Effects of aging and lifelong aerobic exercise on basal and exercise-induced inflammation

Abstract

Age-associated chronic basal inflammation compromises muscle mass and adaptability, but exercise training may exert an anti-inflammatory effect. This investigation assessed basal and exercise-induced inflammation in three cohorts of men: young exercisers [YE; n = 10 men; 25 ± 1 yr; maximal oxygen consumption (V̇o_2max), 53 ± 3 mL·kg⁻¹·min⁻¹; quadriceps area, 78 ± 3 cm²; means ± SE], old healthy nonexercisers (OH; n = 10; 75 ± 1 yr; V̇o_2max, 22 ± 1 mL·kg⁻¹·min⁻¹; quadriceps area, 56 ± 3 cm²), and lifelong exercisers with an aerobic training history of 53 ± 1 yr (LLE; n = 21; 74 ± 1 yr; V̇o_2max, 34 ± 1 mL·kg⁻¹·min⁻¹; quadriceps area, 67 ± 2 cm²). Resting serum IL-6, TNF-α, C-reactive protein, and IGF-1 levels were measured. Vastus lateralis muscle biopsies were obtained at rest (basal) and 4 h after an acute exercise challenge (3 × 10 repetitions, 70% 1-repetition maximum) to assess gene expression of cytokines [IL-6, TNF-α, IL-1β, IL-10, IL-4, interleukin-1 receptor antagonist (IL-1Ra), and transforming growth factor-β (TGF-β)], chemokines [IL-8 and monocyte chemoattractant protein-1 (MCP-1)], cyclooxygenase enzymes [cyclooxygenase-1 and -2 (COX-1 and COX-2, respectively), prostaglandin E₂ synthases [microsomal prostaglandin E synthase 1 (mPGES-1) and cytosolic prostaglandin E₂ synthase (cPGES)] and receptors [prostaglandin E₂ receptor EP3 and EP4 subtypes (EP3 and EP4, respectively), and macrophage markers [cluster of differentiation 16b (CD16b) and CD163], as well as basal macrophage abundance (CD68⁺ cells). Aging led to higher (P ≤ 0.05) circulating IL-6 and skeletal muscle COX-1, mPGES-1, and CD163 expression. However, LLE had significantly lower serum IL-6 levels (P ≤ 0.05 vs. OH) and a predominantly anti-inflammatory muscle profile [higher IL-10 (P ≤ 0.05 vs. YE), TNF-α, TGF-β, and EP4 levels (P ≤ 0.05 vs. OH)]. In OH only, acute exercise increased expression of proinflammatory factors TNF-α, TGF-β, and IL-8 (P ≤ 0.05). LLE had postexercise gene expression similar to YE, except lower IL-10 (P ≤ 0.10), mPGES-1, and EP3 expression (P ≤ 0.05). Thus, although aging led to a proinflammatory profile within blood and muscle, lifelong exercise partially prevented this and generally preserved the acute inflammatory response to exercise seen in young exercising men. Lifelong exercise may positively impact muscle health throughout aging by promoting anti-inflammation in skeletal muscle.

NEW & NOTEWORTHY This study assessed a unique population of lifelong aerobic exercising men and demonstrated that their activity status exerts an anti-inflammatory effect in skeletal muscle and circulation. Furthermore, we provide evidence that the inflammatory response to acute exercise is dysregulated by aging but preserved with lifelong exercise, which might improve skeletal muscle resilience to unaccustomed loading and adaptability into late life.

INTRODUCTION

Chronic low-grade inflammation throughout aging (“inflammaging”) threatens functional capacity, independence, and quality of life in older individuals (20, 21). Inflammaging has also been associated with sarcopenia, muscle atrophy, and accompanying functional deficits in older adults (66, 67, 85), likely through its negative impact on muscle protein balance (33, 70, 84). Further supporting this connection, chronic anti-inflammatory drug consumption in older adults appears beneficial for muscle mass and performance (6, 32, 39).

Skeletal muscle plays an important role in inflammatory signaling: factors such as cytokines, prostaglandins, and chemokines can be both released (37, 52, 55) and taken up (51, 84) by muscle at rest and following exercise. Aging has been shown to alter these factors in skeletal muscle (35, 56), which may contribute to poor communication with inflammatory cells (e.g., macrophages), dysregulated protein balance, and impaired resolution of inflammation following a stimulus, such as muscle loading. In aging individuals, exaggerated or sustained inflammation following exercise may contribute to suboptimal muscle adaptations (17, 38, 68). Our laboratory previously found that older adults taking anti-inflammatory drugs throughout resistance training demonstrated superior muscle growth over placebo (78, 80), suggesting that chronic low-grade muscle inflammation may dysregulate the exercise response and interfere with adaptation. Thus, to reverse age-related muscle mass losses, basal inflammation may need to be controlled (1), or exercise training prescriptions may need to be designed to overcome the basal inflammatory burden (3, 65). Resistance exercise in particular is a potent antisarcopenic stimulus that targets both slow-twitch (oxidative, type I) and fast-twitch (glycolytic, type II) muscle fibers (31, 80), the latter of which are a notable casualty of the aging process (50).

However, although short-term training studies have demonstrated reductions in basal systemic inflammation in older adults (4), this finding is not universal within muscle (17, 81). Thus, rather than attempt to reverse age-related declines in muscle health, the optimal exercise strategy might seek to prevent them. Recent research has examined the impact of lifelong exercise (adherence to a structured exercise program throughout the adult life span into and beyond retirement age; 13, 24). Previous studies have demonstrated superior cardiovascular health and skeletal muscle mass in lifelong aerobic exercisers compared with age-matched nonexercising control subjects (12, 41, 76). Though limited, evidence from 50–65-yr-old populations suggests that these benefits are accompanied by a more favorable circulating inflammatory profile (42, 43) and an improved molecular environment within muscle (64). Thus, the aims of this investigation were to examine skeletal muscle to assess the basal inflammatory profile and inflammatory response to an acute resistance exercise challenge in lifelong exercisers, young exercisers, and old healthy nonexercisers. These objectives were met with a comprehensive analysis of 23 targets related to inflammation in muscle, macrophages, and circulation.

METHODS

Subjects

Old lifelong exercisers (LLE, n = 21 men), old healthy nonexercisers (OH, n = 10), and young exercisers (YE, n = 10) were included in this investigation (Table 1). Subjects were recruited from the greater Muncie, Indiana, area by newspaper advertisements, mailed flyers, and personal interaction. More extensive subject characteristics and more details regarding the recruitment and screening process, along with cardiovascular and skeletal muscle profiles, are presented by our research team elsewhere (13, 24). Enrolled individuals were free from acute or chronic illness (cardiac, pulmonary, liver, or kidney abnormalities, cancer, uncontrolled hypertension, and insulin- or noninsulin-dependent diabetes or other known metabolic disorders), free from orthopedic limitations (including any artificial joints), and did not smoke or participate in other forms of tobacco use. The study was approved by the Institutional Review Board of Ball State University. All study procedures, risks, and benefits were explained to the subjects before they gave written consent to participate.

Table 1.

Subject characteristics

		Lifelong Exercisers
	YE	Combined	LLE-P	LLE-F	OH
n	10	21	14	7	10
Age, yr	25 ± 1*	74 ± 1	74 ± 1	75 ± 2	75 ± 1
Height, cm	181 ± 2	180 ± 2	179 ± 2	182 ± 3	177 ± 2
Weight, kg	75 ± 3	79 ± 2	77 ± 2	83 ± 5	88 ± 3*
BMI, kg/m2	23 ± 1	24 ± 1	24 ± 1	25 ± 1	28 ± 1*
Body fat, %	18 ± 2*	24 ± 1†	22 ± 1‡	27 ± 1	32 ± 1
V̇o2max, mL·kg−1·min−1	53 ± 3*	34 ± 1†	38 ± 1‡	27 ± 2	22 ± 1
Quadriceps size, cm2	78 ± 3*	67 ± 2†	68 ± 2	65 ± 3	56 ± 3
Quadriceps strength, N·m	210 ± 12*	165 ± 6†	166 ± 8	163 ± 10	134 ± 9
Quadriceps power, W	699 ± 30*	370 ± 19	365 ± 13	377 ± 50	318 ± 42
Handgrip strength, kg	51 ± 3	46 ± 2	48 ± 3	43 ± 2	44 ± 1
Steps per day	9,404 ± 635	9,560 ± 619	9,369 ± 725	10,006 ± 1,265	5,813 ± 488*

Values are means ± SE; n = no. of subjects. Additional cardiovascular and skeletal muscle data, as well as details of body fat (dual-energy X-ray absorptiometry), maximal oxygen consumption (V̇o_2max), muscle size (MRI) and function, and steps-per-day measurements are presented by us elsewhere (13, 24). BMI, body mass index; LLE-F, lifelong exercisers-Fitness; LLE-P, lifelong exercisers-Performance; OH, old healthy nonexercisers; YE, young exercisers.

^*P ≤ 0.05 vs. main groups;

^†P ≤ 0.05 vs. OH;

^‡P ≤ 0.05 LLE-P vs. LLE-F.

Exercise history of the subjects was carefully evaluated using a comprehensive questionnaire and confirmed through personal interviews (Table 2). The LLE cohort consisted primarily of cyclists and runners who reported ~50 yr of structured exercise. LLE trained ~5 days and ~7 h per week. Exercise history of LLE subjects was extensively reviewed for frequency, duration, intensity, and athletic achievements. As such, two clear LLE subgroups emerged: one group that participated in lower-intensity training for physical fitness (Fitness, LLE-F; n = 7) and another group that trained more vigorously and often participated in competitive events (Performance, LLE-P; n = 14).

Table 2.

Exercise training histories

		Lifelong Exercisers
	YE	Combined	LLE-P	LLE-F	OH
Total training, yr	5 ± 1*	53 ± 1	53 ± 1	53 ± 3
Competitive focus	Yes		Yes	No
Lifetime average
Frequency, days/wk		4.5 ± 0.2	4.4 ± 0.2	4.6 ± 0.3
Duration, h/wk		7.3 ± 0.5	7.6 ± 0.7	6.6 ± 0.9
Intensity		2.0 ± 0.1	2.1 ± 0.1†	1.8 ± 0.1
Current decade
Frequency, days/wk	5.1 ± 0.2	4.7 ± 0.3	4.5 ± 0.3	4.9 ± 0.7
Duration, h/wk	7.0 ± 0.7	8.1 ± 1.1	8.5 ± 1.4	7.4 ± 1.9
Intensity	2.8 ± 0.1*	2.0 ± 0.1	2.2 ± 0.1†	1.5 ± 0.2

Values are means ± SE. Competitive focus indicates whether exercise training for the purpose of competition was currently or once a primary goal for the majority of the group. Lifetime average reflects current decade exercise habits for young exercisers (YE). Levels of self-reported training intensity were as follows: 1 (light), 2 (moderate), and 3 (hard). In the case that a subject reported >1 training intensity level, values were weighted and averaged (e.g., 80% of training at a 2 and 20% of training at a 3 resulted in an overall intensity of 2.2). More detailed exercise training histories are presented by us elsewhere (24). LLE-F, lifelong exercisers-Fitness; LLE-P, lifelong exercisers-Performance; OH, old healthy nonexercisers.

^*P ≤ 0.05 vs. lifelong exercisers combined;

^†P ≤ 0.05 LLE-P vs. LLE-F.

Serum Inflammatory Markers

A resting, fasted blood draw was taken for measurement of circulating inflammatory markers. Samples were analyzed (LabCorp, Muncie, IN) for serum C-reactive protein (latex immunoturbidimetry), IL-6 and TNF-α (enzyme-linked immunosorbent assay), and IGF-1 (immunochemiluminometric assay).

Acute Exercise Trial and Skeletal Muscle Biopsies

Subjects completed a resistance exercise challenge of the knee extensors, consisting of 3 sets of 10 repetitions at 70% 1-repetition maximum, with 2-min rest between each set. When completed chronically, this acute exercise stimulus elicits significant increases in muscle size and strength in young and old individuals (61, 68, 75, 77, 88). Muscle biopsies (5) of the vastus lateralis were obtained before and 4 h after the resistance exercise challenge. This postexercise time point reflects an optimal time for interrogating expression of numerous intramuscular regulators of muscle adaptation, including cytokine activity (37, 58–60, 62, 89). All biopsies were obtained in the fasted state (≥10 h), after at least 30 min of supine rest. Subjects remained in the laboratory and rested quietly during the 4-h postexercise period. Subjects also refrained from structured exercise and aspirin consumption for 72 h, alcohol consumption for 24 h, and caffeine the morning of the trial.

Following each muscle biopsy, excess blood, visible fat, and connective tissue were removed, and a portion of the muscle (~20 mg) to be used for mRNA analysis was immediately frozen and then stored in liquid nitrogen. Prior to analysis, the muscle was transferred to 0.5 mL of RNAlater-ICE (Ambion, Austin, TX) and stored at −20°C until analysis. A portion of the muscle to be used for immunohistochemistry was oriented longitudinally in a mounting medium (tragacanth gum; Sigma, St. Louis, MO) atop a cork, frozen in isopentane cooled in liquid nitrogen, and subsequently stored in liquid nitrogen until analysis.

Gene Expression Measurements

Inflammatory factors listed in Table 3 were assessed in vastus lateralis skeletal muscle homogenates using real-time quantitative polymerase chain reaction (qPCR). Muscle mRNA analyses were completed on all 41 subjects for basal expression (i.e., preexercise) and on 39 subjects for expression 4 h postexercise (i.e., 2 individuals did not undergo the postexercise biopsy: 1 from LLE and 1 from OH).

Table 3.

Nomenclature, gene information, and mRNA primer characteristics

Common Name	Gene Symbol	Accession No.	Sequence (5′→3′)	Amplicon Size, bp	mRNA Region, bp	Annealing Temperature, °C
Proinflammatory cytokines
IL-1β	IL1B	NM_000576.2	GGATATGGAGCAACAAGTGGTG CGCAGGACAGGTACAGATTCT	113	661–773	61
TNF-α	TNF	NM_000594.3	CCCAGGCAGTCAGATCATCTTCTCGAA CTGGTTATCTCTCAGCTCCACGCCATT	149	390–538	60
Anti-inflammatory cytokines
IL-10	IL10	NM_000572.2	GGCGCTGTCATCGATTTCTTCC GGCTTTGTAGATGCCTTTCTCTTG	101	430–530	60
IL-4	IL4	NM_000589.3b	TCTTCCTGCTAGCATGTGCC TGTTACGGTCAACTCGGTGC	128	100–227	60
IL-1Ra	IL1RN	NM_173842.2c	AGCTGGAGGCAGTTAACATCA ACTCAAAACTGGTGGTGGGG	102	375–476	60
Pleiotropic cytokines
IL-6	IL6	NM_000600.4	CTATGAACTCCTTCTCCACAAGCGCCTT GGGGCGGCTACATCTTTGGAATCTT	127	61–187	60
TGF-β	TGFB1	NM_000660.6	ACCAACTATTGCTTCAGCTCCA GAAGTTGGCATGGTAGCCCT	120	1,683–1,802	60
PGE2/COX pathway components
COX-1v1a	PTGS1	NM_000962	CCCAGGAGTACAGCTACGAGCAGTTCTT CCAGCAATCTGGCGAGAGAAGGCAT	101	1,327–1,427	60
COX-1v2a	PTGS1	NM_080591	GTCCAGTTCCAATACCGCAACCGCAT CCACCGATCTTGAAGGAGTCAGGCAT	92	1,237–1,328	60
COX-2	PTGS2	NM_000963.3	TTGCTGGCAGGGTTGCTGGTGGTA CATCTGCCTGCTCTGGTCAATGGAA	86	1,381–1,466	60
cPGES	PTGES3	NM_006601.6	AGGCCCGCCCACCAGTTCGC AGTCCCTTCGATCGTACCACTTTGCAG	82	254–335	60
mPGES-1	PTGES	NM_004878.4	CGGAAGAAGGCCTTTGCCAACC GGGTAGATGGTCTCCATGTCGTTCC	125	171–295	60
EP3	PTGER3	NM_198715.2d	CTGGTCTCCGCTCCTGATAA TTCAGTGAAGCCAGGCGAAC	132	1,113–1,244	60
EP4	PTGER4	NM_000958.2	GCTCGTGGTGCGAGTATTCGTCAACC TCCAGGGGTCTAGGATGGGGTTCA	122	1,453–1,574	60
Chemokines and macrophage surface markers
IL-8	CXCL8	NM_000584.3	GCTCTGTGTGAAGGTGCAGTTTTGCCAA GGCGCAGTGTGGTCCACTCTCAAT	135	153–287	60
MCP-1	CCL2	NM_002982.3	GCAATCAATGCCCCAGTCAC CTTGAAGATCACAGCTTCTTTGGG	123	152–274	60
CD16b	FCGR3B	NM_001271037.1	CCAGGCCTCGAGCTACTTCA TGCCAAACCGATATGGACTTCT	121	441–561	60
CD163	CD163	NM_004244.5e	CCCAGTGAGTTCAGCCTTTA TCAGCAGCAGTCTTAGGAATC	140	3,600–3,739	60

Top sequence reflects the forward primer, and bottom sequence reflects the reverse primer. CD16b, cluster of differentiation 16b [Fc fragment of IgG receptor IIIb (FCGR3B)]; COX-1, cyclooxygenase-1 [prostaglandin-endoperoxide synthase 1 (PTGS1)]; COX-2, cyclooxygenase-2 [prostaglandin-endoperoxide synthase 2 (PTGS2)]; cPGES, cytosolic prostaglandin E₂ synthase [prostaglandin E synthase 3 (PTGES3)]; EP3, prostaglandin E₂ receptor EP3 subtype [prostaglandin E receptor 3 (PTGER3)]; EP4, prostaglandin E₂ receptor EP4 subtype [prostaglandin E receptor 4 (PTGER4)]; IL-1Ra, interleukin-1 receptor antagonist (IL1RN); IL-8, interleukin-8 [C-X-C motif chemokine ligand 8 (CXCL8)]; MCP-1, monocyte chemoattractant protein-1 [C-C motif chemokine ligand 2 (CCL2)]; mPGES-1, microsomal prostaglandin E synthase 1 [prostaglandin E synthase (PTGES)]; TGF-β, transforming growth factor-β [transforming growth factor-β1 (TGFB1)].

^aOne primer was designed for each variant (v) of COX-1 based on our previous research (87);

^bPrimers detect variant 1 isoform 1 (NM_000589.3);

^cPrimers detect all variants: variant 1 isoform 1 (NM_173842.2), variant 2 isoform 2 (NM_173841.2), variant 3 isoform 3 (NM_000577.4), variant 4 isoform 4 (NM_173843.2), and variant 5 isoform 4 (NM_001318914.1);

^dPrimer detects variant 4 isoform 4 (NM_198714.1), variant 5 isoform 5 (NM_198715.2), variant 6 isoform 6 (NM_198716.1), variant 7 isoform 7 (NM_198717.1), variant 8 isoform 8 (NM_198718.1), variant 9 isoform 4 (NM_198719.1), and variant 11 isoform 4 (NM_001126044.1);

^ePrimer detects both variant 1 isoform a (NM_004244.5) and variant 2 isoform b (NM_203416.3).

Total RNA extraction and quality check.

Total RNA was extracted in TRI Reagent RT (Molecular Research Center, Cincinnati, OH). The quality and integrity (RNA integrity number = 8.34 ± 0.05, mean ± SE) of extracted RNA (94.24 ± 3.97 ng/µL) were evaluated using an RNA 6000 Nano LabChip kit on an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA) as previously described (29, 78).

Real-time polymerase chain reaction.

Oligo(dT) primed first-strand cDNA was synthesized (96–144 ng of total RNA, depending on magnitude of target gene expression) using SuperScript II Reverse Transcriptase (Invitrogen, Carlsbad, CA). For each target, quantification of mRNA levels was performed in duplicate in a 72-well Rotor-Gene Q Centrifugal Real-Time Cycler (Qiagen, Germantown, MD). Ribosomal protein lateral stalk subunit P0 (RPLP0) was selected as a housekeeping/reference gene, as previously done in human muscle (48). RPLP0 expression was similar among the three groups at baseline [threshold cycle (C_T), 19.02 ± 0.03] and stable after exercise (C_T, 19.01 ± 0.03). All primers used in this study were mRNA specific (on different exons and/or crossing over an intron) and designed for qPCR [Vector NTI Advance 9 software (Invitrogen) and Primer Design Tool (Entrez) National Center for Biotechnology Information Primer-Basic Local Alignment Search Tool (BLAST) program] using SYBR Green chemistry (29). Primers details are presented in Table 3. A melting-curve analysis was generated for all PCR runs to validate that only one product was present. For each run, a serial dilution curve was made using cDNA from a known amount (500–2,000 ng) of human skeletal muscle RNA (Ambion) or from human muscle samples collected in our laboratory. The amplification calculated by the Rotor-Gene software was specific and highly efficient (efficiency = 1.02 ± 0.01; R² = 0.98 ± 0.00; slope = 3.29 ± 0.03). Basal gene expression among YE, LLE, and OH was compared using the $2^{-\Delta {\text{C}}_{\text{T}}}$ (arbitrary units) method. Gene expression before and after the resistance exercise challenge was compared using the $2^{-\Delta \Delta {\text{C}}_{\text{T}}}$ (fold change) relative quantification method, as previously described (36, 37, 70, 81). On the basis of the principle of the calculation, it was determined that the preexercise value and the associated variability should be very close to 1 for each group. In the present study, this was true of all genes analyzed, and preexercise $2^{-\Delta \Delta {\text{C}}_{\text{T}}}$ values were not statistically different among the three groups or between subgroups (P > 0.05). Therefore, to simplify interpretation, preexercise expression for each gene is graphically represented as a dotted line at 1.0-fold.

Immunohistochemistry

For histochemical detection of skeletal muscle macrophages, transverse sections (7 µm) for histochemical analysis were cut on a microtome-cryostat (HM 525; Microm, Walldorf, Germany) at −20°C. Prior to staining, sections were air-dried in a humidified chamber for 30 min and then fixed in cold (−20°C) acetone for 10 min and rehydrated in PBS for 5 min. Endogenous peroxidase activity was quenched with 0.3% peroxide. Sections were incubated in anti-cluster of differentiation 68 (CD68) primary antibody (1:100 dilution in PBS, M0718; Dako, Carpinteria, CA) at 4°C overnight (15 h) in a humidified chamber. Sections were treated using HistoStain Kit (Invitrogen, Frederick, MD), visualized using aminoethyl carbazole (AEC) single solution substrate (Invitrogen), and then counterstained with hematoxylin (Gill no. 3; Sigma) for 30 s. All analyses included a CD68 negative control (no primary antibody during incubation) and an internal positive control [skeletal muscle biopsy obtained following a damaging exercise protocol similar to those shown to elicit macrophage infiltration (87)]. Positive and negative control slides were included in all analyses. A sample image from a lifelong exercise subject is shown in Fig. 1. Macrophage abundance is represented as CD68⁺ cells per 100 fibers [the number of CD68⁺ cells relative to the number of muscle fibers assessed (109 ± 4 fibers per subject), multiplied by a factor of 100] and CD68⁺ cell density [number of CD68⁺ cells in an analyzed area of muscle (645,271 ± 6,721 µm² per subject)]. All measurements were completed by two independent investigators and averaged to represent each sample.

Fig. 1.

Representative cross section from vastus lateralis skeletal muscle of a lifelong exerciser. Arrows indicate immunohistochemically stained cluster of differentiation 68-positive (CD68⁺) cells (red); hematoxylin-stained nuclei appear purple.

Statistical Analyses

Data were analyzed with a one-way analysis of variance (ANOVA) to compare subject characteristics, training histories, serum inflammatory factor levels, macrophage parameters, and basal gene expression ($2^{-\Delta {\text{C}}_{\text{T}}}$ method) among the three main groups (YE, LLE, and OH) and between LLE subgroups (LLE-F and LLE-P). A two-way ANOVA (group × time) was completed to evaluate gene expression ($2^{-\Delta \Delta {\text{C}}_{\text{T}}}$ method) in response to exercise among the three main groups and LLE subgroups. Follow-up one-way ANOVAs were used to compare basal gene expression and postexercise expression levels between YE and both old groups combined (LLE and OH). Post hoc comparisons were made with Tukey’s test. Significance was accepted at P ≤ 0.05. Data are presented as means ± SE.

RESULTS

Basal Circulating Inflammatory Factors

Serum concentrations of inflammatory factors are shown in Table 4. IL-6 levels were lower (P ≤ 0.05) in young exerciser (YE) and lifelong exerciser (LLE) men than in old healthy nonexerciser (OH) men. Serum TNF-α and C-reactive protein (CRP) levels were not different among groups (P > 0.05). Both LLE and OH had 43% lower (P ≤ 0.05) IGF-1 levels than YE. However, the Performance subgroup (LLE-P) had higher (P ≤ 0.05) IGF-1 levels (+23%) than the Fitness subgroup (LLE-F). IL-6, TNF-α, and CRP levels were not different (P > 0.05) between LLE-F and LLE-P.

Table 4.

Basal serum concentrations of inflammatory factors and intramuscular macrophage abundance

		Lifelong Exercisers
	YE	Combined	LLE-P	LLE-F	OH
Serum inflammatory factors
IL-6, pg/mL	0.9 ± 0.1	2.0 ± 0.2†	2.1 ± 0.2	1.8 ± 0.4	3.9 ± 1.2*
TNF-α, pg/mL	1.7 ± 0.2	1.7 ± 0.1	1.6 ± 0.1	1.9 ± 0.3	1.3 ± 0.2
CRP, mg/L	0.6 ± 0.2	0.8 ± 0.1	0.8 ± 0.1	0.9 ± 0.3	0.8 ± 0.1
IGF-1, ng/mL	204 ± 14	116 ± 5*	124 ± 6‡	101 ± 5	117 ± 12*
Intramuscular macrophage abundance
CD68+, cells per 100 fibers	7.8 ± 1.4	8.0 ± 0.7	8.3 ± 0.8	7.3 ± 1.4	6.8 ± 1.1
CD68+, cells/mm2	13.0 ± 2.4	12.8 ± 1.2	13.4 ± 1.5	11.2 ± 2.0	11.7 ± 2.0

Values are means ± SE. CD68, cluster of differentiation 68; CRP, C-reactive protein; LLE-F, lifelong exercisers-Fitness; LLE-P, lifelong exercisers-Performance; OH, old healthy nonexercisers; YE, young exercisers.

^*P ≤ 0.05 vs. YE;

^†P ≤ 0.05 vs. OH;

^‡P ≤ 0.05 LLE-P vs. LLE-F.

Basal Muscle Macrophage Abundance

Skeletal muscle macrophage parameters also appear in Table 4. CD68⁺ cells per 100 fibers and CD68⁺ cell density were similar (P > 0.05) across all three groups and between the LLE subgroups.

Basal Skeletal Muscle Inflammation

Basal muscle cytokine expression among YE, LLE, and OH is presented in Fig. 2A (proinflammatory) and Fig. 3A (anti-inflammatory). TNF-α expression tended (P ≤ 0.10) to be lower in OH than in YE (−55%) and was significantly lower (P ≤ 0.05) than in LLE (−62%). LLE also had higher expression (P ≤ 0.05) of anti-inflammatory IL-10 (+130% vs. YE) and transforming growth factor-β (TGF-β; +66% vs. OH). As shown in the LLE subgroup summary (Table 5), higher (P ≤ 0.05) gene expression in LLE-F than in LLE-P may have contributed to these findings in TNF-α (+59%) and TGF-β (+48%). No differences were found in expression of IL-6, IL-1β, IL-4, or interleukin-1 receptor antagonist (IL-1Ra).

Fig. 2.

Basal expression (A) and exercise-induced fold change in expression (B) of proinflammatory cytokines in vastus lateralis skeletal muscle homogenate of young exercisers (YE), lifelong exercisers (LLE), and old healthy nonexercisers (OH). The dashed line at 1.0-fold represents the preexercise fold change for each group, derived from the comparative threshold cycle ($2^{-\Delta \Delta {\text{C}}_{\text{T}}}$) calculation (see methods). AU, arbitrary units; TNF-α, tumor necrosis factor-α. ‡P ≤ 0.05 vs. OH, †P ≤ 0.10 vs. YE, **P ≤ 0.05 vs. preexercise.

Fig. 3.

Basal expression (A) and exercise-induced fold change in expression (B) of anti-inflammatory cytokines in vastus lateralis skeletal muscle homogenate of young exercisers (YE), lifelong exercisers (LLE), and old healthy nonexercisers (OH). The dashed line at 1.0-fold represents the preexercise fold change for each group, derived from the comparative threshold cycle ($2^{-\Delta \Delta {\text{C}}_{\text{T}}}$) calculation (see methods). AU, arbitrary units; IL-1Ra, interleukin-1 receptor antagonist; TGF-β, transforming growth factor-β. *P ≤ 0.05 vs. YE, ‡P ≤ 0.05 vs. OH, †P ≤ 0.10 vs. YE, **P ≤ 0.05 vs. preexercise.

Table 5.

Basal gene expression in LLE subgroups

	LLE-P	LLE-F
Proinflammatory cytokines
IL-6	0.16 ± 0.03	0.17 ± 0.04
TNF-α*	0.21 ± 0.03	0.33 ± 0.06
IL-1β	0.88 ± 0.14	0.66 ± 0.12
Anti-inflammatory cytokines
IL-10	0.24 ± 0.04	0.19 ± 0.04
IL-4	0.87 ± 0.17	0.59 ± 0.32
IL-1Ra	0.83 ± 0.16	1.46 ± 0.60
TGF-β*	60.3 ± 3.7	89.3 ± 16.2
PGE2/COX pathway components
COX-1v1	2.71 ± 0.22	2.90 ± 0.27
COX-1v2	4.89 ± 0.43	5.94 ± 0.60
COX-2†	0.22 ± 0.03	0.13 ± 0.03
cPGES	214 ± 17	225 ± 22
mPGES-1	3.97 ± 0.62	4.75 ± 0.76
EP3	3.36 ± 0.90	4.02 ± 0.83
EP4	9.54 ± 0.58	9.03 ± 0.51
Chemokines and macrophage markers
IL-8	0.22 ± 0.04	0.23 ± 0.05
MCP-1	51.3 ± 5.2	53.5 ± 14.7
CD16b	2.56 ± 0.29	3.02 ± 1.28
CD163	54.1 ± 7.0	41.3 ± 5.9

Values are group means ± SE and are in arbitrary units. LLE, lifelong exercisers; LLE-F, lifelong exercisers-Fitness; LLE-P, lifelong exercisers-Performance. CD, cluster of differentiation; COX-1v1 and COX-1v2, cyclooxygenase-1 variants 1 and 2, respectively; COX-2, cyclooxygenase-2; cPGES, cytosolic prostaglandin E₂ synthase; EP3 and EP4, prostaglandin E₂ receptor EP3 and EP4 subtypes, respectively; IL-1Ra, interleukin-1 receptor antagonist; MCP-1, monocyte chemoattractant protein-1; mPGES-1, microsomal prostaglandin E synthase 1; TGF-β, transforming growth factor-β.

^*P ≤ 0.05 between groups;

^†P ≤ 0.10 between groups.

Within the PGE₂/cyclooxygenase (COX) pathway, expression of both COX-1 variant 2 (COX-1v2; Fig. 4A, +37%) and microsomal prostaglandin E synthase 1 (mPGES-1; Fig. 5A, +69%) was higher (P ≤ 0.05) in the older cohorts (LLE and OH combined) than in YE. The downstream anti-inflammatory receptor prostaglandin E₂ receptor EP4 subtype (EP4) was differentially expressed across the groups (Fig. 5A), with higher (P ≤ 0.05) expression in LLE (+21%) and lower (P ≤ 0.05) expression in OH (−37%) compared with YE. COX-1v1, COX-2, cytosolic prostaglandin E2 synthase (cPGES), and EP3 expression was not different across the groups (P > 0.05). The LLE-P subgroup showed a trend (P ≤ 0.10) for higher expression of COX-2 (+69% vs. LLE-F), although no differences (P > 0.05) were found for COX-1, cPGES, mPGES-1, EP3, or EP4 (Table 5).

Fig. 4.

Basal expression (A) and exercise-induced fold change in expression (B) of cyclooxygenase (COX) enzymes in vastus lateralis skeletal muscle homogenate of young exercisers (YE), lifelong exercisers (LLE), and old healthy nonexercisers (OH). The dashed line at 1.0-fold represents the preexercise fold change for each group, derived from the comparative threshold cycle ($2^{-\Delta \Delta {\text{C}}_{\text{T}}}$) calculation (see methods). AU, arbitrary units; v1, variant 1; v2, variant 2. *P ≤ 0.05 vs. YE, **P ≤ 0.05 vs. preexercise.

Fig. 5.

Basal expression (A) and exercise-induced fold change in expression (B) of PGE₂/cyclooxygenase (COX) pathway components in vastus lateralis skeletal muscle homogenate of young exercisers (YE), lifelong exercisers (LLE), and old healthy nonexercisers (OH). The dashed line at 1.0-fold represents the preexercise fold change for each group, derived from the comparative threshold cycle ($2^{-\Delta \Delta {\text{C}}_{\text{T}}}$) calculation (see methods). AU, arbitrary units; cPGES, cytosolic prostaglandin E₂ synthase; EP3 and EP4, prostaglandin E₂ receptor EP3 and EP4 subtypes, respectively; mPGES-1, microsomal prostaglandin E synthase 1. *P ≤ 0.05 vs. YE, ‡P ≤ 0.05 vs. OH, †P ≤ 0.10 vs. YE, **P ≤ 0.05 vs. preexercise.

Basal expression of chemokines IL-8 and monocyte chemoattractant protein-1 (MCP-1) was similar (P > 0.05) among the three groups (Fig. 6A) and between the LLE subgroups (Table 5). LLE (+98%) and OH (+125%) each had significantly higher (P ≤ 0.05) expression of CD163 than YE (Fig. 6A), and there was no difference between LLE subgroups (Table 5). No differences (P > 0.05) were found in expression of CD16b (Fig. 6A and Table 5).

Fig. 6.

Basal expression (A) and exercise-induced fold change in expression (B) of chemokines and macrophage surface markers in vastus lateralis skeletal muscle homogenate of young exercisers (YE), lifelong exercisers (LLE), and old healthy nonexercisers (OH). The dashed line at 1.0-fold represents the preexercise fold change for each group, derived from the comparative threshold cycle ($2^{-\Delta \Delta {\text{C}}_{\text{T}}}$) calculation (see methods). AU, arbitrary units; CD, cluster of differentiation; MCP-1, monocyte chemoattractant protein 1. *P ≤ 0.05 vs. YE, †P ≤ 0.10 vs. YE and LLE, **P ≤ 0.05 vs. preexercise, §P ≤ 0.10 vs. preexercise.

Effects of Acute Resistance Exercise

After resistance exercise, expression of TNF-α was significantly elevated (P ≤ 0.05) only in the OH group (2.1-fold; Fig. 2B). There was a trend for a larger increase in IL-6 expression in the older groups (P ≤ 0.10), likely explained by a 1.4-fold increase in LLE. IL-1β expression was unaffected by exercise across the groups (P > 0.05). For the anti-inflammatory genes (Fig. 3B), IL-10 expression tended to be higher (P ≤ 0.10) after exercise in YE (2.0-fold) compared with LLE. Exercise led to a 1.7-fold increase (P ≤ 0.05) in TGF-β expression in OH, and TGF-β expression in LLE was significantly lower than in OH postexercise (1.2-fold). IL-1Ra expression was unchanged (P > 0.05) after exercise. IL-4 expression was also unchanged (P > 0.05) across the three groups, although the LLE subgroups (Table 6) showed a 2.5-fold change after exercise (P ≤ 0.05, main effect). No differences (P > 0.05) were found between the LLE subgroups for expression of any cytokines postexercise (Table 6).

Table 6.

Change in gene expression in LLE subgroups

	Time Point	LLE-P	LLE-F
Proinflammatory cytokines
IL-6	Pre	1.24 ± 0.21	1.25 ± 0.32
	Post	1.55 ± 0.28	1.23 ± 0.23
TNF-α	Pre	1.08 ± 0.12	1.12 ± 0.19
	Post	1.25 ± 0.14	0.94 ± 0.11
IL-1β	Pre	1.20 ± 0.21	1.10 ± 0.21
	Post	0.95 ± 0.13	1.23 ± 0.26
Anti-inflammatory cytokines
IL-10	Pre	1.22 ± 0.24	1.10 ± 0.21
	Post	0.88 ± 0.13	1.23 ± 0.25
IL-4	Pre	1.31 ± 0.28	1.84 ± 1.01
	Post*	2.55 ± 0.62	2.45 ± 0.88
IL-1Ra	Pre	1.23 ± 0.25	1.37 ± 0.57
	Post	1.22 ± 0.12	0.76 ± 0.15
TGF-β	Pre	1.03 ± 0.07	1.09 ± 0.20
	Post	1.35 ± 0.16	1.00 ± 0.12
PGE2/COX pathway components
COX-1v1	Pre	1.05 ± 0.90	1.03 ± 0.90
	Post	0.94 ± 0.09	0.82 ± 0.11
COX-1v2	Pre	1.05 ± 0.11	1.03 ± 0.10
	Post†	0.91 ± 0.10	0.78 ± 0.11
COX-2	Pre	1.20 ± 0.20	1.21 ± 0.26
	Post*	1.58 ± 0.22	2.15 ± 0.28
cPGES	Pre	1.04 ± 0.08	1.03 ± 0.10
	Post	1.17 ± 0.06	1.08 ± 0.10
mPGES-1	Pre	1.21 ± 0.23	1.09 ± 0.18
	Post*	0.70 ± 0.11	0.80 ± 0.12
EP3	Pre	1.52 ± 0.51	1.63 ± 0.34
	Post	1.75 ± 0.22	1.62 ± 0.16
EP4	Pre	1.02 ± 0.06	1.01 ± 0.06
	Post	1.11 ± 0.08	1.10 ± 0.11
Chemokines and macrophage markers
IL-8	Pre	1.28 ± 0.28	1.20 ± 0.27
	Post†	1.18 ± 0.18	0.53 ± 0.11
MCP-1	Pre	1.07 ± 0.12	1.17 ± 0.32
	Post	1.30 ± 0.13	1.32 ± 0.20
CD16b	Pre	1.09 ± 0.13	1.48 ± 0.63
	Post†	1.68 ± 0.23	1.64 ± 0.48
CD163	Pre	1.08 ± 0.14	1.05 ± 0.15
	Post*	1.28 ± 0.21	1.57 ± 0.26

Values are means ± SE and represent fold change from preexercise. LLE, lifelong exercisers; LLE-F, lifelong exercisers-Fitness; LLE-P, lifelong exercisers-Performance; Post, postexercise; Pre, preexercise. CD, cluster of differentiation; COX-1v1 and COX-1v2, cyclooxygenase-1 variants 1 and 2, respectively; COX-2, cyclooxygenase-2; cPGES, cytosolic prostaglandin E₂ synthase; EP3 and EP4, prostaglandin E₂ receptor EP3 and EP4 subtypes, respectively; IL-1Ra, interleukin-1 receptor antagonist; MCP-1, monocyte chemoattractant protein-1; mPGES-1, microsomal prostaglandin E synthase 1; TGF-β, transforming growth factor-β.

^*P ≤ 0.05 vs. preexercise;

^†P ≤ 0.10 vs. preexercise.

Expression of COX-1v1 and COX-1v2 was unchanged with exercise in YE, LLE, and OH (Fig. 4B) and in both LLE subgroups (Table 6). There was a trend (P ≤ 0.10, main effect) for the LLE subgroups to have decreased expression of COX-1v2 following exercise (Table 6). Conversely, exercise increased (P ≤ 0.05, main effect) expression of COX-2 in the three groups and the LLE subgroups (~1.9-fold).

Downstream PGE₂/COX pathway components were differentially responsive to exercise across the groups (Fig. 5B). cPGES expression tended to be higher (P ≤ 0.10) after exercise in the aging cohorts combined (1.1-fold) than in YE. LLE had lower (P ≤ 0.05) expression of mPGES-1 (0.8-fold) compared with YE (1.5-fold) following exercise, which was supported by an overall decrease (P ≤ 0.05, main effect) in the LLE subgroups (Table 6). Exercise led to a 3.4-fold increase (P ≤ 0.05) in EP3 expression in YE, and LLE had lower (P ≤ 0.05) EP3 expression than YE postexercise (1.5-fold). EP4 expression increased after exercise for all three groups (P ≤ 0.05, main effect). Expression of cPGES, EP3, and EP4 was not different (P > 0.05) between the subgroups (Table 6).

Postexercise skeletal muscle IL-8 expression (Fig. 6B) tended to be higher (P ≤ 0.10) in OH (2.2-fold) than in YE (0.9-fold) or LLE (1.0-fold). With the LLE subgroups, IL-8 expression tended (P ≤ 0.10, main effect) to decrease overall, primarily driven by a 0.5-fold change in the LLE-F subgroup (Table 6). MCP-1 expression tended to increase overall after exercise (P ≤ 0.10, main effect), but YE showed a significantly (P ≤ 0.05) greater response (1.8-fold increase) than the older groups (1.3-fold increase). Muscle macrophage surface markers CD16b and CD163 (Fig. 6B and Table 6) both approached (P ≤ 0.10, main effect) or attained (P ≤ 0.05, main effect) significantly higher expression after exercise across all groups.

DISCUSSION

This study examined the influences of aging and lifelong exercise on inflammation in circulation and skeletal muscle at baseline and after acute resistance exercise. This investigation arose given the negative impact of chronic basal inflammation on muscle size and function in older individuals (66, 67, 85), along with the established anti-inflammatory benefits of exercise training (4, 42, 43). Findings from this study show that aging led to a proinflammatory profile within the blood and muscle. Lifelong exercise partially protected against this effect and favored a generally anti-inflammatory profile within muscle. A resistance exercise bout was chosen to provide a potent antisarcopenic stimulus and present an unaccustomed exercise challenge to all groups. Our laboratory has historically been interested in resistance training as a tool to combat sarcopenia in aging adults. The present study builds on this tenet, along with previous work from us (78, 80, 81) and others (18, 38) showing that skeletal muscle inflammation may preclude optimal adaptations. We provide evidence that highly aerobically trained older adults display a preserved response to exercise, which may indicate they are better prepared to adapt to resistance training. Further research is needed to understand whether this advantage persists into the ninth decade of life, where exercise adaptations appear to be blunted (23, 61, 68).

Basal Inflammatory Profile

The impressive and unique training history of the lifelong exercisers (LLE) resulted in an anti-inflammatory muscle environment, which complements findings of higher levels of anti-inflammatory factors in circulation of older exercise-trained individuals (28, 43). Relative to OH, higher basal levels of IL-10 (+43%) and TGF-β (+66%) in LLE likely contribute to suppression of the transcription and signaling activity of proinflammatory factors. For example, IL-10 and TGF-β have regulatory relationships with the proinflammatory cytokine TNF-α (7, 15), which may explain the tendency of these cytokines to track together in the present study. Both YE and LLE had higher basal expression of TNF-α than OH, suggesting that exercise training increases its expression. Other studies have also demonstrated higher TNF-α expression in trained compared with untrained individuals, including circulation of older men (41) and skeletal muscle of young men (51). Likewise, short-term (12 wk) resistance training in older adults leads to elevated expression of muscle TNF-α, along with a number of other cytokines (81). The typically proteolytic effects of TNF-α are likely moderated in trained individuals as a result of the overall anti-inflammatory profile.

Despite lower expression of muscle TNF-α, OH had circulating TNF-α levels similar to the other groups. IL-6 was similarly expressed in muscle across the three groups but highest in the circulation of OH. Other studies have also shown an apparent disconnect between muscle and circulating inflammation (44, 72). Thus, assuming that muscle cytokine mRNA is translated and released into circulation similarly across the groups, a source other than muscle [e.g., a differential circulating immune cell population (19) or proinflammatory effects of greater adipose tissue mass (47) in OH] likely contributed to the observed patterns in circulating IL-6 and TNF-α levels. Nevertheless, the deleterious effects of inflammation on skeletal muscle were apparent in a negative association between circulating IL-6 levels and quadriceps muscle cross-sectional area across the groups (Fig. 7). Large-scale studies (66, 67, 85) have often reported similar relationships, positing that sarcopenia may be a long-term consequence of the negative impact of IL-6 on muscle protein metabolism (9, 25, 84). Lifelong exercise apparently exerts a positive influence on this trend, preventing the age-related increase in circulating IL-6 levels and thereby partially attenuating the decrease in whole muscle mass (13). Overall reductions in muscle mass in both older groups might also be related to the age-related decrease in circulating IGF-1. This factor is known to promote anabolism and reduce cytokine production in skeletal muscle (34). Despite a minimally protective effect of higher lifelong training intensity in LLE-P compared with LLE-F, lifelong aerobic exercise was not able to rescue the decline of IGF-1 levels.

Fig. 7.

Association between serum interleukin-6 (IL-6) and quadriceps muscle cross-sectional area (CSA) in young exerciser (YE), lifelong exerciser (LLE), and old healthy nonexerciser (OH) men. This finding is in agreement with the work of others suggesting a negative impact of circulating inflammation on skeletal muscle (41, 66, 67, 85).

Aging may also be accompanied by an increased capacity for production of PGE₂ in skeletal muscle (35). Produced in the COX pathway, this lipid-based inflammatory mediator can promote protein breakdown (63) and activate proinflammatory signaling within muscle (70). Fittingly, long-term use of COX-inhibiting drugs may aid in combatting sarcopenia (32) and skeletal muscle dysfunction during inflammatory conditions (6, 39) in older adults. In the present study, there was an overall effect of aging (LLE and OH combined) in increasing the expression of several components of the PGE₂/COX pathway. Most apparently, aging increased expression of COX-1 and the PGE₂-specific synthase mPGES-1. EP3, a proinflammatory receptor for PGE₂ (22, 46), followed a similar though nonsignificant pattern (+74% in OH vs. YE). Comparable to previous findings at the protein level in sarcopenic muscle (35), the OH group had lower expression of EP4, a purportedly anti-inflammatory PGE₂ receptor. Existing data in human tissues other than skeletal muscle have demonstrated that PGE₂ signaling through EP4 may enhance IL-10 activity (14), reduce chemokine production (74), and inhibit maturation of IL-1β (69). Therefore, age-related increases in proinflammatory flux through the PGE₂/COX pathway could have consequences for regulation of muscle mass (32), protein turnover, and exercise adaptations (78–80).

Despite an apparent age-related increase in PGE₂/COX pathway expression, LLE had higher expression of EP4. The effect of training on increasing expression of this receptor is supported in the literature (81). Given its anti-inflammatory roles, higher EP4 expression in LLE muscle could negatively regulate proinflammatory activity at rest and during periods of heightened PGE₂ availability (e.g., exercise). High basal expression of EP4, along with several anti-inflammatory cytokines, generally supports the idea that lifelong exercise fosters an anti-inflammatory profile, potentially as a positive adaptation to long-term aerobic training.

Subdivision of the LLE group provided insight into the effects of lifetime training intensity on basal inflammation. Within LLE, higher-intensity training in LLE-P reduced basal expression of TNF-α and TGF-β, further supporting the idea that these likely modulate one another’s activity. Higher training intensity also led to upregulation of COX-2. Although not typically measurable at the protein level in healthy human muscle (35, 87), COX-2 expression can be induced during challenging inflammatory conditions (57). Thus, higher expression of COX-2 may indicate an adaptation to intense muscular exercise for many decades in the LLE-P group. Likewise, short-term training has been shown to lead to increased basal muscle COX-2 expression in older adults (78). Despite this potential for heightened COX pathway activity in LLE-P, any increased PGE₂ production would likely result in a downstream anti-inflammatory response due to high EP4 expression in both LLE subgroups.

To provide insight into the capacity of muscle for intercellular signaling with inflammatory cells, basal macrophage abundance and gene expression of muscle chemokines and macrophage surface markers were assessed. Heightened basal expression of chemokines or elevated macrophage abundance in skeletal muscle could indicate the presence of unresolved inflammation (73). Interestingly, a general nonsignificant pattern for reduction in muscle chemokine expression was observed in OH (IL-8, −53%, and MCP-1, −24%, vs. YE), with LLE partially mitigating these effects. Both older groups had higher basal expression of anti-inflammatory (M2) macrophage surface marker CD163. However, because no differences in intramuscular macrophage abundance were detected among the groups, higher CD163 receptor density for a given number of resident macrophages might indicate heightened capacity for CD163-mediated signaling. Given its established roles in anti-inflammatory signal transduction and cytokine production (10, 53), CD163 may partially contribute to the overall anti-inflammatory profile of the LLE muscle. However, this relationship was not seen in OH, which may suggest the presence of regulatory defects between muscle and inflammatory cells (56). Such dysfunctions could contribute to impaired ability to resolve exercise-induced inflammation and create a resting environment that favors muscle atrophy in OH.

Response to Acute Resistance Exercise

Recent evidence suggests that failure to resolve inflammation after exercise may partially limit adaptability to short-term training in older adults (38, 78, 79). We sought insight into this phenomenon by examining the response to an exercise bout previously shown to lead to adaptations when completed chronically (75, 78). The present investigation found an exaggerated response for TNF-α and TGF-β in OH muscle. TNF-α may engage proteolytic pathways within the muscle (e.g., NF-κB and MAPK; 30), and the pleiotropic nature of TGF-β may contribute to chemotaxis of inflammatory cells or aid in tissue repair following mechanical stress (26, 86). Thus, it appears that the bout of muscle loading presented a challenge to the OH muscle, whereas both YE and LLE had adapted through exercise training to better tolerate the exercise stress.

Furthermore, increased expression of cPGES in OH, coupled with higher basal expression of COX pathway components, could enable greater PGE₂ production following exercise. Previous work has demonstrated that PGE₂ leads to transcription of inflammatory and proteolytic factors within muscle (70). Thus, in combination with elevated postexercise expression of muscle cytokines, elevated PGE₂ production capacity within OH muscle could lead to heightened activity of and cross talk between inflammatory signaling pathways. Failure to resolve this might contribute to a sustained proinflammatory environment after exercise. This may partially explain why COX-inhibiting drugs have successfully modulated inflammation and enhanced muscle growth in older adults undergoing resistance training (79, 80).

Production of skeletal muscle chemokines IL-8 and MCP-1 is often increased after exercise to recruit inflammatory cells to the site of insult (16). Given the role of IL-8 in signaling with inflammatory cells (2), the exercise-induced increase in IL-8 levels seen in only OH further suggests that the untrained aging muscle was less prepared than muscle in the trained groups to tolerate the exercise stress. This may be problematic, since other investigations have demonstrated that aging may impair the responsiveness of inflammatory cells after exercise (11, 56). The present study did not demonstrate a difference in postexercise macrophage surface marker gene expression across groups. However, both older groups had lower expression of MCP-1 compared with YE after exercise. Thus, although aging may preserve the ability of resident macrophages to respond to exercise, there may be an age-related disconnect or delay in the ability to recruit more macrophages to aid in resolution of exercise-induced inflammation within muscle tissue. In OH, impairments in intercellular communication might lead to a sustained cytokine environment and contribute to longer-duration impairments in protein balance following exercise.

Only LLE showed a modest elevation (1.4-fold) in IL-6 after exercise. The fact that no change in IL-6 expression was seen in the other groups is not an uncommon finding (38, 83). However, our laboratory has previously demonstrated a dramatic increase in muscle IL-6 expression (791-fold) 4 h following an exercise bout identical to the present study (37). These differences may be due to the familiarity of the stimulus to the resistance-trained subjects in the previous study (37). In support of this, muscle samples from this previous time course investigation were reanalyzed using the present qPCR conditions and confirmed the different findings between these studies. IL-6 is often highly upregulated following endurance exercise (≥1 h; 49, 71) because of its important roles in intracellular and intercellular inflammatory signaling and glucose metabolism (27). Thus, the apparently higher sensitivity of LLE muscle to transcribe IL-6 after exercise may be a product of their decades of aerobic training and/or the pleiotropic nature of IL-6. A better understanding of the capacity of highly trained muscle to mount an inflammatory response might be gained by examining time points outside of the present 4-h window or by imposing a different exercise stimulus (e.g., longer duration or more familiar activity). This could also provide insight into the effect of lifelong training intensity, as the present investigation did not demonstrate any effect of training intensity on IL-6 or any other genes measured following acute exercise.

Some responses in LLE suggest that a heightened threshold may have been established as an adaptation to repeated exposure to the stress of exercise. For example, IL-10, EP3, and mPGES-1 expression was lower after exercise in LLE than in YE, indicating a blunted response. The potent effects of IL-10 in reducing proinflammatory cytokine transcription and signaling (8, 45) are implicated in resolution of acute inflammation. Although no data exist on the typical response to exercise for EP3 and mPGES-1, these proinflammatory PGE₂/COX pathway components may be involved in PGE₂ signaling after exercise to aid in muscle protein turnover necessary for remodeling (54, 82). Further research is necessary to clarify whether differences in postexercise gene expression contribute to differences in responsiveness to a resistance exercise regimen between young and lifelong exercisers.

Summary

This study supports the recent evidence that exercise training is anti-inflammatory. Although aging contributes to the elevation of proinflammatory factors in blood and muscle, lifelong aerobic exercise training partially reduces these effects and promotes an overall anti-inflammatory profile. Lifelong training intensity appears to have a minimal effect on this pattern. Future studies may expand on these findings in a muscle fiber type-specific manner, given that slow- and fast-twitch muscle fibers may have differential inflammatory profiles in healthy young muscle (35, 55, 80). Aging also results in an altered inflammatory response to acute exercise, which may have implications for the ability to increase muscle mass and handle a loading stress. However, this effect is largely rescued by lifelong exercise, with no additional influence of lifelong training intensity. Thus, further investigation into whether lifelong aerobic exercise improves skeletal muscle adaptability to resistance training (i.e., size and strength gains) would provide considerable insight. Additionally, further work is needed to establish whether LLE-induced patterns in the inflammatory profile of older skeletal muscle at baseline and following exercise mirror differences between young exercisers and young untrained individuals. Combined with previous studies on basal inflammation in individuals with a shorter training history (40, 41), this would provide further insight into whether an individual’s age, training status, and/or duration of training interact to provide the anti-inflammatory benefits observed here. The results of the present study help to understand the long-term benefits of exercise for avoidance of a chronic inflammatory state that may contribute to poor health and functional decline in aging adults.

GRANTS

This research was supported by National Institutes of Health Grant AG-038576.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

K.M.L., R.K.P., B.J., U.R., S.W.T., and T.A.T. conceived and designed research; K.M.L., R.K.P., B.J., U.R., S.W.T., and T.A.T. performed experiments; K.M.L., R.K.P., B.J., and T.A.T. analyzed data; K.M.L., R.K.P., B.J., U.R., S.W.T., and T.A.T. interpreted results of experiments; K.M.L., R.K.P., and T.A.T. prepared figures; K.M.L., R.K.P., and T.A.T. drafted manuscript; K.M.L., R.K.P., B.J., U.R., S.W.T., and T.A.T. edited and revised manuscript; K.M.L., R.K.P., B.J., U.R., S.W.T., and T.A.T. approved final version of manuscript.