Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Jun 9.
Published in final edited form as: Exp Gerontol. 2017 Oct 10;100:1–10. doi: 10.1016/j.exger.2017.10.003

Metabolites related to renal function, immune activation, and carbamylation are associated with muscle composition in older adults

Michael S Lustgarten 1,*, Roger A Fielding 1
PMCID: PMC6556217  NIHMSID: NIHMS966603  PMID: 29030163

Abstract

Reduced skeletal muscle density in older adults is associated with insulin resistance, decreased physical function, and an increased all-cause mortality risk. To elucidate mechanisms that may underlie the maintenance of skeletal muscle density, we conducted a secondary analysis of previously published muscle composition and serum metabolomic data in 73 older adults (average age, 78 y). Multivariable-adjusted linear regression was used to examine associations between 321 metabolites with muscle composition, defined as the ratio between normal density (NDM) with low density (LDM) thigh muscle cross sectional area (NDM/LDM). Sixty metabolites were significantly (p ≤ 0.05 and q < 0.30) associated with NDM/LDM. Decreased renal function and the immune response have been previously linked with reduced muscle density, but the mechanisms underlying these connections are less clear. Metabolites that were significantly associated with muscle composition were then tested for their association with circulating markers of renal function (blood urea nitrogen, creatinine, uric acid), and with the immune response (neutrophils/lymphocytes) and activation (kynurenine/tryptophan). 43 significant NDM/LDM metabolites (including urea) were co-associated with at least 1 marker of renal function; 23 of these metabolites have been previously identified as uremic solutes. The neutrophil/lymphocyte ratio was significantly associated with NDM/LDM (β ± SE: −0.3 ± 0.1, p = 0.01, q = 0.04). 35 significant NDM/LDM metabolites were co-associated with immune activation. Carbamylation (defined as homocitrulline/lysine) was identified as a pathway that may link renal function and immune activation with muscle composition, as 29 significant NDM/LDM metabolites were co-associated with homocitrulline/lysine, with at least 2 markers of renal function, and with kynurenine/tryptophan. When considering that elevated urea and uremic metabolites have been linked with an increased systemic microbial burden, that antimicrobial defense can be reduced in the presence of carbamylation, and that adipocytes can promote host defense, we propose the novel hypothesis that the age-related increase in adipogenesis within muscle may be a compensatory antimicrobial response to protect against an elevated microbial burden.

Keywords: Muscle composition, Metabolomics, Renal function, Immune activation, Carbamylation

1. Introduction

Aged muscle is characterized by an increase in fat content and a decrease in skeletal muscle density (Goodpaster et al., 2001), a phenotype that is known as myosteatosis (Borkan et al., 1983). Decreased skeletal muscle density is associated with insulin resistance (Goodpaster et al., 1997), reduced mobility and physical function (Goodpaster et al., 2001; Visser et al., 2005), and with an elevated risk for all-cause mortality in older adults (Miljkovic et al., 2015). Because older adults (70+ years) are the fastest growing subpopulation in the world (Affairs and Division, 2009), the development of an improved understanding about mechanisms related to muscle composition will be important for addressing the public health priority of healthy aging.

Decreased renal function and the immune response have been previously linked with reduced skeletal muscle density. First, adult hemodialysis patients have an increased non-contractile cross-sectional area of the ankle dorsiflexors, when compared with age-matched controls (Johansen et al., 2003). More specifically, proliferation and differentiation of satellite cells are impaired, but fibro/adipogenic progenitor (FAP) cells proliferate in mice that have chronic renal disease (CKD), an effect that increases fibrotic tissue and adipocyte gene expression in muscle (Zhang et al., 2010; Dong et al., 2016). In addition, intramuscular fat is derived from FAPs (Joe et al., 2010; Uezumi et al., 2010).

Second, an elevated neutrophil/lymphocyte ratio, as a marker of the immune response (Zahorec, 2001), has been associated with myosteatosis in patients with colon cancer (Malietzis et al., 2016). It is important to note that an elevated neutrophil/lymphocyte ratio may also be reflective of an increased circulating and/or systemic microbial burden. For example, neutrophils increase in conjunction with decreased lymphocytes in response to lipopolysaccharide (LPS) (Passler et al., 2013), a component of the outer wall of gram-negative bacteria (Rietschel et al., 1994), and in response to infection with gram-positive bacteria (Dolma et al., 2014) or virus (Holub et al., 2012). Similarly, serum levels of LPS-binding protein (LBP) are elevated in association with an increased neutrophil/lymphocyte ratio (Lemesch et al., 2016).

While decreased renal function and the immune response have been previously linked with muscle composition, the mechanisms that connect these pathways in older adults are less clear. One approach that can be used to elucidate mechanisms between muscle composition with renal function or the immune response is mass spectrometry-based metabolomics. An untargeted metabolomic approach aims to characterize and quantify all of the metabolites in a biological sample, thereby providing an analytical description of complex metabolic processes (Fiehn, 2002). With use of this approach, we have identified potential pathways that may underlie the maintenance of body composition, physical function, and inflammation in older adults (Lustgarten et al., 2013, 2014a, 2014b; Lustgarten and Fielding, 2016).

Accordingly, the goal of the present study was to develop an improved understanding about mechanisms that may underlie muscle composition in older adults. To achieve this objective, we conducted a secondary analysis on the muscle composition and serum metabolomic data reported by Chale et al. (2013) and Lustgarten et al. (2013), respectively. Initially, we examined associations between serum metabolites with the ratio between normal density (NDM) with low density (LDM) thigh muscle cross sectional area (NDM/LDM). To investigate potential links between muscle composition with renal function, with the immune response and activation, and with microbial burden, we then examined associations between significant NDM/LDM metabolites with circulating markers of renal function (blood urea nitrogen, creatinine, uric acid, α-klotho), with the immune response (neutrophils/lymphocytes) and activation (kynurenine/tryptophan), and with microbial burden (LPS, LBP).

2. Materials and methods

2.1. Study design and participants

To identify serum metabolites significantly associated with muscle composition, a secondary analysis on the baseline muscle composition data of Chale et al. (2013) in conjunction with the metabolomic data obtained from the baseline serum samples of Chale et al. (2013), as reported by Lustgarten et al. (2013), was performed. Data for 73 community dwelling, overweight, older adults (average BMI, age:27.0 kg/m2, 77.7 y), including 43 women and 30 men, was used. The study was approved by the Tufts University Health Sciences Campus Institutional Review Board.

Inclusion and exclusion criteria were previously reported by Chale et al. (2013). Briefly, all participants were required to be sedentary, defined as the absence of structured exercise during the previous 6 months. Moreover, relevant exclusion criteria were the presence of type I or II diabetes mellitus, and an eGFR < 30 mL/min/1.73 m2. The median eGFR, calculated with use of the MDRD equation, was 73.1 mL/min/1.73 m2 (interquartile range: 60.8, 92), a value that is within the range reported for the 2965 older adults (> 70 y) of NHANES III (Coresh et al., 2003).

2.2. Measurement of LDM, NDM, and whole body fat mass

Values for LDM and NDM, as reported by Chale et al. (2013), were obtained with use of computed tomography (CT) imaging (Siemens Somotom Scanner, Erlangen, Germany) of the non-dominant thigh at the midpoint of the femur. CT scans were analyzed by a technician in a blinded manner with use of SliceOmatic v4.2 software (Montreal, Canada). The mean value of all pixels within the range of 0–34 and 35–100 Hounsfield units (HU) was used to quantify the amount of LDM and NDM, respectively. To account for the quantity of normal density muscle relative to the amount of low density muscle, NDM was divided by LDM (NDM/LDM), as an index of muscle composition. A high NDM/LDM is indicative of good muscle composition, whereas a low NDM/LDM is indicative of poor muscle composition. NDM/LDM was strongly correlated with the mean attenuation value (43.3 ± 4.8) for all pixels within the 0–100 HU range (r = 0.9, p = 2.4E − 25).

Values for whole body fat mass, as reported by Chale et al. (2013), were obtained with use of dual-energy X-ray absorptiometry (DXA; Hologic Inc., Bedford, MA). DXA scan acquisition and analysis was performed according to manufacturer guidelines, with three passes over the subject to acquire the full DXA image. Scans were analyzed using Hologic QDR software version 12.3 in array mode.

2.3. Metabolomic analysis

Baseline serum samples obtained from the fasted subjects of Chale et al. (2013) were sent to Metabolon Inc. (Research Triangle Park, NC) for metabolomic data acquisition, as reported by Lustgarten et al. (2013). Briefly, small molecule metabolites were extracted from serum and the reconstituted extracts were resolved using mass spectrometry platforms, including ultrahigh performance liquid chromatography/tandem mass spectrometry and gas chromatography/mass spectrometry, with details of this platform described by Evans et al. (2009).

2.4. Measurement of blood urea nitrogen (BUN), creatinine, uric acid, LPS, LBP, α-klotho, neutrophils and lymphocytes

Baseline serum samples obtained from the fasted subjects of Chale et al. (2013) were used for measurement of BUN, creatinine, uric acid, LPS, LBP, and α-klotho. BUN, creatinine, and uric acid were measured with use of a clinical chemistry automated analyzer (Olympus AU400, Olympus America Inc., Melville, NY), using reagents, calibrators, and standard operating procedures as specified by the manufacturer. LPS was measured using the endpoint chromogenic LAL assay (Lonza, Switzerland). LBP and α-klotho were measured using ELISA kits (human LBP multispecies reactive ELISA kit, Cell Sciences, MA, USA; human soluble α-klotho assay kit, IBL International, Germany).

Blood levels of neutrophils and lymphocytes were quantified in baseline samples obtained from the fasted subjects of Chale et al. (2013) with use of impedance with hydrofocus cytometry (ABX Pentra 60 C+, HORIBA Medical, Irvine, CA).

2.5. Statistics

Box-Cox normality plots (Wessa, 2015) were used to determine the lambda value that results in the optimal fit against the normal distribution for NDM/LDM, BUN, creatinine, uric acid, α-klotho, homocitrulline/lysine, neutrophils/lymphocytes, LPS, LBP, kynurenine/tryptophan, and phenylalanine/tyrosine. These data were then transformed with use of the following lambda values: NDM/LDM (−0.25), BUN (0.61), creatinine (0.1), uric acid (−0.05), α-klotho (−0.78), homocitrulline/lysine (0), neutrophils/lymphocytes (−0.28), LPS(0.45), LBP (0.6), kynurenine/tryptophan (−0.56), phenylalanine/tyrosine (−0.24). To maintain the directionality of associations following transformations with negative lambda values, the data was multiplied by −1.

Sex, age, and whole body fat mass were each significantly associated with the transformed value for NDM/LDM (β ± SE for sex, age, and whole body fat mass, respectively: 0.1 ± 0.0, p = 0.04; −0.0 ± 0.0, p = 0.04; −0.0 ± 0.0, p = 1.5E − 05). Accordingly, sex, age, and whole body fat mass-adjusted linear regression (SAS Enterprise Guide 4.3) was used to examine the association between NDM/LDM with circulating metabolites. Each model included sex, age, whole body fat mass, and individual metabolites.

To explore potential overlapping mechanisms between muscle composition with renal function, the immune response and activation, and microbial burden, metabolites that were significantly associated with NDM/LDM were then investigated for their sex, age, and whole body fat mass-adjusted association with BUN, creatinine, uric acid, α-klotho, homocitrulline/lysine, neutrophils/lymphocytes, LPS, LBP, and kynurenine/tryptophan.

False discovery rates (Benjamini and Hochberg, 1995) were computed with use of the q-value method (Storey and Tibshirani, 2003) to account for multiple comparisons. Q-values were computed based on 870 p-values, including 9 associations between NDM/LDM with BUN, creatinine, uric acid, α-klotho, homocitrulline/lysine, neutrophils/lymphocytes, LPS, LBP, and kynurenine/tryptophan, 321 associations between serum metabolites with NDM/LDM, 60 associations each for the significant NDM/LDM metabolites with BUN (with the exception of the association between serum urea with BUN), creatinine, uric acid, α-klotho, homocitrulline/lysine, neutrophils/lymphocytes, LPS, LBP, and kynurenine/tryptophan (539 total comparisons), and the association between phenylalanine/tyrosine with kynurenine/tryptophan. Statistical significance for all multivariable-adjusted associations was set at p ≤ 0.05 and q < 0.30, as reported by (Meyers et al., 2010). A q-value of 0.30 indicates that the result is likely to be valid 7 out of 10 times, which we suggest is reasonable in the setting of exploratory discovery.

Stepwise linear regression was used to develop a NDM/LDM predictive model. Sex, age, and whole body fat mass were forced into the stepwise model because of their significant univariate associations with NDM/LDM. The 60 significant NDM/LDM metabolites were then considered as candidate variables, and statistical significance for metabolites to enter and be retained in the model was set at p < 0.05.

3. Results

Subject characteristics (pre-transformation) for NDM, LDM, NDM/LDM, whole body fat mass, BUN, creatinine, uric acid, homocitrulline, lysine, homocitrulline/lysine, neutrophils, lymphocytes, neutrophils/lymphocytes, LPS, LBP, kynurenine, tryptophan, and kynurenine/tryptophan are shown in Table 1. Subject demographics, including gender, age, BMI, number of medical diagnoses, number of medications, and ethnicity are shown in Supplementary Table 1.

Table 1.

Subject characteristics.

Median (interquartile range)
NDM (cm2) 68.5 (53.7, 84.3)
LDM (cm2) 24.8 (19.5, 30.8)
NDM/LDM 2.7 (1.9, 4.1)
Whole body fat mass (kg) 26.4 (19.9, 30.8)
BUN (mg/dL) 18 (16, 24)
Creatinine (mg/dL) 0.9 (0.8, 1.1)
Uric Acid (mg/dL) 5.9 (4.9, 6.5)
Homocitrulline (kilounits) 43 (29, 64)
Lysine (kilounits) 427 (364, 515)
Homocitrulline/lysine 0.10 (0.08, 0.12)
Neutrophils (cells/μL) 2968 (2496, 3758)
Lymphocytes (cells/μL) 1610 (1297, 1991)
Neutrophils/lymphocytes 1.93 (1.87, 1.99)
LPS (EU/mL) 1.5 (0.9, 2.4)
LBP μg/mL) 22.2 (19.3, 28.4)
Kynurenine (kilounits) 528 (490, 591)
Tryptophan (megaunits) 13 (12, 15)
Kynurenine/tryptophan 0.039 (0.038, 0.041)
Open in a new tab

Data for outcome variables are shown with their pre-transformed median and interquartile range values. Mass spectrometry-obtained values for homocitrulline, lysine, kynurenine, and tryptophan are shown with arbitrary units.

3.1. Metabolites significantly associated with NDM/LDM

Significant (p ≤ 0.05 and q < 0.30) and non-significant associations between serum metabolites with NDM/LDM are shown in Table 2 and Supplementary Table 2, respectively. Fifteen metabolites, including amino acids and their degradation products (alanine, asparagine, ergothionine, glutamine, glycine, serine, trans-urocanate), bile acid metabolites (7-α-hydroxy-3-oxo-4-cholestenoate, glycoursodeoxycholate), purines (7-methylguanine, xanthine), the carnitine precursor deoxycarnitine, the heme degradation product biliverdin, the unsaturated fatty acid 10-undecanoate, and the glucocorticoid cortisone were positively associated with NDM/LDM.

Table 2.

Metabolites significantly associated with muscle composition (NDM/LDM).

β ± SE p-Value q-Value
Mannitol − 0.0 ± 0.0 1.0E − 05 0.0001
2-Hydroxyisobutyrate − 0.1 ± 0.0 5.6E − 05 0.0006
Urea − 0.1 ± 0.0 0.0008 0.005
Butyrylcarnitine − 0.1 ± 0.0 0.0009 0.006
Pseudouridine − 0.1 ± 0.0 0.001 0.008
N-acetylthreonine − 0.1 ± 0.0 0.001 0.008
Tiglyl carnitine − 0.1 ± 0.0 0.002 0.01
Erythronate − 0.1 ± 0.1 0.002 0.01
4-Acetamidobutanoate − 0.1 ± 0.0 0.002 0.01
Glutamine 0.2 ± 0.1 0.003 0.01
C-glycosyltryptophan − 0.1 ± 0.0 0.003 0.02
Serine 0.1 ± 0.0 0.004 0.02
Erythritol − 0.1 ± 0.0 0.004 0.02
Glutaroyl carnitine − 0.1 ± 0.0 0.007 0.03
Glycoursodeoxycholate 0.0 ± 0.0 0.009 0.04
Phenylacetylglutamine − 0.0 ± 0.0 0.01 0.04
Dimethylglycine − 0.1 ± 0.0 0.01 0.04
5α-Androstan-3β,17beta-diol disulfate − 0.0 ± 0.0 0.01 0.04
2-Methylbutyroylcarnitine − 0.1 ± 0.0 0.01 0.04
Indolelactate − 0.1 ± 0.0 0.01 0.04
Alanine 0.1 ± 0.0 0.01 0.05
N-acetylalanine − 0.1 ± 0.1 0.01 0.05
Methylglutarylcarnitine − 0.0 ± 0.0 0.01 0.05
Octanoylcarnitine − 0.0 ± 0.0 0.01 0.05
Myo-inositol − 0.1 ± 0.0 0.01 0.05
Arabitol − 0.1 ± 0.0 0.01 0.05
Biliverdin 0.0 ± 0.0 0.01 0.05
Arabonate − 0.0 ± 0.0 0.02 0.05
Isobutrylcarnitine − 0.0 ± 0.0 0.02 0.05
Decanoylcarnitine − 0.0 ± 0.0 0.02 0.06
N-formylmethionine − 0.1 ± 0.0 0.02 0.06
Glycine 0.1 ± 0.0 0.02 0.07
Acetylcarnitine − 0.1 ± 0.0 0.02 0.07
Phenol sulfate − 0.0 ± 0.0 0.02 0.07
N1-methyl-2-pyridone-5-carboxamide − 0.1 ± 0.0 0.02 0.07
Glucuronate − 0.0 ± 0.0 0.02 0.07
Trans-4-hydroxyproline − 0.1 ± 0.0 0.02 0.07
Laurylcarnitine − 0.0 ± 0.0 0.03 0.07
N-acetylmethionine − 0.1 ± 0.0 0.03 0.07
Xanthine 0.0 ± 0.0 0.03 0.07
Deoxycarnitine 0.1 ± 0.1 0.03 0.07
Cis-4-decenoyl carnitine − 0.0 ± 0.0 0.03 0.08
Hexanoylcarnitine − 0.0 ± 0.0 0.03 0.08
Ergothioneine 0.0 ± 0.0 0.03 0.08
Symmetric dimethylarginine − 0.1 ± 0.0 0.03 0.08
Cortisone 0.1 ± 0.0 0.03 0.08
Asparagine 0.1 ± 0.0 0.03 0.09
N6-acetyllysine − 0.1 ± 0.0 0.03 0.09
Tetradecanedioate − 0.0 ± 0.0 0.03 0.09
7-Methylguanine 0.1 ± 0.0 0.04 0.09
Trans-urocanate 0.1 ± 0.0 0.04 0.09
10-Undecanoate 0.1 ± 0.0 0.04 0.09
1,5-Anhydroglucitol − 0.0 ± 0.0 0.04 0.09
Epiandrosterone sulfate − 0.0 ± 0.0 0.04 0.09
7-α-Hydroxy-3-oxo-4-cholestenoate 0.1 ± 0.0 0.04 0.10
Threitol − 0.0 ± 0.0 0.05 0.10
1-Methylurate − 0.0 ± 0.0 0.05 0.10
Androsterone sulfate − 0.0 ± 0.0 0.05 0.10
N-acetylserine − 0.1 ± 0.0 0.05 0.10
N1-methyladenosine − 0.1 ± 0.1 0.05 0.10
Open in a new tab

Sex, age and whole body fat mass-adjusted associations between serum metabolites with NDM/LDM are shown with parameter estimates and standard errors (β ± SE), in order of significance (p-value), and with q-values.

45 metabolites were negatively associated with NDM/LDM, including acylcarnitines (acetylcarnitine, butyrylcarnitine, 2-methylbutroylcarnitine, isobutyrylcarnitine, tiglyl carnitine glutaroylcarnitine, methylglutarylcarnitine, hexanoylcarnitine, octanoylcarnitine, decanoylcarnitine, cis-4-decenoyl carnitine, laurylcarnitine), sugar (1,5 anhydroglucitol), sugar alcohols (arabitol, erythritol, mannitol, myo-inositol, threitol) and acids (arabonate, erythronate, glucuronate), N-acetylated amino acids (4-acetamidobutanoate, N-acetylalanine, N6-acetyllysine, N-acetylmethionine, N-acetylserine, N-acetylthreonine), steroids (5α-androstan-3β, 17β-diol disulfate, androsterone sulfate, epiandrosterone sulfate), methylated amino acids and purines (dimethylglycine, symmetric dimethylarginine, N1-methyladenosine, 1-methylurate), glycosylated pyrimidine and amino acids (pseudouridine, C-glycosyltrytophan), nitrogen excretion (urea, phenylacetylgluta-mine), hydroxylated and dicarboxylic fatty acids (2-hydroxyisobutyrate, tetradecanedioate), gut bacterial metabolism (phenol sulfate), the tryptophan degradation metabolite indolelactate, the NAD degradation product N1-methyl-2-pyridone-5-carboxamide, the collagen-related metabolite trans-4-hydroxyproline, and a cleavage product of chemotactic peptide, N-formylmethionine.

Stepwise linear regression was used to determine a NDM/LDM predictive model. While sex, age, and whole body fat mass explained35.3% of the variability inherent in NDM/LDM, the combination of 4 metabolites (2-hydroxyisobutyrate, mannitol, 7-methylguanine, and 1, 5-anhydroglucitol) explained an additional 31.3%, for a total adjusted R2 equal to 66.6% (Supplementary Table 3).

3.2. Metabolites significantly associated with NDM/LDM and with circulating markers of renal function

BUN, creatinine, and uric acid are circulating markers of renal function (Gowda et al., 2010) whose values are elevated when renal function is reduced. BUN was significantly negatively associated with NDM/LDM (β ± SE: −0.0 ± 0.0, p = 0.004, q = 0.02), whereas creatinine and uric acid were borderline significant (β ± SE: −0.7 ± 0.4, p = 0.06, q = 0.13; −1.7 ± 0.9, p = 0.06, q = 0.13, respectively). To explore potential overlapping pathways between muscle composition with renal function, metabolites that were significantly associated with NDM/LDM were then investigated for their association with BUN, creatinine, and uric acid. 43, 34, and 23 of the 60 significant NDM/LDM metabolites were significantly associated with 1, 2, and 3 circulating markers of renal function, respectively (Table 3). Data for the 17 significant NDM/LDM metabolites that were not significantly associated with any markers of renal function are shown in Supplementary Table 4.

Table 3.

Metabolites significantly associated with NDM/LDM and with circulating markers of renal function.

Vs. BUN Vs. creatinine Vs. uric acid
β ± SE p-Value q-Value p ± SE p-Value q-Value β ± SE p-Value q-Value
2-Hydroxyisobutyrate 1.8 ± 0.3 8.5E − 09 4.4E − 07 0.0 ± 0.0 1.9E − 05 0.0002 0.0 ± 0.0 0.005 0.02
Butyrylcarnitine 0.9 ± 0.3 0.002 0.01 0.0 ± 0.0 0.24 0.33 0.0 ± 0.0 0.006 0.03
Urea – – – 0.0 ± 0.0 1.6E − 06 3.5E − 05 0.0 ± 0.0 0.005 0.02
Pseudouridine 2.6 ± 0.5 2.8E − 06 5.1E − 05 0.1 ± 0.0 5.9E − 12 1.7E − 09 0.0 ± 0.0 2.3E − 06 4.5E − 05
N-acetylthreonine 2.1 ± 0.4 5.4E − 07 1.7E − 05 0.1 ± 0.0 9.3E − 10 6.5E − 08 0.0 ± 0.0 0.003 0.01
Tiglyl carnitine 1.4 ± 0.3 9.7E − 06 0.0001 0.0 ± 0.0 0.02 0.07 0.0 ± 0.0 0.25 0.34
Erythronate 1.5 ± 0.4 0.0001 0.0009 0.1 ± 0.0 4.5E − 12 1.6E − 09 0.0 ± 0.0 0.002 0.01
4-Acetamidobutanoate 2.0 ± 0.4 1.6E − 05 0.0002 0.1 ± 0.0 6.7E − 08 2.7E − 06 0.0 ± 0.0 0.09 0.20
Glutamine − 3.8 ± 0.8 1.4E − 05 0.0002 − 0.1 ± 0.0 0.0009 0.006 − 0.0 ± 0.0 0.0001 0.0009
C-glycosyltryptophan 2.7 ± 0.5 6.4E − 07 1.8E − 05 0.1 ± 0.0 2.5E − 10 2.4E − 08 0.0 ± 0.0 0.01 0.04
Serine − 2.1 ± 0.6 0.001 0.006 − 0.0 ± 0.0 0.01 0.05 − 0.0 ± 0.0 0.37 0.43
Erythritol 1.8 ± 0.4 8.7E − 06 0.0001 0.1 ± 0.0 7.9E − 10 6.4E − 08 0.0 ± 0.0 0.002 0.01
Glutaroyl carnitine 1.3 ± 0.3 8.5E − 05 0.0008 0.0 ± 0.0 0.0002 0.002 0.0 ± 0.0 0.007 0.03
Glycoursodeoxycholate − 0.1 ± 0.1 0.41 0.44 − 0.0 ± 0.0 0.61 0.55 − 0.0 ± 0.0 0.02 0.07
Phenylacetylglutamine 0.7 ± 0.2 0.0002 0.002 0.0 ± 0.0 0.0001 0.0009 0.0 ± 0.0 0.03 0.09
Dimethylglycine 0.9 ± 0.4 0.02 0.06 0.0 ± 0.0 0.0002 0.002 0.0 ± 0.0 0.0002 0.002
2-Methylbutyroylcarnitine 1.7 ± 0.4 2.5E − 05 0.0003 0.0 ± 0.0 0.0005 0.003 0.0 ± 0.0 0.01 0.05
Indolelactate 1.2 ± 0.3 0.0007 0.005 0.0 ± 0.0 0.0001 0.0009 0.0 ± 0.0 0.28 0.36
N-acetylalanine 3.4 ± 0.6 8.8E − 07 2.2E − 05 0.1 ± 0.0 4.3E − 11 6.1E − 09 0.0 ± 0.0 0.0002 0.002
Methylglutarylcarnitine 0.3 ± 0.2 0.08 0.16 0.0 ± 0.0 0.04 0.09 0.0 ± 0.0 0.49 0.49
Octanoylcarnitine 0.4 ± 0.2 0.12 0.21 0.0 ± 0.0 0.06 0.13 0.0 ± 0.0 0.01 0.04
Myo-inositol 2.0 ± 0.3 4.1E − 08 1.9E − 06 0.1 ± 0.0 1.9E − 10 2.1E − 08 0.0 ± 0.0 0.0008 0.00
Arabitol 3.3 ± 0.6 9.5E − 08 3.6E − 06 0.1 ± 0.0 6.4E − 07 1.8E − 05 0.0 ± 0.0 0.0009 0.006
Arabonate 1.1 ± 0.2 2.1E − 05 0.000 0.0 ± 0.0 2.0E − 06 4.2E − 05 0.0 ± 0.0 0.009 0.036
Isobutrylcarnitine 1.2 ± 0.2 4.0E − 06 6.8E − 05 0.0 ± 0.0 4.2E − 06 7.0E − 05 0.0 ± 0.0 0.22 0.32
Decanoylcarnitine 0.2 ± 0.3 0.34 0.41 0.0 ± 0.0 0.28 0.36 0.0 ± 0.0 0.02 0.07
N-formylmethionine 2.0 ± 0.5 2.5E − 05 0.0003 0.1 ± 0.0 1.1E − 09 6.9E − 08 0.0 ± 0.0 0.008 0.03
Glycine − 1.2 ± 0.5 0.02 0.06 − 0.0 ± 0.0 0.62 0.55 − 0.0 ± 0.0 0.0008 0.005
Phenol sulfate 0.5 ± 0.2 0.008 0.03 0.0 ± 0.0 0.04 0.11 0.0 ± 0.0 0.05 0.12
N1-methyl-2-pyridone-5-carboxamide 1.3 ± 0.3 4.4E − 06 7.08E − 05 0.0 ± 0.0 5.6E − 05 0.0006 0.0 ± 0.0 0.006 0.03
Glucuronate 1.3 ± 0.2 4.1E − 07 1.36E − 05 0.0 ± 0.0 0.0003 0.002 0.0 ± 0.0 7.9E − 05 0.0008
N-acetylmethionine 1.0 ± 0.5 0.04 0.10 0.0 ± 0.0 0.004 0.02 0.0 ± 0.0 0.21 0.31
Deoxycarnitine − 0.3 ± 0.8 0.73 0.58 0.0 ± 0.0 0.008 0.03 − 0.0 ± 0.0 0.46 0.47
Cis-4-decenoyl carnitine 0.5 ± 0.3 0.10 0.19 0.0 ± 0.0 0.03 0.07 0.0 ± 0.0 0.01 0.05
Hexanoylcarnitine 0.6 ± 0.3 0.07 0.15 0.0 ± 0.0 0.11 0.21 0.0 ± 0.0 0.02 0.07
Symmetric dimethylarginine 1.7 ± 0.6 0.007 0.03 0.0 ± 0.0 0.003 0.02 0.0 ± 0.0 0.24 0.34
N6-acetyllysine 1.4 ± 0.5 0.02 0.05 0.1 ± 0.0 2.2E − 05 0.0002 0.0 ± 0.0 0.02 0.07
Trans-urocanate − 0.8 ± 0.3 0.01 0.04 − 0.0 ± 0.0 0.12 0.22 − 0.0 ± 0.0 0.17 0.27
10-Undecanoate − 1.0 ± 0.4 0.02 0.06 0.0 ± 0.0 0.69 0.57 0.0 ± 0.0 0.82 0.61
Threitol 1.2 ± 0.2 4.7E − 06 7.3E − 05 0.0 ± 0.0 1.3E − 11 2.4E − 09 0.0 ± 0.0 0.007 0.03
1-Methylurate 0.3 ± 0.3 0.22 0.32 0.0 ± 0.0 0.03 0.08 0.0 ± 0.0 0.12 0.22
N-acetylserine 1.1 ± 0.4 0.01 0.04 0.0 ± 0.0 0.0003 0.002 0.0 ± 0.0 0.03 0.08
N1-methyladenosine 2.4 ± 0.8 0.004 0.02 0.1 ± 0.0 2.2E − 06 4.4E − 05 0.0 ± 0.0 0.04 0.11
Open in a new tab

Metabolites significantly associated with NDM/LDM and with at least one circulating marker of renal function (BUN, creatinine, uric acid) are shown with parameter estimates and standard errors (β ± SE), and with p- and q-values.

Moreover, circulating levels of the secreted protein, α-klotho, are decreased in serum in association with reduced renal function (Shimamura et al., 2012). A link between α-klotho with muscle composition is suggested by the findings that muscle mass is reduced in klotho hypomorphic mice (Kuro-o et al., 1997; Phelps et al., 2016), and that α-klotho overexpression increases satellite cell numbers and their proliferation, and increases myotube growth (Wehling-Henricks et al., 2016). We then explored the association between serum α-klotho with NDM/LDM, and between significant NDM/LDM metabolites with α-klotho. α-Klotho was not significantly associated with NDM/LDM (β ± SE: 1.5 ± 5.7, p = 0.79, q = 0.60), nor was there an overlap between any significant NDM/LDM metabolites with α-klotho (Supplementary Table 5).

3.3. Metabolites significantly associated with NDM/LDM and with a circulating marker of carbamylation

When renal function is reduced, circulating levels of urea are elevated. Urea decomposes to form cyanate, which can combine with the amino acid lysine to form the carbamylated metabolite, homocitrulline (Kraus and Kraus, 2001). The homocitrulline/lysine ratio, as an index of carbamylation, is elevated in CKD patients, when compared with subjects that have normal renal function (Koeth et al., 2013). The importance of carbamylation is further illustrated by the finding that it increases during aging, and the accumulation rate of protein carbamylation is negatively correlated with species life expectancy (Gorisse et al., 2016).

To explore the potential link between carbamylation with muscle composition, we investigated the association between the serum homocitrulline/lysine ratio with NDM/LDM, and between the 60 significant NDM/LDM metabolites with homocitrulline/lysine. Although NDM/LDM was not significantly associated with homocitrulline/lysine (β ± SE: −0.2 ± 0.2, p = 0.25, q = 0.34), 35 significant NDM/LDM metabolites were co-associated with homocitrulline/lysine (Table 4). Significant NDM/LDM metabolites that were not associated with homocitrulline/lysine are shown in Supplementary Table 6.

Table 4.

Metabolites significantly associated with NDM/LDM and with a circulating marker of carbamylation.

β ± SE p-Value q-Value
C-glycosyltryptophan 0.2 ± 0.0 8.8E − 07 2.2E − 05
Erythronate 0.1 ± 0.0 1.6E − 06 3.5E − 05
N1-methyladenosine 0.2 ± 0.0 1.1E − 05 0.0001
N-acetylthreonine 0.1 ± 0.0 1.1E − 05 0.0001
Pseudouridine 0.1 ± 0.0 1.8E − 05 0.0002
2-Hydroxyisobutyrate 0.1 ± 0.0 2.3E − 05 0.0003
N-acetylalanine 0.2 ± 0.0 2.6E − 05 0.0003
2-Methylbutyroylcarnitine 0.1 ± 0.0 0.0002 0.002
4-Acetamidobutanoate 0.1 ± 0.0 0.0002 0.002
Threitol 0.1 ± 0.0 0.0002 0.002
N-formylmethionine 0.i ± 0.0 0.0002 0.002
Glutamine − 0.2 ± 0.0 0.0004 0.003
Myo-inositol 0.1 ± 0.0 0.0004 0.003
Urea 0.1 ± 0.0 0.001 0.007
Phenylacetylglutamine 0.0 ± 0.0 0.001 0.007
Tiglyl carnitine 0.1 ± 0.0 0.001 0.008
Erythritol 0.1 ± 0.0 0.001 0.008
Arabonate 0.0 ± 0.0 0.003 0.01
Glutaroyl carnitine 0.1 ± 0.0 0.003 0.01
Isobutrylcarnitine 0.0 ± 0.0 0.003 0.01
Dimethylglycine 0.1 ± 0.0 0.003 0.01
Serine − 0.1 ± 0.0 0.01 0.05
N-acetylserine 0.1 ± 0.0 0.01 0.05
Methylglutarylcarnitine 0.0 ± 0.0 0.02 0.05
Asparagine − 0.1 ± 0.0 0.02 0.06
Glucuronate 0.0 ± 0.0 0.02 0.07
Indolelactate 0.0 ± 0.0 0.02 0.07
N6-acetyllysine 0.1 ± 0.0 0.02 0.07
Arabitol 0.1 ± 0.0 0.02 0.07
Phenol sulfate 0.0 ± 0.0 0.03 0.07
Cis-4-decenoyl carnitine 0.0 ± 0.0 0.03 0.08
Butyrylcarnitine 0.0 ± 0.0 0.03 0.09
Hexanoylcarnitine 0.0 ± 0.0 0.04 0.09
Glycine − 0.1 ± 0.0 0.04 0.09
Mannitol 0.0 ± 0.0 0.05 0.12
Open in a new tab

Metabolites significantly associated with NDM/LDM and with a circulating marker of carbamylation (homocitrulline/lysine) are shown with parameter estimates and standard errors (β ± SE), and with p- and q-values.

3.4. Metabolites significantly associated with NDM/LDM and with the neutrophil/lymphocyte ratio

An elevated neutrophil/lymphocyte ratio has been reported to be negatively associated with muscle composition in colorectal cancer patients (Malietzis et al., 2016). In the older adults of the present study, the neutrophil/lymphocyte ratio was significantly negatively associated with NDM/LDM (β ± SE: −0.3 ± 0.1, p = 0.01, q = 0.04). Of the 60 significant NDM/LDM metabolites, phenol sulfate (β ± SE:0.0 ± 0.0, p = 0.03, q = 0.08) and methylglutarylcarnitine (β ± SE:0.0 ± 0.0, p = 0.04, q = 0.10) were positively associated, whereas xanthine (β ± SE: −0.1 ± 0.0, p = 0.003, q = 0.01) and 7-α-hydroxy-3-oxo-4-cholestenoate (β ± SE: 0.1 ± 0.0, p = 0.03, q = 0.08) were negatively associated with the neutrophil/lymphocyte ratio. Metabolites that were significantly associated with NDM/LDM but not with the neutrophil/lymphocyte ratio are shown in Supplementary Table 7.

3.5. Metabolites significantly associated with NDM/LDM and with circulating markers of microbial burden

An elevated neutrophil/lymphocyte ratio may be reflective of an increased systemic microbial burden. For example, neutrophils increase in conjunction with decreased lymphocytes in response to LPS (Passler et al., 2013), and the neutrophil/lymphocyte ratio is elevated in association with increased levels of LPS-binding protein (LBP) (Lemesch et al., 2016). LPS decreases muscle protein synthesis and increases protein degradation (Orellana et al., 2006; Liu et al., 2013), an effect that would be expected to reduce the quantity of normal density muscle. Similarly, elevated serum LBP is associated with reduced skeletal muscle density in 60-year old adults (Tilves et al., 2016). We then tested the association between LPS and LBP with NDM/LDM, and with the 60 significant NDM/LDM metabolites. LPS and LBP were not significantly associated with NDM/LDM (β ± SE: −0.0 ± 0.0, p = 0.64, q = 0.55; 0.0 ± 0.0, p = 0.24, q = 0.34, respectively). Of the 60 significant NDM/LDM metabolites, four metabolites were significantly associated with LPS, including butyrylcarnitine (β ± SE:0.2 ± 0.1, p = 0.05, q = 0.12), tiglylcarnitine (β ± SE: 0.2 ± 0.1, p = 0.04, q = 0.10), methylglutarylcarnitine (β ± SE: 0.1 ± 0.0, p = 0.04, q = 0.10), and N6-acetyllysine (β ± SE: 0.3 ± 0.1, p = 0.04, q = 0.10), whereas only glycine was significantly associated with LBP (β ± SE: 1.2 ± 0.5, p = 0.02, q = 0.06). Associations between metabolites that were significantly associated with NDM/LDM but not with LPS or LBP are shown in Supplementary Tables 8 and 9.

3.6. Metabolites significantly associated with NDM/LDM and with a circulating marker of immune activation

To further explore the link between the immune response with muscle composition, the 60 significant NDM/LDM metabolites were then tested for their association with the serum kynurenine/tryptophan ratio. The degradation of tryptophan to form kynurenine is catalyzed by either tryptophan 2,3-dioxygenase (TDO) or indoleamine-pyrrole 2,3-dioxygenase (IDO) (Yasui et al., 1986; Saito et al., 2000). Whereas TDO is produced primarily by the liver, IDO is induced during immune-related conditions (Badawy, 2017). IDO expression is increased in response to interferon-gamma (IFN-γ) (O’Connor et al., 2009), a cytokine that is produced by immune cells, including monocyte-derived macrophages and dendritic cells (Munn et al., 1999; Hwu et al., 2000), in response to bacterial, viral, and parasitic infection (Yoshida et al., 1979, 1981; Pfefferkorn, 1984; Taylor and Feng, 1991). Kynurenine/tryptophan has been previously reported as a marker of IFN-γ (Apalset et al., 2014). To examine if the kynurenine/tryptophan ratio was related to immune activation, we investigated the association between the serum phenylalanine/tyrosine ratio with kynurenine/tryptophan. The phenylalanine/tyrosine ratio is positively correlated with kynurenine/tryptophan when the immune system is activated (Mangge et al., 2013). Phenylanine/tyrosine was significantly positively associated with kynurenine/tryptophan (β ± SE: 0.1 ± 0.0, p = 0.01, q = 0.04).

Although the association between the kynurenine/tryptophan ratio with NDM/LDM was borderline significant (β ± SE: −0.1 ± 0.1, p = 0.06, q = 0.13), 35 significant NDM/LDM metabolites were co-associated with kynurenine/tryptophan (Table 5). Metabolites that were significantly associated with NDM/LDM but not with kynurenine/tryptophan are shown in Supplementary Table 10.

Table 5.

Metabolites significantly associated with NDM/LDM and with a circulating marker of immune activation.

β ± SE p-Value q-Value
C-glycosyltryptophan 0.4 ± 0.1 4.4E − 09 2.5E − 07
Erythritol 0.3 ± 0.1 6.3E − 08 2.7E − 06
Pseudouridine 0.4 ± 0.1 1.2E − 07 4.2E − 06
1-Methylurate 0.2 ± 0.0 8.0E − 07 2.1E − 05
4-Acetamidobutanoate 0.3 ± 0.1 9.2E − 07 2.2E − 05
N1-methyladenosine 0.4 ± 0.1 2.5E − 06 4.7E − 05
N-acetylalanine 0.4 ± 0.1 3.2E − 06 5.6E − 05
N-formylmethionine 0.3 ± 0.1 4.8E − 06 7.3E − 05
Arabonate 0.1 ± 0.0 1.3E − 05 0.0002
Urea 0.2 ± 0.0 2.1E − 05 0.0002
Phenylacetylglutamine 0.1 ± 0.0 2.9E − 05 0.0003
Erythronate 0.2 ± 0.0 5.9E − 05 0.0006
Myo-inositol 0.2 ± 0.0 0.0001 0.0009
Glutaroyl carnitine 0.1 ± 0.0 0.0002 0.002
2-Hydroxyisobutyrate 0.2 ± 0.0 0.0004 0.003
Threitol 0.1 ± 0.0 0.0004 0.003
Glucuronate 0.1 ± 0.0 0.0005 0.003
N6-acetyllysine −0..2 ± 0.1 0.001 0.007
N-acetylserine 0.2 ± 0.0 0.001 0.008
Dimethylglycine 0.2 ± 0.0 0.002 0.009
Isobutrylcarnitine 0.1 ± 0.0 0.003 0.01
N-acetylthreonine 0.2 ± 0.1 0.003 0.01
Methylglutarylcarnitine 0.1 ± 0.0 0.003 0.01
Arabitol 0.2 ± 0.1 0.003 0.01
Serine − 0.2 ± 0.1 0.003 0.02
N1-methyl-2-pyridone-5-carboxamide 0.1 ± 0.0 0.005 0.02
Tiglyl carnitine 0.1 ± 0.0 0.006 0.03
Cis-4-decenoyl carnitine 0.1 ± 0.0 0.007 0.03
Ergothioneine − 0.1 ± 0.0 0.007 0.03
2-Methylbutyroylcarnitine 0.1 ± 0.1 0.009 0.04
Cortisone − 0.2 ± 0.1 0.01 0.05
Butyrylcarnitine 0.1 ± 0.0 0.02 0.06
Hexanoylcarnitine 0.1 ± 0.0 0.03 0.07
Indolelactate 0.1 ± 0.0 0.04 0.10
Glutamine − 0.2 ± 0.1 0.05 0.12
Open in a new tab

Metabolites significantly associated with NDM/LDM and with a circulating marker of immune activation (kynurenine/tryptophan) are shown with parameter estimates and standard errors (β ± SE), and with p- and q-values.

3.7. Co-overlapping associations between significant NDM/LDM metabolites with circulating markers of renal function, immune activation, and carbamylation

Of the 60 significant NDM/LDM metabolites, 29 were co-associated with at least 2 circulating markers of renal function, with kynurenine/tryptophan, and with homocitrulline/lysine, including sugar alcohols and acids (arabitol, arabonate, erythritol, erythronate, glucuronate, myo-inositol, threitol), acylcarnitines (2-methylbutyrylcarnitine, butyrylcarnitine, cis-4-decenoylcarnitine, glutarylcarnitine, isobutyrylcarnitine, tiglylcarnitine), N-acetylated amino acids (4-acetamidobutanoate, N-acetylalanine, N6-acetyllysine, N-acetylserine, N-acetylthreonine), glycosylated pyrimidine or amino acids (pseudouridine, C-glycosyltryptophan), methylated amino acids or purines (dimethylglycine, N1-methyladenosine), amino acids (glutamine, serine), urea metabolism (urea, phenylacetylglutamine) and others (N-for-mylmethionine, 2-hydroxyisobutyrate, indolelactate).

4. Discussion

The goal of the present study was to develop an improved understanding about mechanisms that may underlie the maintenance of muscle composition in older adults. Initially, we identified significant associations for 60 serum metabolites with the ratio between normal density with low density thigh muscle cross sectional area. Further investigation identified significant, co-overlapping associations for 29 of these metabolites with circulating markers of renal function, with immune activation, and with carbamylation.

In terms of the link between renal function with muscle composition, 31 and 23 of the significant NDM/LDM metabolites (including urea) have been previously identified as early stage markers of reduced renal function (Sekula et al., 2016) and as uremic solutes, respectively (Niewczas et al., 2014; Tanaka et al., 2015). In patients with renal disease, elevated circulating levels of urea are associated with an increase in urease-producing intestinal bacteria (Wong et al., 2014), a finding that suggests a role for an altered gut microbiome on reduced muscle composition in older adults. Within the intestine, key protein components of the colonic tight junction are reduced in the presence of urea and uremic metabolites (Vaziri et al., 2012, 2013), but are greatly depleted in the combined presence of urea and urease (Vaziri et al., 2013). It is important to note that although the subjects of present study did not have renal disease, almost half (33/73) had BUN values ≥ 20 mg/dL, which is an amount of urea that significantly reduced tight junction protein levels in isolated human enterocytes (Vaziri et al., 2013). The net effect of elevated urea and uremic metabolites is an increase in gut permeability (Vaziri et al., 2012, 2013). In support of this, 10 significant NDM/LDM metabolites, including deoxycarnitine and seven acylcarnitines, phenylacetylglutamine, and 4-acetamidobutanoate have been reported to be elevated in serum in conjunction with increased gut permeability (Semba et al., 2017). Similarly, the kynurenine/tryptophan ratio, which was borderline significant in its negative association with muscle composition, is elevated in the presence of increased gut permeability (Semba et al., 2016). Collectively, these data suggest that the associations identified between urea and uremic metabolites with muscle composition may be related to an altered gut microbiome, to reduced intestinal tight junction protein levels, and to an increase in intestinal permeability. Future studies aimed at testing this hypothesis are of interest.

Beyond the intestine, bacterial translocation occurs during uremia (de Almeida Duarte et al., 2004), and uremic patients frequently exhibit endotoxemia (Goncalves et al., 2006; Szeto et al., 2008). Furthermore, patients with renal disease have an increased systemic microbial burden, including bloodstream infections with gram-negative and gram-positive bacteria (Berman et al., 2004; Shi et al., 2014), with fungi (Serefhanoglu et al., 2012; Gauna et al., 2013), and viruses (Fabrizi et al., 2000; Gallian et al., 2002). Moreover, levels of circulating microbial products increase during aging. In aged mice, gut permeability was increased in conjunction with an elevated plasma level of muramyl dipeptide, a component of the outer bacterial wall (Thevaranjan et al., 2017). In older adult humans, plasma levels of LPS are elevated, when compared with young subjects (Ghosh et al., 2015). Accordingly, to test the hypothesis that microbial burden was elevated in conjunction with poor muscle composition, we examined the association between LPS and LBP with NDM/LDM. LPS and LBP were not significantly associated with muscle composition, and only five significant NDM/LDM metabolites were co-associated with LPS and LBP. While this suggests that gram-negative bacterial burden and its related metabolic products or pathways may not be involved in mechanisms related to muscle composition in older adults, in contrast, a role for gram-positive bacteria, fungi, or viruses is suggested via significant, co-overlapping associations for five short (C4, C5) and five medium chain (C6–C10) acylcarnitines, pseudouridine, and C-glycosyltryptophan. In support of this, plasma levels of short and medium chain acylcarnitines are elevated in patients infected with gram-positive bacteria, or with fungi (Schmerler et al., 2012). Expression of the proteins responsible for the degradation of the cognate fatty acids for these acylcarnitines, short and medium-chain acyl-coenzyme A dehydrogenase, is decreased in the presence of viral infection (Bose et al., 2014). Moreover, pseudouridine is elevated in virally infected patients (Colonna et al., 1996), and C-glycosyltryptophan is found in viral glycoproteins (Falzarano et al., 2007). When considering that the incidence of bloodstream infections with gram-positive bacteria, with fungi, or viruses are increased by > 5 to15-fold in older adults, when compared with young subjects (Laupland et al., 2008; Zilberberg et al., 2008; Parry et al., 2016), these data may collectively suggest a role for circulating microbial burden on muscle composition in older adults.

If systemic microbial burden is elevated in older adults with poor muscle composition, an immune response and/or activation would be an expected result. In support of this, we identified a significant negative association between the neutrophil/lymphocyte ratio, as a marker of the immune response (Zahorec, 2001), with muscle composition. The neutrophil/lymphocyte ratio has been reported to be elevated in response to infection with gram-positive bacteria (Dolma et al., 2014) or with virus (Holub et al., 2012), and in the presence of bloodstream infections (Yang et al., 2015). Moreover, 35 significant NDM/LDM metabolites were co-associated with kynurenine/tryptophan, as a circulating marker of immune activation (Mangge et al., 2013) and of IFN-γ (Apalset et al., 2014). IFN-γ is produced as an antimicrobial response to bacterial, viral, or parasitic infection (Yoshida et al., 1979, 1981; Pfefferkorn, 1984; Taylor and Feng, 1991), evidence that suggests additional support for the hypothesis that systemic microbial burden may be involved in mechanisms related to muscle composition in older adults. Interestingly, peripheral blood mononuclear cells produce 7-fold more IFN-γ following exposure to gram-positive, when compared against gram-negative bacteria (Skovbjerg et al., 2010), a finding that may drive the search for such pathogens.

Additional support for the hypothesis that antimicrobial defense and immune activation may be related to muscle composition is suggested by co-overlapping associations between significant NDM/LDM metabolites with kynurenine/tryptophan. In terms of antimicrobial defense, eight acylcarnitines were significantly associated with both NDM/LDM and kynurenine/tryptophan-the antimicrobial activity of alkyl containing, quaternary amines (as found in acylcarnitines) is well established, with the proposed bactericidal mechanism involving adsorption onto the negatively charged bacterial cell surface, diffusion through the cell wall, binding to and disruption of the cytoplasmic membrane, and eventual cell death (Gilbert and Moore, 2005). Moreover, urea and erythritol have antimicrobial activity against gram-negative and gram-positive bacteria (Weinstein and Mc, 1946), dimethylglycine induces membrane damage and is bactericidal against Escherichia coli (Vanhauteghem et al., 2012), and arabitol exhibits antiviral activity (Xu et al., 2016). In terms of immunity, carnitines support the production of CD4+ and CD8+ T cells during infection, and increase lymphocyte proliferation and polymorphonuclear chemotaxis (De Simone et al., 1982; Jirillo et al., 1993). Glutamine is involved in T cell growth and proliferation (Wang et al., 2011). Serine directly modulates adaptive immunity by controlling T cell proliferative capacity (Ma et al., 2017). Myo-inositol improves the ability of macrophages to eliminate antibiotic-resistant E. coli (Chen et al., 2015). N-formylmethionine is a potent chemoattractant for neutrophils, which are involved in the host’s resistance to infection (Schiffmann et al., 1975).

Investigating further, a role for carbamylation on muscle composition is suggested by the finding that 35 significant NDM/LDM metabolites were co-associated with homocitrulline/lysine. Carbamylation is the process by which urea decomposes to form cyanate, and cyanate combines with the amino acid lysine, thereby forming homocitrulline. The homocitrulline/lysine ratio, as an index of carbamylation (Koeth et al., 2013), is strongly correlated with urea (Pietrement et al., 2013), is elevated in CKD patients, (Koeth et al., 2013), and increases during aging (Gorisse et al., 2016). In the presence of carbamylated lysine, the production of IL-10, IFN-γ, and TGFβ are increased (Mydel et al., 2010; Wang et al., 2016). IL-10 and TGFβ are involved in mechanisms that allow FAPs to differentiate and proliferate (Lemos et al., 2015), thereby increasing muscle fibrosis and/or adipogenesis (Vidal et al., 2008; Lemos et al., 2015). Similarly, IFN-γ polarizes M1 macrophages, which are involved in mechanisms that allow FAPs to differentiate, or to undergo apoptosis (Wehling-Henricks et al., 2016). Moreover, in light of the hypothesis that systemic microbial burden may be elevated in older adults with poor muscle composition, in contrast, carbamylation decreases antimicrobial efficacy: immunoglobin carbamylation abrogates complement activation (Koro et al., 2014), and carbamylation of the antimicrobial peptide, LL-37 (cathelicidin), decreases its ability to kill gram-negative and gram-positive bacteria (Koro et al., 2016). Interestingly, an increase in adipogenesis within muscle may be one means for improving antimicrobial defense. In support of this, when exposed to the gram-positive bacterium, Staphylococcus auerus, preadipocytes differentiate into adipocytes in conjunction with an 80-fold increased production of cathelicidin antimicrobial peptide (Zhang et al., 2015). When considering that the incidence of bloodstream infections with S. auerus is ~15-fold elevated in adults older than 70, when compared with younger adults (Laupland et al., 2008), we propose these data may suggest an antimicrobial role for FAP-derived adipocytes.

In sum, we posit that these data suggest roles for an altered gut microbiome, increased gut permeability and circulating microbial burden, the immune response and activation, and carbamylation on mechanisms related to muscle composition in older adults. Moreover, we propose the novel hypothesis that increased adipogenesis within aged muscle may be a compensatory antimicrobial response to protect against an increased systemic microbial burden. Because this was an association-based study, causality between these pathways on muscle composition cannot be determined, but future studies aimed at testing these hypotheses are of interest.

5. Limitations

Our study has several limitations. First, to account for the relatively small sample size (n = 73), validation in a larger cohort is important. Second, when considering the link between muscle composition with BMI (Goodpaster et al., 2000), our findings in the overweight subjects of the present study may not be generalizable to older adults that have a normal weight BMI. Third, all measurements were performed at one time point-studies aimed at examining if these metabolites are markers of longitudinal changes in muscle composition are of interest. Fourth, whether these metabolites influence muscle composition, or muscle composition influences the serum levels of these metabolites, is currently unknown. Alternatively, our findings may be related to variables (e.g. diet) that can influence muscle composition, analytes (e.g. BUN) and metabolites. For example, dietary protein intake is significantly associated with the quantity of lean mass (Houston et al., 2008), and is strongly correlated (r = 0.98) with urea production (Young et al., 2000). However, in disagreement with this hypothesis, although dietary protein intake was significantly associated with NDM/LDM (β ± SE: 0.1 ± 0.0, p = 0.02; n = 66), it was not associated with BUN (β ± SE: 0.8 ± 0.5, p = 0.13), and only 2 of the 35 metabolites that were significantly associated with both NDM/LDM and BUN were associated with protein intake (data not shown).

Supplementary Material

Supplemental

Funding

This work was supported by the Dairy Research Institute, the Boston Claude D. Pepper Older Americans Independence Center (P30-AG013679), DOD contract #W911SR06C0001, and in part by the U.S. Department of Agriculture, under agreement No. 58-1950-4-003. Any opinions, findings, conclusion, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Dept. of Agriculture.

Footnotes

Conflict of interest

The researchers declare no conflict of interest.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.exger.2017.10.003.

References

  1. Affairs, D.o.E.a.S., Division, P., 2009. World Population Ageing. United Nations, New York. [Google Scholar]
  2. Apalset EM, Gjesdal CG, Ueland PM, Midttun O, Ulvik A, Eide GE, et al. , 2014. Interferon (IFN)-gamma-mediated inflammation and the kynurenine pathway in relation to bone mineral density: the Hordaland Health Study. Clin. Exp. Immunol 176(3), 452–460. 10.1111/cei.12288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Badawy AA, 2017. Tryptophan availability for kynurenine pathway metabolism across the life span: control mechanisms and focus on aging, exercise, diet and nutritional supplements. Neuropharmacology 112 (Pt B), 248–263. 10.1016/j.neuropharm.2015.11.015. [DOI] [PubMed] [Google Scholar]
  4. Benjamini Y, Hochberg Y, 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B 57, 289–300. [Google Scholar]
  5. Berman SJ, Johnson EW, Nakatsu C, Alkan M, Chen R, LeDuc J, 2004. Burden of infection in patients with end-stage renal disease requiring long-term dialysis. Clin. Infect. Dis 39 (12), 1747–1753. 10.1086/424516. [DOI] [PubMed] [Google Scholar]
  6. Borkan GA, Hults DE, Gerzof SG, Robbins AH, Silbert CK, 1983. Age changes in body composition revealed by computed tomography. J. Gerontol 38 (6), 673–677. [DOI] [PubMed] [Google Scholar]
  7. Bose SK, Kim H, Meyer K, Wolins N, Davidson NO, Ray R, 2014. Forkhead box transcription factor regulation and lipid accumulation by hepatitis C virus. J. Virol. 88 (8), 4195–4203. 10.1128/JVI.03327-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chale A, Cloutier GJ, Hau C, Phillips EM, Dallal GE, Fielding RA, 2013. Efficacy of whey protein supplementation on resistance exercise-induced changes in lean mass, muscle strength, and physical function in mobility-limited older adults. J. Gerontol. A Biol. Sci. Med. Sci 68 (6), 682–690. 10.1093/gerona/gls221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chen XH, Zhang BW, Li H, Peng XX, 2015. Myo-inositol improves the host’s ability to eliminate balofloxacin-resistant Escherichia coli. Sci. Rep 5, 10720 10.1038/srep10720. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Colonna A, Guadagnino V, Maiorano A, Stamile E, Costa C, 1996. Pseudouridine for monitoring interferon treatment of patients with chronic hepatitis C. Eur. J. Clin. Chem. Clin. Biochem 34 (9), 697–700. [DOI] [PubMed] [Google Scholar]
  11. Coresh J, Astor BC, Greene T, Eknoyan G, Levey AS, 2003. Prevalence of chronic kidney disease and decreased kidney function in the adult US population: Third National Health and Nutrition Examination Survey. Am. J. Kidney Dis. 41 (1), 1–12. 10.1053/ajkd.2003.50007. [DOI] [PubMed] [Google Scholar]
  12. de Almeida Duarte JB, de Aguilar-Nascimento JE, Nascimento M, Nochi RJ Jr., 2004. Bacterial translocation in experimental uremia. Urol. Res 32 (4), 266–270. [DOI] [PubMed] [Google Scholar]
  13. De Simone C, Ferrari M, Lozzi A, Meli D, Ricca D, Sorice F, 1982. Vitamins and immunity: II. Influence of L-carnitine on the immune system. Acta Vitaminol. Enzymol 4 (1–2), 135–140. [PubMed] [Google Scholar]
  14. Dolma T, Mukherjee R, Pati BK, De UK, 2014. Acute phase response and neutrophils: lymphocyte ratio in response to astaxanthin in Staphylococcal mice mastitis model. J. Vet. Med 2014, 147652 10.1155/2014/147652. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dong J, Dong Y, Chen Z, Mitch WE, Zhang L, 2016. The pathway to muscle fibrosis depends on myostatin stimulating the differentiation of fibro/adipogenic progenitor cells in chronic kidney disease. Kidney Int. 10.1016/j.kint.2016.07.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Evans AM, DeHaven CD, Barrett T, Mitchell M, Milgram E, 2009. Integrated, nontargeted ultrahigh performance liquid chromatography/electrospray ionization tandem mass spectrometry platform for the identification and relative quantification of the small-molecule complement of biological systems. Anal. Chem 81 (16), 6656–6667. 10.1021/ac901536h. [DOI] [PubMed] [Google Scholar]
  17. Fabrizi F, Martin P, Dixit V, Brezina M, Cole MJ, Vinson S, et al. , 2000. Biological dynamics of viral load in hemodialysis patients with hepatitis C virus. Am. J. Kidney Dis. 35 (1), 122–129. 10.1016/S0272-6386(00)70310-6. [DOI] [PubMed] [Google Scholar]
  18. Falzarano D, Krokhin O, Van Domselaar G, Wolf K, Seebach J, Schnittler HJ, et al. , 2007. Ebola sGP—the first viral glycoprotein shown to be C-mannosylated. Virology 368 (1), 83–90. 10.1016/j.virol.2007.06.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fiehn O, 2002. Metabolomics—the link between genotypes and phenotypes. Plant Mol. Biol 48 (1–2), 155–171. [PubMed] [Google Scholar]
  20. Gallian P, Biagini P, Attoui H, Cantaloube JF, Dussol B, Berland Y, et al. , 2002. High genetic diversity revealed by the study of TLMV infection in French hemodialysis patients. J. Med. Virol 67 (4), 630–635. 10.1002/jmv.10150. [DOI] [PubMed] [Google Scholar]
  21. Gauna TT, Oshiro E, Luzio YC, Paniago AM, Pontes ER, Chang MR, 2013Bloodstream infection in patients with end-stage renal disease in a teaching hospital in central-western Brazil. Rev. Soc. Bras. Med. Trop 46 (4), 426–432. 10.1590/0037-8682-0060-2013. [DOI] [PubMed] [Google Scholar]
  22. Ghosh S, Lertwattanarak R, Garduno Jde J, Galeana JJ, Li J, Zamarripa F, et al. , 2015. Elevated muscle TLR4 expression and metabolic endotoxemia in human aging. J. Gerontol. A Biol. Sci. Med. Sci 70 (2), 232–246. 10.1093/gerona/glu067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gilbert P, Moore LE, 2005. Cationic antiseptics: diversity of action under a common epithet. J. Appl. Microbiol 99 (4), 703–715. 10.1111/j.1365-2672.2005.02664.x. [DOI] [PubMed] [Google Scholar]
  24. Goncalves S, Pecoits-Filho R, Perreto S, Barberato SH, Stinghen AE, Lima EG, et al. , 2006. Associations between renal function, volume status and endotoxaemia in chronic kidney disease patients. Nephrol. Dial. Transplant 21 (10), 2788–2794. 10.1093/ndt/gfl273. [DOI] [PubMed] [Google Scholar]
  25. Goodpaster BH, Thaete FL, Simoneau JA, Kelley DE, 1997. Subcutaneous abdominal fat and thigh muscle composition predict insulin sensitivity independently of visceral fat. Diabetes 46 (10), 1579–1585. [DOI] [PubMed] [Google Scholar]
  26. Goodpaster BH, Kelley DE, Thaete FL, He J, Ross R, 2000. Skeletal muscle attenuation determined by computed tomography is associated with skeletal muscle lipid content. J. Appl. Physiol. (1985) 89 (1), 104–110. [DOI] [PubMed] [Google Scholar]
  27. Goodpaster BH, Carlson CL, Visser M, Kelley DE, Scherzinger A, Harris TB, et al. , 2001. Attenuation of skeletal muscle and strength in the elderly: the Health ABC Study. J. Appl. Physiol. (1985) 90 (6), 2157–2165. [DOI] [PubMed] [Google Scholar]
  28. Gorisse L, Pietrement C, Vuiblet V, Schmelzer CE, Kohler M, Duca L, et al. , 2016. Protein carbamylation is a hallmark of aging. Proc. Natl. Acad. Sci. U. S. A 113 (5), 1191–1196. 10.1073/pnas.1517096113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Gowda S, Desai PB, Kulkarni SS, Hull VV, Math AA, Vernekar SN, 2010Markers of renal function tests. N. Am. J. Med. Sci 2 (4), 170–173. [PMC free article] [PubMed] [Google Scholar]
  30. Holub M, Beran O, Kaspříková K, Chalupa P, 2012. Neutrophil to lymphocyte count ratio as a biomarker of bacterial infections. Cent. Eur. J. Med 7 (2), 258–261. 10.2478/s11536-012-0002-3. [DOI] [Google Scholar]
  31. Houston DK, Nicklas BJ, Ding J, Harris TB, Tylavsky FA, Newman AB, et al. , 2008. Dietary protein intake is associated with lean mass change in older, community-dwelling adults: the Health, Aging, and Body Composition (Health ABC) Study. Am. J. Clin. Nutr 87 (1), 150–155. [DOI] [PubMed] [Google Scholar]
  32. Hwu P, Du MX, Lapointe R, Do M, Taylor MW, Young HA, 2000. Indoleamine 2,3-dioxygenase production by human dendritic cells results in the inhibition of T cell proliferation. J. Immunol 164 (7), 3596–3599. [DOI] [PubMed] [Google Scholar]
  33. Jirillo E, Altamura M, Marcuccio C, Tortorella C, De Simone C, Antonaci S, 1993. Immunological responses in patients with tuberculosis and in vivo effects of acetyl-L-carnitine oral administration. Mediat. Inflamm 2 (7), S17–20. 10.1155/S0962935193000699. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Joe AW, Yi L, Natarajan A, Le Grand F, So L, Wang J, et al. , 2010. Muscle injury activates resident fibro/adipogenic progenitors that facilitate myogenesis. Nat. Cell Biol. 12 (2), 153–163. 10.1038/ncb2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Johansen KL, Shubert T, Doyle J, Soher B, Sakkas GK, Kent-Braun JA, 2003. Muscle atrophy in patients receiving hemodialysis: effects on muscle strength, muscle quality, and physical function. Kidney Int. 63 (1), 291–297. 10.1046/j.1523-1755.2003.00704.x. [DOI] [PubMed] [Google Scholar]
  36. Koeth RA, Kalantar-Zadeh K, Wang Z, Fu X, Tang WH, Hazen SL, 2013. Protein carbamylation predicts mortality in ESRD. J. Am. Soc. Nephrol 24 (5), 853–861. 10.1681/ASN.2012030254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Koro C, Bielecka E, Dahl-Knudsen A, Enghild JJ, Scavenius C, Brun JG, et al. , 2014. Carbamylation of immunoglobulin abrogates activation of the classical complement pathway. Eur. J. Immunol 44 (11), 3403–3412. 10.1002/eji.201444869. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Koro C, Hellvard A, Delaleu N, Binder V, Scavenius C, Bergum B, et al. , 2016. Carbamylated LL-37 as a modulator of the immune response. Innate Immun. 22 (3), 218–229. 10.1177/1753425916631404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Kraus LM, Kraus AP Jr., 2001. Carbamoylation of amino acids and proteins in uremia. Kidney Int. Suppl 78, S102–107. 10.1046/j.1523-1755.2001.59780102.x. [DOI] [PubMed] [Google Scholar]
  40. Kuro-o M, Matsumura Y, Aizawa H, Kawaguchi H, Suga T, Utsugi T, et al. , 1997. Mutation of the mouse klotho gene leads to a syndrome resembling ageing. Nature 390 (6655), 45–51. 10.1038/36285. [DOI] [PubMed] [Google Scholar]
  41. Laupland KB, Ross T, Gregson DB, 2008. Staphylococcus aureus bloodstream infections: risk factors, outcomes, and the influence of methicillin resistance in Calgary, Canada, 2000–2006. J. Infect. Dis 198 (3), 336–343. 10.1086/589717. [DOI] [PubMed] [Google Scholar]
  42. Lemesch S, Ribitsch W, Schilcher G, Spindelbock W, Hafner-Giessauf H, Marsche G, et al. , 2016. Mode of renal replacement therapy determines endotoxemia and neutrophil dysfunction in chronic kidney disease. Sci. Rep 6, 34534 10.1038/srep34534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Lemos DR, Babaeijandaghi F, Low M, Chang CK, Lee ST, Fiore D, et al. , 2015. Nilotinib reduces muscle fibrosis in chronic muscle injury by promoting TNF-mediated apoptosis of fibro/adipogenic progenitors. Nat. Med 21 (7), 786–794. 10.1038/nm.3869. [DOI] [PubMed] [Google Scholar]
  44. Liu Y, Chen F, Odle J, Lin X, Zhu H, Shi H, et al. , 2013. Fish oil increases muscle protein mass and modulates Akt/FOXO, TLR4, and NOD signaling in weanling piglets after lipopolysaccharide challenge. J. Nutr 143 (8), 1331–1339. 10.3945/jn.113.176255. [DOI] [PubMed] [Google Scholar]
  45. Lustgarten MS, Fielding RA, 2016. Metabolites associated with circulating Interleukin-6 in older adults. J. Gerontol. A Biol. Sci. Med. Sci 10.1093/gerona/glw039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Lustgarten MS, Price LL, Phillips EM, Fielding RA, 2013. Serum glycine is associated with regional body fat and insulin resistance in functionally-limited older adults. PLoS One 8 (12), e84034 10.1371/journal.pone.0084034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Lustgarten MS, Price LL, Chale A, Fielding RA, 2014a. Metabolites related to gut bacterial metabolism, peroxisome proliferator-activated receptor-alpha activation, and insulin sensitivity are associated with physical function in functionally-limited older adults. Aging Cell 13 (5), 918–925. 10.1111/acel.12251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Lustgarten MS, Price LL, Chale A, Phillips EM, Fielding RA, 2014b. Branched chain amino acids are associated with muscle mass in functionally limited older adults. J. Gerontol. A Biol. Sci. Med. Sci 69 (6), 717–724. 10.1093/gerona/glt152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Ma EH, Bantug G, Griss T, Condotta S, Johnson RM, Samborska B, et al. , 2017. Serine is an essential metabolite for effector T cell expansion. Cell Metab 10.1016/j.cmet.2016.12.011. [DOI] [PubMed] [Google Scholar]
  50. Malietzis G, Johns N, Al-Hassi HO, Knight SC, Kennedy RH, Fearon KC, et al. , 2016. Low muscularity and myosteatosis is related to the host systemic inflammatory response in patients undergoing surgery for colorectal cancer. Ann. Surg 263 (2), 320–325. 10.1097/SLA.0000000000001113. [DOI] [PubMed] [Google Scholar]
  51. Mangge H, Schnedl WJ, Schröcksnadel SSG, Murr C, and Fuchs D, 2013. Immune activation and inflammation in patients with cardiovascular disease are associated with elevated phenylalanine-to-tyrosine ratios. Pteridines 24 (1), 51–55. 10.1515/pterid-2013-0002. [DOI] [Google Scholar]
  52. Meyers KJ, Chu J, Mosley TH, Kardia SL, 2010. SNP-SNP interactions dominate the genetic architecture of candidate genes associated with left ventricular mass in African-Americans of the GENOA study. BMC Med. Genet 11, 160 10.1186/1471-2350-11-160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Miljkovic I, Kuipers AL, Cauley JA, Prasad T, Lee CG, Ensrud KE, et al. , 2015. Greater skeletal muscle fat infiltration is associated with higher all-cause and cardiovascular mortality in older men. J. Gerontol. A Biol. Sci. Med. Sci 70 (9), 1133–1140. 10.1093/gerona/glv027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Munn DH, Shafizadeh E, Attwood JT, Bondarev I, Pashine A, Mellor AL, 1999. Inhibition of T cell proliferation by macrophage tryptophan catabolism. J. Exp. Med 189 (9), 1363–1372. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Mydel P, Wang Z, Brisslert M, Hellvard A, Dahlberg LE, Hazen SL, et al. , 2010. Carbamylation-dependent activation of T cells: a novel mechanism in the pathogenesis of autoimmune arthritis. J. Immunol 184 (12), 6882–6890. 10.4049/jimmunol.1000075. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Niewczas MA, Sirich TL, Mathew AV, Skupien J, Mohney RP, Warram JH, et al. , 2014. Uremic solutes and risk of end-stage renal disease in type 2 diabetes: metabolomic study. Kidney Int. 85 (5), 1214–1224. 10.1038/ki.2013.497. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. O’Connor JC, Lawson MA, Andre C, Moreau M, Lestage J, Castanon N, et al. , 2009. Lipopolysaccharide-induced depressive-like behavior is mediated by indoleamine 2,3-dioxygenase activation in mice. Mol. Psychiatry 14 (5), 511–522. 10.1038/sj.mp.4002148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Orellana RA, O’Connor PM, Bush JA, Suryawan A, Thivierge MC, Nguyen HV, et al. , 2006. Modulation of muscle protein synthesis by insulin is maintained during neonatal endotoxemia. Am. J. Physiol. Endocrinol. Metab 291 (1), E159–166. 10.1152/ajpendo.00595.2005. [DOI] [PubMed] [Google Scholar]
  59. Parry HM, Zuo J, Frumento G, Mirajkar N, Inman C, Edwards E, et al. , 2016. Cytomegalovirus viral load within blood increases markedly in healthy people over the age of 70 years. Immun. Ageing 13 (1). 10.1186/s12979-015-0056-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Passler T, Chamorro MF, Riddell KP, Edmondson MA, van Santen E, Cray C, et al. , 2013. Evaluation of methods to improve the diagnosis of systemic inflammation in alpacas. J. Vet. Intern. Med 27 (4), 970–976. 10.1111/jvim.12102. [DOI] [PubMed] [Google Scholar]
  61. Pfefferkorn ER, 1984. Interferon gamma blocks the growth of Toxoplasma gondii in human fibroblasts by inducing the host cells to degrade tryptophan. Proc. Natl. Acad. Sci. U. S. A 81 (3), 908–912. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Phelps M, Stuelsatz P, Yablonka-Reuveni Z, 2016. Expression profile and over-expression outcome indicate a role for betaKlotho in skeletal muscle fibro/adipogenesis. FEBS J. 283 (9), 1653–1668. 10.1111/febs.13682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Pietrement C, Gorisse L, Jaisson S, Gillery P, 2013. Chronic increase of urea leads to carbamylated proteins accumulation in tissues in a mouse model of CKD. PLoS One 8(12), e82506 10.1371/journal.pone.0082506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Rietschel ET, Kirikae T, Schade FU, Mamat U, Schmidt G, Loppnow H, et al. , 1994. Bacterial endotoxin: molecular relationships of structure to activity and function. FASEB J. 8 (2), 217–225. [DOI] [PubMed] [Google Scholar]
  65. Saito K, Fujigaki S, Heyes MP, Shibata K, Takemura M, Fujii H, et al. , 2000. Mechanism of increases in L-kynurenine and quinolinic acid in renal insufficiency. Am. J. Physiol. Renal Physiol 279 (3), F565–572. [DOI] [PubMed] [Google Scholar]
  66. Schiffmann E, Corcoran BA, Wahl SM, 1975. N-formylmethionyl peptides as chemoattractants for leucocytes. Proc. Natl. Acad. Sci. U. S. A 72 (3), 1059–1062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Schmerler D, Neugebauer S, Ludewig K, Bremer-Streck S, Brunkhorst FM,Kiehntopf M, 2012. Targeted metabolomics for discrimination of systemic in-flammatory disorders in critically ill patients. J. Lipid Res. 53 (7), 1369–1375. 10.1194/jlr.P023309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Sekula P, Goek ON, Quaye L, Barrios C, Levey AS, Romisch-Margl W, et al. , 2016. A metabolome-wide association study of kidney function and disease in the general population. J. Am. Soc. Nephrol 27 (4), 1175–1188. 10.1681/ASN.2014111099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Semba RD, Shardell M, Trehan I, Moaddel R, Maleta KM, Ordiz MI, et al. , 2016. Metabolic alterations in children with environmental enteric dysfunction. Sci. Rep 6, 28009 10.1038/srep28009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Semba RD, Trehan I, Li X, Moaddel R, Ordiz MI, Maleta KM, et al. , 2017. Environmental enteric dysfunction is associated with carnitine deficiency and altered fatty acid oxidation. EBioMedicine 17, 57–66. 10.1016/j.ebiom.2017.01.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Serefhanoglu K, Timurkaynak F, Can F, Cagir U, Arslan H, Ozdemir FN, 2012. Risk factors for candidemia with non-albicans Candida spp. in intensive care unit patients with end-stage renal disease on chronic hemodialysis. J. Formos. Med. Assoc 111 (6), 325–332. 10.1016/j.jfma.2011.03.004. [DOI] [PubMed] [Google Scholar]
  72. Shi K, Wang F, Jiang H, Liu H, Wei M, Wang Z, et al. , 2014. Gut bacterial translocation may aggravate microinflammation in hemodialysis patients. Dig. Dis. Sci 59 (9), 2109–2117. 10.1007/s10620-014-3202-7. [DOI] [PubMed] [Google Scholar]
  73. Shimamura Y, Hamada K, Inoue K, Ogata K, Ishihara M, Kagawa T, et al. , 2012. Serum levels of soluble secreted alpha-Klotho are decreased in the early stages of chronic kidney disease, making it a probable novel biomarker for early diagnosis. Clin. Exp. Nephrol 16 (5), 722–729. 10.1007/s10157-012-0621-7. [DOI] [PubMed] [Google Scholar]
  74. Skovbjerg S, Martner A, Hynsjo L, Hessle C, Olsen I, Dewhirst FE, et al. , 2010. Gram-positive and gram-negative bacteria induce different patterns of cytokine production in human mononuclear cells irrespective of taxonomic relatedness. J. Interf. Cytokine Res. 30 (1), 23–32. 10.1089/jir.2009.0033. [DOI] [PubMed] [Google Scholar]
  75. Storey JD, Tibshirani R, 2003. Statistical significance for genomewide studies. Proc. Natl. Acad. Sci. U. S. A 100 (16), 9440–9445. 10.1073/pnas.1530509100. (1530509100 [pii]). [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Szeto CC, Kwan BC, Chow KM, Lai KB, Chung KY, Leung CB, et al. , 2008. Endotoxemia is related to systemic inflammation and atherosclerosis in peritoneal dialysis patients. Clin. J. Am. Soc. Nephrol 3 (2), 431–436. 10.2215/CJN.03600807. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Tanaka H, Sirich TL, Plummer NS, Weaver DS, Meyer TW, 2015. An enlarged profile of uremic solutes. PLoS One 10 (8), e0135657 10.1371/journal.pone.0135657. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Taylor MW, Feng GS, 1991. Relationship between interferon-gamma, indoleamine 2,3-dioxygenase, and tryptophan catabolism. FASEB J. 5 (11), 2516–2522. [PubMed] [Google Scholar]
  79. Thevaranjan N, Puchta A, Schulz C, Naidoo A, Szamosi JC, Verschoor CP, et al. , 2017. Age-associated microbial dysbiosis promotes intestinal permeability, systemic inflammation, and macrophage dysfunction. Cell Host Microbe 21 (4), 455–466. 10.1016/j.chom.2017.03.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. Tilves CM, Zmuda JM, Kuipers AL, Nestlerode CS, Evans RW, Bunker CH, et al. , 2016. Association of lipopolysaccharide-binding protein with aging-related adiposity change and prediabetes among African ancestry men. Diabetes Care 39 (3), 385–391. 10.2337/dc15-1777. [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Uezumi A, Fukada S, Yamamoto N, Takeda S, Tsuchida K, 2010. Mesenchymal progenitors distinct from satellite cells contribute to ectopic fat cell formation in skeletal muscle. Nat. Cell Biol. 12 (2), 143–152. 10.1038/ncb2014. [DOI] [PubMed] [Google Scholar]
  82. Vanhauteghem D, Janssens GP, Lauwaerts A, Sys S, Boyen F, Kalmar ID, et al. , 2012. Glycine and its N-methylated analogues cause pH-dependent membrane damage to enterotoxigenic Escherichia coli. Amino Acids 43 (1), 245–253. 10.1007/s00726-011-1068-y. [DOI] [PubMed] [Google Scholar]
  83. Vaziri ND, Yuan J, Rahimi A, Ni Z, Said H, Subramanian VS, 2012Disintegration of colonic epithelial tight junction in uremia: a likely cause of CKD-associated inflammation. Nephrol. Dial. Transplant 27 (7), 2686–2693. 10.1093/ndt/gfr624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  84. Vaziri ND, Yuan J, Norris K, 2013. Role of urea in intestinal barrier dysfunction and disruption of epithelial tight junction in chronic kidney disease. Am. J. Nephrol 37(1), 1–6. 10.1159/000345969. [DOI] [PMC free article] [PubMed] [Google Scholar]
  85. Vidal B, Serrano AL, Tjwa M, Suelves M, Ardite E, De Mori R, et al. , 2008. Fibrinogen drives dystrophic muscle fibrosis via a TGFbeta/alternative macrophage activation pathway. Genes Dev. 22 (13), 1747–1752. 10.1101/gad.465908. [DOI] [PMC free article] [PubMed] [Google Scholar]
  86. Visser M, Goodpaster BH, Kritchevsky SB, Newman AB, Nevitt M, Rubin SM, et al. , 2005. Muscle mass, muscle strength, and muscle fat infiltration as predictors of incident mobility limitations in well-functioning older persons. J. Gerontol. A Biol. Sci. Med. Sci 60 (3), 324–333. [DOI] [PubMed] [Google Scholar]
  87. Wang R, Dillon CP, Shi LZ, Milasta S, Carter R, Finkelstein D, et al. , 2011. The transcription factor Myc controls metabolic reprogramming upon T lymphocyte activation. Immunity 35 (6), 871–882. 10.1016/j.immuni.2011.09.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  88. Wang Z, DiDonato JA, Buffa J, Comhair SA, Aronica MA, Dweik RA, et al. , 2016. Eosinophil peroxidase catalyzed protein carbamylation participates in asthma. J. Biol. Chem 291 (42), 22118–22135. 10.1074/jbc.M116.750034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  89. Wehling-Henricks M, Li Z, Lindsey C, Wang Y, Welc SS, Ramos JN, et al. , 2016. Klotho gene silencing promotes pathology in the mdx mouse model of Duchenne muscular dystrophy. Hum. Mol. Genet 10.1093/hmg/ddw111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  90. Weinstein L, Mc DA, 1946. The action of urea and some of its derivatives on bacteria; bacteriostatic and bactericidal effects of urea and urethane. J. Immunol 54, 117–130. [PubMed] [Google Scholar]
  91. Wessa P, 2015. Box-Cox Normality Plot (v1.1.11) in Free Statistics Software (v1.1.23-r7). Office for Research Development and Education.
  92. Wong J, Piceno YM, Desantis TZ, Pahl M, Andersen GL, Vaziri ND, 2014. Expansion of urease- and uricase-containing, indole- and p-cresol-forming and contraction of short-chain fatty acid-producing intestinal microbiota in ESRD. Am. J. Nephrol 39 (3), 230–237. 10.1159/000360010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  93. Xu ML, Wi GR, Kim HJ, Kim HJ, 2016. Ameliorating effect of dietary xylitol on human respiratory syncytial virus (hRSV) infection. Biol. Pharm. Bull 39 (4), 540–546. 10.1248/bpb.b15-00773. [DOI] [PubMed] [Google Scholar]
  94. Yang M, Li L, Su N, Lin J, Wang J, 2015. Dynamic monitoring of the neutrophil/lymphocyte ratio could predict the prognosis of patients with bloodstream infection. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue 27 (6), 471–476. 10.3760/cma.j.issn.2095-4352.2015.06.011. [DOI] [PubMed] [Google Scholar]
  95. Yasui H, Takai K, Yoshida R, Hayaishi O, 1986. Interferon enhances tryptophan metabolism by inducing pulmonary indoleamine 2,3-dioxygenase: its possible occurrence in cancer patients. Proc. Natl. Acad. Sci. U. S. A 83 (17), 6622–6626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  96. Yoshida R, Urade Y, Tokuda M, Hayaishi O, 1979. Induction of indoleamine 2,3-dioxygenase in mouse lung during virus infection. Proc. Natl. Acad. Sci. U. S. A 76 (8), 4084–4086. [DOI] [PMC free article] [PubMed] [Google Scholar]
  97. Yoshida R, Imanishi J, Oku T, Kishida T, Hayaishi O, 1981. Induction of pulmonary indoleamine 2,3-dioxygenase by interferon. Proc. Natl. Acad. Sci. U. S. A 78 (1), 129–132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  98. Young VR, El-Khoury AE, Raguso CA, Forslund AH, Hambraeus L, 2000. Rates of urea production and hydrolysis and leucine oxidation change linearly over widely varying protein intakes in healthy adults. J. Nutr 130 (4), 761–766. [DOI] [PubMed] [Google Scholar]
  99. Zahorec R, 2001. Ratio of neutrophil to lymphocyte counts—rapid and simple parameter of systemic inflammation and stress in critically ill. Bratisl. Lek. Listy 102 (1), 5–14. [PubMed] [Google Scholar]
  100. Zhang L, Wang XH, Wang H, Du J, Mitch WE, 2010. Satellite cell dysfunction and impaired IGF-1 signaling cause CKD-induced muscle atrophy. J. Am. Soc. Nephrol 21(3), 419–427. 10.1681/ASN.2009060571. [DOI] [PMC free article] [PubMed] [Google Scholar]
  101. Zhang LJ, Guerrero-Juarez CF, Hata T, Bapat SP, Ramos R, Plikus MV, et al. , 2015. Innate immunity. Dermal adipocytes protect against invasive Staphylococcus aureus skin infection. Science 347 (6217), 67–71. 10.1126/science.1260972. [DOI] [PMC free article] [PubMed] [Google Scholar]
  102. Zilberberg MD, Shorr AF, Kollef MH, 2008. Secular trends in candidemia-related hospitalization in the United States, 2000–2005. Infect. Control Hosp. Epidemiol 29(10), 978–980. 10.1086/591033. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental
NIHMS966603-supplement-Supplemental.docx (71.8KB, docx)

RESOURCES