- Email: [email protected]
Mouse model of skeletal muscle adiposity: A glycerol treatment approach
Biochemical and Biophysical Research Communications 396 (2010) 767–773
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Mouse model of skeletal muscle adiposity: A glycerol treatment approach Didier F. Pisani a,*, Cynthia D.K. Bottema b, Catherine Butori c, Christian Dani a, Claude A. Dechesne a a
University of Nice Sophia-Antipolis, CNRS, UMR6543, Institute of Developmental Biology and Cancer, Nice, France School of Animal and Veterinary Sciences, University of Adelaide, Roseworthy SA 5371, Australia c Laboratory of Clinical and Experimental Pathology, CHU Nice, Hôpital Pasteur, Nice, France b
a r t i c l e
i n f o
Article history: Received 3 May 2010 Available online 10 May 2010 Keywords: Adipocyte White adipose tissue Muscle regeneration Myonecrosis
a b s t r a c t Fat cell accumulation in skeletal muscle is a major characteristic of various disorders, such as obesity, sarcopenia and dystrophies. Moreover, these fat cells could be involved in muscle homeostasis regulation as previously described for adipocytes in bone marrow. Despite recent advances on the topic, no clearly characterized mouse model is currently available to study fat accumulation within skeletal muscle. Here, we report a detailed characterization of a mouse model of skeletal muscle fat cell accumulation after degeneration induced by intra-muscular injection of glycerol. Information is provided on the kinetics of degeneration/fat deposition, including the quantity of fat deposited based on various parameters such as glycerol concentration, age, sex and strain of mice. Finally, these fat cells are characterized as true white adipocytes morphologically and molecularly. Our study shows that the mouse adipocyte accumulation within skeletal muscle after glycerol degeneration is a reproducible, transposable and easy model to use. This mouse model should allow a more comprehensive understanding of the impact of adipocyte accumulation in skeletal muscle pathophysiology. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Ectopic fat cell deposition in skeletal muscle is a characteristic of various disorders, such as obesity, sarcopenia and particularly muscular dystrophies. The latter refer to an heterogeneous group of genetically inherited disorders, characterized by a progressive fibro-adipose degeneration of skeletal muscles, and leading to a clinical phenotype consisting in muscular weakness and atrophy [1–3]. Fat cell accumulation has even been shown to be an accurate assessment of Duchenne muscular dystrophy (DMD) severity [4]. Nevertheless, little is known about origin and functionality of these fat cells. Several mouse models have been developed to study muscle dystrophies and to advance cellular, genetic and pharmaceutical therapies. A common model is the mdx mouse, which is a strain of mouse displaying a silencing of the dystrophin gene and thus, common features with DMD. Skeletal muscles of this mouse undergo continuous cycles of necrosis/regeneration, with the potential of regeneration decreasing with age. Moreover, older mdx mice suffer from muscle fibrosis, particularly in the diaphragm, but without associated adiposity unlike human dystrophic muscles.
* Corresponding author. Address: ISBDC, CNRS UMR6543, Faculté de Médecine, Tour Pasteur, 11ème étage, 28 avenue Valombrose, 06107 Nice Cedex 2, France. Fax: +33 4 93377058. E-mail address: [email protected] (D.F. Pisani). 0006-291X/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2010.05.021
Existing mouse strains with dystrophic muscles accompanied of fat cell accumulation are also available. Non-transgenic mouse models generally display transient and/or low amount of fat cells within skeletal muscle [5–7]. Spontaneously or experimentally muted mice are not widely usable because of their specific genetic defects. Among them, only mice defective for the chemokine (C–C motif) receptor 2 (ccr2) [8,9] or mice with the rostrocaudal muscular dystrophy [10] display a large adipocyte infiltration in skeletal muscle. To the contrary, mice deficient for alpha-, beta- or gamma-sarcoglycan [11–13], or mice deficient for desmin after muscle injury [14] present very limited fat deposition. Three studies published this year in the field of skeletal muscle adiposity have used a mouse model of intra-muscular injection of glycerol [15–17]. Even though this model is classic in acute renal failure studies [18], it had only been reported in three papers in skeletal muscle context, firstly in rabbit [19] and then in mouse [20,21] before this last year. Glycerol injection induces myofiber damages, by plasma membrane disruption [22], followed by the appearance of clear areas of muscle regeneration accompanied by significant fat deposition between muscle fibers. This model of skeletal muscle adiposity could be a new reference model in skeletal muscle physiopathology. Nevertheless, this model has not been clearly characterized yet in terms of the kinetics of the appearance, quantity and characterization of these neo-fat cells. In this study, we improve our understanding of this ‘‘glycerol model” by studying quantitatively the fat cells deposition in
768
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
correlation with various physiological parameters and by clearly characterizing these cells as adipocytes. 2. Material and methods
a pipette. Homogenate was then centrifuged 10 min at 135g and supernatant was collected and centrifuged one more time. Final supernatant containing adipocytes was filtered through 70 lm nylon membranes, and retained adipocytes were used for oil red O staining or total RNA preparation.
2.1. Reagents and antibodies 2.5. Oil red O staining Serum, buffer and trypsin were purchased from Lonza Verviers (Verviers, Belgium). Polyclonal rabbit anti-FABP4 (Fatty acid binding protein-4) was a gift from DA Bernlohr (University of Minnesota). Polyclonal guinea pig antibody against perilipin (RDI-PROGP29) was purchased from Research Diagnostic Inc., Flanders, USA. Alexa FluorÒ 594 secondary antibodies were purchased from Molecular Probes. DAPI (40 ,6-diamidino-2-phenylindole dihydrochloride) was purchased from Sigma–Aldrich Chimie (Lyon, France). 2.2. Mouse model All animal experiments were done in accordance with our institutional ethic committee guidelines (Comité régional d’éthique en matière d’expérimentation animale – Côte d’Azur). B6D2 (Charles River laboratories, L’Arbresle, France), and Rag2/cc/ immunodeficient mice [28] were maintained and experiments were performed at the local Center Antoine Lacassagne animal facility under specific pathogen free conditions. The mice were anesthetized with Imalgene/Rampun and then the 2 tibialis anterior (TA) muscles were injected with 25 lL of HBSS (hepes buffer saline solution) containing various percentage of glycerol. Mice were sacrificed at indicated time, and the TA muscles were sampled and immediately subjected to primary adipocyte purification or treated for paraffin embedding. 2.3. Embedding and staining of muscle sections Dissected TA muscle or subcutaneous white adipose tissue were fixed over-night in 4% parafolmadehyde (PAF) at RT and then paraffin-embedded. 5 lm sections were cut from the embedded blocks and put on superfrost slides. Before analysis sections were dewaxed as follows: twice in 100% xylene for 10 min, twice in 100% ethanol for 7 min, and serially in 90%, 80% and 70% ethanol for 5 min. Finally, slides were rehydrated in H2O for 10 min. For histological staining, 5 lm muscle sections were stained with hematoxylin-eosin-safranin and mounted with EntellanÒ (Merck Chemical, Darmstadt, Germany). For immunostaining, 5 lm sections were pretreated in boiling citrate buffer (10 mM, pH6.0) for 6 min. Cooled sections were first rinsed in water and in PBS for 5 min, and then permeabilized in PBS 0.2% Triton X-100 (PBT) at RT for 20 min. Sections were saturated in PBT 3% BSA (30 min, RT) and incubated with primary antibody (1 h, RT) and secondary antibody (45 min, RT). Sections were finally mounted with MowiolÒ (Merck Chemical). Visualization was performed with an Axiovert microscope and pictures were captured and analyzed with AxioVision software (Carl Zeiss, Jena, Germany). 2.4. Adipocyte preparation Freshly dissected mouse muscle was washed in PBS and then finely dissected in 1–2mm3 pieces. Muscle pieces were placed in 10 mL of wash medium (DMEM/F12 50/50 v/v, gentamycin) with 150 lL of Liberase Blendzyme3 (Roche Diagnostics, Basel, Switzerland) for 1 h at 37 °C, and then treated with 0.25% trypsin for additional 20 min. Enzymatic activities were stopped by adding 1.1 mL of new-born calf serum and tissue was completely disrupted using
Fresh primary adipocytes were placed in a 25 cm2 flask filled with wash medium. After 24 h, adherent adipocytes on the upper face of the flask were fixed 10 min in PAF (paraformaldehyde) 4% and then stained with oil red O (60% of a stock solution at 0.5% wt/vol in isopropanol and 40% distilled water) for 15 min. 2.6. RNA extracts and RT-PCR Total RNA was extracted using TRI-Reagent (Euromedex, Souffelweyersheim, France) according to the manufacturer’s instructions. RNA was treated with DNase I (Promega, Madison, WI) for 30 min before reverse transcription. First strand cDNA was generated on 2 lg RNA with M-MLV-RT (Promega) in presence of 12.5 ng/l random primers for 2 h at 37 °C. A 30 cycles PCR amplification was performed using 0.5 lL of cDNA with GoTaq polymerase (Promega), 2 lM of each dNTP and 200 nM of each primer. 18S rRNA was amplified (15 cycles) as the reference gene. Forward and reverse primer sequences were respectively: 50 -TGCACA GCTCCGTGTACTTC-30 and 50 -CACCTGCACAGAGTCGTCAT-30 for adipsin, 50 -AAAGGGACAGAAATGGACAC-30 and 50 -TTGAGACAGCCG AGGAAG-30 for CIDEA (cell death-inducing DNA fragmentation factor-a-like effector A), 50 -TCCTCGAAGGTTTACAAAATGTGT-30 and 50 -GCCTCTTCCTTTGGCTCATGCCCT-30 for FABP4, 50 -GGCCA GGCTGCCAGAATT-30 and 50 -GATCTGCCCCCCAGTTTGAT-30 for leptin, 50 -TTGCCCTAAGGACCCCTGAA-30 and 50 -TGGATCACCACGA AGGTCTTG-30 for lipoprotein lipase, 50 -CGACACACCATGCAGACC ACAGCA-30 and 50 -CGTAACACCCTTCAGGGCATCGGA-30 for perilipin, 50 -GCCACCACAGAAAGCTTGTC-30 and 50 -CGGTCCTTCCTTGG TGTACA-30 for UCP1 (uncoupling protein 1), 50 -GTTGGTGGAGCGA TTTGTCT-30 and 50 -GGCCTCACTAAACCATCCAA-30 for 18S rRNA. 2.7. Statistical analysis Differences between the groups were evaluated for significance using unpaired Student’s t-test. A p value <0.05 was considered significant. The data are presented as the mean ± SEM of measurements from independent skeletal muscle samples and p values are indicated. 3. Results and discussion 3.1. Kinetics of the glycerol-injected muscle model Induced regeneration of mouse skeletal muscle has been largely characterized using various protocols such as myonecrotic injection (cardiotoxin, notexin or bupivacaïne) or surgery (denervation, ischemia or cryo-injury). Generally, these methods induce fiber necrosis within the first days after injection, followed by a gradual muscle regeneration which is completed after 2 weeks. On the basis of previous published protocols of the glycerol mouse model, we have performed a kinetic analysis of the effects of glycerol on skeletal muscle (Fig. 1). Left tibialis anterior (TA) muscles of 3 month-old mice (B6D2) were injected with 25 lL of 50% glycerol containing HBSS. To serve as controls, the right TA muscles were injected with only HBSS. Mice were sacrificed 3, 6, 14 and 28 days after injection. Histological analysis of longitudinal sections from injected TA muscles showed that glycerol induced rapidly muscle
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
769
Fig. 1. Kinetic analysis of glycerol induced muscle regeneration. Histological modifications in skeletal muscle were analyzed by hematoxylin-eosin-safranin staining after intra-muscular injection of 25 lL of glycerol 50% (HBSS, v/v) in the TA of 3 month-old B6D2 female. Muscles were sampled at indicated times and paraffin-embedded before analysis. Scale bar = 100 lm.
necrosis, with a maximum at day 6. Two weeks after injection, muscle was regenerating. After another 6 days, regeneration was generally completed similar to other classical approaches. Four weeks after glycerol injection, the muscle had recovered morphological integrity with large areas of regeneration characterized by centronucleated fibers. Clear areas of fat cell deposition were visible after 2 weeks and persisted after 4 weeks. These fat cells were generally distributed all along the muscle between the fibers (Fig. 1). As expected, no regeneration or fat cell deposition were observed in the control muscles (data not shown). Intra-muscular injection of glycerol seemed to delay regeneration of muscle fibers compared to other methods of muscle regeneration. This delay can be explained by (i) a delay in the injury, (ii) a delay in the post-necrosis cleaning or (iii) a delay in the myogenic process. Necrosis is still present at day 3 post-injury and culminates at day 6. This timing is comparable to other injuries and thus cannot explain a delay in the regeneration. On other hand, in several mouse models, a correlation has been described between muscle adiposity and inhibition of macrophages. In addition, intra-
muscular injection of clodronate-containing liposomes [6,23] or the antagonist M-CSF [24] in ccr2/ mice [8,9] results in a depletion in muscular activated macrophages which is concomitant with a decrease in muscle regeneration potential. For the glycerol model, we speculate that glycerol has a negative effect on macrophage function, explaining both fat cell deposition and the delay in regeneration. 3.2. Quantitative characterization of the glycerol-injected muscle model No quantitative data are available for glycerol induced muscle adiposity. Herein various parameters were quantified 4 weeks after injury (Fig. 2). Quantification was done on the largest fat area of the stained muscle sections normalized by the whole muscle area of the section. Previously published protocols used an intra-muscular injection of 25 lL of glycerol 50% (HBSS, v/v) in young mice. First, we have used various concentrations of glycerol (10–75%) to induce injury
770
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
Fig. 2. Involvement of various parameters into fat area size. Fat area was calculated using numerized pictures of hematoxylin-eosin-safranin staining of mouse TA sections 4 weeks after treatment. Fat area was expressed as % of whole area muscle section. (A) Fat area according to various concentrations of injected glycerol (HBSS, v/v) in 3 month-old B6D2 female. (B and C) Fat area after injection of 25 lL of glycerol 50% (v/v in HBSS) according to age of B6D2 female (B) or sex (male and female) (C). (D) Fat area after injection of 25 lL of glycerol 50% (HBSS, v/v) in 3 month-old B6D2 or Rag2/cc/ female. (E) Histological analysis of TA sections stained by hematoxylin-eosin-safranin. Analysis was performed 4 weeks after intra-muscular injection of 25 lL of glycerol 50% (HBSS, v/v) in 3 and 12 month-old B6D2 female mice. Scale bar = 100 lm.
in 3 month-old B6D2 females (Fig. 2A). Best results were obtained with glycerol concentrations of 25% and 50%, without a significant difference between the two conditions (3.22 ± 1.67 and 2.34 ± 0.33, respectively, see Fig. 2E as example for 50% glycerol). Lower or higher concentrations of glycerol resulted in similar but smaller sized fat areas (0.67 ± 0.24 and 0.97 ± 0.18, respectively). Glycerol (10%) was not sufficient to cause good regeneration/fat deposition phenomena and 75% of glycerol did not diffuse into the muscle properly due to the high viscosity. Altogether these results show
that a concentration of 50% glycerol is the best percentage, as there was larger fat areas and better reproducibility of the experiments compared to other concentrations. Thus, we compared the other parameters using an intra-muscular injection of 25 lL of 50% glycerol (HBSS, v/v). Protocols previously published used young or adult mice. Here we tested the impact of the age on the fat areas 4 weeks after an intra-muscular injection of 25 lL of 50% glycerol (HBSS, v/v) in B6D2 females. No difference was found between 3 (young) and 6
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
771
Fig. 3. Histological and immunohistochemical analysis of neo-formed fat cells in the TA of 3-month-old B6D2 mice female 4 weeks after injection of 25 lL of glycerol 50% (HBSS, v/v). (A) Morphology of fat cells evaluated on transversal (upper panel) or longitudinal (lower panel) muscle sections stained with hematoxylin-eosin-safranin. (B) Morphology of fat cells evaluated by differential interference contrast (DIC) on DAPI stained sections. (C) Analysis of adipocyte specific protein expression in fat cells. Perilipin and FABP4 were revealed by Alexa594 red staining. Morphology was assessed by DIC and nucleus by DAPI. (D) Analysis of perilipin and FABP4 expression on serial sections to assess the co-expression of the 2 proteins by fat cells. (E) Perilipin and FABP4 protein expression was assessed on mouse white adipose tissue sections as positive control. Scale bar = 20, 100 or 200 lm.
772
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
(adult) month-old mice (2.34 ± 0.33 and 2.35 ± 0.12, respectively) (Fig. 2B). However, significantly larger fat areas were found in 1 year-old (aging) mice (21.08 ± 9.48) (Fig. 2B). A clear difference in fat deposition between young and aging muscles was observed and seemed to be due to an increase in the number of fat cells within a cluster and not by an increase in the number of clusters (Fig. 2E). To determine if this difference is sex dependent, we injected glycerol in male and female of the same age (Fig. 2C). Interestingly, no difference existed in 3 month-old mice (2.34 ± 0.33 for females and 2.76 ± 0.6 for males), but aging females displayed a significantly larger fat areas after glycerol injury (21.08 ± 9.48 for females and 2.30 ± 0.47 for males). The fact that aging female muscle is more permissive to fat cell deposition, compare to male muscle, could imply hormonal regulation. Even if aging mouse females do not display a true menopause like humans, hormonal deregulation occurs (i.e. FSH increasing with estrogen and oxytocyne decreasing) and involves various physical phenomena including muscle loss and excess fat deposition [25]. Aging mice are not necessarily a convenient model for routine study, but it would be interesting to specifically study the endocrine regulation of the fat cell deposition. Finally, glycerol injury was studied between immunocompetent (B6D2) and immunodepressed (Rag2/cc/) mice (Fig. 2D). We have previously published results on these immunodepressed mice treated with glycerol, but not quantitatively [16]. In order to compare the immunodepressed model with results obtained here, we injected 3 month-old Rag2/cc/ female with 25 lL of 50% glycerol (HBSS, v/v). Fat areas obtained after 4 weeks in the B6D2 and Rag2/cc/ mice were equivalent (2.34 ± 0.33 and 3.91 ± 1.66, respectively). Thus, the glycerol model can be transposed into immunodepressed mice, and the deprivation in B, T and NK lymphocytes appears to have no impact on fat cells neo-formation within the muscle.
plasm of adipocytes (Fig. 3E). As expected, we found expression of these two proteins, with characteristic thin staining around lipid droplets for perilipin and a more diffuse staining for FABP4 (Fig. 3C). Moreover, using serial sections, strict co-expression of perilipin and FABP4 was observed in these neo-fat cells (Fig. 3D). Altogether the morphology and protein expression of the fat cells characterized them as true adipocytes. To confirm this, however, we performed an ex vivo approach. Fat cells were isolated from the rest of the skeletal muscle by collagenase treatment and differential centrifugation (Fig. 4A). White and brown adipocytes were also isolated as controls from subcutaneous fat. Isolated primary fat cells were first stained with oil red O. A clear red coloration of lipid containing droplets was found and confirmed the uniloculated status of the cells (Fig. 4A). Secondly, we have prepared total RNA from the cells and analyzed the mRNA expression of several specific adipocyte markers (Fig. 4B). As observed by the immunohistochemistry, the fat cells isolated from skeletal muscle expressed FABP4 and perilipin mRNA. Moreover, these cells
3.3. Characterization of neo-formed fat cells in the glycerol-injected muscle model In human and mouse skeletal muscle, fat cells appear after necrosis or defective regeneration of the muscular fibers. It can be postulated that these cells simply fill the space in the muscle. However, fat cells are also present within the muscle of aged and obese persons [26]. In these cases, fat cells may correspond to ectopic fat deposition. Other tissues display fat cell deposition, such as bone marrow. In this tissue, recent studies demonstrated the functionality of these cells, which play a key role in the control of haematopoiesis and are not present just to fill any empty spaces [27]. These cells are true adipocytes and regulate haematopoiesis, essentially by adipokine secretion, including TNFa, adiponectine, leptin and neuropillin-1. We characterized the fat cells in the mouse model herein by glycerol intra-muscular injection. We used the TA muscles of 3 month-old B6D2 female 4 weeks after intra-muscular injection of 25 lL of 50% glycerol (HBSS, v/v). Muscles were sampled and either embedded in paraffin to perform histological/histochemical analysis or used to isolate primary fat cells for molecular analysis. Histological analysis demonstrated that the fat cells present into injured muscle were uniloculated round cells displaying a thin cytoplasm. As seen with hematoxylin-eosin-safranin staining, empty round cells were present in regeneration area (characterized by centronucleated myofibers) (Fig. 3A). This morphology was confirmed by differential interference contrast (DIC) analysis of sections pre-stained with DAPI to detect fat cell nuclei (Fig. 3B). To assess functionality of these fat cells, the cells were analyzed by immunostaining two major proteins of adipocytes, perilipin and FABP4. Perilipin is a protein coating the lipid droplet membrane and FABP4 is a key lipid carrier present in the cyto-
Fig. 4. Ex vivo molecular analysis of neo-formed fat cells in the TA of 3 month-old B6D2 female 4 weeks after injection of 25 lL of glycerol 50% (HBSS, v/v). (A) Fat cells were purified from white adipose tissue as control or degenerated skeletal muscles and visualized by light microscopy. Oil red O staining was performed to detect filled lipid vesicles. (B) RT-PCR analysis of various adipocyte markers was performed on RNA extracted from muscle purified fat cells. RNA extracted from adipocytes purified from white adipose tissue of the same animals was used as positive control. 18S rRNA was amplified as RT-PCR control. Scale bar = 20 or 100 lm.
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
expressed adipsin (a serine protease predominantly expressed by white adipocytes), lipoprotein lipase (triglyceride hydrolyzing enzyme found in white adipocytes) and leptin (appetite repressing hormone secreted specifically by white adipocytes). Notably, these cells did not express UCP1 and CIDEA two specific markers of brown adipocytes. The in vivo and ex vivo results allowed us to define the skeletal muscle neo-fat cells as true adipocytes. These cells displayed a clear uniloculated and round morphology, and expressed all the major markers of white adipocytes. 4. Conclusion In this work, we clearly define the optimal parameters for studies on fat cell deposition in skeletal muscle using the glycerol mouse model. Routine protocols include the injection of 25 lL of 50% glycerol in the TA muscles of 3 month-old mice, independently to mouse strain. To obtain stable and reproducible fat deposition in the TA muscle, analysis should be performed 4 weeks after injection. This protocol allows a 2–5% of fat area to obtain. If larger fat areas are required, aged female mice would be the best adaptation of the protocol. Neo-formed fat cells within skeletal muscle after glycerol injury correspond to true white adipocytes, and we postulate that these cells are able to influence skeletal muscle micro-environment by secretion of various cytokines and by modification of the basal lamina. In conclusion, this mouse model is of great interest for various studies, such as (i) defining the features of multipotent cells injected into muscle adipogenic environments, (ii) characterizing the involvement of adipocyte deposition in the decrease of muscle quality associated with aging, dystrophy and metabolic pathologies, and (iii) describing the involvement of adipocyte deposition in insulin-resistant skeletal muscle. Acknowledgments We acknowledge technical assistance of Cendrine Dubaud and Franck Paput with animal care. References [1] A.E. Emery, The muscular dystrophies, Lancet 359 (2002) 687–695. [2] E.M. McNally, P. Pytel, Muscle diseases: the muscular dystrophies, Annu. Rev. Pathol. 2 (2007) 87–109. [3] O. Ozsarlak, E. Schepens, P.M. Parizel, J.W. Van Goethem, F. Vanhoenacker, A.M. De Schepper, J.J. Martin, Hereditary neuromuscular diseases, Eur. J. Radiol. 40 (2001) 184–197. [4] T.A. Wren, S. Bluml, L. Tseng-Ong, V. Gilsanz, Three-point technique of fat quantification of muscle tissue as a marker of disease progression in Duchenne muscular dystrophy: preliminary study, AJR Am. J. Roentgenol. 190 (2008) W8–W12. [5] A. de Castro Rodrigues, J.C. Andreo, G.M. Rosa Jr., N.B. dos Santos, L.H. Moraes, J.R. Lauris, Fat cell invasion in long-term denervated skeletal muscle, Microsurgery 27 (2007) 664–667. [6] M. Summan, G.L. Warren, R.R. Mercer, R. Chapman, T. Hulderman, N. Van Rooijen, P.P. Simeonova, Macrophages and skeletal muscle regeneration: a clodronate-containing liposome depletion study, Am. J. Physiol. Regul. Integr. Comp. Physiol. 290 (2006) R1488–R1495. [7] A. Wagatsuma, Upregulation of gene encoding adipogenic transcriptional factors C/EBPalpha and PPARgamma2 in denervated muscle, Exp. Physiol. 91 (2006) 747–753.
773
[8] V. Contreras-Shannon, O. Ochoa, S.M. Reyes-Reyna, D. Sun, J.E. Michalek, W.A. Kuziel, L.M. McManus, P.K. Shireman, Fat accumulation with altered inflammation and regeneration in skeletal muscle of CCR2/ mice following ischemic injury, Am. J. Physiol. Cell Physiol. 292 (2007) C953–C967. [9] G.L. Warren, T. Hulderman, D. Mishra, X. Gao, L. Millecchia, L. O’Farrell, W.A. Kuziel, P.P. Simeonova, Chemokine receptor CCR2 involvement in skeletal muscle regeneration, FASEB J. 19 (2005) 413–415. [10] R.B. Sher, C. Aoyama, K.A. Huebsch, S. Ji, J. Kerner, Y. Yang, W.N. Frankel, C.L. Hoppel, P.A. Wood, D.E. Vance, G.A. Cox, A rostrocaudal muscular dystrophy caused by a defect in choline kinase beta, the first enzyme in phosphatidylcholine biosynthesis, J. Biol. Chem. 281 (2006) 4938–4948. [11] L. Cordier, A.A. Hack, M.O. Scott, E.R. Barton-Davis, G. Gao, J.M. Wilson, E.M. McNally, H.L. Sweeney, Rescue of skeletal muscles of gamma-sarcoglycandeficient mice with adeno-associated virus-mediated gene transfer, Mol. Ther. 1 (2000) 119–129. [12] F. Duclos, V. Straub, S.A. Moore, D.P. Venzke, R.F. Hrstka, R.H. Crosbie, M. Durbeej, C.S. Lebakken, A.J. Ettinger, J. van der Meulen, K.H. Holt, L.E. Lim, J.R. Sanes, B.L. Davidson, J.A. Faulkner, R. Williamson, K.P. Campbell, Progressive muscular dystrophy in alpha-sarcoglycan-deficient mice, J. Cell Biol. 142 (1998) 1461–1471. [13] M. Durbeej, K.P. Campbell, Muscular dystrophies involving the dystrophin– glycoprotein complex: an overview of current mouse models, Curr. Opin. Genet. Dev. 12 (2002) 349–361. [14] O. Agbulut, Z. Li, S. Perie, M.A. Ludosky, D. Paulin, J. Cartaud, G. Butler-Browne, Lack of desmin results in abortive muscle regeneration and modifications in synaptic structure, Cell Motil. Cytoskeleton 49 (2001) 51–66. [15] A.W. Joe, L. Yi, A. Natarajan, F. Le Grand, L. So, J. Wang, M.A. Rudnicki, F.M. Rossi, Muscle injury activates resident fibro/adipogenic progenitors that facilitate myogenesis, Nat. Cell Biol. [16] D.F. Pisani, C.A. Dechesne, S. Sacconi, S. Delplace, N. Belmonte, O. Cochet, N. Clement, B. Wdziekonski, A.P. Villageois, C. Butori, C. Bagnis, J.P. Di Santo, J.Y. Kurzenne, C. Desnuelle, C. Dani, Isolation of a highly myogenic CD34-negative subset of human skeletal muscle cells free of adipogenic potential, Stem Cells. [17] A. Uezumi, S.I. Fukada, N. Yamamoto, S. Takeda, K. Tsuchida, Mesenchymal progenitors distinct from satellite cells contribute to ectopic fat cell formation in skeletal muscle, Nat. Cell Biol. [18] J.D. Sraer, L. Moulonguet-Doleris, F. Delarue, J. Sraer, R. Ardaillou, Prostaglandin synthesis by glomeruli isolated from rats with glycerolinduced acute renal failure, Circ. Res. 49 (1981) 775–783. [19] H. Kawai, H. Nishino, K. Kusaka, T. Naruo, Y. Tamaki, M. Iwasa, Experimental glycerol myopathy: a histological study, Acta Neuropathol. 80 (1990) 192–197. [20] S.T. Abraham, C. Shaw, Increased expression of deltaCaMKII isoforms in skeletal muscle regeneration: implications in dystrophic muscle disease, J. Cell Biochem. 97 (2006) 621–632. [21] N. Arsic, S. Zacchigna, L. Zentilin, G. Ramirez-Correa, L. Pattarini, A. Salvi, G. Sinagra, M. Giacca, Vascular endothelial growth factor stimulates skeletal muscle regeneration in vivo, Mol. Ther. 10 (2004) 844–854. [22] B. Eisenberg, R.S. Eisenberg, Transverse tubular system in glycerol-treated skeletal muscle, Science 160 (1968) 1243–1244. [23] W. Shen, Y. Li, J. Zhu, R. Schwendener, J. Huard, Interaction between macrophages, TGF-beta1, and the COX-2 pathway during the inflammatory phase of skeletal muscle healing after injury, J. Cell Physiol. 214 (2008) 405– 412. [24] M. Segawa, S. Fukada, Y. Yamamoto, H. Yahagi, M. Kanematsu, M. Sato, T. Ito, A. Uezumi, S. Hayashi, Y. Miyagoe-Suzuki, S. Takeda, K. Tsujikawa, H. Yamamoto, Suppression of macrophage functions impairs skeletal muscle regeneration with severe fibrosis, Exp. Cell Res. 314 (2008) 3232–3244. [25] G.I. Perez, A. Jurisicova, L. Wise, T. Lipina, M. Kanisek, A. Bechard, Y. Takai, P. Hunt, J. Roder, M. Grynpas, J.L. Tilly, Absence of the proapoptotic Bax protein extends fertility and alleviates age-related health complications in female mice, Proc. Natl. Acad. Sci. USA 104 (2007) 5229–5234. [26] M.J. Delmonico, T.B. Harris, M. Visser, S.W. Park, M.B. Conroy, P. VelasquezMieyer, R. Boudreau, T.M. Manini, M. Nevitt, A.B. Newman, B.H. Goodpaster, Longitudinal study of muscle strength, quality, and adipose tissue infiltration, Am. J. Clin. Nutr. 90 (2009) 1579–1585. [27] O. Naveiras, V. Nardi, P.L. Wenzel, P.V. Hauschka, F. Fahey, G.Q. Daley, Bonemarrow adipocytes as negative regulators of the haematopoietic microenvironment, Nature 460 (2009) 259–263. [28] R.N. Cooper, A. Irintchev, J.P. Di Santo, M. Zweyer, J.E. Morgan, T.A. Partridge, G.S. Butler-Browne, V. Mouly, A. Wernig, A new immunodeficient mouse model for human myoblast transplantation, Hum. Gene Ther. 12 (2001) 823– 831.
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Mouse model of skeletal muscle adiposity: A glycerol treatment approach Didier F. Pisani a,*, Cynthia D.K. Bottema b, Catherine Butori c, Christian Dani a, Claude A. Dechesne a a
University of Nice Sophia-Antipolis, CNRS, UMR6543, Institute of Developmental Biology and Cancer, Nice, France School of Animal and Veterinary Sciences, University of Adelaide, Roseworthy SA 5371, Australia c Laboratory of Clinical and Experimental Pathology, CHU Nice, Hôpital Pasteur, Nice, France b
a r t i c l e
i n f o
Article history: Received 3 May 2010 Available online 10 May 2010 Keywords: Adipocyte White adipose tissue Muscle regeneration Myonecrosis
a b s t r a c t Fat cell accumulation in skeletal muscle is a major characteristic of various disorders, such as obesity, sarcopenia and dystrophies. Moreover, these fat cells could be involved in muscle homeostasis regulation as previously described for adipocytes in bone marrow. Despite recent advances on the topic, no clearly characterized mouse model is currently available to study fat accumulation within skeletal muscle. Here, we report a detailed characterization of a mouse model of skeletal muscle fat cell accumulation after degeneration induced by intra-muscular injection of glycerol. Information is provided on the kinetics of degeneration/fat deposition, including the quantity of fat deposited based on various parameters such as glycerol concentration, age, sex and strain of mice. Finally, these fat cells are characterized as true white adipocytes morphologically and molecularly. Our study shows that the mouse adipocyte accumulation within skeletal muscle after glycerol degeneration is a reproducible, transposable and easy model to use. This mouse model should allow a more comprehensive understanding of the impact of adipocyte accumulation in skeletal muscle pathophysiology. Ó 2010 Elsevier Inc. All rights reserved.
1. Introduction Ectopic fat cell deposition in skeletal muscle is a characteristic of various disorders, such as obesity, sarcopenia and particularly muscular dystrophies. The latter refer to an heterogeneous group of genetically inherited disorders, characterized by a progressive fibro-adipose degeneration of skeletal muscles, and leading to a clinical phenotype consisting in muscular weakness and atrophy [1–3]. Fat cell accumulation has even been shown to be an accurate assessment of Duchenne muscular dystrophy (DMD) severity [4]. Nevertheless, little is known about origin and functionality of these fat cells. Several mouse models have been developed to study muscle dystrophies and to advance cellular, genetic and pharmaceutical therapies. A common model is the mdx mouse, which is a strain of mouse displaying a silencing of the dystrophin gene and thus, common features with DMD. Skeletal muscles of this mouse undergo continuous cycles of necrosis/regeneration, with the potential of regeneration decreasing with age. Moreover, older mdx mice suffer from muscle fibrosis, particularly in the diaphragm, but without associated adiposity unlike human dystrophic muscles.
* Corresponding author. Address: ISBDC, CNRS UMR6543, Faculté de Médecine, Tour Pasteur, 11ème étage, 28 avenue Valombrose, 06107 Nice Cedex 2, France. Fax: +33 4 93377058. E-mail address: [email protected] (D.F. Pisani). 0006-291X/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2010.05.021
Existing mouse strains with dystrophic muscles accompanied of fat cell accumulation are also available. Non-transgenic mouse models generally display transient and/or low amount of fat cells within skeletal muscle [5–7]. Spontaneously or experimentally muted mice are not widely usable because of their specific genetic defects. Among them, only mice defective for the chemokine (C–C motif) receptor 2 (ccr2) [8,9] or mice with the rostrocaudal muscular dystrophy [10] display a large adipocyte infiltration in skeletal muscle. To the contrary, mice deficient for alpha-, beta- or gamma-sarcoglycan [11–13], or mice deficient for desmin after muscle injury [14] present very limited fat deposition. Three studies published this year in the field of skeletal muscle adiposity have used a mouse model of intra-muscular injection of glycerol [15–17]. Even though this model is classic in acute renal failure studies [18], it had only been reported in three papers in skeletal muscle context, firstly in rabbit [19] and then in mouse [20,21] before this last year. Glycerol injection induces myofiber damages, by plasma membrane disruption [22], followed by the appearance of clear areas of muscle regeneration accompanied by significant fat deposition between muscle fibers. This model of skeletal muscle adiposity could be a new reference model in skeletal muscle physiopathology. Nevertheless, this model has not been clearly characterized yet in terms of the kinetics of the appearance, quantity and characterization of these neo-fat cells. In this study, we improve our understanding of this ‘‘glycerol model” by studying quantitatively the fat cells deposition in
768
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
correlation with various physiological parameters and by clearly characterizing these cells as adipocytes. 2. Material and methods
a pipette. Homogenate was then centrifuged 10 min at 135g and supernatant was collected and centrifuged one more time. Final supernatant containing adipocytes was filtered through 70 lm nylon membranes, and retained adipocytes were used for oil red O staining or total RNA preparation.
2.1. Reagents and antibodies 2.5. Oil red O staining Serum, buffer and trypsin were purchased from Lonza Verviers (Verviers, Belgium). Polyclonal rabbit anti-FABP4 (Fatty acid binding protein-4) was a gift from DA Bernlohr (University of Minnesota). Polyclonal guinea pig antibody against perilipin (RDI-PROGP29) was purchased from Research Diagnostic Inc., Flanders, USA. Alexa FluorÒ 594 secondary antibodies were purchased from Molecular Probes. DAPI (40 ,6-diamidino-2-phenylindole dihydrochloride) was purchased from Sigma–Aldrich Chimie (Lyon, France). 2.2. Mouse model All animal experiments were done in accordance with our institutional ethic committee guidelines (Comité régional d’éthique en matière d’expérimentation animale – Côte d’Azur). B6D2 (Charles River laboratories, L’Arbresle, France), and Rag2/cc/ immunodeficient mice [28] were maintained and experiments were performed at the local Center Antoine Lacassagne animal facility under specific pathogen free conditions. The mice were anesthetized with Imalgene/Rampun and then the 2 tibialis anterior (TA) muscles were injected with 25 lL of HBSS (hepes buffer saline solution) containing various percentage of glycerol. Mice were sacrificed at indicated time, and the TA muscles were sampled and immediately subjected to primary adipocyte purification or treated for paraffin embedding. 2.3. Embedding and staining of muscle sections Dissected TA muscle or subcutaneous white adipose tissue were fixed over-night in 4% parafolmadehyde (PAF) at RT and then paraffin-embedded. 5 lm sections were cut from the embedded blocks and put on superfrost slides. Before analysis sections were dewaxed as follows: twice in 100% xylene for 10 min, twice in 100% ethanol for 7 min, and serially in 90%, 80% and 70% ethanol for 5 min. Finally, slides were rehydrated in H2O for 10 min. For histological staining, 5 lm muscle sections were stained with hematoxylin-eosin-safranin and mounted with EntellanÒ (Merck Chemical, Darmstadt, Germany). For immunostaining, 5 lm sections were pretreated in boiling citrate buffer (10 mM, pH6.0) for 6 min. Cooled sections were first rinsed in water and in PBS for 5 min, and then permeabilized in PBS 0.2% Triton X-100 (PBT) at RT for 20 min. Sections were saturated in PBT 3% BSA (30 min, RT) and incubated with primary antibody (1 h, RT) and secondary antibody (45 min, RT). Sections were finally mounted with MowiolÒ (Merck Chemical). Visualization was performed with an Axiovert microscope and pictures were captured and analyzed with AxioVision software (Carl Zeiss, Jena, Germany). 2.4. Adipocyte preparation Freshly dissected mouse muscle was washed in PBS and then finely dissected in 1–2mm3 pieces. Muscle pieces were placed in 10 mL of wash medium (DMEM/F12 50/50 v/v, gentamycin) with 150 lL of Liberase Blendzyme3 (Roche Diagnostics, Basel, Switzerland) for 1 h at 37 °C, and then treated with 0.25% trypsin for additional 20 min. Enzymatic activities were stopped by adding 1.1 mL of new-born calf serum and tissue was completely disrupted using
Fresh primary adipocytes were placed in a 25 cm2 flask filled with wash medium. After 24 h, adherent adipocytes on the upper face of the flask were fixed 10 min in PAF (paraformaldehyde) 4% and then stained with oil red O (60% of a stock solution at 0.5% wt/vol in isopropanol and 40% distilled water) for 15 min. 2.6. RNA extracts and RT-PCR Total RNA was extracted using TRI-Reagent (Euromedex, Souffelweyersheim, France) according to the manufacturer’s instructions. RNA was treated with DNase I (Promega, Madison, WI) for 30 min before reverse transcription. First strand cDNA was generated on 2 lg RNA with M-MLV-RT (Promega) in presence of 12.5 ng/l random primers for 2 h at 37 °C. A 30 cycles PCR amplification was performed using 0.5 lL of cDNA with GoTaq polymerase (Promega), 2 lM of each dNTP and 200 nM of each primer. 18S rRNA was amplified (15 cycles) as the reference gene. Forward and reverse primer sequences were respectively: 50 -TGCACA GCTCCGTGTACTTC-30 and 50 -CACCTGCACAGAGTCGTCAT-30 for adipsin, 50 -AAAGGGACAGAAATGGACAC-30 and 50 -TTGAGACAGCCG AGGAAG-30 for CIDEA (cell death-inducing DNA fragmentation factor-a-like effector A), 50 -TCCTCGAAGGTTTACAAAATGTGT-30 and 50 -GCCTCTTCCTTTGGCTCATGCCCT-30 for FABP4, 50 -GGCCA GGCTGCCAGAATT-30 and 50 -GATCTGCCCCCCAGTTTGAT-30 for leptin, 50 -TTGCCCTAAGGACCCCTGAA-30 and 50 -TGGATCACCACGA AGGTCTTG-30 for lipoprotein lipase, 50 -CGACACACCATGCAGACC ACAGCA-30 and 50 -CGTAACACCCTTCAGGGCATCGGA-30 for perilipin, 50 -GCCACCACAGAAAGCTTGTC-30 and 50 -CGGTCCTTCCTTGG TGTACA-30 for UCP1 (uncoupling protein 1), 50 -GTTGGTGGAGCGA TTTGTCT-30 and 50 -GGCCTCACTAAACCATCCAA-30 for 18S rRNA. 2.7. Statistical analysis Differences between the groups were evaluated for significance using unpaired Student’s t-test. A p value <0.05 was considered significant. The data are presented as the mean ± SEM of measurements from independent skeletal muscle samples and p values are indicated. 3. Results and discussion 3.1. Kinetics of the glycerol-injected muscle model Induced regeneration of mouse skeletal muscle has been largely characterized using various protocols such as myonecrotic injection (cardiotoxin, notexin or bupivacaïne) or surgery (denervation, ischemia or cryo-injury). Generally, these methods induce fiber necrosis within the first days after injection, followed by a gradual muscle regeneration which is completed after 2 weeks. On the basis of previous published protocols of the glycerol mouse model, we have performed a kinetic analysis of the effects of glycerol on skeletal muscle (Fig. 1). Left tibialis anterior (TA) muscles of 3 month-old mice (B6D2) were injected with 25 lL of 50% glycerol containing HBSS. To serve as controls, the right TA muscles were injected with only HBSS. Mice were sacrificed 3, 6, 14 and 28 days after injection. Histological analysis of longitudinal sections from injected TA muscles showed that glycerol induced rapidly muscle
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
769
Fig. 1. Kinetic analysis of glycerol induced muscle regeneration. Histological modifications in skeletal muscle were analyzed by hematoxylin-eosin-safranin staining after intra-muscular injection of 25 lL of glycerol 50% (HBSS, v/v) in the TA of 3 month-old B6D2 female. Muscles were sampled at indicated times and paraffin-embedded before analysis. Scale bar = 100 lm.
necrosis, with a maximum at day 6. Two weeks after injection, muscle was regenerating. After another 6 days, regeneration was generally completed similar to other classical approaches. Four weeks after glycerol injection, the muscle had recovered morphological integrity with large areas of regeneration characterized by centronucleated fibers. Clear areas of fat cell deposition were visible after 2 weeks and persisted after 4 weeks. These fat cells were generally distributed all along the muscle between the fibers (Fig. 1). As expected, no regeneration or fat cell deposition were observed in the control muscles (data not shown). Intra-muscular injection of glycerol seemed to delay regeneration of muscle fibers compared to other methods of muscle regeneration. This delay can be explained by (i) a delay in the injury, (ii) a delay in the post-necrosis cleaning or (iii) a delay in the myogenic process. Necrosis is still present at day 3 post-injury and culminates at day 6. This timing is comparable to other injuries and thus cannot explain a delay in the regeneration. On other hand, in several mouse models, a correlation has been described between muscle adiposity and inhibition of macrophages. In addition, intra-
muscular injection of clodronate-containing liposomes [6,23] or the antagonist M-CSF [24] in ccr2/ mice [8,9] results in a depletion in muscular activated macrophages which is concomitant with a decrease in muscle regeneration potential. For the glycerol model, we speculate that glycerol has a negative effect on macrophage function, explaining both fat cell deposition and the delay in regeneration. 3.2. Quantitative characterization of the glycerol-injected muscle model No quantitative data are available for glycerol induced muscle adiposity. Herein various parameters were quantified 4 weeks after injury (Fig. 2). Quantification was done on the largest fat area of the stained muscle sections normalized by the whole muscle area of the section. Previously published protocols used an intra-muscular injection of 25 lL of glycerol 50% (HBSS, v/v) in young mice. First, we have used various concentrations of glycerol (10–75%) to induce injury
770
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
Fig. 2. Involvement of various parameters into fat area size. Fat area was calculated using numerized pictures of hematoxylin-eosin-safranin staining of mouse TA sections 4 weeks after treatment. Fat area was expressed as % of whole area muscle section. (A) Fat area according to various concentrations of injected glycerol (HBSS, v/v) in 3 month-old B6D2 female. (B and C) Fat area after injection of 25 lL of glycerol 50% (v/v in HBSS) according to age of B6D2 female (B) or sex (male and female) (C). (D) Fat area after injection of 25 lL of glycerol 50% (HBSS, v/v) in 3 month-old B6D2 or Rag2/cc/ female. (E) Histological analysis of TA sections stained by hematoxylin-eosin-safranin. Analysis was performed 4 weeks after intra-muscular injection of 25 lL of glycerol 50% (HBSS, v/v) in 3 and 12 month-old B6D2 female mice. Scale bar = 100 lm.
in 3 month-old B6D2 females (Fig. 2A). Best results were obtained with glycerol concentrations of 25% and 50%, without a significant difference between the two conditions (3.22 ± 1.67 and 2.34 ± 0.33, respectively, see Fig. 2E as example for 50% glycerol). Lower or higher concentrations of glycerol resulted in similar but smaller sized fat areas (0.67 ± 0.24 and 0.97 ± 0.18, respectively). Glycerol (10%) was not sufficient to cause good regeneration/fat deposition phenomena and 75% of glycerol did not diffuse into the muscle properly due to the high viscosity. Altogether these results show
that a concentration of 50% glycerol is the best percentage, as there was larger fat areas and better reproducibility of the experiments compared to other concentrations. Thus, we compared the other parameters using an intra-muscular injection of 25 lL of 50% glycerol (HBSS, v/v). Protocols previously published used young or adult mice. Here we tested the impact of the age on the fat areas 4 weeks after an intra-muscular injection of 25 lL of 50% glycerol (HBSS, v/v) in B6D2 females. No difference was found between 3 (young) and 6
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
771
Fig. 3. Histological and immunohistochemical analysis of neo-formed fat cells in the TA of 3-month-old B6D2 mice female 4 weeks after injection of 25 lL of glycerol 50% (HBSS, v/v). (A) Morphology of fat cells evaluated on transversal (upper panel) or longitudinal (lower panel) muscle sections stained with hematoxylin-eosin-safranin. (B) Morphology of fat cells evaluated by differential interference contrast (DIC) on DAPI stained sections. (C) Analysis of adipocyte specific protein expression in fat cells. Perilipin and FABP4 were revealed by Alexa594 red staining. Morphology was assessed by DIC and nucleus by DAPI. (D) Analysis of perilipin and FABP4 expression on serial sections to assess the co-expression of the 2 proteins by fat cells. (E) Perilipin and FABP4 protein expression was assessed on mouse white adipose tissue sections as positive control. Scale bar = 20, 100 or 200 lm.
772
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
(adult) month-old mice (2.34 ± 0.33 and 2.35 ± 0.12, respectively) (Fig. 2B). However, significantly larger fat areas were found in 1 year-old (aging) mice (21.08 ± 9.48) (Fig. 2B). A clear difference in fat deposition between young and aging muscles was observed and seemed to be due to an increase in the number of fat cells within a cluster and not by an increase in the number of clusters (Fig. 2E). To determine if this difference is sex dependent, we injected glycerol in male and female of the same age (Fig. 2C). Interestingly, no difference existed in 3 month-old mice (2.34 ± 0.33 for females and 2.76 ± 0.6 for males), but aging females displayed a significantly larger fat areas after glycerol injury (21.08 ± 9.48 for females and 2.30 ± 0.47 for males). The fact that aging female muscle is more permissive to fat cell deposition, compare to male muscle, could imply hormonal regulation. Even if aging mouse females do not display a true menopause like humans, hormonal deregulation occurs (i.e. FSH increasing with estrogen and oxytocyne decreasing) and involves various physical phenomena including muscle loss and excess fat deposition [25]. Aging mice are not necessarily a convenient model for routine study, but it would be interesting to specifically study the endocrine regulation of the fat cell deposition. Finally, glycerol injury was studied between immunocompetent (B6D2) and immunodepressed (Rag2/cc/) mice (Fig. 2D). We have previously published results on these immunodepressed mice treated with glycerol, but not quantitatively [16]. In order to compare the immunodepressed model with results obtained here, we injected 3 month-old Rag2/cc/ female with 25 lL of 50% glycerol (HBSS, v/v). Fat areas obtained after 4 weeks in the B6D2 and Rag2/cc/ mice were equivalent (2.34 ± 0.33 and 3.91 ± 1.66, respectively). Thus, the glycerol model can be transposed into immunodepressed mice, and the deprivation in B, T and NK lymphocytes appears to have no impact on fat cells neo-formation within the muscle.
plasm of adipocytes (Fig. 3E). As expected, we found expression of these two proteins, with characteristic thin staining around lipid droplets for perilipin and a more diffuse staining for FABP4 (Fig. 3C). Moreover, using serial sections, strict co-expression of perilipin and FABP4 was observed in these neo-fat cells (Fig. 3D). Altogether the morphology and protein expression of the fat cells characterized them as true adipocytes. To confirm this, however, we performed an ex vivo approach. Fat cells were isolated from the rest of the skeletal muscle by collagenase treatment and differential centrifugation (Fig. 4A). White and brown adipocytes were also isolated as controls from subcutaneous fat. Isolated primary fat cells were first stained with oil red O. A clear red coloration of lipid containing droplets was found and confirmed the uniloculated status of the cells (Fig. 4A). Secondly, we have prepared total RNA from the cells and analyzed the mRNA expression of several specific adipocyte markers (Fig. 4B). As observed by the immunohistochemistry, the fat cells isolated from skeletal muscle expressed FABP4 and perilipin mRNA. Moreover, these cells
3.3. Characterization of neo-formed fat cells in the glycerol-injected muscle model In human and mouse skeletal muscle, fat cells appear after necrosis or defective regeneration of the muscular fibers. It can be postulated that these cells simply fill the space in the muscle. However, fat cells are also present within the muscle of aged and obese persons [26]. In these cases, fat cells may correspond to ectopic fat deposition. Other tissues display fat cell deposition, such as bone marrow. In this tissue, recent studies demonstrated the functionality of these cells, which play a key role in the control of haematopoiesis and are not present just to fill any empty spaces [27]. These cells are true adipocytes and regulate haematopoiesis, essentially by adipokine secretion, including TNFa, adiponectine, leptin and neuropillin-1. We characterized the fat cells in the mouse model herein by glycerol intra-muscular injection. We used the TA muscles of 3 month-old B6D2 female 4 weeks after intra-muscular injection of 25 lL of 50% glycerol (HBSS, v/v). Muscles were sampled and either embedded in paraffin to perform histological/histochemical analysis or used to isolate primary fat cells for molecular analysis. Histological analysis demonstrated that the fat cells present into injured muscle were uniloculated round cells displaying a thin cytoplasm. As seen with hematoxylin-eosin-safranin staining, empty round cells were present in regeneration area (characterized by centronucleated myofibers) (Fig. 3A). This morphology was confirmed by differential interference contrast (DIC) analysis of sections pre-stained with DAPI to detect fat cell nuclei (Fig. 3B). To assess functionality of these fat cells, the cells were analyzed by immunostaining two major proteins of adipocytes, perilipin and FABP4. Perilipin is a protein coating the lipid droplet membrane and FABP4 is a key lipid carrier present in the cyto-
Fig. 4. Ex vivo molecular analysis of neo-formed fat cells in the TA of 3 month-old B6D2 female 4 weeks after injection of 25 lL of glycerol 50% (HBSS, v/v). (A) Fat cells were purified from white adipose tissue as control or degenerated skeletal muscles and visualized by light microscopy. Oil red O staining was performed to detect filled lipid vesicles. (B) RT-PCR analysis of various adipocyte markers was performed on RNA extracted from muscle purified fat cells. RNA extracted from adipocytes purified from white adipose tissue of the same animals was used as positive control. 18S rRNA was amplified as RT-PCR control. Scale bar = 20 or 100 lm.
D.F. Pisani et al. / Biochemical and Biophysical Research Communications 396 (2010) 767–773
expressed adipsin (a serine protease predominantly expressed by white adipocytes), lipoprotein lipase (triglyceride hydrolyzing enzyme found in white adipocytes) and leptin (appetite repressing hormone secreted specifically by white adipocytes). Notably, these cells did not express UCP1 and CIDEA two specific markers of brown adipocytes. The in vivo and ex vivo results allowed us to define the skeletal muscle neo-fat cells as true adipocytes. These cells displayed a clear uniloculated and round morphology, and expressed all the major markers of white adipocytes. 4. Conclusion In this work, we clearly define the optimal parameters for studies on fat cell deposition in skeletal muscle using the glycerol mouse model. Routine protocols include the injection of 25 lL of 50% glycerol in the TA muscles of 3 month-old mice, independently to mouse strain. To obtain stable and reproducible fat deposition in the TA muscle, analysis should be performed 4 weeks after injection. This protocol allows a 2–5% of fat area to obtain. If larger fat areas are required, aged female mice would be the best adaptation of the protocol. Neo-formed fat cells within skeletal muscle after glycerol injury correspond to true white adipocytes, and we postulate that these cells are able to influence skeletal muscle micro-environment by secretion of various cytokines and by modification of the basal lamina. In conclusion, this mouse model is of great interest for various studies, such as (i) defining the features of multipotent cells injected into muscle adipogenic environments, (ii) characterizing the involvement of adipocyte deposition in the decrease of muscle quality associated with aging, dystrophy and metabolic pathologies, and (iii) describing the involvement of adipocyte deposition in insulin-resistant skeletal muscle. Acknowledgments We acknowledge technical assistance of Cendrine Dubaud and Franck Paput with animal care. References [1] A.E. Emery, The muscular dystrophies, Lancet 359 (2002) 687–695. [2] E.M. McNally, P. Pytel, Muscle diseases: the muscular dystrophies, Annu. Rev. Pathol. 2 (2007) 87–109. [3] O. Ozsarlak, E. Schepens, P.M. Parizel, J.W. Van Goethem, F. Vanhoenacker, A.M. De Schepper, J.J. Martin, Hereditary neuromuscular diseases, Eur. J. Radiol. 40 (2001) 184–197. [4] T.A. Wren, S. Bluml, L. Tseng-Ong, V. Gilsanz, Three-point technique of fat quantification of muscle tissue as a marker of disease progression in Duchenne muscular dystrophy: preliminary study, AJR Am. J. Roentgenol. 190 (2008) W8–W12. [5] A. de Castro Rodrigues, J.C. Andreo, G.M. Rosa Jr., N.B. dos Santos, L.H. Moraes, J.R. Lauris, Fat cell invasion in long-term denervated skeletal muscle, Microsurgery 27 (2007) 664–667. [6] M. Summan, G.L. Warren, R.R. Mercer, R. Chapman, T. Hulderman, N. Van Rooijen, P.P. Simeonova, Macrophages and skeletal muscle regeneration: a clodronate-containing liposome depletion study, Am. J. Physiol. Regul. Integr. Comp. Physiol. 290 (2006) R1488–R1495. [7] A. Wagatsuma, Upregulation of gene encoding adipogenic transcriptional factors C/EBPalpha and PPARgamma2 in denervated muscle, Exp. Physiol. 91 (2006) 747–753.
773
[8] V. Contreras-Shannon, O. Ochoa, S.M. Reyes-Reyna, D. Sun, J.E. Michalek, W.A. Kuziel, L.M. McManus, P.K. Shireman, Fat accumulation with altered inflammation and regeneration in skeletal muscle of CCR2/ mice following ischemic injury, Am. J. Physiol. Cell Physiol. 292 (2007) C953–C967. [9] G.L. Warren, T. Hulderman, D. Mishra, X. Gao, L. Millecchia, L. O’Farrell, W.A. Kuziel, P.P. Simeonova, Chemokine receptor CCR2 involvement in skeletal muscle regeneration, FASEB J. 19 (2005) 413–415. [10] R.B. Sher, C. Aoyama, K.A. Huebsch, S. Ji, J. Kerner, Y. Yang, W.N. Frankel, C.L. Hoppel, P.A. Wood, D.E. Vance, G.A. Cox, A rostrocaudal muscular dystrophy caused by a defect in choline kinase beta, the first enzyme in phosphatidylcholine biosynthesis, J. Biol. Chem. 281 (2006) 4938–4948. [11] L. Cordier, A.A. Hack, M.O. Scott, E.R. Barton-Davis, G. Gao, J.M. Wilson, E.M. McNally, H.L. Sweeney, Rescue of skeletal muscles of gamma-sarcoglycandeficient mice with adeno-associated virus-mediated gene transfer, Mol. Ther. 1 (2000) 119–129. [12] F. Duclos, V. Straub, S.A. Moore, D.P. Venzke, R.F. Hrstka, R.H. Crosbie, M. Durbeej, C.S. Lebakken, A.J. Ettinger, J. van der Meulen, K.H. Holt, L.E. Lim, J.R. Sanes, B.L. Davidson, J.A. Faulkner, R. Williamson, K.P. Campbell, Progressive muscular dystrophy in alpha-sarcoglycan-deficient mice, J. Cell Biol. 142 (1998) 1461–1471. [13] M. Durbeej, K.P. Campbell, Muscular dystrophies involving the dystrophin– glycoprotein complex: an overview of current mouse models, Curr. Opin. Genet. Dev. 12 (2002) 349–361. [14] O. Agbulut, Z. Li, S. Perie, M.A. Ludosky, D. Paulin, J. Cartaud, G. Butler-Browne, Lack of desmin results in abortive muscle regeneration and modifications in synaptic structure, Cell Motil. Cytoskeleton 49 (2001) 51–66. [15] A.W. Joe, L. Yi, A. Natarajan, F. Le Grand, L. So, J. Wang, M.A. Rudnicki, F.M. Rossi, Muscle injury activates resident fibro/adipogenic progenitors that facilitate myogenesis, Nat. Cell Biol. [16] D.F. Pisani, C.A. Dechesne, S. Sacconi, S. Delplace, N. Belmonte, O. Cochet, N. Clement, B. Wdziekonski, A.P. Villageois, C. Butori, C. Bagnis, J.P. Di Santo, J.Y. Kurzenne, C. Desnuelle, C. Dani, Isolation of a highly myogenic CD34-negative subset of human skeletal muscle cells free of adipogenic potential, Stem Cells. [17] A. Uezumi, S.I. Fukada, N. Yamamoto, S. Takeda, K. Tsuchida, Mesenchymal progenitors distinct from satellite cells contribute to ectopic fat cell formation in skeletal muscle, Nat. Cell Biol. [18] J.D. Sraer, L. Moulonguet-Doleris, F. Delarue, J. Sraer, R. Ardaillou, Prostaglandin synthesis by glomeruli isolated from rats with glycerolinduced acute renal failure, Circ. Res. 49 (1981) 775–783. [19] H. Kawai, H. Nishino, K. Kusaka, T. Naruo, Y. Tamaki, M. Iwasa, Experimental glycerol myopathy: a histological study, Acta Neuropathol. 80 (1990) 192–197. [20] S.T. Abraham, C. Shaw, Increased expression of deltaCaMKII isoforms in skeletal muscle regeneration: implications in dystrophic muscle disease, J. Cell Biochem. 97 (2006) 621–632. [21] N. Arsic, S. Zacchigna, L. Zentilin, G. Ramirez-Correa, L. Pattarini, A. Salvi, G. Sinagra, M. Giacca, Vascular endothelial growth factor stimulates skeletal muscle regeneration in vivo, Mol. Ther. 10 (2004) 844–854. [22] B. Eisenberg, R.S. Eisenberg, Transverse tubular system in glycerol-treated skeletal muscle, Science 160 (1968) 1243–1244. [23] W. Shen, Y. Li, J. Zhu, R. Schwendener, J. Huard, Interaction between macrophages, TGF-beta1, and the COX-2 pathway during the inflammatory phase of skeletal muscle healing after injury, J. Cell Physiol. 214 (2008) 405– 412. [24] M. Segawa, S. Fukada, Y. Yamamoto, H. Yahagi, M. Kanematsu, M. Sato, T. Ito, A. Uezumi, S. Hayashi, Y. Miyagoe-Suzuki, S. Takeda, K. Tsujikawa, H. Yamamoto, Suppression of macrophage functions impairs skeletal muscle regeneration with severe fibrosis, Exp. Cell Res. 314 (2008) 3232–3244. [25] G.I. Perez, A. Jurisicova, L. Wise, T. Lipina, M. Kanisek, A. Bechard, Y. Takai, P. Hunt, J. Roder, M. Grynpas, J.L. Tilly, Absence of the proapoptotic Bax protein extends fertility and alleviates age-related health complications in female mice, Proc. Natl. Acad. Sci. USA 104 (2007) 5229–5234. [26] M.J. Delmonico, T.B. Harris, M. Visser, S.W. Park, M.B. Conroy, P. VelasquezMieyer, R. Boudreau, T.M. Manini, M. Nevitt, A.B. Newman, B.H. Goodpaster, Longitudinal study of muscle strength, quality, and adipose tissue infiltration, Am. J. Clin. Nutr. 90 (2009) 1579–1585. [27] O. Naveiras, V. Nardi, P.L. Wenzel, P.V. Hauschka, F. Fahey, G.Q. Daley, Bonemarrow adipocytes as negative regulators of the haematopoietic microenvironment, Nature 460 (2009) 259–263. [28] R.N. Cooper, A. Irintchev, J.P. Di Santo, M. Zweyer, J.E. Morgan, T.A. Partridge, G.S. Butler-Browne, V. Mouly, A. Wernig, A new immunodeficient mouse model for human myoblast transplantation, Hum. Gene Ther. 12 (2001) 823– 831.