T/T homozygosity of the tenascin-C gene polymorphism rs2104772 negatively influences exercise-induced angiogenesis

Abstract

Background: Mechanical stress, including blood pressure related factors, up-regulate expression of the pro-angiogenic extracellular matrix protein tenascin-C in skeletal muscle. We hypothesized that increased capillarization of skeletal muscle with the repeated augmentation in perfusion during endurance training is associated with blood vessel-related expression of tenascin-C and would be affected by the single-nucleotide polymorphism (SNP) rs2104772, which characterizes the non-synonymous exchange of thymidine (T)-to-adenosine (A) in the amino acid codon 1677 of tenascin-C.

Methods: Sixty-one healthy, untrained, male white participants of Swiss descent performed thirty 30-min bouts of endurance exercise on consecutive weekdays using a cycling ergometer. Genotype and training interactions were called significant at Bonferroni-corrected p-value of 5% (repeated measures ANOVA).

Results: Endurance training increased capillary-to-fiber-ratio (+11%), capillary density (+7%), and mitochondrial volume density (+30%) in m. vastus lateralis. Tenascin-C protein expression in this muscle was confined to arterioles and venules (80% of cases) and increased after training in A-allele carriers. Prior to training, volume densities of subsarcolemmal and myofibrillar mitochondria in m. vastus lateralis muscle were 49% and 18%, respectively, higher in A/A homozygotes relative to T-nucleotide carriers (A/T and T/T). Training specifically increased capillary-to-fiber ratio in A-nucleotide carriers but not in T/T homozygotes. Genotype specific regulation of angiogenesis was reflected by the expression response of 8 angiogenesis-associated transcripts after exercise, and confirmed by training-induced alterations of the shear stress related factors, vimentin and VEGF A.

Conclusion: Our findings provide evidence for a negative influence of T/T homozygosity in rs2104772 on capillary remodeling with endurance exercise.

Fig 1. Assessed myocellular parameters.

Electron micrograph indicating the assessed ultrastructural parameters in vastus lateralis muscle of a subject before endurance training.

Fig 2. Analysis of SNP rs2104772 by High-Resolution Melt analysis (HRM).

A-C) Line graph showing the detected or derived fluorescence intensity for the analyzed SNP (i.e. melting curves). The displayed examples included measurements for A/A homozygotes (red line, n = 5)) and T/T homozygotes (blue line, n = 3) relative to the heterozygote A/T (green line, n = 12). Every sample was analyzed in duplicate. A) Raw data of the pre-melt, melt, and post-melt regions. B) Normalized data derived from the raw data plots. C) Melting curves derived after normalization versus the A/A genotype. D-F) Sequence analysis of the three identified genotypes in chromatograms presenting the forward sequence. Arrows link the single-nucleotide polymorphism (SNP) position 44513. The presence of 'W' in A/T genotype denotes heterozygosis for the SNP where a double-peak is present at position 44513 for both sequenced alleles (arrow, nucleotides A and T simultaneously).

Fig 3. rs140772 affects changes in muscle capillarization with endurance training.

Box whisker plot visualizes the medians ± standard errors (central lines and boxes, respectively) and minima/maxima (whiskers) of the fold changes in capillary-to-fiber ratio post- vs. pre-training for the three rs2104772 genotypes. A/A (n = 12), A/T (n = 38), and T/T (n = 11). *, P < 0.05 vs. T/T, ANOVA with Fisher’s post-hoc test. #, P < 0.05 for post vs. pre for the indicated comparison.

Fig 4. Tenascin-C protein in vastus lateralis muscle with endurance training.

(A) Immunoblot (Top part of panel) showing the detection of tenascin-C protein in the three rs2104772 genotypes before and after endurance training. The position corresponding to the large tenascin-C isoform at 230 kDa is indicated. The bottom part of the panel shows the corresponding actin band (loading control) on the Ponceau-S-stained membrane before immunoblotting. For image assembly see S2, S3 and S4 Figs. (B) Box whisker plot visualizing medians ± standard errors (central lines and boxes, respectively) and minima/maxima (whiskers) of actin-related Tenascin- content pre and post training in the respective genotypes. In total biopsies from 18 subjects were assessed: A/A (n = 4), A/T (n = 10), and T/T (n = 4). *, p < 0.05 vs. T/T pre; $, p < 0.05 vs. pre. ANOVA with Fisher’s post-hoc test.

Fig 5. Tenascin-C protein expression in vastus lateralis muscle.

A,B) Controls showing Tenascin-C specific straining (orange) of a large blood vessel (arteriole) in a muscle cross-section of an untrained participant after incubation with antibody MA3 (A) compared to incubation with a non-specific pre-immune serum (B). C) Tenascin-C (green, antibody B28.13) and CD31 (red) immuno-fluorescent staining, as well as co-localized tenascin-C and CD31 staining (yellow) in a section from a trained subject. A large blood vessel (most likely a venule) is identified based on CD31-staining and its thickened vessel wall. D,E) Tenascin-C staining (antibody MA3) in vastus lateralis muscle of a same participant before (D) and after (E) endurance training. Nuclei are stained in blue. Arrows and arrowheads point to tenascin-C staining in capillary structures and larger blood vessels (arterioles and venules). Bar = 50 μm.

Fig 6. rs140772-dependent alterations in transcript expression.

Heat map visualizing fold changes in the expression of the 34 gene transcripts, which demonstrated genotype specific alterations 24 h post exercise. The identified transcripts demonstrated level alterations during the course of the first 24 h after exercise and showed genotype specific differences in fold changes 24 h post exercise vs. pre-exercise for the normalized expression levels (significance analysis of microarrays, SAM). Fold changes are given in color coding: Red: up, blue: down. Arrows indicates the transcripts, which are associated with angiogenesis.

Fig 7. Vimentin and VEGF A protein in vastus lateralis muscle with endurance training.

(A, B) Immunoblot showing the small (32 kDa) and large (45 kDa) vimentin isoforms (A) and VEGF A monomers and dimer (B) in one participant for each genotype before and after endurance training. The position of the actin band on the Ponceau-S-stained membrane before immunoblotting, which served as a loading control, is indicated. For image assembly see S5, S6, S7, S8, S9 and S10 Figs. (C, D) Box whisker plot visualizing the median ± standard error (central lines and boxes, respectively) and minima/maxima (whiskers) of the fold changes in vimentin (C) and VEGF A content (D). In total biopsies from 18 subjects were assessed: A/A (n = 4), A/T (n = 10), and T/T (n = 4). *, p < 0.05 vs. T/T same time point; $, p < 0.05 vs. pre. Repeated ANOVA with Fisher’s post-hoc test.