Sage Journals: Discover world-class research

Abstract

Inbred mice are a unique model system for studying aging because of the genetic homogeneity within inbred strains, the short life span of mice relative to humans, and the rich array of analytic tools that are available. A large-scale aging study was conducted on 28 inbred strains representing great genetic diversity to determine, via histopathology, the type and diversity of spontaneous diseases that aging mice develop. A total of 20 885 different diagnoses were made, with an average of 12 diagnoses per mouse in the study. Eighteen inbred strains have had their genomes sequenced, and many others have been partially sequenced to provide large repositories of data on genetic variation among the strains. This vast amount of genomic information can be utilized in genome-wide association studies to find candidate genes that are involved in the pathogenesis of spontaneous diseases. As an illustration, this article presents a genome-wide association study of the genetic associations of age-related intestinal amyloidosis, which implicated 3 candidate genes: translocating chain-associated membrane protein 1 (Tram1); splicing factor 3b, subunit 5 (Sf3b5); and syntaxin 11 (Stx11). Representative photomicrographs are available on the Mouse Tumor Biology Database and Pathbase to serve as a reference when evaluating inbred mice used in other genetic or experimental studies to rule out strain background lesions. Many of the age-related mouse diseases are similar, if not identical, to human diseases; therefore, the genetic discoveries have direct translational benefit.

Keywords

aging intestinal amyloidosis genetics genome-wide association study mice

Laboratory mice are important tools for studying complex genetics and disease mechanisms for almost all human diseases.⁵² High-resolution genetic variation data for the commonly used inbred strains (http://www.sanger.ac.uk/sanger/Mouse_SnpViewer/rel-1410) as well as the ability to manipulate the genome, particularly with the new CRISPR/Cas9 technology,^15,21,22,88 make the mouse the tool of choice for a wide range of biomedical studies. Common causes of human morbidity and mortality are often associated with age-related degenerative or neoplastic diseases, and with an increasingly long-lived human (as well as domestic animal) population, the economic and social impacts of these diseases are ever increasing.⁸⁰ The relatively short life span of laboratory mice has enabled many studies of inbred strains and hybrid stocks to investigate aging.^34,38,91 However, most of these studies focused on the genetics of life span rather than the underlying diseases commonly associated with aging, which usually have a direct impact on life span. Such diseases of aging can have a cumulative effect, where the complications are associated either with many minor degenerative diseases that eventually overwhelm the body’s homeostatic mechanisms resulting in death or with catastrophic diseases, such as metastatic cancer, which ultimately kill the patient. Paradigms are changing from an aging research focused strictly on expanding life span to improving the quality of life independent of longevity that is the health span. The relationship of these events and the genetics underlying their onset can be dissected with these large inbred mouse population data sets.

Historically, research focused on lesions in aging mice in strains that were commonly used in biomedical research at the time. This provided background data intended as an aid for interpreting experimental results but also provided insight into the aging process itself as well as how inbred strains differed from one another. A limited number of inbred strains (BALB/cStCrlf, C3H/Nctr, DBA/2NNia), hybrid stocks (B6C3F1, C57BL/6N X C3H/HeN; B6D2F1, C57BL/6NNia X DBA/2NNia; or B6C3F1, C57BL/6NNia X C3H/NNia, in which females from the first strain are bred with males from the second strain) and a variety of other strains and stocks were used for various types of aging studies.^11,14,34,43 More recently, studies focused on general types of lesions in aging mice used in genetic engineering experiments, both inbred and those on segregating backgrounds.⁹⁰ Over time, some of the diseases change due to improvements in husbandry, pathogen status, and genetic quality control. Helicobacter hepaticus was considered part of the normal flora prior to 1994, but it is now recognized an opportunistic pathogen and has been eliminated from many research and vendor colonies.^84,85 Klebsiella oxytoca, another organism long ignored, is now considered an opportunistic pathogen, especially in strains that either have mutations in their toll-like receptor 4 (Tlr4) gene or are immunodeficient.¹⁰ Mammary cancer was once common in many mouse strains. The Jackson Laboratory was founded to use mice affected with mammary cancer to discover the underlying cause, which proved to be the Bittner agent, later called the mouse mammary tumor virus.^18,71 Mouse mammary tumor virus has been eradicated from most vendor colonies, resulting in a decrease in mammary cancer, especially from C3H substrains.⁵¹

Understanding the complex genetics of chronic diseases of aging in humans as a result of studies with mice will ultimately provide biomarkers that can be used to guide lifestyle changes or pharmacologic intervention. If done early, such modifications may prolong health span, improve end-of-life care, and ultimately reduce health care costs in humans.

We report here an overview of detailed histopathology-based investigations of 28 inbred strains of mice at various ages in cross-sectional and longitudinal studies. While >30 strains were initially used, most survived to the 12-month time point, but not all survived to the planned age for cross-sectional testing that is 20 months of age. Data were captured with the Mouse Disease Information System,^61,73 which provided a computable system for compiling and analyzing the subsequent massive data sets based on genome-wide association studies (GWASs). Representative photomicrographs from this and other studies can be accessed on public repositories (Mouse Tumor Biology Database, http://tumor.informatics.jax.org/ and Pathbase, http://www.pathbase.net).^{3,31,58,59,61} These data provide a unique resource and background information that can be an aid to correct interpretation of lesions seen in old mice.

Materials and Methods

Mice

The following inbred strains, obtained from 1 vendor (The Jackson Laboratory, Bar Harbor, ME, USA), were selected for their interstrain genetic diversity:⁵³ 129S1/SvImJ, A/J, AKR/J, BALB/cByJ, BTBRT ⁺tf/J, BUB/BnJ, C3H/HeJ, C57BL/10J, C57BL/6J, C57BLKS/J, C57BR/cdJ, C57L/J, CAST/EiJ, CBA/J, DBA/2J, FVB/NJ, KK/HIJ, LP/J, MRL/MpJ, NOD.B10Sn-H2^b/J (a congenic strain with the NOD genetic background but with a histocompatibility locus from a diabetes-resistant strain), NON/ShiLtJ, NZO/HlLtJ, NZW/LacJ, P/J, PL/J, PWD/PhJ, RIIIS/J, SJL/J, SM/J, SWR/J, and WSB/J. Three of the strains were not evaluated because of lymphoma development (AKR/J) confirming a well-documented strain-specific phenotype,^68,69 lack of sufficient mice for analysis (CAST/EiJ), or self-mutilations resulting in euthanasia for humane purposes (SJL/J). Therefore, GWAS analyses were performed in 28 strains of mice.

Mice were maintained in a humidity-, temperature-, and light cycle– (12:12 hours) controlled vivarium under specific pathogen–free conditions (http://jaxmice.jax.org/genetichealth/index.html). Mice were housed in double-pen polycarbonate cages (330-cm² floor area), with 4 mice per pen. Mice were allowed free access to autoclaved food (NIH 31, 6% fat; LabDiet 5K52, Purina Mills, St Louis, MO, USA) and acidified water (pH 2.8–3.2).

All work was done with the approval of The Jackson Laboratory Animal Care and Use Committee (approval No. 06005). Mouse handling and care were followed according to the Public Health Service animal welfare policies.

Aging Colonies

The study was divided into 2 parts: (1) a cross-sectional study to evaluate onset and diversity of diseases at 12 and 20 months of age; (2) a longitudinal, or life span, investigation, to determine how long each strain survived. Mice in the longitudinal study were maintained until they became moribund, at which point they were euthanized by CO₂ asphyxiation.

Mice in the cross-sectional study were euthanized in groups by strain during a 4-month period at 12 months ± 28 days and 20 months ± 28 days of age. Blood was collected at the time of necropsy from the submandibular vein for hematology (Siemens Advia 2120 Hematology Analyzer, Siemens Healthcare Diagnostics, Newark, DE, USA) and blood chemistry (Beckman Coulter 600 Pro Chemistry Analyzer, Beckman Coulter Corp, Brea, CA, USA). Just prior to necropsy, body mass was determined with a PIXImus machine (Lunar PIXImus, GE Medical Systems, Hermosa Beach, CA, USA). All mice were subjected to a complete necropsy.⁶⁶ All physiologic data, including clinical pathology data, are available on the Mouse Phenome Database (http://phenome.jax.org/).

The Cross-sectional Study

Two cohorts were maintained from all the strains described above—each consisting of 15 females and 15 males and each with a designated survival end date: 12 months and 20 months. Mice that survived to their respective time points were euthanized by CO₂ asphyxiation, and a complete necropsy was performed.⁶⁶

The Longitudinal Life Span Study

A cohort of 32 females and 32 males for each of the 28 strains was initially set up. A second cohort of 32 females was subsequently added, totaling 96 mice/strain, which was calculated to provide enough statistical power to detect a 10% difference in life span.⁹¹ Complete necropsies⁶⁶ were done on moribund mice but not on mice found dead, as autolysis precluded a meaningful interpretation.

Tissue Collection

Tissues were collected in a systematic manner and fixed by immersion in Fekete’s acid alcohol formalin solution.⁶⁶ In our experience, Fekete’s acid alcohol formalin solution provides the optimal fixation for histology and especially for antigen preservation for immunohistochemistry of mouse tissues.⁴⁰ After overnight fixation, tissues were transferred to 70% ethanol and later trimmed, processed routinely, embedded in paraffin, sectioned at 6 μm, and stained with hematoxylin and eosin. Pancreata were fixed attached to the intestines. Additional samples of pancreata were removed without any attached tissues other than mesentery, fixed in Bouin’s solution, washed in running water overnight, processed as above, but stained with aldehyde fuchsin to evaluate pancreatic islets for insulin production. Skulls and the vertebral column were also fixed in Bouin’s solution, allowed to fix and decalcify for a week, and then processed as above to generate hematoxylin and eosin–stained slides. The neurologic tissues were evaluated separately and are not reported here.⁷⁴ Additional special stains—such as toluidine blue for mast cell tumors, phosphotungstic acid hematoxylin for rhabdomyosarcomas, or alizarin red or von Kossa for mineralization—were used as needed to aid in reaching a definitive diagnosis. More extensive workups were done on specific diseases, which will be or were reported separately. Twenty-four slides were generated per mouse, with an average of about 3 tissues per slide.

These included

“Swiss Rolls”⁴² of the duodenum, jejunum, ileum, and colon (with anus and perineal skin)

Longitudinal section of the stomach with esophagus and cecum (inflated with fixative)

Cross sections of the left lateral and medial lobes of liver (to include the gall bladder), spleen, right kidney with adrenal glands, the left kidney was trimmed lengthwise so that it could be differentiated from the right kidney, the reproductive organs (testis, epididymis, accessory sex organs for males; ovary, uterine tube, uterus, mammary glands for females), the preputial gland for males / the clitoral gland for females, the salivary gland cluster with cervical lymph nodes, the heart, the esophagus and trachea with thyroid and parathyroid glands, and the tongue

Longitudinal sections of the center of the lobes of both lungs, dorsal skin, ear skin (pinna), ventral skin, muzzle skin, eyelid, and left kidney

Longitudinal section of hind leg, including stifle/knee joint

Longitudinal section of front leg, including shoulder and elbow joints

Longitudinal section of hind foot (soft tissues, bone, and nail unit/footpad)

Longitudinal section of front foot (soft tissues, bone, and nail unit/footpad)

Longitudinal and cross sections of the lumbar spine

Longitudinal and cross sections of the tail

Longitudinal and cross sections of the skull

Sections of the lower jaw to include teeth⁹

Bones fixed in Fekete’s acid alcohol formalin solution were processed in Cal-Ex (Fisher, Pittsburgh, PA, USA). Bones fixed in Bouin’s solution had no additional decalcification done.

Data Management

Diagnoses were entered into an electronic medical record system consisting of a relational database and a user interface for the pathologist to seamlessly enter specific findings for each mouse. This database, the Mouse Disease Information System,^73,78 uses the Mouse Anatomy Ontology to code for the organ²⁵ and the Mouse Pathology Ontology⁶¹ to code for the disease process (diagnosis). A severity score was recorded (0, normal; 1, mild; 2, moderate; 3, severe; or 4, extreme) to account for differences in the size, severity, and biological behavior of lesions.

GWASs and Variant Identification

Phenotype-genotype association analysis was performed via the expedited efficient mixed-model association algorithm.²⁹ Frequencies and mean severities of the selected disease phenotypes in 28 strains of mice were used as input phenotype data. Four million single-nucleotide polymorphisms (SNPs), which are freely available at http://mouse.cs.ucla.edu/mousehapmap/full.html, were used as input genotype data. Each SNP was evaluated individually and recorded as –log(P values) as the strength of the genotype-phenotype associations in Manhattan plots. SNPs were annotated with the Variant Effect Predictor by Ensembl (http://useast.ensembl.org/info/docs/variation/vep/index.html). Coding region SNPs known to change the amino acid sequence of the protein (ie, nonsynonymous SNPs) were tested for functional importance via the PolyPhen-2 algorithm (http://genetics.bwh.harvard.edu/pph2/).¹ Identified genes were evaluated by extensive literature searches and evaluations of protein-protein interactions and potential biological pathways.

Results

Numbers of Mice Examined

While most inbred mouse strains survived to 12 months of age, it became rapidly apparent with the longitudinal study that few mice would reach 24 months of age, the initial goal for the older cohort. As such, the second group was only aged to 20 months of age. There was wide variation in the numbers of individual mice that reached 20 months of age, which was reported previously.⁷⁴ All AKR/J mice died after the 12 months of age exclusively due to development of lymphomas, a known strain-specific lesion due to an endogenous retrovirus.^68,69 All AKR/J mice had structural abnormalities of their hair shafts, unrelated to age, due to a mutation in the Soat1 gene.^23,86 SJL/J fought extensively, especially but not exclusively the males, requiring many to be euthanized. As a result, no SJL/J mice were available for necropsy at 20 months of age. Other diseases, such as alopecia areata, were found in old mice in strains previously identified to develop the disease.^39,75

A total of 1742 full necropsies were performed. Others were found dead and not necropsied due to advanced autolysis. A total of 20 885 different diagnoses were made, with an average of 12 diagnoses per mouse in the study. A summary of the most common lesions identified by strain, excluding neurologic lesions, is shown in Tables 1 and 2. Detailed summaries of all lesions or specific cancer types by strain are published elsewhere.^2,74

Table 1.

Condensed Summary of Diseases in 20-Month-Old Female Mice Representing 28 Inbred Strains of Mice.^a

^aFrequently diagnosed diseases in almost all mice of a strain are in darker gray tones. Intermediate grey tones indicate that about half the mice were affected. Light gray tones indicate that at least 1 mouse was affected in the group. Data are shown only for lesions that affected at least 1 strain at an intermediate level. Neurologic diseases are not reported here.

very common diagnosis; moderately common diagnosis; occasional diagnosis.

Table 2.

Condensed Summary of Diseases in 20-Month-Old Male Mice Representing 28 Inbred Strains of Mice.^a

^aFrequently diagnosed diseases in almost all mice of a strain are in darker gray tones. Intermediate gray tones indicate that about half the mice were affected. Light gray tones indicate that at least 1 mouse was affected in the group. Data are shown only for lesions that affected at least 1 strain at an intermediate level. Neurologic diseases are not reported here.

very common diagnosis; moderately common diagnosis; occasional diagnosis.

Diseases and Causes of Death

The mice selected at 12 and 20 months of age were, for the most part, in reasonably good health and were euthanized by CO₂ asphyxiation. However, detailed histologic analyses of all organ systems revealed a wide variety of chronic diseases, the most frequent of which are listed in Tables 1 and 2. These tables compress variations in disease presentations (eg, dystrophic cardiac calcinosis, which includes epicardial and myocardial mineralization and fibrosis [Berndt A, Li Q, Sundberg BA, et al, “Mouse Genome-wide Association Study Identifies Polymorphisms on Chromosomes 4, 11 and 15 for Age-Related Cardiac Fibrosis,” unpublished data]) that are defined in greater detail in other focused studies. Neurologic lesions were reviewed by others and are not reported here.

Some lesions were found in all or almost all mice regardless of strain or sex, such as oral foreign body impactions (hair impaction secondary to grooming and food material impacted around teeth or in the lingual gland duct, causing acute to chronic inflammation and, in some cases, avulsion of teeth; Fig. 1),³⁶ degenerative changes of the preputial and clitoral glands (ductal ectasia, sebocyte atrophy, and various degrees of acute to chronic inflammation or abscessation),⁶³ temporomandibular joint disease in very old mice in the longitudinal study (Figs. 2, 3), or pancreatic islet hyperplasia.⁹ Other diseases, especially cancer, were for the most part not consistent in any one strain, with the following exceptions (Figs. 4–8): lymphomas in AKR/J,^68,69 adrenal cortical adenomas in C3H/HeJ (Fig. 4),²⁶ and rhabdomyosarcomas in very old A/J mice (Figs. 7, 8).⁶⁵ Mast cell tumors are extremely rare spontaneous findings in inbred mice, limited in this study to single cases in FVB/NJ (Figs. 5, 6) and NZW/LacJ mice. Detailed summaries for cancer diagnoses by strain are reported elsewhere.² Eye lesions have been well described in laboratory mice, and these were similar in the aged mice.⁶⁷

Figure 1. Periodontal disease, 648-day-old male 129S1/SvLmJ mouse. Note the foreign body (hairs) between the oral mucosa and molar, resulting in acute inflammation of the underlying alveolar bone. Hematoxylin and eosin (HE). Figure 2. Temporomandibular joint disease in a 636-day-old female A/J. The articular cartilage is fragmented, filling the interarticular space with “joint mice.” This lesion is common in most old inbred mouse strains. HE. Figure 3. Higher magnification of boxed area in Figure 2 to illustrate fragmentation of the cartilage. HE. Figure 4. Adrenocortical adenoma, 619-day-old male C3H/HeJ mouse. HE. Figure 5. Mast cell tumor, skin, 639-day-old male FVB/NJ mouse. Figure 5. (continued) Toluidine blue. Figure 6. Higher magnification of the boxed area in Figure 5, illustrating the metachromatic cytoplasmic granules. Toluidine blue. Figure 7. Rhabdomyosarcoma, epaxial muscles, 718-day-old female A/J mouse. HE. Figure 8. Higher magnification of the boxed area in Figure 7, illustrating the pleomorphic features of the neoplastic cells. HE. Figures 9–11. Amyloidosis, ileum. Figure 9. 945-day-old male C57BL/6J mouse. Figure 10. 637-day-old female C57BL/10J mouse. Figure 11. 630-day-old female SWR/6J mouse. A, HE; B, Congo red; C, Congo red with polarized light. The deposits stain orange-red with Congo red and have birefringence characteristic of amyloid. Note the high degree of histologic similarity among the 3 mouse strains, which is a critical feature for grouping strains for genome-wide association studies.

GWAS Analysis

Not surprising, because of how inbred strains were developed,⁵³ groups of related strains often had similar or identical diseases. Due to the large amount of data collected, it is not possible to include detailed analyses on all diseases in an overview of this kind; therefore, individual workups integrating GWAS analyses with detailed phenotyping are being reported elsewhere (Berndt A, Li Q, Sundberg BA, et al, “Mouse Genome-wide Association Study,” unpublished data).^5

–9 To illustrate how these computational methods can be applied to aging disease data sets, amyloidosis of the small intestine, well known in some C57BL/6 substrains,^37,55 is used as an example. Amyloidosis of the small intestine was found in about half of the strains in this study (Figs. 9–12), not just C57BL/6J. Genome-wide scans were performed with the expedited efficient mixed-model association algorithm based on a 4 million–SNP panel from the National Institute of Environmental Health Sciences. Twenty-month-old mice, females and males, were evaluated and scored for small intestinal amyloidosis in 28 inbred strains.⁷⁴ Frequency of intestinal amyloidosis lesions was most strongly associated (P < 10⁻⁹) with genome variations on chromosomes 1 and 10: Chr 1 at 13.549419 Mb (MM38-1-13549419:NES15822440; P = 4.86·10⁻¹⁰) and Chr 10 at 12.92 Mb (NES11975634; P = 4.86325E-10), 12.95 Mb (NES11974532; P = 4.86325E-10), 13.023 Mb (NES11972003; P = 4.86325E-10), and 13.024 Mb (NES11971958; P = 4.86325E-10), which contained 3 candidate genes: Tram1, Stx11, and Sf3b5 (Table 3).

Figure 12.

Frequency (percentage of affected mice by strain) of amyloidosis of the small intestine in 28 inbred strains of mice at 20 months of age. F, female; M, male. Note that the C57 strains and substrains are commonly affected.

Table 3.

Candidate Genes for Amyloidosis of the Small Intestine as Identified by Genome-wide Association Studies of 28 Aged Inbred Mouse Strains.^a

Gene Symbol: Name
CHR	Position	P Value	rs No.	Relative Location
Tram1: translocating chain–associating membrane protein 1^b
1	13549419	4.86325E-10	MM38-1-13549419: NES15822440	Upstream
Stx11: syntaxin 11^c
10	12927772	4.96325E-10	NES11975634	Upstream
10	12955618	4.96325E-10	NES11974532	Intronic
Sf3b5: splicing factor 3b, subunit 5^d
10	13023341	4.96325E-10	NES11972003	Downstream
10	13024458	4.96325E-10	NES11971958	Downstream

Abbreviation: CHR, chromosome.

^aMouse Genome Informatics (January 22, 2015).

^bMutant mice: Tram1^{Gt(A065A02)Wrst} , Tram1^{Gt(BC0375)Wtsi} , Tram1^{Gt(OST15511)Lex} , Tram1^{Gt(XR0010)Wtsi} , targeted ES cells only.

^cMutant mice: Stx11^tm1.2Ics , hemophagocytic lymphohistiocytosis after LCMV infection.⁶⁴

^dMutant mice: Sf3b5^{tm1(KOMP)Vlcg} , targeted ES cells only.

Discussion

While the data gathered here are enormously valuable for understanding the types and diversity of diseases found in commonly used inbred mouse strains, one of the most valuable outcomes was an understanding of how to set up these types of histopathology-based studies of aging. Debates continue regarding whether (1) to focus on moribund mice to establish a mean life span with detailed histopathology near the time of death or (2) to collect samples from mice at specific ages or time points. Based on experience doing large-scale disease surveillance based on limited, lesion-specific diagnostic workups rather than complete necropsies, disease trends can be found in young mice up to 8 months of age.⁷⁶ However, for very detailed multiorgan histologic studies, only small numbers of old mice are practical due to animal maintenance costs and the perceived high cost of histopathology.⁷⁹ Diseases were identified in 12-month-old mice that were also found in 20-month-old group but less frequently. Some diseases became obvious only at a very old age. For example, rhabdomyosarcomas changed from a rare to a common disease in A/J mice >20 months of age.⁶⁵ By contrast, AKR/J mice are highly prone to lymphomas and rarely survive to 20 months of age, and those that do have few, if any, other lesions.^68,69 Evaluating mice at standardized ages does provide a cross-sectional profile of all strains at a particular age; however, it is not the best way to determine what diseases are life-threatening and potentially the ultimate cause of death in any particular strain. Aging mice, until they become moribund, provide a higher chance of discovering strain-specific disease predisposition and correlating these diseases to mean life span. However, many of these mice die before they can be necropsied, making generation of population data more difficult. Regardless, both approaches yield valuable information.

Another concern is whether histopathology is a useful, cost-effective tool for evaluating phenotypes or if physiologic (in vivo) phenotyping⁶⁰ (ie, panels of functional diagnostic tests) provide an adequate and more cost-effective means of evaluating mouse health.⁴¹ Histopathology, when done by experienced personnel, has for >150 years been the stand-alone standard for arriving at a diagnosis. By contrast, physiologic phenotyping approaches reflect human clinical practice and, in the context of model organisms, are often applied later in hypothesis-driven approaches to understanding phenotypes as confirmatory assays selected on the basis of histopathology findings. A problem with physiologic assays is that they provide data out of context. For example, high red blood cell counts were observed in some members of a cohort of a single strain (SM/J females; Table 1). While statistical analyses would suggest that this might be an error in the hematologic assay and exclude the data, the histologic diagnosis that affected mice had an erythropoietic neoplasm revealed the real significance of these data. Histopathology in most cases can be used independently to arrive at a definitive answer, often with direct correlation with a human disease.

The value of these histopathologic data sets is dependent on the skill of the pathologist or investigator evaluating the study. For example, teratomas^19,20 and sebaceous gland tumors^44,92 reported in aged mutant mice were actually misdiagnoses of common, nonneoplastic, normal “aging changes” of the preputial and clitoral glands (Tables 1 and 2), organs found in mice but not humans.⁶³ Hair shaft impaction around the incisors and especially the molars, causing foreign body periodontitis, was supposedly associated with ulcerative dermatitis in C57BL/6 strains and hybrids¹⁶ but is found in almost all mice regardless of their hair coat or cutaneous ulcers^36,77 and was commonly diagnosed in C57BL/6J and many other strains in this aging study in mice with no concurrent alopecia or other skin condition. Many other examples are abundant in the literature.²⁸

While the interest of pathologists is and should be in the variety of types of lesions that old mice develop, especially in strains beyond those commonly used, it is fascinating to see that groups of inbred strains, sometimes closely related but not necessarily, have essentially the same diseases while others do not. As an example, amyloidosis of the small intestine is a well-described finding in aging mice, particularly in some of the C57BL/6 substrains.^37,55 This diagnosis was confirmed in the study reported here. By treating this as a binary trait—that is, either a strain had the disease or it did not—GWAS analyses identified 3 candidate genes for amyloidosis of the intestine (Tram1, Sf3b5, and Stx11).

Only 1 of the 3 candidate genes identified has a known role in inflammation or amyloidosis (Stx11), and this is discussed below. An approach to further establishing whether genes associated with particular phenotypes may have a hitherto unknown role is to consider known protein-protein interactions and network membership—the so-called guilt-by-association approach.⁴⁷ Based on the modularity of disease phenotypes,⁵⁰ this approach has now become widely used. Using protein-protein and genetic interaction data, we explored the immediate interactome surrounding our candidate genes and report significant molecular or regulatory associations that might explain their genetic association with amyloidosis.⁹³

Tram1 presents an interesting candidate for involvement in amyloidosis and amyloid-accumulating diseases such as Alzheimer disease. So far Tram1 has not been implicated genetically in amyloid-associated degenerative disease; however, a search of BioGRID¹³ for interacting partners presents an abundance of evidence for its potential involvement (Figure 13, Suppl. Table 1). TRAM1 is an integral transmembrane glycoprotein of 36 kDa, which forms part of the translocon in the endoplasmic reticulum, is involved in membrane insertion of nascent proteins, and has a role in the disposal of misfolded proteins in the endoplasmic reticulum.⁴⁵ This is a mechanism common to amyloid accumulation in many tissues irrespective of the protein involved. TRAM1 interacts directly with amyloid precursor protein (APP),⁴⁵ as shown in an interactome screen in Alzheimer disease, as well as with the SEC61P complex. It has been shown to bind SEC61B by direct crosslinking.³⁰ SEC61B has been shown to be a critical protein in the cytoplasmic export of amyloid beta.³⁰ In Alzheimer disease and other amyloidosis syndromes, the misfolded peptide accumulates both in extracellular plaques and in the cytosol. Exported proteins are normally degraded in the cytosol by protease-dependent and protease-independent pathways. Other interacting partners of TRAM1, such as BCAP31 and DERL1, are also involved in protein translocation and endoplasmic reticulum processing and degradation, which is a precursor for amyloid accumulation.^{27,49,62,82,83}

Figure 13.

Interaction diagrams from BioGRID for STX11 and TRAM1. BioGRID 3.3.123 searched 44 772 publications for 823 466 raw protein and genetic interactions from major model organism species. The diagram shows physical interactions as edges connecting gene products (ie, protein-protein interactions) for all of the interspecies interactions, self-interaction, and interactions between interactors. The raw data and evidence are shown in Supplemental Tables 1 and 2.

TRAM1 also interacts with HM13.⁴⁹ Previously known as presenilin-like protein 3, HM13 is a GxGD class of proteinase of the same family as the presenilins and acts as a transmembrane ion channel.³² The presenilins are responsible for cleavage of the cleavage of APP and are implicated in Alzheimer disease,⁸⁷ although direct involvement of HM13 in other amyloid accumulation diseases has not yet been shown.

TRAM1 functions as a cofactor to promote efficient cytomegalovirus US2- and US11-dependent dislocation of major histocompatibility complex class I heavy chains and seems to hijack the translocation complex, also involving DERL1.⁴⁹ Recently, a suggestive link has been made between Alzheimer disease and cytomegalovirus susceptibility.³⁵ It may be interesting to pursue the relationship between cytomegalovirus infection and apparently sporadic amyloidosis, with the hypothesis that subvention of the TRAM1-associated translocation complex may compete for translocation of other proteins, such as amyloid beta retrotranslocated protein.

Finally, the direct interaction with VKORC1, the vitamin K epoxide reductase complex subunit 1,⁵⁷ is intriguing. While VKORC1 does have direct interaction with TRAM1, APLP2, and BCAP31, the gene has never been directly linked to amyloid accumulation, being a primary determinant of blood-clotting efficiency and sensitivity to warfarin.⁵⁴ However, the Vkorc1^tm1.1Jobg knockout mouse shows phenotypes consistent with cerebral amyloid angiopathy as well as the expected clotting defects.⁷⁰

In humans, SF3B5 interacts with survival motor neuron domain containing 1 (SMNDC1), which is genetically associated with Crohn disease and therefore may be involved in the site specificity of intestinal amyloidosis.⁸⁹ Targeted embryonic stem cells have been made for Sf3b5; however, live mice have not yet been produced. It is possible that they will be created as part of the International Mouse Phenotyping Consortium on the C57BL/6N background, but even in those studies, mice will be aged only to 15 weeks, too early to detect any effect on intestinal amyloidosis.

Stx11 is a member of the soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) family, which function as executors of exocytosis during CTL cell killing and are important for the fusion of lytic granules at the immunologic synapse.²⁴ In humans, a deficiency in syntaxin 11 (STX11) causes familial hemophagocytic lymphohistiocytosis type 4 (FHL4).^12,94 Genetically engineered Stx11^tm1.2Ics null mice on a C57BL/6 background had a normal immune system when compared with wild-type controls with no evidence of amyloidosis, but this was not the focus of the study. Disease similar to FHL4 in humans could be reproduced in these mutant mice by infection with lymphocytic choriomeningitis virus.⁶⁴

A search of BioGRID (Fig. 13, Suppl. Table 2) revealed that the immediate interactome of STX11 contains several genes of potential interest, including amyloid beta A4 protein.⁴⁶ The interactions with Vamp2 (synaptobrevin), Snap23 (synaptosomal-associated protein 23), and Psma3 (Proteasome subunit alpha type-3), a component of the proteasome, reflect the involvement of STX11 in vesicle transport.⁷² However, their interactions are modified by amyloid resulting in modulation of neurotransmitter release.⁵⁶

Interestingly STX11 interacts with TNFRSF21,⁸¹ a member of the TNF-receptor superfamily also called death receptor 6 (DR6), which has an important role in inflammation and both B- and T-cell function and proliferation⁴ and functions with APP in axonal branching control.⁴⁸ Two allelic mutations exist in mice (Stx11^tm1.2Ics and Stx11^tm1Bupa , as well as several ES cell lines, MGI:2151075). These have been mainly studied from the standpoint of immune and vascular function or neurobiology, so it is not clear whether extra–CNS amyloidosis would have been detected and reported.

Overall, the associations that we have found between TRAM1 and STX11 and their interacting partners suggest possible novel pathogenetic roles for these proteins based on their activity as part of larger complexes and networks. That there may be several predisposing factors to the pathogenesis of intestinal amyloidosis is consistent with what we know about its pathobiology, and the analysis offered here provides new candidates for further mechanistic investigation.

This study demonstrated how the combination of defined and extensive genetic variation in inbred strains, with detailed phenotyping, can generate hypotheses about the genetic basis for relative predisposition to the phenotypes or diseases of interest. The genes that we have identified therefore provide hypotheses about the underlying pathogenetic process. Association does not necessarily imply causation, and as with all GWAS analyses, hypothesis-driven experimentation is now needed to support or reject the genetically derived hypothesis. The causative genes associated with identified SNPs are not necessarily the closest, and it is very rare that the SNPs themselves are the cause of the altered gene function.¹⁷ Use of informatics, annotations, and extensive literature searches can provide clues regarding which associations reflect true involvement,³³ but we are far from having enough phenotype and genotype data for this to be useful in all cases. Until we have all of the systematic phenotyping from the International Mouse Phenotyping Consortium completed, the gaps and biases in data will render any prioritization approach suboptimal but valuable. Finally, using whole genome variant analysis instead of imputed SNP sets may eventually provide targeted identification of true genomic variants.

Footnotes

Acknowledgement

P.N.S. thanks the Warden, Fellows, and Scholars of Robinson College, Cambridge, for travel support.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: JPS, KAS, and VEK have a research contract with Bioniz, LLC, and a research grant from NAAF for AA preclinical trials. All other authors state that they have no conflicts of interest.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by grants from the Ellison Medical Foundation, Parker B. Francis Foundation, National Institutes of Health (AG25707, for the Shock Aging Center; CA89713, for the Mouse Tumor Biology Database). The Jackson Laboratory Shared Scientific Services were supported in part by a Basic Cancer Center Core Grant from the National Cancer Institute (CA34196, to The Jackson Laboratory).

Supplemental material for this article is available on the Veterinary Pathology website at .

References

Adzhubei

Schmidt

Peshkin

. A method and server for predicting damaging missense mutations. Nat Methods. 2010;7(4):248–249.

Begley

Sundberg

Berndt

. Diversity of spontaneous neoplasms in commonly used inbred strains of laboratory mice. In: Hedrich

, ed. The Laboratory Mouse. 2nd ed. London, UK: Academic Press; 2012:409–424.

Begley

Krupke

Neuhauser

. Identifying mouse models for skin cancer using the Mouse Tumor Biology Database. Exp Dermatol. 2014;23(10):761–763.

Benschop

Wei

. Tumor necrosis factor receptor superfamily member 21: TNFR-related death receptor-6, DR6. Adv Exp Med Biol. 2009;647:186–194.

Berndt

Ackert-Biknell

Silva

. Genetic determinants of fibro-osseous lesions in aged inbred mice. Exp Mol Pathol. 2015;100(1):92–100.

Berndt

Cario

Silva

. Identification of Fat4 and Tsc22d1 as novel candidate genes for spontaneous pulmonary adenomas. Cancer Res. 2011;71(17):5779–5791.

Berndt

Leme

Williams

. Comparison of unrestrained plethysmography and forced oscillation for identifying genetic variability of airway responsiveness in inbred mice. Physiol Genomics. 2011;43(1):1–11.

Berndt

Potter

. A single nucleotide polymorphism in the Abcc6 gene associates with connective tissue mineralization in mice similar to targeted models for pseudoxanthoma elasticum. J Invest Dermatol. 2013;133(3):833–836.

Berndt

Sundberg

Silva

. Phenotypic characterization of the KK/HlJ inbred mouse strain. Vet Pathol. 2014;51(4):846–857.

10.

Bleich

Kirsch

Sahly

. Klebsiella oxytoca: opportunistic infections in laboratory rodents. Lab Anim. 2008;42(3):369–375.

11.

Bronson

. Rate of occurrence of lesions in 20 inbred and hybrid genotypes of rats and mice sacrificed at 6 month intervals during the first years of life. In: Harrison

, ed. Genetic Effects on Aging II. Caldwell, NJ: Telford Press; 1990:279–358.

12.

Bryceson

Rudd

Zheng

. Defective cytotoxic lymphocyte degranulation in syntaxin-11 deficient familial hemophagocytic lymphohistiocytosis 4 (FHL4) patients. Blood. 2007;110(6):1906–1915.

13.

Chatr-Aryamontri

Breitkreutz

Oughtred

. he BioGRID interaction database: 2015 update. Nucleic Acids Res. 2015;43(database issue):D470–D478.

14.

Committee on Animal Models for Research on Aging. Mammalian Models for Research on Aging. Washington, DC: National Academy Press; 1981:25–75.

15.

Cong

Ran

Cox

. Multiplex genome engineering using CRISPR/Cas systems. Science. 2013;339:819–823.

16.

Duarte-Vogel

Lawson

. Association between hair-induced oronasal inflammation and ulcerative dermatitis in C57BL/6 Mice. Comp Med. 2011;61(1):13–19.

17.

Edwards

Beesley

French

. Beyond GWASs: illuminating the dark road from association to function. Am J Hum Genet. 2013;93(7):779–797.

18.

Fekete

. Comparative morphological study of the mammary gland in a high and a low tumor strain of mice. Am J Pathol. 1938;14:557–583.

19.

Pelicano

Liu

. The circadian gene Period2 plays an important role in tumor suppression and DNA damage response in vivo. Cell. 2002;111:41–50.

20.

Pelicano

Liu

. Erratum. Cell. 2002;111:1055.

21.

Sander

Reyon

. Improving CRISPR-Cas nuclease specificity using truncated guide RNAs. Nature Biotechnology. 2014;32:279–284.

22.

Gaj

Gersbach

III . CFB: ZFN, TALEN, and CRISPR/Cas-based methods for genome engineering. Trends Biotechnol. 2013;31(7):397–405.

23.

Giehl

Potter

. Hair interior defect (hid) in AKR/J mice maps to mouse Chromosome 1. Clin Exp Dermatol. 2009;34:509–517.

24.

Halimani

Pattu

Marshall

. Syntaxin11 serves as a t-SNARE for the fusion of lytic granules in human cytotoxic T lymphocytes. Eur J Immunol. 2014;44:573–584.

25.

Hayamizu

Mangan

Corradi

. The adult mouse anatomical dictionary: a tool for annotating and integrating data. Genome Biol. 2005;6:R29.

26.

Husler

Beamer

Boggess

. Neoplastic and hyperplastic lesions in aging C3H/HeJ mice. J Exp Anim Sci. 1998;38:165–180.

27.

Huttunen

Guénette

Peach

. HtrA2 regulates beta-amyloid precursor protein (APP) metabolism through endoplasmic reticulum-associated degradation. J Biol Chem. 2007;282(38):28285–28295.

28.

Ince

Ward

Valli

. Do-it-yourself (DIY) pathology. Nat Biotechnol. 2008;26(9):978–979.

29.

Kang

Sul

Service

. Variance component model to account for sample structure in genome-wide association studies. Nat Genet. 2010;42:348–354.

30.

Knight

High

. Membrane integration of Sec61alpha: a core component of the endoplasmic reticulum translocation complex. Biochem J. 1998;331(1):161–167.

31.

Krupke

Begley

Sundberg

. The mouse tumor biology database. Nature Rev Cancer. 2008;8(6):459–465.

32.

Kuo

. Presenilin-like GxGD membrane proteases have dual roles as proteolytic enzymes and ion channels. J Biol Chem. 2015;290(10):6419–6427.

33.

Leiserson

Eldridge

Ramachandran

. Network analysis of GWAS data. Curr Opin Genet Dev. 2013;23(6):602–610.

34.

Lipman

. Pathobiology of aging rodents: inbred and hybrid models. Exp Gerontol. 1997;32(1–2):215–228.

35.

Lurain

Hanson

Martinson

. Virological and immunological characteristics of human cytomegalovirus infection associated with Alzheimer disease. J Infect Dis. 2013;208(4):564–572.

36.

Mahler

Rozell

Mahler

. Pathology of the gastrointestinal tract of genetically engineered and spontaneous mutant mice. In: Ward

Mahler

Maronpot

Sundberg

, eds. Pathology of Genetically Engineered Mice. Ames, IA: Iowa State University Press; 2000:269–297.

37.

Martin

Kirkwood

TBL

Potten

. Age changes in stem cells of murine small intestinal crypts. Exp Cell Res. 1998;241:316–323.

38.

McClearn

McCarter

. Heterogeneous stocks and selective breeding in aging research. ILAR J. 2011;52(1):16–23.

39.

McElwee

Boggess

Miller

. Spontaneous alopecia areata-like hair loss in one congenic and seven inbred laboratory mouse strains. J Invest Dermatol Symp Proc. 1999;4:202–206.

40.

Mikaelian

Nanney

Parman

. Antibodies that label paraffin-embedded mouse tissues: a collaborative endeavor. Toxicol Pathol. 2004;32:1–11.

41.

Moeller

Hirose

Mueller

. Inbred mouse strains reveal biomarkers that are pro-longevity, anti-longevity or role switching. Aging Cell. 2014;13(4):729–738.

42.

Moolenbeek

Ruitenberg

. The “Swiss roll”: a simple technique for histological studies of the rodent intestine. Lab Anim. 1981;15(1):57–59.

43.

Myers

. Review of disease patterns and life span in aging mice: genetic and environmental interactions. In: Bergsma

Harrison

Paul

, eds. Genetic Effects on Aging. New York, NY: Alan R. Liss Inc; 1978:41–53.

44.

Nakamura

Fukami

. Phospholipase C-delta1 is required for skin stem cell lineage commitment. EMBO J. 2003;22:2981–2991.

45.

Oresic

Tortorella

. TRAM1 is involved in disposal of ER membrane degradation substrates. Exp Cell Res. 2010;316:2113–2122.

46.

Oláh

Vincze

Virók

. Interactions of pathological hallmark proteins: tubulin polymerization promoting protein/p25, beta-amyloid, and alpha-synuclein. J Biol Chem. 2011;286(39):34088–34100.

47.

Oliver

. Guilt-by-association goes global. Nature. 2000;403:601–603.

48.

Olsen

Kallop

McLaughlin

. Genetic analysis reveals that amyloid precursor protein and death receptor 6 function in the same pathway to control axonal pruning independent of β-secretase. J Neuroscience. 2014;34(19):6438–6447.

49.

Oresic

Tortorella

. TRAM1 participates in human cytomegalovirus US2- and US11-mediated dislocation of an endoplasmic reticulum membrane glycoprotein. J Biol Chem. 2009;284:5905–5914.

50.

Oti

Brunner

. The modular nature of genetic diseases. Clin Genet. 2007;71(1):1–11.

51.

Percy

Barthold

. Pathology of Laboratory Rodents and Rabbits. Ames, IA: Iowa State University Press; 2007.

52.

Peters

Robledo

Bult

. The mouse as a model for human biology; a resource guide for complex trait analysis. Nat Rev Genet. 2007;8(1):58–69.

53.

Petkov

Ding

Cassell

. An efficient SNP system for mouse genome scanning and elucidating strain relationships. Genome Res. 2004;14:1806–1811.

54.

Rosand

Hylek

O’Donnell

. Warfarin-associated hemorrhage and cerebral amyloid angiopathy: a genetic and pathologic study. Neurology. 2000;55(7):947–951.

55.

Rowlatt

Chesterman

Sheriff

. Lifespan, age changes and tumour incidence in ageing C57BL mouse colony. Laboratory Animals. 1976;10:419–442.

56.

Russell

Semerdjieva

Empson

. Amyloid-beta acts as a regulator of neurotransmitter release disrupting the interaction between synaptophysin and VAMP2. PLos One. 2012;7:e43201.

57.

Schaafhausen

Rost

Oldenburg

. Identification of VKORC1 interaction partners by split-ubiquitin system and coimmunoprecipitation. Thromb Haemost. 2011;105(2):285–294.

58.

Schofield

Bard

JBL

Boniver

. Pathbase: a new reference resource and database for laboratory mouse pathology. Radiat Prot Dosimetry. 2004;112(4):525–528.

59.

Schofield

Bard

JBL

Booth

. Pathbase: a database of mutant mouse pathology. Nucleic Acids Res. 2004;32:D512–D515.

60.

Schofield

Dubus

Klein

. Pathology of the laboratory mouse: an international workshop on challenges for high throughput phenotyping. Toxicol Pathol. 2011;39(3):559–562.

61.

Schofield

Gruenberger

Sundberg

. Pathbase and the MPATH ontology: community resources for mouse histopathology. Vet Pathol. 2010;47(6):1016–1020.

62.

Schrötter

Pfeiffer

Magraoui

. The amyloid precursor protein (APP) family members are key players in S-adenosylmethionine formation by MAT2A and modify BACE1 and PSEN1 gene expression-relevance for Alzheimer’s disease. Mol Cell Proteomics. 2012;11(11):1274–1288.

63.

Seely

Boorman

. Mammary gland and specialized sebaceous glands (Zymbal, preputial, clitoral, anal). In: Maronpot

Boorman

Gaul

, eds. Pathology of the Mouse: Reference and Atlas. Vienna, IL: Cache River Press; 1999:613–635.

64.

Sepulveda

Debeurme

Ménasché

. Distinct severity of HLH in both human and murine mutants with complete loss of cytotoxic effector PRF1, RAB27A, and STX11. Blood. 2013;121(4):595–603.

65.

Sher

Cox

Mills

. Rhabdomyosarcomas in aging A/J mice. PLos One. 2011;6(8):e23498.

66.

Silva

Sundberg

. Necropsy methods. In: Hedrich

, ed. The Laboratory Mouse. 2nd ed. London, UK: Academic Press; 2012:779–806.

67.

Smith

John

SWM

Nashina

. Systematic Evaluation of the Mouse Eye: Anatomy, Pathology, and Biomethods. Boca Raton, FL: CRC Press; 2002.

68.

Sommer

Lerchl

. The risk of lymphoma in AKR/J mice does not rise with chronic exposure to 50 Hz magnetic fields (1 microT and 100 microT). Radiat Res. 2004;162(2):194–200.

69.

Sommer

Streckert

Hansen

. No effects of GSM-modulated 900 MHz electromagnetic fields on survival rate and spontaneous development of lymphoma in female AKR/J mice. BMC Cancer. 2004;4:77.

70.

Spohn

Kleinridders

Wunderlich

. VKORC1 deficiency in mice causes early postnatal lethality due to severe bleeding. Thromb Haemost. 2009;101(6):1044–1050.

71.

Staff of the Roscoe B. Jackson Memorial Laboratory. The existence of non-chromosomal influence in the incidence of mammary tumors in mice. Science. 1933;78:465–466.

72.

Stelzl

Worm

Lalowski

. A human protein-protein interaction network: a resource for annotating the proteome. Cell. 2005;122(6):957–968.

73.

Sundberg

Schofield

Gruenberger

. A data capture tool for mouse pathology phenotyping. Vet Pathol. 2009;46(6):1230–1240.

74.

Sundberg

Berndt

Sundberg

. The mouse as a model for understanding chronic diseases of aging: the histopathologic basis of aging in inbred mice. Pathobiol Aging Age-related Dis. 2011;1(1):7179.

75.

Sundberg

Cordy

King

. Alopecia areata in aging C3H/HeJ mice. J Invest Dermatol. 1994;102:847–856.

76.

Sundberg

Ichiki

. Common diseases found in inbred strains of laboratory mice. In: Sundberg

Ichiki

, eds. Handbook on Genetically Engineered Mice. Boca Raton, FL: CRC Press; 2005:223–229.

77.

Sundberg

Rozell

Everts

. Association between hair-induced oronasal inflammation and ulcerative dermatitis in C57BL/6 mice. Comp Med. 2011;61(3):204–205.

78.

Sundberg

Schofield

. Integrating mouse anatomy and pathology ontologies into a diagnostic/phenotyping database: tools for record keeping and teaching. Mammalian Genome. 2008;19(6):413–419.

79.

Treuting

Snyder

Ikeno

. Improving reproducibility and translation of aging studies using rodents: the vital role of pathology. Vet Pathol. In press.

80.

Vaupel

. Biodemography of human ageing. Nature. 2010;464(7288):536–542.

81.

Vinayagam

Stelzl

Foulle

. A directed protein interaction network for investigating intracellular signal transduction. Sci Signal. 2011;4(189):rs8.

82.

Wang

Heath-Engel

Zhang

. BAP31 interacts with Sec61 translocons and promotes retrotranslocation of CFTRDeltaF508 via the derlin-1 complex. Cell. 2008;133(6):1080–1092.

83.

Wang

Nguyen

Breckenridge

. Uncleaved BAP31 in association with A4 protein at the endoplasmic reticulum is an inhibitor of Fas-initiated release of cytochrome c from mitochondria. J Biol Chem. 2003;278(16):14461–14468.

84.

Ward

Anver

Haines

. Chronic active hepatitis in mice caused by Helicobacter hepaticus . Am J Pathol. 1994;145:959–968.

85.

Ward

Fox

Anver

. Chronic active hepatitis and associated liver tumors in mice caused by a persistent bacterial infection with a novel Helicobacter species. J Natl Cancer Inst. 1994;86:1222–1227.

86.

Potter

Silva

. Mutations in sterol O-acyltransferase 1 (Soat1) results in hair interior defects in AKR/J mice. J Invest Dermatol. 2010;130(11):2666–2668.

87.

Xia

Wolfe

. Intramembrane proteolysis by presenilin and presenilin-like proteases. J Cell Sci. 2003;116:2839–2844.

88.

Yang

Wang

Shivalila

. One-step generation of mice carrying reporter and conditional alleles by CRISPR/cas-mediated genome engineering. Cell. 2013;154:1370–1379.

89.

Yang

S-K

Hong

Zhao

. Genome-wide association study of Crohn’s disease in Koreans revealed three new susceptibility loci and common attributes of genetic susceptibility across ethnic populations. Gut. 2014;63:80–87.

90.

Yuan

Paigen

LLPB

. Mice as a mammalian model for research on the genetics of aging. ILAR J. 2011;52(1):4–15.

91.

Yuan

Tsaih

Petkova

. Aging in inbred strains of mice: study design and interim report on median lifespans and circulating IGF1 levels. Aging Cell. 2009;8(3):277–287.

92.

Zanesi

Fidanza

Fong

. The tumor spectrum in FHIT-deficient mice. Proc Natl Acad Sci U S A. 2001;98(18):10250–10255.

93.

Zhou

Menche

Barabasi

. Human symptoms-disease network. Nature Commun. 2014;5:4212.

94.

zur Stadt

Schmidt

Kasper

. Linkage of familial hemophagocytic lymphohistiocytosis (FHL) type-4 to chromosome 6q24 and identification of mutations in syntaxin 11. Hum Mol Genet. 2005;14(6):827–834.

Approaches to Investigating Complex Genetic Traits in a Large-Scale Inbred Mouse Aging Study

Abstract

Keywords

Materials and Methods

Mice

Aging Colonies

The Cross-sectional Study

The Longitudinal Life Span Study

Tissue Collection

Data Management

GWASs and Variant Identification

Results

Numbers of Mice Examined

Diseases and Causes of Death

GWAS Analysis

Discussion

Footnotes

Acknowledgement

Declaration of Conflicting Interests

Funding

References