Aging‐affiliated post‐translational modifications of skeletal muscle myosin affect biochemical properties, myofibril structure, muscle function, and proteostasis

Abstract The molecular motor myosin is post‐translationally modified in its globular head, its S2 hinge, and its thick filament domain during human skeletal muscle aging. To determine the importance of such modifications, we performed an integrative analysis of transgenic Drosophila melanogaster expressing myosin containing post‐translational modification mimic mutations. We determined effects on muscle function, myofibril structure, and myosin biochemistry. Modifications in the homozygous state decreased jump muscle function by a third at 3 weeks of age and reduced indirect flight muscle function to negligible levels in young flies, with severe effects on flight muscle myofibril assembly and/or maintenance. Expression of mimic mutations in the heterozygous state or in a wild‐type background yielded significant, but less severe, age‐dependent effects upon flight muscle structure and function. Modification of the residue in the globular head disabled ATPase activity and in vitro actin filament motility, whereas the S2 hinge mutation reduced actin‐activated ATPase activity by 30%. The rod modification diminished filament formation in vitro. The latter mutation also reduced proteostasis, as demonstrated by enhanced accumulation of polyubiquitinated proteins. Overall, we find that mutation of amino acids at sites that are chemically modified during human skeletal muscle aging can disrupt myosin ATPase, myosin filament formation, and/or proteostasis, providing a mechanistic basis for the observed muscle defects. We conclude that age‐specific post‐translational modifications present in human skeletal muscle are likely to act in a dominant fashion to affect muscle structure and function and may therefore be implicated in degeneration and dysfunction associated with sarcopenia.


DNA constructs
The Drosophila P-element-containing Mhc genomic construct pwMhc2 (Swank et al., 2000) was digested with Eag I to produce two subclones.The pwMhc-5' subclone contains an 11.3 kb fragment in pCasper.The pMhc-3' subclone contains a 12.5 kb fragment in pBluescriptKS (Stratagene, La Jolla, CA).These served as substrates for site-directed mutagenesis.
For production of N81T and N81A mutant transgenes, pwMhc-5' was digested with Xho I and Avr II.A 6.8 kb fragment was gel isolated and ligated into pLitmus 28I (New England Biolabs).This subclone was digested with Pst I and Avr II to yield a 4.3 kb fragment that was ligated into pLitmus 28I.The resulting subclone was digested with Pst I and Age I, yielding a 1.7 kb fragment that was ligated into pLitmus 28I.The subsequent subclone was subjected to sitedirected mutagenesis using the QuickChange II kit (Agilent) and primer 5'-TGCTCCAGCAAGTGACCCCCCCGAA -3' containing the N81T nucleotide coding change (underlined).For N81A, 5'-TGCTCCAGCAAGTGGCCCCCCCGAA -3' containing the N81A mutation (underlined) was used.The mutated subclone fragments were sequentially cloned back into the intermediate subclones from which they originated.The resulting plasmids were digested with Eag I.The 12.5-kb Eag I fragment of pMhc-3' was ligated into the mutated subclones, to yield pwMhcN81T and pwMhcN81A.
Mimic mutations N81T, R908E and N1168D were also cloned into the pUASattB vector for transgenic insertion using the PhiC31 integrase system (Bischof et al., 2007).To produce the pUASattBMhc control plasmid, a 4.7 kb PCR fragment was generated from the start of exon 2 (5' non-coding transcribed region) through exon 7d.The positive primer introduced Eco RI and Bsi WI restriction sites (underlined) at the 5' end of exon 2 (bold) 5'-CCGGAATTCCGTACGGAAGTTTTGGGCTCACGACGC -3'.A negative PCR primer in exon 7d (5'-GCTGGAATTCCTCACCGTCATCCATGTTGGGTAC -3') contains a natural Eco RI site (underlined).Following PCR, the 4.7 kb Eco RI fragment was ligated into Eco RI-digested pUASattB and pLitmus 28i vectors to yield pUASattB4.7RI and pLitMhc4.7RI.pUASattB4.7RIwas digested with Avr II and Eag I and a 4.4 kb fragment from pwMhc5' was ligated into it to yield pUASattBMhc5'.The 12.5-kb Eag I fragment from pMhc-3' was ligated into this clone to yield pUASattBMhc.For introduction of N81T, a 1.7 kb Mfe I-Sgr AI fragment from pwMhcN81T-5' was ligated into pLitMhc4.7RI.A 4.4 kb Bsi WI-Avr II fragment from this subclone was ligated into the Bsi WI-Sgr AI site of pUASattBMhc5' to yield pUASattBN81T-5'.Ligation of the 12.5-kb Eag I fragment from pMhc-3' into pUASattBN81T-5' yielded pUASattB-N81T.For R908E and N1168D, the 12.5-kb Eag I fragment from pMhcR908E-3' or pMhcN1168D-3' was ligated into pUASattBMhc5' to yield pUASattB-R908E and pUASattB-N1168D.For all full-length clones, the entire coding region and all ligation sites were confirmed by DNA sequencing (Eton Bioscience, San Diego, CA).

Transgenic line validation
RT-PCR (New England Biolabs Protoscript cDNA synthesis kit) was used to confirm that the Mhc transcripts from each homozygous transgenic line were spliced correctly and contained the appropriate site-directed nucleotide changes.LiCl2 extraction was employed to prepare RNA from upper thoraces of two-day-old adult female transgenic flies (Becker et al., 1992).For cDNA synthesis, 3 μmol of reverse primer was mixed with 0.5 μg of RNA from each transgenic line.To assess splicing of alternative exons 3, 7 and 9, a reverse primer for exon 10 (5'-TCGAACGCAGAGTGGTCAT -3') and a forward primer for exon 2 (5'-TGGATCCCCGACGAGAAGGA-3') were used.For alternative exons 11 and 15, a reverse primer for exon 16 (5'-GGGTGACAGACGCTGCTTGGT -3') and a forward primer for exon 10 (5'-GTTCCCCAAGGCCTCCGATCA -3') were used.PCR was performed using 1 μl of cDNA and 3 μmol of each primer pair using the following conditions: 60 s at 94 o C, 30 cycles of: 30 s at 94 o C, 30 s at 55 o C and 2 min at 68 o C. RT-PCR products were sequenced by Eton Bioscience.
We determined myosin expression levels relative to actin accumulation for each homozygous transgenic line in an Mhc 10 background by SDS polyacrylamide gel electrophoresis and densitometry (O'Donnell et al., 1989).For transgenic pUASattB lines, inserts crossed into the Mhc 10 background were crossed with flies containing a fln-Gal4 construct (http://flybase.org/reports/FBtp0097341) that had been recombined onto the Mhc 10 second chromosome, in order to drive Mhc expression.Upper thoraces from five two-day-old female flies were homogenized in 60 µl SDS gel buffer.Five µl of sample were loaded on a 9% polyacrylamide gel; this was repeated five different times, each time using a freshly prepared sample.Protein accumulation was determined using Coomassie blue stained gels that were digitally scanned and analyzed on NIH Image J software (https://imagej.nih.gov/ij/).

Flight and jump assays
Transgenic lines were assayed for flight ability by determining upward (U), horizontal (H), downward (D) or no flight (N) from a release point 20 cm high inside a Plexiglas box with a light source at the top (Drummond et al., 1991).Flight assays were performed at 22°C on ~100 flies for each transgenic line.Flight index was calculated as 6U/T + 4H/T +2D/T + 0N/T, where T is the total number of flies tested (Tohtong et al., 1995).Flies were grouped into cohorts of 10-20; each cohort average value served as a single data point.
The jump ability of 20 homozygous flies from each line was tested at 22°C after surgically removing wings from newly eclosed female flies and allowing a 48-hour recovery time.Flies were encouraged to jump from a pedestal 9.5 cm high by tapping a paint brush on its edge to cause vibration.The horizontal jump distance was measured and noted on a paper marked with concentric rings surrounding the pedestal (Eldred et al., 2010).The greatest three jump distances out of ten jumps per fly were averaged and used as a single data point.

Electron and light microscopy
Transmission electron microscopy was performed as previously described (O'Donnell & Bernstein, 1988).Cross-and longitudinal-sections were obtained from females, with at least three different organisms examined for each transgenic line.Myofibrils shown in each panel are representative of the population at that given developmental stage.For light microcopy, 1.0-μm thick sections were taken from blocks prepared for electron microscopy.Slides containing the thick sections were placed on an 80°C heating block to remove moisture.Sections were then stained with 1% of toluidine blue on the heating block for 20 s.Sections were then rinsed with water and allowed to air dry.
Following centrifugation (5 min, 15,000 x g, ~13000 rpm in a Beckman FA241.5Pfixed angle rotor), the pellet was resuspended in YMG with 2% Triton-X detergent.The supernatant was discarded after another centrifugation (as above) and permeabilized fibers were washed free of detergent in YMG without glycerol so that myosin could be extracted for 15 min in 82.5 µl of myosin extraction buffer (1 M KCl, 50 mM KPi, pH 6.8, 5 mM MgCl2, 0.5 mM EGTA, 16.4 mM Na-pyrophosphate, 20 mM DTT, protease inhibitor tablet).The high salt extract was centrifuged (as previously) to remove insoluble material and myosin was precipitated on ice overnight after dilution to 40 mM KCl and 10 mM DTT.After centrifugation at 100,000 x g (43,000 rpm) for 20 min (Beckman TLA 100.3 fixed angle rotor), the pellet was dissolved in 13.5 µl Wash B (2.4 M KCl, 0.5 mM EGTA, 0.14 M histidine, 20 mM DTT, 90 mM KOH, protease inhibitor tablet) on ice for 30 min.Myosin remaining complexed with actin was precipitated by slowly adding 94.5 µL of 10 mM DTT to decrease the KCl concentration to 0.3 M. The sample was then centrifuged at 60,000 x g (33,300 rpm) for 25 min.The supernatant was removed, diluted 10-fold with 10 mM DTT, incubated on ice for 1 h and centrifuged at 100,000 x g (43,000 rpm) for 25 min.Following centrifugation, pelleted myosin was dissolved in myosin storage buffer (20 mM MOPS, 0.5 M KCl, 20 mM DTT, 2 mM MgCl2) on ice for 30 min.The concentration of myosin was determined by spectrophotometry using the equation: [(A280-A310)/0.53]* dilution factor, where 0.53 is the extinction coefficient for Drosophila myosin.Addition of myosin storage buffer allowed concentration adjustment as needed.
Reactions were incubated for 10 min.Upon immediate transfer to 500 μL 0.039% malachite green (in 1.1% ammonium molybdate, 1N HCl, and 0.0005% NP-40 alternative), binding of inorganic phosphate yielded green complexes (Littlefield et al., 2003).Each colorimetric reaction was stopped after 2 min with 50 μL of 34% sodium citrate, pH 2.0 (we note that 34% anhydrous citric acid, pH 1, stops the reaction most effectively).Controls included phosphate standards of 0-8000 pmol, a blank (no actin or myosin), and determination of actin ATPase activity without myosin.All samples were run in duplicate.
Spectrophotometer readings on 500 μL of malachite green reactions taken at A650 were averaged between technical replicates and interpolated within the standard curve using GraphPad Prism (GraphPad Software Inc., La Jolla, CA).Picomoles of inorganic phosphate generated by samples without myosin were subtracted from paired samples with myosin.
Nanomoles of phosphate generated per min per µg of myosin were converted to the reaction rate in s -1 .Values for Vmax, and Km were determined (after subtracting basal ATPase activity) by plotting actin-activated myosin ATPase activity vs. actin concentration according to Michaelis-Menten kinetics.

In vitro motility assay
Actin sliding velocity arising from myosin interaction was determined by computational analysis of in vitro fluorescent optics videos (Kron & Spudich, 1986).Nitrocellulose-coated coverslips for flow cells were prepared by putting ten drops of 1% nitrocellulose in amyl acetate (Ladd Research) on the surface of deionized water in a petri dish.After 10 s, coverslips were immersed on one side to bind nitrocellulose through water surface tension.They were dried nitrocellulose-side up.Two 0.005 in thick, 1 mm wide Artus Motor Mount Shims were affixed to a clean microscope slide using Loctite 532 adhesive (Ellsworth Adhesives).When the adhesive is still malleable, the coverslip was pressed nitrocellulose-side down onto the shims.The resulting flow cell was placed under UV light for 20 min to cure the adhesive.

Flow
Photonics) and the Fiji processing package of Image J (https://imagej.net/software/fiji/) was used to track smooth movements of the front end of filaments.