All animal studies described were performed according to the official regulations effective in the Canton of Basel-City, Switzerland. CMKLR1 knockout mice were backcrossed to C57BL/6 mice from Charles River Laboratories for >10 generations. C57BL/6 mice and CMKLR1 KO mice on the same C57BL/6 background were bred at Charles River Laboratories. Male animals were used for all studies. 8, 12 and over 15 weeks old mice were denoted as “young”, “adult” and “full-grown”, respectively. All animals were acclimatized to the research facility in Basel (Switzerland) for 7 days. They were housed at 25 °C with a 12:12 h light-dark cycle and were fed a standard laboratory diet containing 18.2% protein and 3.0% fat with an energy content of 15.8 MJ/kg (Nafag, product # 3890, Kliba, Basel, Switzerland). Food and water were provided ad libitum.

For studies with the CMKLR1 antagonist 2-(anaphthoyl) ethyltrimethylammonium iodide (α-NETA), full-grown C57BL/6 mice were subcutaneously injected with 10 mg/kg α-NETA formulated in 10% sulfobutyl ether β-cyclodextrin. Muscle samples were harvested 24 h after the injection.

Evoked isometric force was measured non-invasively via electrical stimulation of the hind leg through transcutaneous electrodes as previously described (Summermatter et al., 2017). In brief, the animals were anesthetized, the foot was placed on a homemade pedal connected to a force transducer, and the force generated in response to electrical stimulation through the skin was recorded. Various frequencies ranging from 10 to 120 Hz were sequentially applied, and tetanic forces recorded at each frequency.

Animals were placed on a horizontal rotatable metal grid approximately 30 cm over a cushion. The grid was flipped upside down, and the time taken until the animals fell onto the cushion recorded.

Gait parameters of freely moving mice were measured using the Catwalk gait analysis system (Noldus Information Technology, Netherlands) as described previously (Parvathy and Masocha, 2013). Briefly, each mouse was placed individually in the CatWalk walkway and allowed to freely walk along an illuminated walkway glass plate. Animal contact with the glass plate resulted in changes in light intensity, which was captured by a camera positioned under the walkway. Mice were habituated to the Catwalk walkway prior to the start of the study. Gait parameters at young and adult phase were measured. Three runs for each animal were recorded per time point for further analysis by the software (CatWalk XT 10.6, Noldus Information Technology, Netherlands). The CatWalk calculates different parameters based on footprints, or on time recorded.

Total RNA was extracted from skeletal muscle using TRIzol reagent (Invitrogen). Reverse transcription was performed with random hexamers on 1 μg of total RNA using a high-capacity reverse transcription kit (Applied Biosystems), and the reaction mixture was diluted 20-fold. RT-PCRs were performed in duplicates in 384-well plates on an AB7900HT cycler (Applied Biosystems) using specific TaqMan probes (Applied Biosystems). Data were normalized to two housekeeping genes using the ΔΔCT threshold cycle (CT) method. Fluorescence was measured at the end of each cycle, and after 40 reaction cycles, a profile of fluorescence versus cycle number was obtained. Automatic settings were used in most cases to determine the CT. The comparative method using 2^−ΔΔCT was applied to determine the relative expression. All results are expressed as fold changes over controls.

Tibialis Anterior muscles were harvested and processed for paraffin embedding. For histology, muscles were cut into 10 μm cross-sections. The immunostaining was performed in the automate Ultra Discovery according to the manufacturer’s recommendations (Roche Diagnostics, Rotkreuz, Switzerland). Briefly, sections were de-waxed and antigen retrieval was obtained after incubation in the cell conditioner 1 (Roche) for 12 min at 95°C. After blocking in 6% normal goat serum and washing in the reaction buffer, slides were incubated with the primary antibody solution against Pax7 (ab187339, Abcam) for 1 h at room temperature to detect satellite cells. After a post-fixation in 0.05% glutaraldehyde for 8 min, antibody detection was performed using an anti-rabbit-HQ (Roche) incubated for 12 min, followed by an anti-HQ-HRP (Roche) incubated for 12 min. After several washes, the sections were counterstained with Hematoxylin II and Post Counterstained with Bluing reagents (Roche). After washing and dehydration steps samples were mounted with Pertex.

Glass slides were scanned with a Philips Ultra-Fast Scanner (Philips AG, Horgen, Switzerland) at 400x magnification resulting in image pixel dimensions of 0.25 µm × 0.25 µm. For the quantitative assessment of Pax7 positive nuclei, the HALO image analysis platform (Indica Labs, Albuquerque, NM, United States) was used. The image analysis was performed in two steps: first detecting the regions of interest (ROIs) in the tissue (excluding areas that may result in artifacts, e.g., tissue folds) and subsequently detecting nuclei positive and negative for PAX7. For the ROI detection a machine learning based classifier (DenseNet, HALO AI) has been devised and trained to recognize three different classes: muscle tissue, exclusion areas and whitespace. In a second step, nuclei positive and negative for PAX7 have been detected in the previously defined region of muscle tissue using the HALO module Multiplex IHC v3.0.1.

Statistical analyses were performed as indicated using GraphPad Prism version 7.03 (GraphPad Software, Inc., La Jolla, CA). Differences were considered significant when the probability value was <0.05.

Chemerin is a secreted protein that induces mitochondrial autophagy and insulin resistance in skeletal muscle (Sell et al., 2009; Becker et al., 2010; Xie et al., 2015). Since CMKLR1 is the most expressed chemerin receptor in muscle (https://www.ebi.ac.uk/gxa/home), we evaluated whether blocking CMKLR1 has an impact on skeletal muscle function. We found that young (8-weeks-old) CMKLR1-deficient mice were able to generate approximately 24% higher isometric force than their aged-matched controls (Figure 1A). This adaptation was observed exclusively following electrical stimulation of the hind limb muscles at high frequencies, which reflects maximal strength. By contrast, no differences in force were detected at low frequency stimulations.

As the mice grew older, CMKLR1 knockout mice retained their elevated maximal strength, while control animals continuously caught up. In fact, the difference diminished by a factor of two during adulthood (12 weeks of age) (Figure 1B) and disappeared completely once the animals were full-grown (after 15 weeks of age) (Figure 1C). The observed disparity in maximal strength was not due to altered body weight, which was indistinguishable between CMKLR1 knockout and control mice at any age (Supplementary Table S1). These data indicate that blockade of the chemerin receptor CMKLR1 accelerates the gain in maximal isometric strength during physiological muscle growth.

We next evaluated whether the absence of CMKLR1 modulates other parameters relevant for physical activity. To this end, we assessed strength endurance by using an inverted grip strength test. Intriguingly, inhibition of chemerin signaling reduced the latency to when an animal fell by 89, 90 and 74% in young, adult, and full-grown mice, respectively (Figures 1D,E). Abrogating CMKLR1 signaling thus lastingly impairs strength endurance performance.

The ability to generate and maintain force can significantly affect gait patterns. Since both strength parameters were altered concomitantly in young and adult mice, we assessed gait parameters in these two groups. In young animals lacking a functional chemerin receptor, the print area of the left and right hind legs was markedly larger compared to controls, reaching levels comparable to adult control and CMKLR1 knockout mice (Figure 2A). In line with the larger print area, the maximum contact area, which is the area of the foot touching the ground during the time of maximum contact, was also larger in young CMKLR1 knockout mice as compared to control animals (Figure 2B). We were unable to observe any significant differences between adult control and CMKLR1 knockout mice for any of the parameters tested. Furthermore, no significant differences were detected for the front legs (Figures 2C,D), or for any other gait-related key parameter (Supplementary Table S2). The observed changes in gait parameters demonstrate that young mice deficient in CMKLR1 adopt a walking pattern that resembles adult control mice.

While the changes in maximal strength and gait occur specifically during the growth phase, the reduction in strength endurance extends beyond the growth phase. We therefore sought to understand this prolonged re-programming induced by blockade of CMKLR1. Satellite cells play a central role in controlling skeletal muscle physiology, and depletion of resident muscle stem cells negatively impacts the volume of physical activity (Englund et al., 2020). We thus performed molecular and histological analyses on skeletal muscle of full-grown animals to evaluate the contribution of satellite cells to the observed phenotype. Gastrocnemius muscle of mice deficient in CMKLR1 showed a pattern of myogenesis markers that was distinct from controls: significantly reduced CD34 (Figure 3A), unaltered MyoD and Myf5 (Figures 3B,C), and significantly increased myogenin mRNA levels (Figure 3D). Since CD34 is a marker of satellite cell quiescence, and its expression is reduced upon satellite cell activation (Beauchamp et al., 2000; Zammit et al., 2006), these data suggest that loss of CMKLR1 promotes the exit of satellite cells from quiescence. In accordance with this, early markers of myogenic commitment and differentiation such as Myf5 tended to be increased (Figure 3C), while myogenin as a late-stage marker for muscle differentiation (Zammit et al., 2006) was significantly increased (Figure 3D). This transcriptional profile pointed towards activation and subsequent myogenic commitment of satellite cells in the absence of an intact signaling via CMKLR1. Moreover, when we analyzed the number of Pax7 positive satellite cells with a machine-learning-based approach (Figures 3E,F), we found that CMKLR1 knockout mice displayed a significantly lower number of these cells (Figure 3G). Collectively, these data underpin that knockout of CMKLR1 drives satellite cells from quiescence into myogenesis.

To gain further insights into the adaptive molecular mechanisms that persist beyond the growth phase, we measured the mRNA expression of key components regulating muscle contraction. Inhibition of CMKLR1 markedly enhanced the expression of genes involved in calcium regulation in full-grown animals. In gastrocnemius muscle, blockade of CMKLR1 increased the mRNA expression of the ryanodine receptor, indicating a higher capacity to release calcium from the sarcoplasmic reticulum to initiate muscle contraction, and of SERCA1, which is responsible for subsequent calcium reuptake (Figures 4A,B). In addition, the expression of genes, such as calsequestrin 1 and parvalbumin, involved in fast muscle relaxation, which is characteristic of a resistance-trained, but fatigue-prone muscle, were significantly elevated in the absence of functional CMKLR1 (Figures 4C,D). In contrast, the other isoforms of these calcium-handling genes did not significantly differ between the two genotypes (Figures 4E,F). Hence, the targeted transcriptional profiling of skeletal muscle of CMKLR1 knockout mice further corroborates a role for CMKLR1 in regulating transcriptional adaptations that favor an accelerated development of maximal but hamper the development of strength endurance.

As a final verification that abrogating CMKLR1 alters inflammation in our tissue of interest, we measured the gene expression of inflammatory markers in skeletal muscle. Blockade of CMKLR1 tended to decrease the muscle mRNA expression of TNFα, and significantly reduced the levels of CD68 and RELA (p65) (Supplementary Figures S1A–C). These data underscore an anti-inflammatory effect of CMKLR1 inhibition in skeletal muscle.

So far, no approved drugs to suppress CMKLR1 activity are available commercially. However, a small molecule with the structure 2-(anaphthoyl) ethyltrimethylammonium iodide (α-NETA) has been found to suppress CMKLR1 signaling in preclinical studies (Graham et al., 2014). We thus evaluated the effect of α-NETA on markers of myogenesis, calcium regulation and inflammation. Administration of α-NETA significantly decreased CD34 and MyoD expression in skeletal muscle (Supplementary Figures S2A,B) without altering any of the other markers (Supplementary Figures S2C–M).