Dietary tributyrin supplementation and submaximal exercise promote activation of equine satellite cells

Abstract

Postexercise skeletal muscle repair is dependent on the actions of satellite cells (SCs). The signal(s) responsible for activation of these normally quiescent cells in the horse remain unknown. The objective of the experiment was to determine whether submaximal exercise or tributyrin (TB) supplementation is sufficient to stimulate SC activation. Adult geldings were fed a control diet (n = 6) or a diet containing 0.45% TB (n = 6). After 30 d, the geldings performed a single bout of submaximal exercise. Middle gluteal muscle biopsies and blood were collected on days −1, 1, 3, and 5 relative to exercise. Diet had no effect on any parameter of physical performance. Total RNA isolated from the gluteal muscle of TB fed geldings contained greater (P < 0.05) amounts of myogenin mRNA than controls. Satellite cell isolates from TB supplemented horses had a greater (P = 0.02) percentage of proliferating cell nuclear antigen immunopositive (PCNA+) SC than controls after 48 h in culture. Submaximal exercise was sufficient to increase (P < 0.05) the percentage of PCNA(+) cells in all isolates obtained during recovery period. No change in the amount of gluteal muscle Pax7 mRNA, a lineage marker of SCs, occurred in response to either diet or exercise. Our results indicate that both submaximal exercise and TB prime SCs for activation and cell cycle reentry but are insufficient to cause an increase in Pax7 expression during the recovery period.

Introduction

Physical exercise in people causes some amount of damage that is reflective of intensity and duration. Muscle damage in humans performing eccentric (lengthening) exercise includes disruption of the fiber associations with one another and ultrastructural changes such as Z-disk streaming (Fridén and Lieber, 2001). The appearance of creatine kinase and myoglobin in the blood is supportive of damage; however, changes in sarcomere structures and altered contractile protein expression may represent an adaptive response to work (hypertrophy) and not myotrauma (Hyldahl and Hubal, 2014). Following an exercise regimen that results in substantial (>50%) loss of strength or force, neutrophil and macrophage accumulation occurs within the first 48 h, and increased cytokine gene expression in the muscle is noted (Peake et al., 2017). A return to full muscle power may not occur for 5 to 7 d. Within 72 h, satellite cell (SC) numbers per fiber peak suggesting sufficient damage occurred to initiate a muscle regenerative response in healthy men (Snijders et al., 2015).

A similar pattern of events is noted in horses following a bout of exercise. Unfit horses worked to fatigue upregulate expression of metabolic genes within the first 4 h of the recovery period indicating that glycogen replenishment, mitochondria biogenesis, and energy restoration within the muscle fiber precede contractile machinery repair (Eivers et al., 2010, 2012; Valberg et al., 2018). In comparison to humans, the muscle recovery period in horses is protracted. Glycogen repletion may take up to 3 d to return to pre-exercise levels, substantially different from the less than 24-h time frame found in people, and is unresponsive to glucose delivery post-exercise (Nout et al., 2003; Rosset et al., 2017). Exercise to fatigue results in an increase Pax7 expression, the lineage marker of SCs, after day 3 with peak SC activity evident at day 7 of the recovery period (Kawai et al., 2013). The reasons for the delayed metabolic and repair efforts in equine skeletal muscle remain poorly defined.

Tributyrin (TB) is formed by acylation of the hydroxy groups of glycerol with butyric acid. The prodrug can lead to increased levels of butyrate in the intestine and systemic blood supply (Conley et al., 1998). Elevated butyrate serves as a histone deacetylase (HDAC) inhibitor that can act as an intestinal anti-inflammatory, stimulate colonocyte differentiation, and offer protective effects against liver damage in rodents (Miyoshi et al., 2011; Leonel et al., 2013; Nguyen et al., 2017; Biondo et al., 2019). Treatment of mouse SCs with sodium butyrate, and other class I and II HDAC inhibitors, supports larger myotube formation in vitro (Iezzi et al., 2004). Pigs fed TB demonstrate an increase in myogenin mRNA expression in vivo and a propensity for SC fusion in vitro, both indicative of enhanced muscle cell differentiation (Murray et al., 2018). Altered metabolism in response to butyrate includes increased expression of PGC1α leading to mitochondrial biogenesis and a greater percentage of type I and IIA fibers in mouse muscles (Gao et al., 2009). Increased oxidative fibers in skeletal muscle are supportive of improved athletic performance in Thoroughbred racehorses (McGivney et al., 2012; Rooney et al., 2017). Thus, TB may indirectly affect SC dynamics through altered fiber niche properties that ultimately lead to a reduction in postexercise recovery time.

The objective of the experiment was to determine the impact of exercise and TB supplementation on equine SC activation.

Materials and Methods

All animal procedures were reviewed and approved by the Virginia Tech Institutional Animal Care and Use Committee (18–049).

Diet and Exercise

Twelve Thoroughbred geldings (7.4 ± 0.6 yr of age; 506 ± 8 kg) were housed in dry lot paddocks in groups of 3 with ad libitum access to water and mineral blocks. Mixed grass hay was supplied to the paddocks at 1.5% of BW. Horses were assigned randomly to either a control diet (CON, Purina Animal Nutrition, Gray Summit, MO; n = 6) or a similar diet formulated to contain 0.45% TB (AviPremium D, Vetagro, Chicago, IL; n = 6) fed individually at 0.5% of BW. Forage and concentrate samples were taken weekly and stored at −20 °C. Feed samples were pooled and a subsample used for proximate analysis (Eurofins Food Chemistry Testing, Madison, WI; Table 1). After 30 d, the horses performed a single bout of submaximal exercise on a free-stall exerciser (Equigym, Paris, KY). In brief, the exercise test consisted of 4 repetitions of walk (2 m/s, 3 min), trot (5 m/s, 5 min), and canter (7 m/s, 7 min) in alternating directions. Polar heart rate monitors (V800, Polar Electro Inc., Bethpage, NY) were used to continuously record heart rate during and immediately after exercise. Venous blood was analyzed before and immediately after exercise for lactate concentration (MAK064, MilliporeSigma, Burlington, MA) and calculation of hematocrit.

Table 1.

Chemical composition of feedstuffs, as-fed basis

	Concentrate
Component	Control	Tributyrin	Hay
Moisture, %	11.8	12.0	11.0
Crude protein, %	16.3	15.7	12.7
Acid detergent fiber, %	10.4	10.8	30.8
Neutral detergent fiber, %	21.6	21.4	9.2
Starch, %	19.4	18.9	0.33
Fat, %	7.2	6.8	2.4
Ash, %	7.5	7.7	6.4
Calcium, ppm	14,800	12,900	5,600
Copper, ppm	79.6	75.8	6.0
Iron, ppm	896	860	59
Magnesium, ppm	3,710	3,840	2,980
Manganese, ppm	237	231	76.1
Phosphorus, ppm	6,580	6,140	3,150
Potassium, ppm	11,700	11,500	17,700
Sodium, ppm	2,030	2,000	218
Zinc, ppm	326	307	20.4
Calculated DE, Mcal/kg DM	3.28	3.27	2.22

Muscle Biopsies and SC Isolation

Animals were sedated (xylazine; 1 mg/kg; Bimeda-MTC Animal Health, Cambridge, ON), and the skin atop the middle gluteal muscle was shaved and surgically sterilized. A local anesthetic (lidocaine; Aspen Veterinary Resources, Liberty, MO) was administered subcutaneously, and a small incision (~1 cm) was created. Biopsies were retrieved from the middle gluteus muscle using a Bard Vacora device equipped with a sterile 10-gauge needle and under vacuum. A single biopsy (~100 mg) was flash frozen in liquid nitrogen for RNA isolation. Approximately 200 mg of muscle tissue was dissected free of connective tissue, finely minced, and incubated in 1 mg/mL protease (P5147; MilliporeSigma) for 40 min at 37 °C. The slurry was passed through a 70-µm nylon filter and muscle fragments retained. The fiber fragments were vortexed for 2 min in PBS and passed through a 40-µm nylon filter and supernatants retained. The final cell pellet was collected by centrifugation, and equal amounts of cell suspension were seeded into gelatin (EMD Millipore, MilliporeSigma) coated 24-well Corning tissue culture plates (4 wells in total; Corning, Tewksbury, MA). Cells were maintained in growth permissive media (GM) composed of Dulbecco’s modified Eagle medium supplemented with 20% fetal bovine serum, 4 ng/mL FGF2, 1% penicillin-streptomycin, and 0.2% gentamicin at 37 °C in a humidified environment continually purged with 5% CO₂. All cell culture media and supplements were purchased from Invitrogen (Thermofisher, Waltham, MA).

Activation Assay

After 18 h in culture, SCs were washed extensively with PBS containing 5% FBS, 5% penicillin-streptomycin and 0.2% gentamicin and refed GM. Cells were fixed with fresh 4% paraformaldehyde (PFA; Polysciences, Warrington, PA) in PBS (PFA) for 10 min after 48 and 72 h in culture. The times were selected based on previous work demonstrating activation of adult SC after 72 h in culture as determined by thymidine analog incorporation (Brandt et al., 2018). Fixed cells were incubated with 3% BSA in PBS containing 0.1% Triton X-100 for 20 min at room temperature to block nonspecific antigen sites. Antiproliferating cell nuclear antigen (PCNA) was diluted 1:500 in blocking solution and applied to the fixed cells for 1 h at room temperature. After washing with PBS, the cells were further incubated with goat anti-mouse IgG-Alexafluor488 (Invitrogen) diluted 1:250 in PBS containing 10 µg/mL 2-(4-carbamimidoylphenyl)-1H-indole-6-carboximidamide dihydrochloride (DAPI). Following multiple washes with PBS, immune complexes were visualized with an epifluorescent microscope (Nikon Eclipse TS200) equipped with a cooled capture device camera (CoolSNAP HQ) with shutter speed and image digitization controlled with NIS Elements software (Nikon). Representative images from 8 microscope fields at 200-fold magnification were analyzed for PCNA and Hoechst positive nuclei. Percent PCNA(+) was calculated as total PCNA(+)/total DAPI(+) × 100.

RNA Isolation and Quantitative PCR

Approximately 30 mg of frozen tissue was added to Trizol Reagent (2 mL). Tissue was homogenized using a handheld rotor-stator polytron (Kinematica, Bohemia, NY) in bursts at maximum speed for 60 s. After a 5-min incubation at room temperature, chloroform was added as per manufacturer’s recommendation. Samples were centrifuged for 15 min at 12,000 × g at 4 °C. The upper phase was combined with 0.8 M sodium acetate and 100% ice-cold ethanol and precipitated at −80 °C overnight. Precipitates were pelleted by centrifugation for 30 min in a refrigerated centrifuge at 20,000 × g and washed twice with ice-cold 70% ethanol. RNA quantity and quality was assessed spectrophotometrically (NanoDrop, ND-1000, ThermoFisher). Genomic DNA contaminants were removed by digestion (DNase I, ThermoFisher). Fifty nanograms of total RNA was reverse transcribed (High Capacity cDNA Reverse Transcription kit, ThermoFisher) in a final volume of 20 µL. Amplification was performed using 5 ng of cDNA, SYBR chemistry (Power SYBR Green PCR, ThermoFisher) and gene-specific primers for myogenin (F-TCACGGCTGACCCTACAGATG; R-GGTGATGCTGTCCACAATGG), Pax7 (F-CATCGGCGGCAGCAA; R-TCCTCGATCTTTTTCTCCACATC), or glyceraldehyde phosphate dehydrogenase (F-CAAGGCTGTGGGCAAGGT; R-GGAAGGCCATGCCAGTGA) in an Eppendorf Realplex thermocycler (Eppendorf, Hamburg, Germany). The optimum thermal cycling parameters included 95 °C for 10 min, 40 cycles of 95 °C for 15 s and 62 °C for 1 min. A melt curve was generated by 95 °C for 15 s, 62 °C for 15 s followed by 1.75 °C/min for 20 min. All reactions contained a single product as determined by melt-curve analysis. Primer efficiencies ranged from 87% to 107%. Glyceraldehyde phosphate dehydrogenase was used as a housekeeping gene for normalization as neither diet nor exercise affected its expression. Fold change was calculated using the 2^−ΔΔCt method and reported as relative expression (Brandt et al., 2018).

Statistical Analysis

Data were analyzed as a 2-way analysis of variance with repeated measures and Tukey’s post hoc multiple comparisons. The model included the fixed effects of diet and exercise recovery time (Prism GraphPad, San Diego, CA). Significance was established as P < 0.05.

Results

Tributyrin supplementation to adult Thoroughbreds for 30 d caused an increase (P < 0.05) in myogenin expression in the middle gluteal muscle (Fig. 1A). These results are consistent with published effects in growing pigs indicating that the dietary supplementation was sufficient to elicit a similar response (Murray et al., 2018).

Figure 1.

Tributyrin and submaximal exercise affect gluteal muscle gene expression and satellite cell activity. Horses were fed a control (CON) or tributyrin (TB) diet for 30 d prior to retrieval of a middle gluteal muscle biopsy. Quantitative PCR was performed for myogenin expression (A). *Significance at P < 0.05. Relative amounts of Pax7 mRNA were measured before (Pre) and 1, 3 and 5 d after the performance of a submaximal exercise test (B). Satellite cells isolated before (Pre) and 1, 3, and 5 d post-exercise were cultured for 48 h in vitro, fixed, and immunostained for proliferating cell nuclear antigen (PCNA). Data reported as PCNA(+)/total nuclei × 100. TB increased (P < 0.05) the percentage of PCNA(+) cells (C). Percentage of PCNA(+) cells in SC isolates during the recovery period (D). ^a,bDifferent superscripts denote significance at P < 0.05.

The submaximal exercise test was sufficient to increase (P < 0.05) plasma lactate from 0.70 ± 0.04 to 1.35 ± 0.12 mmol/L and packed cell volume from 43.9 ± 0.4% to 49.7 ± 1.9% pre- and post-exercise, respectively. Neither exercise nor diet affected muscle Pax7 mRNA expression before (day −1) or after (days 1, 3, 5) performance of the exercise test (Fig. 1B).

Satellite cell isolates from TB supplemented horses contained a greater (P < 0.05) percentage of PCNA(+) cells than CON contemporaries after 48 h in culture (Fig. 1C). Submaximal exercise caused an increase (P < 0.05) in activated SCs isolated at days 1, 3, and 5 of the recovery period (Fig. 1D). No interactions of diet and exercise were noted.

Discussion

Tributyrin, a butyrate prodrug, affects monogastric gastrointestinal (GI) tract health by acting as an anti-inflammatory as well as altering GI microbiota (Dong et al., 2016; Bedford et al., 2017; Gu et al., 2017; Nguyen et al., 2017). Butyrate acts as a class I and II HDAC inhibitor thus, promoting an open chromatin configuration and gene expression. The class II HDACs, HDAC4 and HDAC7, physically associate with myocyte enhancer factor 2, a cooperative partner of the myogenic regulatory factors (MRFs), to inhibit muscle gene transcription (Dressel et al., 2001; Chan et al., 2003). In contrast, an indirect association of HDAC1 with the MRF, MyoD, serves to suppress proliferation and facilitate myoblast fusion and transcription of genes indicative of terminal differentiation (Mal and Harter, 2003; Galatioto et al., 2010). Although TB offers GI benefits, the butyrate precursor also improves muscle deposition and affects SC biology (Murray et al., 2018). Dietary supplementation with TB causes an increase in myogenin gene expression in pigs, as observed in the middle gluteal muscle herein. The absence of initial myogenin expression levels is a limitation, but it is unlikely that the TB horses would have segregated uniquely into the treatment group. Given the enormity of myonuclei in the biopsy sample, it is likely that the increase in myogenin mRNA is reflective of fiber transcriptional outputs. Elevated expression of myogenin in mouse muscle causes a shift from glycolytic to oxidative metabolism with no change in myosin isoform expression profiles or fiber types (Hughes et al., 1999). Sodium butyrate supplementation for 8 wk resulted in an increase in muscle mitochondria biogenesis and type I oxidative fibers in pigs; myogenin mRNA or protein was not measured (Zhang et al., 2019). In adult mice, HDAC4 indirectly increases myogenin expression that, in turn, upregulates expression of type IIA myosin and nicotinic acetylcholine receptor while suppressing myosin type IIB and phosphofructokinase gene expression (Tang et al., 2009). Interestingly, the gastrocnemius of old mice contains greater amounts of HDAC4 in comparison to their younger counterparts, and levels of the protein in the muscle are further increased following dietary butyrate supplementation (Walsh et al., 2015). The short-chain fatty acid blunted muscle atrophy and promoted oxidative metabolism without affecting either myogenin expression or glycolytic enzyme activity. Myogenin expression was measured only as a marker of TB effects in the present study. The increased expression of the myogenic transcription factor suggests that acetylation status of histone H3 is altered by TB conversion to butyrate. Future efforts, however, should focus on the effects of dietary butyrate supplementation as an effector of oxidative metabolism and fiber type as both parameters are predictive of athletic performance in horses (Rivero and Hill, 2016; Avila et al., 2018).

Satellite cells are the requisite stem and progenitor cell population responsible for repair and regeneration of skeletal muscle (Seale et al., 2000; von Maltzahn et al., 2013). Following strenuous exercise in mice and humans, these cells exit their normally quiescent state becoming bioactive presumably to aid in the repair process (Peake et al., 2017; Joanisse et al., 2018). The niche-localized growth factor directing activation in rodents is hepatocyte growth factor (HGF; Anderson, 2016). Inhibition of HGF suppresses G₀ exit in vitro, whereas supplementation of culture media with the peptide or direct delivery of HGF into muscle promotes SC activation (Allen et al., 1995; Tatsumi et al., 1998; Sheehan et al., 2000). The identity of microenvironment factors that stimulate entry and exit of G₀ in equine SCs remains largely unexplored. Our results demonstrate an effect of submaximal exercise on SC bioactivity. Within 24 h post-exercise, a 50% increase in the percentage of PCNA immunopositive cells traversing G₁ is observed using the indirect activation assay. Tributyrin supplementation also stimulated an increase in the number of PCNA-expressing SCs. Although both exercise and diet affected SC biology, it remains unclear if either caused an increase in activation leading to self-renewal and/or proliferation. No change in muscle Pax7 mRNA content following exercise suggests that the cells detected insufficient damage or mitogenic signals to fully re-engage the cell cycle. However, the definitive increase in PCNA(+) cells in vitro suggests that the SCs have experienced G(alert), a reversible state of quiescence dependent on HGF signals, or a similar priming event (Rodgers et al., 2014, 2017). Although previous work from our lab reported that HGF failed to shorten the initial lag period, an indicator of quiescence, the role of the growth factor as the stimulus for G(alert) remains unresolved (Brandt et al., 2018). Discerning a population containing both Pax7 and phosphoS6 following exercise would establish the existence of the primed state and serve as a useful tool for defining HGF function during equine SC activation.

In summary, both TB and submaximal exercise are effective stimulants of SC activity. The ability of TB to possibly promote an alert phase of quiescence suggests that the compound may be a useful dietary aid that promotes muscle repair and speeds the postexercise recovery process.