were applied in the homozygosity evaluation.

Recessive inheritance validation

During breeding, total numbers of the AA pheno-type and normal F2, F4, F6, F8, F10 mice were calcul-ated. The practical ratio of mutant mice to normal mice and theoretical ratio derived from recessive inheri-tance were compared. Then, the practical number and percen-tage of AA phenotype G3 mice were compared with theoretical values derived from recessive inherita-nce. G3 mice were bred through the backcrossing of G2 (hybrids of B6.KM-mice and B6 mice) and G1 mice (B6.KM-AA mice).

Developments of congenic B6.KMmice

Hair growth in litters from birth to 12 weeks of age was observed, and animal weight from birth to 8 weeks of age were recorded. Three-week-old males and females were caged separately and gains in mass were recorded and statistical analysis was done by taking normal B6 mice bred by our animal center as controls.

Hair observation of congenic B6.KMmice

Hair samples from dorsal areas of the should blades of six 8-week-old B6.KMmice (3 males, 3 females) and six B6 mice (3 males, 3 females) were mounted on dimethyl benzene marinated glass slides, sealed with neutral balsam and then observed under a microscope.

Major organs and pathology of skin tissues of B6.KMmice

Mice used in the hair observation were euthanatized by cervical dislocation. Major organs, including brain, heart, liver, spleen, lung, kidney, thymus, adrenal gland, testicle,appendix testis,uterus and ovary were fixed by 10% formalin. After dehydration, infiltration, embedding, sectioning and Hematoxylin-Eosin (HE) staining they were observed under a light scope.

Three males and 3 females of B6.KMmice and B6 mice by birth, 2 weeks, 4 weeks, 6 weeks, 8 weeks and 12 weeks of age were euthanatized by cervical dislocation. Dorsal skin samples were fixed using 10% formalin. After dehydration, infiltration, embedding, sectioning and HE staining skin tissue pathology was observed under a light scope.

Immunohistochemistry staining

Paraffin sections of skin tissues were used in horseradish peroxidase conjugated CD4and CD8im-munohistochemical experiment. Working concentrations of CD4, CD8and rat IgG (HRP) were 1:50, 1:50 and 1:100, respectively. Sections were incubated in primary antibody at 4 °C overnight, and then in secondary antib-ody at 37 °C for 60 min, followed with chromogenic staining by 3, 3’-diaminobenzidine (DAB) and Hemato-xylin counterstaining. Phosphate Buffered Saline (PBS) was used as the negative control of primary antibody.

Analysis of peripheral blood CD4and CD8T lymphocytes

Retro-orbital venous blood samples (200 μL) were collected from three male and three female B6.KMmice and B6 mice at 8 weeks of age, and then transferred into anticoagulant sodium heparin tubes. Tubes were gently converted to prevent coagulating and were kept on ice. Next, 100 μL blood samples were transferred into polystyrene tubes with CD3-PE/Cy7 (5 μL), CD4-PE (5 μL) and CD8-FITC (2 μL) monoclonal antibodies. Mean-while, negative and blank controls were set. Blood sam-ples were incubated in the dark at 4 °C for 30 min and then 2 mL 1× FACS lysing solution was added. After incubation in the dark at 4 °C for 15 min, samples were centrifuged at 4 °C for 5 min at 1 500 rpm. Supernatants were discarded and cell pellets were rinsed with ice-cold PBS 1-2 times, then centrifuged again at 4°C for 5 min at 1 500 rpm. Supernatants were discarded and cell pellets were resuspended with 400 μL PBS. Five thousand cells/tube went through flow cytometry and numbers of CD3, CD4and CD8cells were recorded. Results were analyzed using Cell Quest.

Statistical analysis

All statistical analysis was conducted via SPSS17.0 (SPSS Inc, Chicago, IL). Heritability patterns were determined usingsquare tests. All other data were analyzed using independent-sample-tests. All results are expressed as mean±

RESULTS

Breeding of congenic B6.KMmice

AAmice came from white KM mice and were characterized by black hair and a close appearance to B6 mice. After two years and 10 generations of backcro-ssing/intercrossing with B6 mice, we obtained F10 AA mice. This AA mutant mice strain was then conserved by inbred breeding and named B6.KMaccording to naming regulations of gene introgession. After 10 gener-ations of crossing, theoretically, the genetic background of B6.KMmice is 96.88% (1-(50%)), consistent with that of B6 mice. Moreover, homozygosity detection based on 19 microsatellites (D1Mit416, D1Mit180, D2Mit62, D2Mit249, D4Mit214, D4Mit55, D5Mit356, D5Mit409, D7Mit230, D7Mit318, D7Mit203, D10Mit70, D11Mit229, D15Mit171, D15Mit29, D17Mit123, D17-Mit224, D18Mit187 and D18Mit149) showed a single band for each locus of all mice, indicating the B6 genetic background of this congenic B6.KMmice strain (Figure 1).

B6.KMrecessive inheritance confirmation

During the breeding of B6.KMmice, among the 235 individuals obtained from F2, F4, F6, F8 and F10 generations, 49 were with AA phenotype and 186 were normal.-square values andvalues were calculated by comparing mice numbers with recessive inheritance patterns of a single gene (χ=0.753,=0.385). Therefore, the single recessive inheritance of themutant gene was confirmed. The G2 hybrids of B6. KMand B6 mice were all normal, whereas, 24 AA phenotype mice were bred among 54 hybrids of G2 and B6.KMmice. The single recessive inheritance of B6.KMmice was also confirmed (χ=0.12,=0.729).

Figure 1 Homozygosity detection of microsatellite loci

Lane 1: B6D2F1 control; Lane 2-12: B6.KMmice.

Development observation of B6.KMmice

Hair growth in juvenile B6.KMmice was normal. Patchy hair loss developed by 4 weeks of age on the head and face. Symptoms were more obvious with increasing age and were severe by 12 weeks of age, characterized by sparse, short and thin hair (Figure 2). However, hair structures were normal with aligned internal medulla (not shown in the figure). From birth to 4-weeks-old, B6.KMmice were lighter than control B6 mice. No significant difference in mass between 5-week-old B6.KMmales and controls was observed, whereas, female B6.KMmice were consistently and significantly lighter than controls until puberty. These patterns indicate retarded development inB6.KMmice compared with B6 mice (Figure 3).

Figure 2 Hair growth in B6.KMmice at different developmental stages

A, B, C, D, E, F and G: Developmental stages of birth, 1-week-, 2-week-, 4-week-, 6-week-, 8-week- and 12-week-old, respectively.

Major organs, pathology and immunology of skin tissues

From birth to 2 weeks of age, no differences in skin tissue betweenB6.KMand B6 mice were found.Lymphocyte infiltration could be seen around hair folli-cles in B6.KMmice by 4 weeks of age and worsened with disease progression. Lymphocyte infiltration in hair follicles was obvious, accompanied with hair follicle dystrophy and number reduction. Those symptoms were most severe in 12-week-old B6.KMmice and could not be seen in control B6 mice. Monoclonal immunohi-stochemistry showed CD8T lymphocyte positive expre-ssion around hair follicles and the root sheath derived from skins of B6.KMmice, whereas, only small CD4T lymphocytes expression was found around hair follicles (Figure 4)

Figure 3 Body weight gain curvature of congenic B6.KMmice and B6 mice

No significant abnormality was found in paraffin sections of major organs of B6.KMmice, including brain, heart, liver, spleen, lung, kidney, thymus, adrenal gland and intestine (not shown).

Analysis of CD4and CD8T lymphocytes in periph-eral blood of B6.KM

No significant differences in peripheral blood total lymphocytes between B6.KMmice and normal control B6 mice were found. Compared with normal control B6 mice,B6.KMmice had higher expression of CD8T lymphocytes, but lower expression of CD4T lymphocytes (Figure 5, Table 1).

Table 1 Peripheral blood CD4+ and CD8+ T lymphocytes of B6.KM-AA mice

DISCUSSION

AlthoughAAmice were first identified and isolated over 10 years ago, research progression has been inconsistent and not well documented.AAmice were originally from white KM mice, but until now, their black hair and body appearance were quite close to B6 mice. The uncertainty in genetic background has hind-ered studies and the efficacy of this mouse model. The most important component of our study is the successful introduction of themutant gene into a B6 background. After 10 generations of continuous introduction over two years, theoretically, the genetic background consistency of B6.KMwith B6 has reached 96.88% (1-(50%)). In general, an inbred strain needs 20 generations of intercrossing and the allele homozygosity should reach 98.6%. In our case, taking into account B6-like charac-teristicsAAmice originally had, we assume the allele homozygosity of B6.KMis over 96.88% and is relatively stable. Allele homozygosity detection conduc-ted with microsatellite markers indicates the single genetic background of B6.KMmice. More importa-ntly, this purified genetic background does not influence the development of the AA phenotype in B6.KMmice, but clarifies the single genetic recessive inherita-nce of themutant gene, which can be used in model characteristic analysis and related genetic chromosome localization.

Figure 4 Lymphocyte infiltrations in hair follicles of B6.KMmice

A, E: Dorsal skin of 8-week-old B6.KMmice, HE staining, lymphocyte infiltrations could be seen peri- and intra- hair follicles; B: Dorsal skin of 8-week-old B6 mice, HE staining, no lymphocytes infiltrations could be seen around hair follicles; C, F: CD4immunohistochemistry staining of the dorsal skin of 8-week-old B6.KMmice, yellow-brown positive CD4T-lymphocytes could be seen around hair follicles; D, G: CD8immunohistochemistry staining of the dorsal skin of 8-week-old B6.KMmice, yellow-brown positive CD8T-lymphocytes could be seen around and inside of hair follicles.

Figure 5 Two dimensional lattice chart of peripheral blood CD4and CD8Tlymphocytes

A: B6.KMmice; B: B6 mice.

Symptoms of patchy hair loss initiated by 4 weeks of age in B6.KMmice and get worse with age. Hair follicle dystrophy and reduction as well as CD4and CD8T lymphocyte infiltrations around hair follicles are more obvious with disease progression. Low expression of CD4T lymphocytes, but high expression of CD8T lymphocytes in peripheral blood of B6.KMmice compared with normal control mice indicates immune dysfunction. It is believed that the loss of hair follicles is mediated by T lymphocyte infiltration. The mechanisms of AA are yet to be determined, but most researchers believe it is a systemic autoimmune disorder. Studies on C3H/HeJ mice claimed that the main reason for AA is because CD4and CD8T lymphocytes attack the body’s own anagen hair follicles and suppress or stop hair growth, a process which TNF-α, IL-6, IL-12, IFN-γ, IL-10, IL-4 and many other cell factors also play a role in (McElwee et al, 1998). McElwee et al (1996) and Freyschmidt-Paul et al (2000) found that using monocl-onal antibodies to reduce counts of CD4and CD8T lymphocytes and inhibit inflammatory cell infiltrations restrains the development of AA. Perret et al (1984) found that the percentage of CD8T lymphocytes in peripheral blood of human AA patients was significantly low, whereas around hair follicles the ratio of CD4/CD8is only 4:1. Chen et al (2005) adopted immu-nohistochemistry techniques to evaluate T lymp-hoc-ytes in skin lesions of human AA patients and found that during the active phase and stable phase, cell infiltrations of CD4T lymphocytes are significantly higher than CD8T lymphocytes around hair follicles and vessels. Huang et al (2007) reported that in both minor and severe cases of AA, expressions of CD4and CD8T lymphocytes are all higher than those of control groups. Our results presented here support the idea that AA is related to cellular immunity and T lymphocyte infiltration in hair follicles plays an important role in disease progression. However, further questions such as the disorder in CD4/CD8lympho-cytes remain unanswered.

Over 14 mice strains have been reported with sparse hair phenotype (Wu et al, 2009; Zang et al, 2009; Li et al, 199; Tian et al, 2004), but experience congenital hair loss like in thesnthrmice strain bred in our lab. Therefore, their hair loss symptoms are neither required nor progressive as in human AA (Wu et al, 2010; Wu et al, 2009). C3H/HeJ mice are a widely applied AA animal model. Sundberg et al (2003) identified the pathogenic QTL locus on the 17chromosome of C3H/HeJ mice, but relevant functional research is difficult. C3H/HeJ mice have also been used in medication research and development. It has been reported that SADBE (squaric acid dibutylester), DPCP(diphenylcyclopropenone), glucocorticosteroid, cyclosporine (CsA),tacrolimus (FK506) and minoxidil all promote hair regeneration (Sun et al, 2008). However, although the pathology in C3H/HeJ mice is similar to that in humans, the low incidence (20%) and uncertainty in identifying pathog-enic genes have hampered its application as a human AA mouse model. Moreover, the homologies of pathogenic AA genes between humans and mice are pivotal in the establishment of an animal model (Mao & Zheng, 2006; Peters et al, 2007). Themutant gene in our study is a spontaneous mutation and characterized by single genetic recessive inheritance. Disease progression is close to that in human AA patients, which is beneficial for future pathogenic gene localization and colonial identification. In summary, the B6.KMmice bred in our study are a promising AA animal model, and can be used in studies into the molecular mechanisms of AA, observations of medication efficacy and strain conserv-ation of economic fur animals.

Chen LF, Shi WP, Liu JH, Chen XM, Qin XW, Hao SY, Zheng XG, Shang YL, Wang YF. 2005. A study on infiltrating T-lymphocytic phenotype and Fas expression in skin lesion area of pelade patients.34(1): 9-11. (in Chinese)

Freyschmidt-Paul P, Seiter S, Z?ller M, K?nig A, Ziegler A, Sundberg JP, Happle R, Hoffmann R. 2000. Treatment with an anti-CD44v10-specific antibody inhibits the onset of alopecia areata in C3H/HeJ mice.115(4): 653-657.

Huang WN, Hou XZ, Lu HQ, Lu HM, Xu X. 2007. Studies on the relationship between T lymphocyte subtypes and alopecia areata.14(4): 208-210. (in Chinese)

Li SR, Wang DP, Lan H, Zang WY, Ge BS, Li JQ, Wang CE, Lu YF, Lu YY, Li RF. 1999. A new mouse mutant gene mapping.44(7): 50-55. (in Chinese)

Mao L, Zheng WJ. 2006. Combining comparative genomics with de novo motif discovery to identify human transcription factor DNA-binding motifs.7(Suppl 4): S21.

McElwee KJ, Boggess D, King LE Jr, Sundberg JP. 1998. Experimental induction of alopecia areata-like hair loss inmice using full-thickness skin grafts.111(5): 797-803.

McElwee KJ, Pickett P, Oliver RF. 1996. The DEBR rat, alopecia areata and autoantibodies to the hair follicle.134(1): 55-63.

Michie HJ, Jahoda CA, Oliver RF, Johnson BE. 1991. The DEBR rat: an animal model of human alopecia areata.125(2): 94-100.

McElwee KJ, Hoffmann R. 2002. Alopecia areata-animal models.27(5): 410-417.

McElwee KJ, Spiers EM, Oliver RF. 1996.depletion of CD8T cells restores hair growth in the DEBR model for alopecia areata.135(2): 211-217.

Perret C, Wiesner-Menzel L, Happle R. 1984. Immunohistochemical analysis of T-cell subsets in the peribulbar and intrabulbar infiltrates of alopeciaareata.64(1): 26-30.

Peters LL, Robledo RF, Bult CJ, Churchill GA, Paigen BJ, Svenson KL. 2007. The mice as a model for human biology: a resource guide for complex trait analysis.8(1): 58-69.

Sundberg JP, Cordy WR, King LE Jr. 1994. Alopecia areata in aging C3H/HeJ mice.102(6): 847-856.

Sundberg JP, Boggess D, Silva KA, McElwee KJ, King LE, Li R, Churchill G, Cox GA. 2003. Major locus on mouse chromosome 17 and minor locus on chromosome 9 are linked with alopecia areata in C3H/HeJmice.120(5): 771-775.

Sundberg JP, King LE Jr. 1996. Mouse mutations as animal models and biomedical tools for dermatological research.106(2): 368-376.

Sun J, Silva1 KA, McElwee KJ, King LE Jr, Sundberg JP. 2008. The C3H?HeJ mouse and DEBR rat models for alopecia areata: review of preclinical drug screening approaches and results.17(10): 793-805.

Tian M, Xiong YL, Wang WY, Zhang YD. 2004. Kunming rhinos in mice and its molecular genetic.49(1): 74-80. (in Chinese)

Wu BJ, Mao HH, Zhu H, Yan ZF, Yang L, Sun Q, Xu XM, Xue ZF, Li HD. 2003. PCR conditions and application of 39 mouse microsatellites.11(4): 216-220. (in Chinese)

Wu BJ, Mao HH, Zeng YM, Yin LJ, Yin XS, Yang WW, Kang XD, Liu GJ, Yu LP, Gu ME, Wu PL. 2009. Fine Mapping and Identifying the Mutation Gene of-Scant Hair Mouse.30(3): 267-275. (in Chinese)

Wu BJ, Zeng YM, Mao HH, Yin LJ, Zhu J, Yang W W, Yin XS, Wu PL, Zhang WD. 2010. Mapping of genetic modifiers of Plcd1 in scant hair mice (snthr).55(35): 4026-4031.

Zang WQ, Yang X, Wang T, Xuan XY, Li M, Du Y. 2009. Orientation and analysis of Yuyi hairless mice hairless gene mutation point.44(6): 1144-1145. (in Chinese)

25 November 2013; Accepted: 25 December 2013

This work was supported by the Public Program of the Science Technology Department, Zhejiang (2011C37077) and the National Natural Science Foundation of China (31071092).

, E-mail: baojinwu@163.com