Identification of vacuolar autophagic aggregates in the skeletal muscles of inbred C57BL/6NCrl mice

Introduction

Animal-based research is vital to understanding many diseases and key biological mechanisms that are the basis of neuroscience, physiology and toxicology.^1,2 Laboratory mice are the most common animals used in biomedical research, since their anatomical, physiological and genetic features present remarkable advantages for scientists.^3,4 Their small size and short reproductive cycle makes their husbandry and breeding relatively easy. Moreover, murine genome sequencing has revealed they are genetically very close to humans.^1,2,5,6 Most importantly, modern inbreeding techniques produce lines and strains of mice that are genetically uniform and homozygous at virtually all the loci.^5,6 Every inbred mouse is essentially a genetically identical clone of its parents and siblings, allowing for the perpetual propagation of genetically identical animals.^5,6 However, even fully inbred strains may develop variations due to genetic contamination or accumulation of new, spontaneous or artificially induced mutations.^2,7–9 This study aimed to report and describe a disorder affecting the skeletal muscles of an inbred C57BL/6NCrl colony. Characterised by the large sarcoplasmic vacuoles in the muscle fibres of male mice, the affected mice showed an increase in size and number of vacuoles with age. In this study, a thorough diagnostic process was directed to uncover the nature of the vacuole contents and to investigate the potential pathogenesis underlying their accumulation. Histological and immunohistochemical analyses oriented our etiopathogenetic hypothesis to dysregulation of the autophagy machinery while ruling out a morphologically very similar condition characterised by the accumulation of tubular aggregates (TAs).^10,11

Autophagy is a well-recognised cellular mechanism fundamental to eliminating and recycling waste material through the degradation of cellular components within lysosomes.^12–14 In normal skeletal muscles, lysosomes or autophagic vacuoles are morphologically unremarkable, but they can become prominent in certain muscle diseases known as autophagic vacuolar myopathies (AVMs).^15–17 Among AVMs, there is a group of myopathies known as autophagic vacuoles with sarcolemmal features (AVSFs) in which the vacuoles are lined by sarcolemmal proteins such as dystrophin. AVSFs define a group of diseases, including Danon disease, X-linked myopathy with excessive autophagy (XMEA), infantile autophagic vacuolar myopathy, adult-onset autophagic vacuolar myopathy with multiorgan involvement and X-linked congenital autophagic vacuolar myopathy.^15–17 We believe it is essential to report every new spontaneous disorder affecting inbred mice colonies to support the scientific community in distinguishing experimentally induced lesions from those occurring naturally.

Methods

C57BL/6NCrl mice were originally purchased from Charles River (Charles River Italia, Lecco, Italy) and housed in the Istituto di Ricerche Genetiche ‘Gaetano Salvatore’ (IRGS, Ariano Irpino, Avellino, Italy) facility to establish an inbred colony of mice. Animal care and use, and all the procedures, were in accordance with the Guide for the Care and Use of Laboratory Animals. ¹⁸ All animal procedures were reviewed and approved by the local Ethics Committee ‘Comitato Etico per la Sperimentazione Animale’ of IRSG and conformed to the regulations and guidelines of Italy and the European Union. All efforts were made to minimise animal suffering. Specific pathogen-free conditions were guaranteed by group housing two to three mice of the same sex and age in individually ventilated cages (Tecniplast, Varese, Italy). Mice were housed on a 12-hour/12-hour light/dark cycle, with a light phase from 7:00 a.m to 7:00 p.m. They received a pelleted, autoclavable rodent diet (long-term maintenance diet; cat. no. 4RF18; Mucedola srl, Settimo Milanese, Italy) and acidified water (HCl, pH 3.5) ad libitum. A constant room temperature (22 ± 2°C) and relative humidity (55 ± 10%) were maintained. Based on FELASA recommendations, the breeding colony was monitored annually. ¹⁹ Mice were visually inspected daily by an animal caretaker to assess their general condition and identify major clinical abnormalities. Each animal was clinically examined by the facility veterinarian at least once a week before euthanasia. For breeding, two females were housed with one male per cage and kept at no more than 10 months of age or until there was a decreased reproductive activity. Over 20 months, 50 C57BL/6NCrl mice (32 males and 18 females) were euthanised by cervical dislocation according to the American Veterinary Medical Association guidelines. ²⁰ Based on age and sex, animals were divided into four groups: group A comprised male mice aged up to 3 months, group B comprised male mice aged up to 10 months, group C comprised female mice aged up to 3 months and group D comprised female mice aged up to 10 months. Since the animals bred in the facility were commonly used to develop mutant mice, a complete post-mortem examination ²¹ and histological evaluation were performed to ascertain the phenotypes and pathological changes in the background control strain. Tissue samples were preserved in 10% neutral buffered formalin (code no. 05-01007Q; Bio-Optica, Milan, Italy), dehydrated and embedded in paraffin (code no. 06-7920; Bio-Optica, Milan, Italy). Furthermore, samples from the triceps brachii, tibialis anterior and gastrocnemius muscles were collected and snap-frozen in isopentane precooled in liquid nitrogen within two hours of sampling.

Results

Histopathology and IHC

Histological examination of formalin-fixed and paraffin-embedded sections of selected organs showed no evident pathological changes. In 29/32 (90.6%) male C57BL/6NCrl mice, morphological analysis of fresh frozen muscle samples revealed the presence of basophilic (with H&E stain; Figure 1(a) and (b)) or red purple (with ET stain) vacuoles in both the intermyofibrillar network and in contact with the sarcolemma (Figure 1(d) and (e)). The severity of vacuole accumulation (percentage of fibres affected) varied with the age and sex of the animal: group A mice had a lower percentage (25%) of affected fibres and smaller vacuoles compared to group B mice, which had a higher percentage (>50%) of affected fibres. The skeletal muscles of age-matched female mice (groups C and D) showed no abnormalities or vacuole accumulation (Figure 1(c) and (f)). There was a mild to moderate variation in fibre size, with many hypotrophic and some slightly hypertrophic fibres. However, although the vacuoles severely affected several fibres, there was no fibre necrosis, inflammatory infiltrates or increased endomysial connective or adipose tissue proliferation.

OPEN IN VIEWER

Figure 1.

Histological findings from the skeletal muscles of C57BL/6NCrl mice. H&E staining shows the presence of basophilic vacuoles in affected animals in (a) group A and (b) group B, staining blue or light violet. Skeletal muscles of age-matched female mice from (c) groups C and D show no pathological alterations. The skeletal muscles of affected mice from (d) group A and (e) group B have vacuoles staining red or purple with Engel’s trichrome and (f) The skeletal muscles from adult female unaffected littermate (group D). Group A: male mice aged up to 3 months; group B: male mice aged up to 10 months; group C: female mice aged up to 3 months; group D: female mice aged up to 10 months. H&E: hematoxylin and eosin; scale bar = 50 μm.

The Mann–Whitney test showed statistically significant differences in vacuole accumulation between (a) the group A and group B mice (p = 0.001) and (b) male mice (groups A and B) and age-matched female littermates (groups C and D; p = 0.001; Figure 2(a)). Significant differences were observed in the degree of muscle atrophy between mice aged up to 10 months (groups B and D) and sex-matched mice (p = 0.001; groups A and C; Figure 2(b)). The groups showed no significant differences in the degree of muscular inflammation.

OPEN IN VIEWER

Figure 2.

Statistical analysis of vacuole accumulation and muscle atrophy in the skeletal muscles of C57BL/6NCrl mice. (a) Based on the Mann–Whitney test, mice in groups A and B show significant differences in vacuole accumulation. Moreover, vacuole accumulation in the male mice (groups A and B) also significantly differs from that in age-matched female littermates (groups C and D) and (b) Significant differences were observed in the degree of muscle atrophy between mice in groups B and D compared to sex-matched mice in groups A and C (***p < 0.001). Group A: male mice aged up to 3 months; group B: male mice aged up to 10 months; group C: female mice aged up to 3 months; group D: female mice aged up to 10 months.

In the affected skeletal muscles, the mosaic pattern of fibre distribution revealed no selective fibre type atrophy or selectivity for vacuolar changes. The non-vacuolated regions had typical myofibrillar architecture, and some focal densities with NADH-TR probably corresponded to the reticulum modifications (Figure 3(a)). SDH and COX staining showed several areas with loss in mitochondrial oxidative enzyme activities, giving the fibres a moth-eaten aspect (Figure 3(b) and (c)), suggesting a mild alteration in the mitochondrial number and distribution. ATPase staining revealed stainless vacuoles in dark brown type II muscle fibres (Figure 3(d)). Histochemical staining demonstrated the vacuoles were negative for PAS and Congo red stain, ruling out glycogen and amyloid accumulation (Figure 3(e) and (f)). Moreover, the vacuoles appeared as dark brown deposits with non-specific esterase staining and light orange deposits with acid phosphatase staining (Figure 3(g) and (h)).

OPEN IN VIEWER

Figure 3.

Mitochondrial enzyme activity and vacuole contents in the affected skeletal muscles of C57BL/6NCrl mice. (a) Nicotinamide adenine dinucleotide dehydrogenase (NADH), (b) succinic dehydrogenase (SDH) and (c) cytochrome oxidase (COX) staining show several fibres with loss of mitochondrial oxidative enzyme activity (arrows), suggesting mild mitochondrial disruption, small decrease in mitochondrial functionality and slight variation in mitochondrial number and distribution, respectively. (d) ATPase 10.4 staining shows stainless vacuoles in dark brown type II fibres (arrows). Vacuoles are negative for both (e) periodic acid–Shiff and (f) Congo red stains, ruling out glycogen and amyloid accumulation, respectively. Vacuoles are positive for (g) esterase and (h) acid phosphatase stains. Scale bar = 50 μm.

Dystrophin immunolabelling showed normal sarcolemmal staining in all the skeletal muscles. Dystrophin staining was also observed in many vacuolar membranes in the affected mice (Figure 4(a)). Moreover, vacuoles in the affected skeletal muscles were positive for lysosome-associated membrane protein 2 (LAMP2) but negative for the micro-tubular marker calsequestrin (CASQ1; Figure 4(b) and (c)).

OPEN IN VIEWER

Figure 4.

Immunohistochemical staining of vacuolar proteins in the affected skeletal muscles of C57BL/6NCrl mice. Vacuoles are positive for (a) dystrophin (brownish, arrowhead) and (b) lysosomal marker lysosome-associated membrane protein 2 (LAMP2; arrows) but negative for (c) microtubular marker calsequestrin (CASQ1; arrows) immunostaining. Scale bar = 50 μm.

Double colour immunofluorescence

Co-localisation of two autophagy markers (LC3 and LAMP2) within the vacuoles was evaluated by double indirect immunofluorescence on sections of freshly frozen affected muscles. Co-labelling with fluorescein isothiocyanate (FITC)-conjugated LC3 (green) and tetramethylrhodamine (TRITC)-conjugated LAMP2 (red) antibodies resulted in yellow fluorescence (Figure 5(a)–(c)). In contrast, no co-localisation was seen with FITC-conjugated LC3 and TRITC-conjugated VMA21 staining (Figure 5(e)–(g)). Double-staining image analysis using two-dimensional plots of staining intensity and co-localisation revealed increased autophagic vacuoles within the affected skeletal muscles (Figure 5(d)). Moreover, the green signal corresponding to LC3 was most intense in vacuole-burden-affected skeletal muscles, whereas the red signal corresponding to VMA21 was absent (Figure 5(h)).

OPEN IN VIEWER

Figure 5.

Double labelling immunofluorescence. Co-localisation of (a) LC3 seen as green fluorescein isothiocyanate (FITC) immunofluorescence and (b) LAMP2 visible as red tetramethylrhodamine (TRITC) immunofluorescence is evident by the presence of (c) yellow fluorescence in the autophagic vacuoles within muscle fibres of the affected mice. (d) Two-dimensional plots of the staining intensity and co-localisation reveal an increase in both green and red signals, suggesting an increase in autophagic vacuoles within the affected skeletal muscles. (e) LC3 (green FITC immunofluorescence) and (f) absence of red TRITC labelling of VMA21 indicate that (g) LC3 and VMA 21 do not co-localise within the vacuoles and (h) Two-dimensional plots of the staining intensity and co-localisation reveal a green signal but no red signal.

TEM

Transverse ultra-thin muscle sections from affected mice were examined by TEM to investigate the ultrastructural features of vacuoles within the muscle fibres (Figure 6(a)–(c)). Numerous double membrane-bound autophagic vacuoles were observed, resulting in moderate disorganisation of myofibrils. Smaller vacuoles were present inside the myofibres and between myofibrils, while larger vacuoles were observed under the sarcolemma, where they appeared to open and extrude their contents into the extracellular space. Mitochondria occasionally presented moderate to severe damage with a partly cleared matrix or matrix devoid of electron-dense material and a severely dilated external membrane detached from the condensed inner membrane. These abnormalities were often associated with vacuoles showing membrane blebbing of possibly autophagic nature containing heterogeneous material, including small round dense bodies, granular material, membrane whorls (myelin figures) and degenerated mitochondria.

OPEN IN VIEWER

Figure 6.

Transmission electron microscopy of the skeletal muscles of C57BL/6NCrl mice. (a) and (b) Electron micrographs of skeletal muscle biopsies from an affected mouse show round vacuoles (arrows) lined by a single (pre-autophagosomes) or double (autophagosomes) membrane containing granular, weakly electro-dense material (residual bodies, asterisk) and (c) Membranous whorls (myelin-like figures) can be seen within the vacuoles (arrow). Scale bar = 500 nm.

Mutation identification and mRNA quantification

The murine VMA21 gene was sequenced to identify genetic alterations. In mice, VMA21 is located on chromosome X, with three exons and a transcript length of 4195 base pairs (Figure 7(a)). The human and murine gene sequences were aligned using the ClustalW2 multiple sequence alignment program for DNA (www.ebi.ac.uk/Tools/msa/clustalw2/) to identify homology in the six single nucleotides carrying the substitutions described for human XMEA patients (Figure 7(b)). In particular, we sequenced the intron–exon junctions for exons 2 and 3, involved in mRNA splicing efficiency, and the initial sequence of the 3′-UTR, involved in mRNA stability. We found substitutions in two of the six single nucleotides described for XMEA patients (Figure 7(b)). RT-PCR detected a significant (p = 0.0017, unpaired t-test) reduction in VMA21 mRNA expression in affected male mice compared to unaffected female littermates. There was no significant difference in muscle VMA21 mRNA levels between age-and sex-matched C57BL/6 control mice (Figure 7(c)).

OPEN IN VIEWER

Figure 7.

Mouse VMA21 gene sequencing. (a) Schematic diagram of the murine VMA21 gene, with the exons (rectangles). The human and mouse gene sequences were aligned to identify homology in the six single nucleotides carrying the substitutions described for human XMEA patients. Bullets indicate the positions of the corresponding mutations in the human gene. (b) Arrows indicate the polymerase chain reaction (PCR) primer positions. The human and murine gene alignment in the intron–exon junctions for exons 2 and 3 are indicated in red and black, respectively, and the 3′-UTR is indicated in blue. The positions of the corresponding mutations in the human gene are underlined and (c) The relative VMA21 mRNA expression measured by reverse transcription PCR in the skeletal muscles is shown as a ratio to GAPDH. Data from replicate analysis are presented as the mean ± SD. p = 0.0017 (unpaired t-test).

Discussion

In this study, we report a spontaneously occurring disorder in the skeletal muscles of male mice in a C57BL/6NCrl inbred colony. All biopsy samples had almost identical morphological features, characterised by numerous large vacuoles within the muscle fibres, in both the subsarcolemmal spaces and the intermyofibrillary network. Aggregates were observed mainly in the type II muscle fibres of male mice, increasing with age, both in size and in number. Clinically, we did not observe any muscle weakness or loss of ambulation, and there was no evidence of cardiac or respiratory involvement. Moreover, post-mortem evaluation and histological analysis excluded the presence of pathological accumulations or relevant lesions in other tissues and organs. Our investigation aimed to determine the nature of the vacuole contents and explore the pathogenesis underlying their accumulation. We found mild to moderate changes in fibre size, with many hypotrophic and some slightly hypertrophic fibres. However, there were no significant myopathic features, such as necrotic and regenerative fibres or inflammation. Furthermore, based on histochemical staining, the vacuoles were negative for glycogen and lipids, ruling out myopathy associated with pathological accumulations. However, they were positive for non-specific esterase and showed increased acid phosphatase activity, indicative of lysosomal accumulation.

Results from morphological and molecular analyses pointed to dysregulation of the autophagy machinery, leading to the formation and accumulation of autophagic vacuoles. Autophagy is a highly regulated process characterised by non-specific self-degradation of cytoplasmic macromolecules and organelles via the lysosomal system.^12–14 The three different mechanisms that deliver cellular cargo to the lysosomes include macroautophagy, microautophagy and chaperone-mediated autophagy.^12,13 Macroautophagy (herein referred to as autophagy) is the main lysosomal pathway for the turnover and renewal of cytoplasmic components. It involves the following phases: induction or initiation and cargo selection, vesicle nucleation and expansion, lysosome targeting, lysosome docking and autophagosome-lysosome fusion, vesicle breakdown, and recycling. Activation of specific autophagy effectors, including Beclin-1 (Atg6 orthologue) and LC3 (microtubule-associated protein 1 light chain 3), generates a phagophore that wraps around a portion of the cytoplasm and soluble proteins, aggregates or organelles to form a double membrane-bounded structure called ‘autophagosome’. The autophagosome fuses with lysosomes to form an autophagolysosome in which the lysosome hydrolases degrade the cytoplasmic cargo, and the products are ready to be recycled to synthesise new molecules. ¹³ Defects in the autophagic machinery have been associated with numerous pathological conditions such as cancer, autoimmunity and neurodegeneration.^12,29–32 We confirmed the increased accumulation of autophagic vacuoles in the skeletal muscles of the male C57BL/6NCrl mice by LC3 and LAMP2 immunoreactivity. In addition, electron microscopy revealed membrane-bound autophagic vacuoles often associated with moderate disorganisation of myofibrils, mitochondrial alterations and electron-dense membranous whorls (myelin figures). In humans, the accumulation of autophagic vacuoles in the skeletal muscles is a distinctive and pathognomonic morphological hallmark of a group of debilitating muscle disorders known as AVMs.^15–17,33 The AVMs include AVSFs, characterised by autophagic vacuoles, lined with sarcolemmal proteins such as dystrophin 1. ³⁴ These include Danon disease caused by the lack of LAMP2,^35,36 and X-linked myopathy with excessive autophagy triggered by mutations in VMA21, an essential assembly chaperone of V-ATPase.^15,29,37 The V-ATPase proton pump is vital and ubiquitous to all mammalian cells and is composed of over 13 subunits. Under normal conditions, V-ATPase uses ATP to pump protons and acidify organelles, including lysosomes, autolysosomes, the Golgi apparatus and others.^33,37 Decreased V-ATPase activity alters the lysosomal pH (from 4.7 to 5.2), ²⁹ partially blocking the final degradative phase of macroautophagy and the subsequent feedback up-regulation of the initial stages of autophagy with autophagosome generation and accumulation.^29,38 The autophagic vacuoles observed in the skeletal muscles mice in this study shared several morphological features with those classically described in XMEA patients (sarcolemmal features of vacuolar membranes), suggesting similar causative mechanisms. Thus, we investigated the expression of VMA21 protein by immunofluorescence and VMA21 mRNA expression by RT-PCR. Compared to age-matched female littermates, male mice showed lower levels of VMA21 protein and mRNA in the muscle tissues.

The formation and accumulation of debris-filled autophagosomes in the mice reflect the induction of the autophagic processes, albeit with a decreased capacity to degrade and recycle cellular components and organelles, consistent with the findings in XMEA patients. Mice carrying spontaneous mutations have often been considered unique animal models for human genetic diseases. In contrast to targeted mutations, spontaneous mutations often do not lead to complete loss of function, mimicking the subtler missense mutations of naturally occurring human inherited diseases. ³⁹ Traditionally, changes or alterations in the observable phenotypes are beneficial and practical for detecting spontaneous mutations in an inbred mouse colony. Spontaneous mutations are usually first identified based on relevant phenotypes such as modifications in coat colour (e.g. yellow, leaden), growth defects (e.g. dwarf, pigmy), abnormal morphology (e.g. limb deformity, legless), or alterations in behaviour or motor coordination (e.g. ataxia, circling). ⁶ Sequencing and aligning the murine VMA21 gene with its human counterpart showed substitutions in two of the six single nucleotide sequences described for patients with XMEA, suggesting a different mutation in mice. However, single mutations in the murine VMA21 gene could potentially decrease VMA21 mRNA splicing efficiency ³⁸ and protein production.

Autophagic vacuoles in mice were PAS negative and LAMP2 positive, excluding AVMs such as Pompe disease and Danon disease. TAs are another critical differential diagnosis for vacuole accumulation in skeletal muscle fibres. TAs are accumulations of densely packed tubules located between the myofibrils, particularly beneath the sarcolemma.^10,11,40,41 These aggregates usually increase in size and number with age and are typically localised in type II muscle fibres of male mice. TAs have been observed in several mouse myopathies, and they have also been reported in the skeletal muscles of normal inbred mouse strains.^11,40,42 It is generally agreed that TAs arise from the dilated terminal cisternae of the sarcoplasmic reticulum (SR), and their formation requires two temporally distinct steps operating via different mechanisms. ⁴¹ The SR Ca2+ binding protein calsequestrin (CASQ) initially accumulates at the I band level, causing free SR cisternae to swell. The SR sacs enlarged at the I band level extend into multiple longitudinally oriented tubules with a full complement of sarco(endo)plasmic reticulum Ca2+ ATPases (SERCA) in the membrane and CASQ in the lumen. These tubules gradually acquire a regular cylindrical shape and uniform size in concert with partial crystallisation of SERCA. ⁴¹ In our mice, TAs were excluded because vacuoles were negative for calsequestrin staining. Moreover, ultrastructure analysis of longitudinal and transverse muscle sections did not show the typical pathognomonic honeycomb-like structure consisting of closely packed clusters of tubules.

In conclusion, we describe a spontaneous condition in a C57BL/6NCrl inbred mouse colony with morphological and molecular features consistent with autophagosome accumulation, possibly due to impaired autophagy. This novel disorder may potentially represent a spontaneous animal model that could be utilised to further explore and dissect the role of autophagy in muscle physiology and pathology, allowing for new diagnostic and therapeutic advances. This study also aimed to highlight further the importance of a complete necropsy and histopathological analysis of inbred strains to define strain-specific and/or spontaneous lesions better, clarify the underlying mechanisms and aetiologies and comprehend if a specific phenotype results from experimental procedures or reflects a naturally occurring disease.

Supplemental Material

Supplemental Material