Patient cohort

This study was approved by the local research ethics committee (Newcastle and North Tyneside 2, 17-NE-0015) and performed in accordance with the Declaration of Helsinki. As a separate, new study, skeletal muscle samples were obtained by percutaneous biopsy of tibialis anterior under local anaesthesia from males with (n = 30) and without (n = 15) HIV. Some were recruited in conjunction with the Pharmacokinetics and Clinical Observations in People over Fifty (POPPY) study³⁹. All participants were 50 years or older. In addition, participants supplied 30 ml of whole blood. Men with HIV were able to participate if they had been on ART for at least 6 months and had a suppressed plasma HIV-1 viral load (

All participants completed a standardised interview, and clinical information (including, for those with HIV, CD4+ lymphocyte count, HIV diagnosis details, and ART history) was collected and confirmed through medical records, where available. The presence or absence of comorbidities was confirmed through medical records. Written consent was obtained from all participants.

The presence or absence of the following comorbidities was assessed through patient questionnaire: stroke and CVD, neuropathy, diabetes, dementia, cancer, renal disease, fractures, hepatitis, peripheral vascular disease, joint disease or replacements, osteoporosis, and falls. Clinical HIV data were captured by patient questionnaire and case note review and included current CD4 count, months since diagnosis, months on ART, and months untreated.

Assessment of ageing phenotype

Frailty was assessed as previously described using the original five Fried’s frailty phenotype (FFP) criteria³. Participants with one or two criteria were classified as pre-frail. Physical function was determined through a standardised short physical performance battery (SPPB) consisting of a repeat chair stand, standing balance test, hand grip assessment, and 4 m walk, with scoring described previously in ref. ⁴⁰.

Sarcopenia was classified based on muscle mass (DXA), strength (grip strength), and physical performance (SPPB)². Participants were classified as having pre-sarcopenia if they had abnormal results for muscle mass only; sarcopenia if they had abnormal results for muscle mass, as well as either muscle strength (classified as weak grip strength) or physical performance; and severe sarcopenia if they had abnormal results from all three criteria, as based on EWGSOP2 classifications.

Metabolic equivalent (MET) expenditure per week was calculated as a surrogate for physical activity, with participants classified as either ‘inactive’, ‘minimally active’, or ‘HEPA active’. Results were calculated as previously described in ref. ⁴¹.

Skeletal muscle histology and immunofluorescence

Transverse snap-frozen percutaneous biopsy sections (10 µm) were subjected to an array of immunofluorescence assessments in order to analyse mitochondrial RC CI and IV (CIV) deficiency, mitochondrial outer membrane (MOM) protein VDAC1 abundance, intramyocellular lipid accumulation (IMCL), Pax7⁺ SC frequency, fibre type composition, and lipofuscin granule deposition. In addition, histology was performed to determine skeletal muscle fibrosis, as well as myofibre degeneration and regeneration.

Specifically, CI and CIV deficiency were quantified using a multiplex immunofluorescence assay as described previously in ref. ⁴², with markers for laminin (myofibre boundary) (Sigma, #L9393); VDAC1 (mitochondrial outer membrane marker) (Abcam, #ab14734); NDUFB8 (mitochondrial RC complex I) (Abcam, #ab110242); and MTCO1 (mitochondrial RC complex IV) (Abcam, #ab14705). Secondary antibodies included AlexaFluor 405 (Laminin) (Invitrogen, #A31556); AlexaFluor 546 (VDAC1) (Thermo Fisher Scientific, #A21143); Anti-IgG1-Biotin (Jackson Laboratories, #115-065-205) plus Strep-conjugated AlexaFluor 647 (both NDUFB8) (Thermo Fisher Scientific, #S-32357); AlexaFluor 488 (Life Technologies, #A21131). For each batch of skeletal muscle analysed, additional cryosections were stained in parallel using only the primary antibody for laminin, along with all the secondary antibodies (termed ‘no primary controls’, NPC). In order to minimise batch effects, all batches of cryosections analysed comprised a mixture of all experimental groups, NPC, and calibrator samples. For analysis, fibres were segmented using the laminin signal, before the average fluorescence intensity for NDUFB8, MTCO1, and VDAC1 was automatically quantified using in-house software (MATLAB 2015a). NDUFB8 and MTCO1 intensity was then expressed relative to VDAC1 before the abundance of each respective protein was expressed as z-scores based on the expression from young healthy calibrator samples using an in-house R Shiny script. Myofibres with NDUFB8 or MTCO1 z-scores −3 as ‘normal’. Finally, the percentage of myofibres with either NDUFB8 or MTCO1 deficiency was log₁₀ transformed. Fluorescence images for RC complex abundance, as well as all other immunofluorescence imaging, were acquired on a Zeiss Axio Imager M1 at 20× magnification and tiled and analysed on Zen 2011 Blue Edition (Zeiss).

IMCL was determined through BODIPY (493/503) (Thermo Fisher Scientific) staining and imaging. After air-drying at room temperature (RT) and fixation with 3.7% formaldehyde (ChemCruz) for 30 min each respectively, sections were permeabilised with 0.25% Triton x-100 (Thermo Fisher Scientific) for 5 min then incubated with antibody against laminin (Sigma, #L9393) for 60 min at RT before incubation with AlexaFluor 405 (Invitrogen, #A31556) secondary antibody and BODIPY (493/503) (Thermo Scientific Fisher, #D3822) for 90 min in a dark humidified chamber at RT. Finally, sections were mounted in Molwoil 4–88 (Sigma) and stored at −20 °C. For image analysis, individual myofibres were qualitatively classified as having IMCL or being ‘normal’ depending on BODIPY staining coverage and fluorescence intensity.

For the quantification of Pax7⁺ SC, 10 µm cryo-sections were air-dried for 1 h at RT before fixation in cold 4% PFA for 4 min. Next, sections were permeabilised for 10 min at RT with 0.2% Trion-x 100 before blocking with 5% NGS for 1 h at RT, followed by overnight incubation with Pax7 primary antibody (diluted in 10% NGS) (DSHB) at 4 °C. The following day, sections were incubated with AlexaFluor 488 (Thermo Fisher Scientific, #A-21242) secondary antibody for 2 h at RT before incubation with Hoerst (1:1200) (abcam) for 15 min at RT and finally mounting with Prolong Gold Antifade Mountant (Thermo Fisher Scientific). Stained sections were imaged before the frequency of Pax7⁺ SCs per fibre was quantified following confirmation of Pax7 and nuclei colocalisation.

To quantify the proportion of skeletal muscle fibre types I, IIa, and IIx, air-dried cryo-sections were subjected to multiplex immunofluorescence staining with targets for the respective fibre types. After air-drying of cryo-sections for 1 h at RT, sections underwent fixation with 4% paraformaldehyde (PFA) for 1 h at RT before blocking with 10% NGS (Abcam) again for 1 h at RT, followed by incubation with a primary antibody cocktail with markers for laminin (Sigma, #L9393), BA-F8 (type I) (DSHB, #10572253), SC-71 (type IIa) (DSHB, #2147165), and 6H1 (type IIx) (DSHB, #2314830) overnight at 4 °C. Next, sections were incubated with a secondary antibody cocktail consisting of AlexaFluor 405 (laminin) (Invitrogen, #A31556); AlexaFluor 488 (BA-F8) (Invitrogen, #A31141); AlexaFluor 546 (SC-71) (Invitrogen, #A21123); and AlexaFluor 647 (6H1) (Invitrogen, #A21238) for 2 h at 4 °C. Sections were then mounted using ProLong Gold Antifade Mountant (Thermo Fisher Scientific) and stored at −20 °C prior to imaging. Following imaging, the number and proportion of each respective fibre type were quantified based on fluorescence staining profile.

Next, in order to quantify lipofuscin granule accumulation, cryo-sections were air-dried for 1 h and immediately mounted on coverslips. Sections were imaged at 546 nm and 647 nm wavelengths at 20× magnification and tiled before the frequency and area of autofluorescent lipofuscin granules were quantified on the Columbus Image Data Storage and Analysis System software based on colocalisation of both respective channels.

To quantify degeneration and regeneration of muscle fibres, 10 µm cryo-sections were subjected to Haematoxylin and Eosin (H&E) staining, while fibrosis was assessed by Masson’s trichrome staining. With regards to H&E histochemistry, 10 µm cryo-sections were removed from −80 °C and air-dried for 1 h at RT. The following sections underwent fixation with cold 4% PFA for 3 min. Next, sections were stained with Haematoxylin for 10 min, then rinsed clear in H₂O before being washed in Scott’s tap water for 1 min, followed by Eosin for 1 min. Finally, sections were rehydrated through an EtOH gradient (10 dips 70% EtOH, 10 dips 95% EtOH, 20 dips 100% EtOH) followed by 2 changes of 20 dips in Histoclear, mounting with DPX (Sigma), and storage at RT.

For Masson’s trichrome histochemistry, 10 µm cryo-sections were removed from −80 °C and air-dried for 1 h at RT before fixation as described previously, followed by further fixation in Bouin’s Fluid (Sigma) at 60 °C for 30 min before being rinsed clear with H₂O. Next, sections were stained with Weigert’s Iron Haematoxylin (Abcam) for 5 min, rinsed clear with H₂O, then differentiated in phoshotunsic acid solution (Abcam) for 10 min. After rinsing with H₂O, sections were incubated with alanine blue (Abcam) for 7 min followed by differentiation with acetic acid (Abcam) for 3 min. Finally, sections were rehydrated by an EtOH gradient and mounted as described above.

For imaging, both H&E and Masson’s trichome-stained sections were imaged using a Zeiss Axio Imager M1 and Zen 2011 (blue edition) with a chromatic digital camera (AxioCam MRm) at 10x magnification. Regenerated myofibres were determined by the presence of central nuclei, whilst degenerated myofibres were qualitatively determined as described previously in ref. ⁴³. Briefly, this included either loss of cross-striations, cytoplasmic eosinophilia, fibre fragmentation or rupture, or sarcoplasmic vacuolisation. Fibrosis was quantified as the percentage area covered by collagen (blue staining).

Cytokine and myokine quantification

A bespoke panel of inflammatory cytokines, as well as putative myokines and mitokines, was measured using the MSD U-PLEX (MCP-4, Leptin, TNFα, IL-10, IL-6, IL-8, IL-15, IP-10, and CRP) and V-PLEX (FGF21 and FGF23) assays (Meso Scale Discovery) according to the manufacturer’s instructions, with signal intensity read on the MESO QuickPlex SQ 120 (Meso Scale Discovery).

Statistical analysis

Statistical analysis was performed in Prism v5.04, IBM SPSS Statistics v23, Columbus Image Data Storage and Analysis System, and Microsoft Excel 2016. Shapiro–Wilk tests were used to determine normality within datasets, with a p ≥ 0.05 indicating normality. One-way ANOVA or Kruskal–Wallis test and unpaired t-tests or Mann–Whitney test were used to compare differences between groups. Fisher’s exact test was used to determine differences between categorical factors. Spearman’s correlation analysis was performed to compare continuous variables. To account for multiple testing, a Bonferroni correction was performed within families of related assays.

Specifically, proportions between groups (men with and without HIV) were compared by Fisher’s Exact test.

Univariate analysis between continuous variables for clinical HIV data, characteristics related to physical function (including grip strength, waist circumference, fat and lean mass, BMI), and muscle pathophysiology parameters was performed using Spearman’s correlation.

Groupwise comparison of continuous variables (including biomarkers) was performed by the Mann–Whitney test. One-way ANOVA (normally distributed) or Kruskal–Wallis test (non-normally distributed) was performed in order to determine the associations between the categorical measures of physical function (frailty phenotype, sarcopenia, SPPB, and MET) with clinical HIV data.

Bonferroni correction for multiple testing was performed. Bonferroni-corrected p values were calculated for families of related analyses, as follows. For correlation analyses of HIV clinical data, a Bonferroni-corrected p value p value p value 

As this was an exploratory study to describe histochemical changes in skeletal muscle, we additionally focused on the mechanistic consistency of observed associations.