Key words
Cathelicidin antimicrobial peptide - diet - bariatric surgery - adipose tissue - adipocyte
Introduction and Aims
Antimicrobial peptides (AMPs) as part of the innate immune system are able to kill
bacteria directly via membrane permeation and thus act somewhat like a molecular
hole puncher [1]. However, AMPs have been
increasingly recognized as pleiotropic and immunomodulating molecules regulating
cell migration, wound healing, proliferation, differentiation, and angiogenesis.
CAMP (Cathelicidin antimicrobial peptide, also named LL-37 or CAP18) represents the
only human member of the Cathelicidin family and it is expressed in activated cells
and organs such as epithelial cells, keratinocytes, neutrophils, monocytes, mast
cells, skin, gastrointestinal and genito-urinary tract [1]. CAMP is composed of an N-terminal signal
peptide, a conserved Cathelin-like domain, and a C-terminal antimicrobial domain
that forms the mature antimicrobial peptide after cleavage by proteinase-3 [2]. CAMP has been regarded as a peptide of the
innate immune system linking host defense and inflammation with angiogenesis and
atherosclerosis [3].
Of note, Zhang et al. [4] reported for the
first time in 2015 a fascinating link between CAMP and adipose tissue defense
against Staphylococcus aureus (S. aureus). The authors demonstrated
that proliferation and expansion of adipose tissue occurs after infection of
subdermal adipose tissue by S. aureus
[4]. Interestingly, this type of infection was more severe in murine
models with impaired adipogenesis [4].
Inhibition of adipogenesis decreased CAMP expression and CAMP knockout mice had a
decreased ability to fight against S. aureus growth [4]. Taken together, host defense against S.
aureus was mediated by CAMP secreted from adipocytes [4]. A subsequent review article by Alcorn and
Kolls commented on these results by using the term “killer
fat”
[5]. Gram positive
bacteria such S. aureus and their respective lipopeptides are sensed and
recognized by the toll-like receptors TLR-1/-2 and TLR-2/-6. Most
recently, we could demonstrate that the TLR-2-CAMP-pathway is expressed and highly
inducible in adipocytes [6]. During hormonally
induced differentiation of murine 3T3-L1 pre-adipocytes into mature adipocytes,
TLR-2 and CAMP were significantly upregulated in parallel. In mature adipocytes and
also in pre-adipocytes, CAMP and TLR-2 expression were significantly increased upon
stimulation with the TLR-1/-2 agonist Pam3Cys and the TLR-2/-6
agonist MALP-2, respectively [6]
[7]. Taken together, CAMP is a
differentiation-dependent secretory peptide of adipocytes fighting against adipose
tissue infections. Thus, the regulation of CAMP in the context of obesity and during
weight loss is of clinical interest.
It was our aim to
-
quantify the serum levels of human CAMP in obese patients undergoing either
low calorie diet (LCD) or bariatric surgery (BS) before and after weight
loss in two large and well-characterized clinical cohorts
-
to investigate CAMP gene expression in subcutaneous and in visceral adipose
tissue of patients undergoing bariatric surgery
-
to correlate serum concentrations and gene expression levels with
anthropometric and biochemical parameters
-
to investigate the effects of metabolic factors and sex hormones on CAMP
expression in vitro in adipocytes
Material and Methods
Serum samples and specimens from subcutaneous (abdominal) and visceral
(intra-abdominal) adipose tissue were collected from the ROBS (Research in Obesity
and Bariatric Surgery) study cohort. ROBS is an open-label, non-randomized,
monocentric, prospective and observational (explorative and confirmatory) study of
patients routinely undergoing either bariatric surgery (gastric sleeve or Roux-en-Y
gastric bypass) or a low calorie formula diet (LCD) in the tertiary care center at
the University of Giessen, Germany. The detailed information about this study cohort
can be drawn from a recent publication [8].
Briefly, patients were treated by a multidisciplinary team of physicians and
professionals from Internal Medicine, Endocrinology/Diabetology,
Metabolic/Visceral Surgery, Psychosomatic Medicine/Psychotherapy,
Nutritional Science/Dietetics, and Sports Medicine at the Obesity Centre at
the University of Giessen, Germany. The study was approved by the local ethical
committee at the University of Giessen, Germany (AZ 101/14). All patients
gave informed consent and were informed about the aim of the study. Data
anonymization and privacy policy were accurately applied. Obese patients with a BMI
>40 kg/m2 or with a BMI
>35 kg/m2 and coexisting type 2
diabetes were consecutively admitted for bariatric surgery from January 2015 to
February 2020. Exclusion criteria were: pregnancy, evidence of or suspicion on
underlying endocrine diseases, untreated bulimia nervosa and binge eating behavior,
use of illicit drugs, neoplasm, severe psychiatric disorders, psychosis, and
psychopathologic instability.
Serum measurement of CAMP concentrations
The concentration of serum CAMP in each sample was measured in duplicates by
ELISA (Hycultec GmbH, Beutelsbach, Germany) and is expressed as
mean±standard deviation. The intra- and inter-assay variation
coefficient of the ELISA was <20%. The lower detection limit was
100 pg/ml.
Adipocyte cell culture and stimulation experiments
Murine 3T3-L1-pre-adipocytes [9] were
cultured at 37°C and 5% CO2 in DMEM
(Dulbecco’s Modified Eagle Medium, Biochrom AG, Berlin, Germany) that
was supplemented with 10% newborn calf serum (NCS; Sigma-Aldrich,
Deisenhofen, Germany) and 1% penicillin/streptomycin (PAN,
Aidenbach, Germany). The cells were then differentiated into mature adipocytes
at confluence by DMEM/F12/glutamate-medium (Lonza, Basel,
Switzerland) supplemented with 20 μM 3-isobutylmethylxanthine
(Serva, Heidelberg, Germany), 1 μM corticosterone,
100 nM insulin, 200 μM ascorbate,
2 μg/ml apo-transferrin, 5% fetal calf serum
(FCS), 1 μM biotin, 17 μM pantothenate (all from
Sigma Aldrich, Deisenhofen, Germany), 1%
penicillin/streptomycin, and 300 μg/ml
Pedersen-fetuin (MP Biomedicals, Illkirch, France) [10]
[11] for 9 days using a slightly modified protocol as reported in the
literature [9]
[12]
[13]
[14]
[15]. The cells were supplied with
adipogenesis-inducing medium from day 0 until day 8 of differentiation, when
cell culture medium was switched to serum-free conditions (with elevation of
insulin supplementation to a concentration of 1 μM). At day 9,
mature adipocytes were adjusted to serum-free
DMEM/F12/glutamate-medium lacking insulin for
3–5 h. With the start of stimulation experiments, cells were
supplied with fresh serum-free medium. Cell phenotype was controlled by
light-microscopy (appearance of extensive accumulation of lipid droplets).
Mature adipocytes at day 9 of differentiation were used for stimulation
experiments following overnight incubation under serum-free culture conditions.
Mature adipocytes were cultured in serum-free media containing differing glucose
(5.6 and 25 mM) and insulin doses (0.2 and 2 nM) for
18 h. Further stimulation experiments were conducted with the sex
hormones estradiol and testosterone (0.5 and 2 μM, each) and
with GLP-1 (100 nM), all purchased by Sigma-Aldrich, Germany. Estradiol
was dissolved in H2O and testosterone was dissolved in ethanol. GLP-1
was dissolved in H2O. The following bile acids were tested and
purchased from Sigma-Aldrich, Germany: tauromuricholic acid (TMCA; 1 and
10 μM), taurohyodeoxycholic acid (THDCA; 10 μM),
cholic acid (CA; 10 and 100 μM), deoxycholic acid (DCA; 1 and
10 μM), and taurodeoxycholic acid (TDCA; 1 and
10 μM). Experiments were performed in mature adipocytes
overnight (18 h) under serum-free culture conditions. LDH (lactate
dehydrogenase) concentration was measured in the supernatants of all stimulation
experiments (Cytotoxicity Detection Kit, Roche, Mannheim, Germany) in order to
exclude any unexpected cytotoxic effects.
Isolation of mRNA from human tissues
Total mRNA was isolated from frozen human subcutaneous and visceral adipose
tissue. Human subcutaneous and visceral adipose tissue was resected in obese
patients undergoing bariatric surgery and participating in the ROBS study cohort
[8]. Gene expression was quantified by
reverse transcription of 300 ng RNA (QuantiTect Reverse Transcription
Kit from Qiagen, Hilden, Germany) and subsequent real-time PCR (RT-PCR) (iTaq
Universal SYBR Green Supermix, CFX Connect RT-PCR system; Bio-Rad, Munich,
Germany) of the corresponding cDNA as mentioned below in detail.
Quantitative real-time PCR analysis of human and murine mRNA
expression
Gene expression of human CAMP during hormonally induced differentiation of 3T3-L1
pre-adipocytes as well as in murine and human adipose tissue depots was
quantified by reverse transcription of isolated RNA and subsequent RT-PCR. The
following primer sequences were used:
Murine Cathelicidin:
5′-GTGGCTGGCCTGGAGAAGAT-3′/5′-TTGTCATCTACGGGCACAAAG-3′
Human Cathelicidin:
5′-TAGATGGCATCAACCAGCGG-3′/5′-CTGGGTCCCCATCCATCGT-3′
Expression levels of the target gene were normalized to the gene expression of
GAPDH as a house keeping gene as done by other experienced groups [16]. The primer-pairs used were:
5′-TGTCCGTCGTGGATCTGAC-3′/5′-AGGGAGATGCTCAGTGTTGG-3′
(mouse),
5′-GAGTCCACTGGCGTCTTCAC-3′/5′-CCAGGGGTGCTAAGCAGTT-3′
(human). All oligonucleotides used were purchased from Metabion (Martinsried,
Germany).
Data base and statistical analysis
For explorative data analysis, a statistical software package (SPSS 22.0) was
used. CAMP serum concentrations and gene expression levels did not follow a
Gaussian distribution. Non-parametric numerical parameters were analyzed by the
Mann–Whitney U-test (for 2 unrelated samples), the
Kruskal–Wallis test (>2 unrelated samples), the Wilcoxon test
(for 2 related samples) or the Friedman test (>2 related samples).
Correlation analysis was performed by using the Spearman test. Partial
correlation analysis was applied to control for possible covariates. A p-value
below 0.05 (two tailed) was considered as statistically significant. In the
figures, the bars are showing the mean values and the whiskers are giving the
SEM (standard error of the mean).
Results
Serum CAMP concentrations in severely obese patients undergoing either a low
calorie formula diet (LCD) or bariatric surgery (BS)
Since reference values of human serum CAMP concentrations have not yet been
evaluated in the literature systematically, basal CAMP concentrations were
measured in two large cohorts of patients undergoing either BS (n=156;
122 females, 34 males; mean age: 40.1±11.0 years; mean BMI:
53.6±6.6 kg/m2; BMI range:
40.9–83.7 kg/m2) or LCD (n=79; 51
females, 28 males; mean age: 42.0+12.1 years; mean BMI:
44.0±5.5 kg/m2; BMI range:
31.9–59.2 kg/m2). The detailed study
characteristics can be drawn from a recent publication [8]. [Table
1] summarizes mean values±SEM/SD, median, range,
variance and gender-specific concentrations of CAMP at a glance. The maximum
range of CAMP over the two cohorts was 9.15–90.37 ng/ml.
Females had significantly lower basal CAMP concentrations when compared to males
in both cohorts (*p=0.033 and °p=0.006,
respectively). We could demonstrate that CAMP serum levels during weight loss
were strongly influenced by the clinical approach of weight reduction. The time
course of CAMP serum concentrations over the time is depicted in [Fig. 1a, b]. In patients during LCD ([Fig. 1a]), CAMP concentrations remained
unchanged over the time. In contrast, CAMP levels increased significantly and in
a stepwise manner (p<0.001) from V0 to V3 ([Fig. 1b]) after BS. Between V3 and V12, a
significant decline occurred (p=0.02) but levels at V12 remained
increased when compared to V0 pre-study levels (p<0.001).
Fig. 1 Serum CAMP concentrations during weight loss in patients
undergoing low calorie diet (panel a) or bariatric surgery (panel
b). Serum CAMP concentrations (ng/ml) were measured by ELISA
over 12 months. CAMP: Cathelicidin antimicrobial peptide; n.s.: Not
significant; V: Clinical visits (months). For statistical analysis, the
Friedman test was used.
Table 1 Explorative data analysis of CAMP concentrations in
human sera of patients undergoing either bariatric surgery (BS) or
low calorie diet (LCD).
|
CAMP (ng/ml)
|
Bariatric surgery (BS) (n=156)
Roux-en-Y
gastric bypass 121 (77.6%)
Sleeve
gastrectomy 35 (22.4%)
|
Low calorie diet (LCD) (n=79)
|
|
V0 (n=156)
|
V12 (n=91)
|
V0 (n=79)
|
V12 (n=66)
|
|
Mean±SEM
|
39.48±1.10
|
46.02±1.66
|
42.35±3.64
|
29.31±1.65
|
|
SD
|
13.76
|
15.81
|
32.37
|
13.42
|
|
Median
|
37.25
|
42.70
|
30.44
|
26.40
|
|
Range
|
14.35–91.04
|
9.61–98.69
|
10.91–190.37
|
9.15–94.16
|
|
Variance
|
189.47
|
249.94
|
1047.55
|
180.04
|
|
Mean±SEM females
|
38.25±1.18* (n=122)
|
44.39±1.76 (n=69)
|
37.62±4.55° (n=51)
|
28.49±1.61 (n=46)
|
|
Mean±SEM males
|
43.91±2.68* (n=34)
|
51.15±3.94 (n=22)
|
50.97±5.84° (n=28)
|
31.20±4.05 (n=20)
|
CAMP was measured by ELISA. V: Visit; V0: Prior to BS or LCD; V12: 12
Months after BS or LCD; SEM: Standard error of the mean; SD: Standard
deviation. A sum of 121 (77.6%) patients underwent Roux-en-Y
bastric bypass and 35 (22.4%) underwent sleeve gastrectomy.
Females had significantly lower basal CAMP concentrations when compared
to males in both cohorts (*p=0.033 and
°p=0.006 respectively; Mann–Whitney U-test).
Statistical analysis (Friedman test) and graphical illustration of CAMP
concentrations in those patients who could completely finish all visits
until month 12 are shown in [Fig.
1a], b.
Correlation analysis of CAMP serum concentrations with weight loss associated
changes in anthropometric and clinical parameters
We aimed to test whether changes in anthropometric and clinical parameters were
correlated to basal CAMP levels as well as to changes of CAMP levels (Δ
CAMP) over the time. At first, basal CAMP levels were correlated with changes in
anthropometric and clinical parameters upon weight loss after 12 months in the
subgroups of BS and LCD ([Table 2], panel
a). There was a positive correlation (p=0.025;
rho=+ 0.252) of ΔHbA1c levels with
basal CAMP concentrations in BS patients but not in LCD patients. Next, changes
of CAMP concentrations (Δ CAMP) over the time were correlated with
changes in anthropometric and clinical parameters ([Table 2], panel b). There was a
positive correlation of Δ CAMP with Δ body fat (%) in BS
patients and with Δ body weight in LCD patients.
Table 2 Correlation analysis of basal CAMP serum
concentrations with weight loss associated changes in anthropometric
and clinical parameters (a). Correlation analysis of weight
loss associated changes in CAMP serum concentrations with
anthropometric/clinical parameters (b).
|
Correlations with CAMP levels
|
Bariatric surgery (BS) (n=91)
|
Low calorie diet (LCD) (n=66)
|
|
Parameter
|
n
|
rho
|
p
|
n
|
rho
|
p
|
|
a:
|
|
Δ body weight (kg)
|
87
|
+0.021
|
0.845
|
65
|
−0.068
|
0.590
|
|
Δ BMI (kg/m2)
|
87
|
+0.100
|
0.356
|
65
|
+0.008
|
0.947
|
|
Δ body fat (%)
|
80
|
+0.111
|
0.327
|
64
|
−0.054
|
0.675
|
|
Δ WHR
|
72
|
+0.200
|
0.092
|
59
|
+0.184
|
0.164
|
|
Δ HbA1c (%)
|
79
|
+0.252
|
0.025*
|
64
|
+0.109
|
0.389
|
|
Δ Total cholesterol (mg/dl)
|
81
|
–0.084
|
0.458
|
64
|
+0.173
|
0.173
|
|
Δ LDL cholesterol (mg/dl)
|
81
|
+0.108
|
0.336
|
64
|
+0.196
|
0.121
|
|
Δ HDL cholesterol (mg/dl)
|
81
|
−0.029
|
0.796
|
64
|
+0.064
|
0.617
|
|
Δ Triglycerides (mg/dl)
|
81
|
−0.032
|
0.778
|
64
|
−0.049
|
0.703
|
|
Δ CRP (mg/dl)
|
86
|
+0.003
|
0.975
|
64
|
+0.001
|
0.991
|
|
b:
|
|
Δ body weight (kg)
|
87
|
+0.027
|
0.801
|
65
|
+0.256
|
0.039*
|
|
Δ BMI (kg/m2)
|
87
|
−0.010
|
0.928
|
65
|
+0.222
|
0.075
|
|
Δ body fat (%)
|
80
|
+0.277
|
0.013*
|
64
|
+0.232
|
0.065
|
|
Δ WHR
|
72
|
+0.008
|
0.950
|
59
|
+0.136
|
0.303
|
|
Δ HbA1c (%)
|
79
|
−0.129
|
0.258
|
64
|
+0.026
|
0.838
|
|
Δ Total cholesterol (mg/dl)
|
81
|
+0.081
|
0.473
|
64
|
+0.114
|
0.371
|
|
Δ LDL cholesterol (mg/dl)
|
81
|
+0.016
|
0.890
|
64
|
+0.066
|
0.602
|
|
Δ HDL cholesterol (mg/dl)
|
81
|
−0.113
|
0.314
|
64
|
−0.071
|
0.580
|
|
Δ Triglycerides (mg/dl)
|
81
|
+0.155
|
0.168
|
64
|
+0.109
|
0.392
|
|
Δ CRP (mg/dl)
|
86
|
+0.001
|
0.999
|
64
|
+0.197
|
0.119
|
Basal CAMP levels were correlated with changes in anthropometric and
clinical parameters upon weight loss after 12 months in the subgroups of
BS (n=91) and LCD (n=66) subgroups. Changes (Δ)
between V0 and V12 were correlated with basal CAMP concentrations by
using the Spearman test and significant correlations are indicated by
* (p<0.05). Changes in systemic CAMP
levels (Δ CAMP) as well as in anthropometric and clinical
parameters upon weight loss after 12 months were correlated by using the
Spearman test. Significant correlations are indicated by
* (p<0.05).
Correlation analysis of CAMP mRNA expression in subcutaneous and visceral
adipose tissue with anthropometric and biochemical parameters
In our large cohort of patients undergoing BS (n=156), we had the
possibility to analyze mRNA expression quantitatively in subcutaneous and
visceral adipose tissue. At first, basal CAMP serum levels prior to surgery were
correlated with clinical parameters in all BS patients ([Table 3], panel a). Basal CAMP
levels were positively correlated with resistin (p=0.003;
rho=+ 0.242) and triglycerides (p=0.005;
rho=+ 0.239), and negatively with HDL cholesterol
(p=0.038; rho=− 0.176). In subcutaneous adipose
tissue ([Table 3], panel b), CAMP
expression was negatively correlated with adiponectin (p=0.003;
rho=− 0.24). In visceral adipose tissue ([Table 3], panel c), CAMP expression
was positively correlated with BMI (p=0.02;
rho=+ 0.186), HbA1c (p=0.031;
rho=+ 0.185), and triglycerides (p=0.026;
rho=+ 0.188). Taken together, CAMP serum levels
and/or CAMP expression in adipose tissue correlate positively with
unfavorable metabolic factors (BMI, HbA1c, triglycerides, resistin)
and negatively with favorable factors (HDL cholesterol, adiponectin).
Table 3 Correlation analysis of basal CAMP serum
concentrations (a), mRNA expression levels in subcutaneous
(b), and visceral (c) adipose tissue with basal
anthropometric and clinical parameters.
|
a:
|
|
Serum CAMP
|
Bariatric surgery (BS) (n=156)
|
|
Parameter
|
n
|
rho
|
p
|
|
Age
|
156
|
−0.085
|
0.292
|
|
Body weight
|
156
|
+0.145
|
0.071
|
|
BMI
|
156
|
+0.025
|
0.754
|
|
Body fat (%)
|
131
|
−0.089
|
0.314
|
|
WHR
|
135
|
+0.053
|
0.541
|
|
HbA1c
|
136
|
−0.100
|
0.245
|
|
Total cholesterol
|
139
|
+0.004
|
0.960
|
|
LDL cholesterol
|
139
|
+0.041
|
0.631
|
|
HDL cholesterol
|
139
|
−0.176
|
0.038*
|
|
Triglycerides
|
139
|
+0.239
|
0.005*
|
|
CRP
|
155
|
−0.059
|
0.462
|
|
Adiponectin
|
150
|
−0.062
|
0.452
|
|
Leptin
|
150
|
−0.076
|
0.353
|
|
Resistin
|
150
|
+0.242
|
0.003*
|
|
b:
|
|
CAMP expression in subcutaneous AT
|
Bariatric surgery (BS) (n=156)
|
|
Parameter
|
n
|
rho
|
p
|
|
Age
|
156
|
−0.006
|
0.944
|
|
Body weight
|
156
|
+0.038
|
0.166
|
|
BMI
|
156
|
+0.090
|
0.264
|
|
Body fat (%)
|
131
|
−0.053
|
0.549
|
|
WHR
|
135
|
+0.268
|
0.002
|
|
HbA1c
|
136
|
+0.096
|
0.266
|
|
Total cholesterol
|
139
|
−0.096
|
0.261
|
|
LDL cholesterol
|
139
|
−0.021
|
0.808
|
|
HDL cholesterol
|
139
|
−0.165
|
0.052
|
|
Triglycerides
|
139
|
+0.120
|
0.160
|
|
CRP
|
155
|
+0.072
|
0.373
|
|
Adiponectin
|
150
|
−0.240
|
0.003*
|
|
Leptin
|
150
|
−0.160
|
0.050
|
|
Resistin
|
150
|
−0.101
|
0.219
|
|
c:
|
|
CAMP expression in visceral AT
|
Bariatric surgery (BS) (n=156)
|
|
Parameter
|
n
|
rho
|
p
|
|
Age
|
156
|
+0.015
|
0.855
|
|
Body weight
|
156
|
+0.110
|
0.173
|
|
BMI
|
156
|
+0.186
|
0.020*
|
|
Body fat (%)
|
131
|
+0.014
|
0.876
|
|
WHR
|
135
|
−0.077
|
0.375
|
|
HbA1c
|
136
|
+0.185
|
0.031*
|
|
Total cholesterol
|
139
|
−0.001
|
0.995
|
|
LDL cholesterol
|
139
|
−0.041
|
0.635
|
|
HDL cholesterol
|
139
|
−0.036
|
0.672
|
|
Triglycerides
|
139
|
+0.188
|
0.026*
|
|
CRP
|
155
|
+0.045
|
0.582
|
|
Adiponectin
|
150
|
−0.057
|
0.487
|
|
Leptin
|
150
|
0
|
1.000
|
|
Resistin
|
150
|
−0.017
|
0.837
|
CAMP serum levels and mRNA expression in subcutaneous and visceral
adipose tissue were correlated with anthropometric and clinical
parameters in BS patients (n=156). Significant correlations are
indicated by * (p<0.05). The Spearman test
was used. AT: Adipose tissue; BMI: Body mass index; CRP: C-reactive
protein; WHR: Waist-hip ratio.
CAMP gene expression is higher in subcutaneous adipose tissue when compared
to visceral adipose tissue but not related to serum CAMP concentrations
Since data on CAMP gene expression in human adipose tissue compartments are
scarce, we analyzed gene expression by quantitative real time PCR in
subcutaneous and visceral adipose tissue of a large cohort of patients
undergoing BS (n=156). As depicted in [Fig. 2a], CAMP gene expression was significantly higher
(p<0.001) in subcutaneous adipose tissue (~about 20%).
Moreover, we aimed to investigate whether adipose tissue gene expression is
directly correlated to the respective serum CAMP concentration in these
patients. As shown in [Fig. 2b,c],
neither subcutaneous nor visceral adipose tissue gene expression was directly
correlated with serum CAMP levels.
Fig. 2 Gene expression analysis of CAMP in adipose tissue
compartments. The mRNA expression of CAMP was quantified by real time
PCR (in relation to GAPDH) in subcutaneous (sc) and visceral (vis)
adipose tissue (AT) of n=156 patients undergoing bariatric
surgery (BS). In parallel, serum CAMP concentrations were measured by
ELISA in these patients prior to surgery. GAPDH,
glyceraldehyde-3-phosphate dehydrogenase. Panel a: CAMP gene
expression is higher in subcutaneous than in visceral adipose tissue
(p<0.001; Wilcoxon test). Panel b: Subcutaneous adipose
tissue CAMP expression is not a predictor of serum CAMP concentrations
(Spearman test). Panel c: Visceral adipose tissue CAMP expression
is not a predictor of serum CAMP concentrations (Spearman test). Panel
d: Gender-specific analysis of CAMP gene expression in
adipose tissue compartments. In contrast to visceral adipose tissue,
CAMP gene expression is lower in females when compared to males
(p=0.01) in subcutaneous adipose tissue (Mann–Whitney U
test).
Females have lower CAMP gene expression levels than males in subcutaneous
adipose tissue but not in visceral adipose tissue
Since females are characterized by lower serum CAMP concentrations ([Table 1]), we intended to clarify whether
there exist gender-specific effects of gene expression in adipose tissue
compartments. In contrast to visceral adipose tissue, females were characterized
by significantly lower (p=0.01) CAMP gene expression levels
(~about 20%) in subcutaneous adipose tissue ([Fig. 2d]).
Effect of glucose and insulin concentrations on CAMP expression in vitro in
differentiated 3T3-L1 adipocytes
Since glucose and insulin concentrations are crucial metabolic parameters in the
context of obesity, we aimed to test their effects on CAMP expression in
adipocytes in vitro under standardized conditions. [Fig. 3a] summarizes the effects of varying
glucose and insulin concentrations on CAMP expression in differentiated 3T3-L1
adipocytes. Supraphysiologic glucose concentrations reduced CAMP expression by a
nonsignificant trend when compared to cells exposed to low/normal
glucose concentrations. Insulin in the higher dose also reduced CAMP expression,
although this reduction closely missed statistical significance. However, the
combined exposure to high glucose and high insulin concentrations showed the
most impressive effect regarding the suppression of CAMP expression
(p=0.004) by ~about 50%. These data might argue for a
synergistic inhibitory effect of glucose and insulin on CAMP expression in
adipocytes and for a potential impact of hyperglycemic hyperinsulinemia (as seen
in type 2 diabetes and obesity) on the regulation of CAMP expression.
Fig. 3 Regulation of CAMP gene expression in vitro in
differentiated 3T3-L1 adipocytes. CAMP mRNA was measured by quantitative
real-time PCR. CAMP, Cathelicidin anti-microbial peptide; GAPDH,
glyceraldehyde-3-phosphatedehydrogenase. Panel a: High
concentrations of glucose and insulin synergistically suppress the gene
expression of CAMP in adipocytes in vitro. Differentiated 3T3-L1
adipocytes were incubated under varying glucose and insulin
concentrations. Cells were exposed to low/normal (low Glc/LGlc) glucose
concentrations (5.6 mM) and to supraphysiologic (high Glc/HGlc)
glucose concentrations (25 mM) alone or in combination with low
(0.2 nM) and high (2.0 nM) insulin (Ins) concentrations
for 18 h. n=10–12 wells were investigated per
experimental setting. The combined exposure to high glucose and high
insulin concentrations significantly reduced CAMP gene expression
(p=0.004; Kruskal–Wallis test). Panel b: Effects
of bile acids on CAMP expression in adipocytes in vitro. Differentiated
3T3-L1 adipocytes were exposed to a broad panel of bile acids.
Tauromuricholic acid (TMCA; 1 and 10 μM),
taurohyodeoxycholic acid (THDCA; 10 μM), cholic acid
(CA; 10 and 100 μM), deoxycholic acid (DCA; 1 and
10 μM), and taurodeoxycholic acid (TDCA; 1 and
10 μM) were investigated. °p<0.05;
* p<0.01 (Kruskal–Wallis test). n=6
wells were investigated per experimental setting. Panel c: Effect
of GLP-1 on CAMP expression in adipocytes in vitro. GLP-1 significantly
(p=0.007) inhibited CAMP gene expression (Kruskal–Wallis
test). GLP-1: Glucagon-like peptide-1; Ctrl.: Control.
n=8–12 wells were investigated per experimental setting.
Panel d: Effects of testosterone and estradiol on CAMP expression
in adipocytes in vitro. Neither testosterone nor estradiol significantly
(Kruskal–Wallis test) influenced CAMP expression. Test.:
Testosterone; Est.: Estradiol; n=6–12 wells were
investigated per experimental setting.
Effects of bile acids and GLP-1 on the regulation of CAMP expression in
differentiated adipocytes in vitro
Since the increase of CAMP concentrations during weight loss could exclusively be
documented after bariatric surgery ([Fig.
1b]) but not under low calorie diet ([Fig. 1a]), effects independent of weight loss but induced by the
surgical procedure itself might account for this observation. In this context,
it seems important to remember that an elevation of systemic bile acids has been
regularly reported after bariatric surgery [17]
[18]. Thus, we aimed to test
the hypothesis that bile acids might regulate CAMP expression directly in
adipocytes in vitro. [Fig. 3b] summarizes
the effects of different bile acids on CAMP expression in adipocytes. TMCA and
THDCA significantly induced CAMP expression (p<0.05 and p<0.01,
respectively) whereas CA, DCA, and TDCA had no effect. Thus, systemic bile acids
might represent potential mediators standing behind the increasing CAMP levels
after BS.
Incretin hormones such as GLP-1 and others are commonly increased after BS [19]
[20]. GLP-1 significantly (p=0.007) reduced CAMP expression in
adipocytes ([Fig. 3c]) in vitro. Thus, it
could be speculated that the decline of CAMP concentrations after BS between V3
and V12 ([Fig. 1b]) might be regulated by
incretin hormones such as GLP-1. Interestingly, former experiments using
100 nM GLP-1 could demonstrate an inhibition of progranulin gene
expression in adipocytes [21]. Progranulin
also plays a role in adipose tissue innate immunity.
Since females were shown to have lower CAMP serum concentrations ([Table 1]) and lower CAMP gene expression
levels in subcutaneous adipose tissue ([Fig.
2d]), sex hormones might regulate the expression of CAMP in
adipocytes. Thus, differentiated 3T3-L1 adipocytes were incubated under
increasing doses of testosterone and estradiol. However, neither testosterone
nor estradiol were able to modulate CAMP expression significantly ([Fig. 3d]). Thus, other (systemic)
mechanisms seem to be responsible for gender-specific effects on CAMP
expression.
Discussion
The adipocytic innate immune system seems to orchestrate the defense of local
(subcutaneous) adipose tissue compartments against Gram positive bacteria such as
Staphylococci
[4]
[5]. This mechanism might represent an
evolutionarily conserved machinery that is highly reasonable and meaningful since
adipocytes and adipose tissue often reside at tissue interfaces involved in
complicated local infections such as bowel perforation (mesenteric adipose tissue),
acute pancreatitis (retroperitoneal adipose tissue), chronic inflammatory bowel
disease (mesenteric adipose tissue), arthritis (periarticular adipose tissue),
ophthalmitis (retro-orbital adipose tissue), soft tissue infection (subcutaneous
adipose tissue), and many others [22]. Thus,
detailed knowledge of the physiological and pathophysiological regulation of CAMP is
important and of translational impact. The clinical perspective [5] derived from the data published by Zhang et
al. [4] raises the possibility that adipocytes
use CAMP in order to fight against Gram positive bacterial infection of adipose
tissue. Soft tissue infections in obesity are a common clinical problem involving
subcutaneous adipose tissue.
In our experiments in vitro, high glucose plus high insulin concentrations resemble
the metabolic situation of poorly controlled type 2 diabetes mellitus and/or
obesity (hyperglycemic hyperinsulinemia). We could demonstrate that glucose and
insulin synergistically decrease adipocytic CAMP expression. Based on these
observations, one could speculate that high rates of soft tissue infections in
obesity and/or diabetes mellitus type 2 might be worsened by decreased CAMP
expression in local subcutaneous adipose tissue. Future translational and clinical
studies should be performed in order to test the suitability of CAMP as a diagnostic
marker of soft tissue infection and treatment. Moreover, CAMP might has the
potential as an endogenous and inducible anti-microbial molecule (drug target).
We could demonstrate that serum CAMP levels increase significantly after BS but not
during LCD. Thus, mechanisms other than weight reduction seem to have caused this
effect. Bile acids have been reported to increase regularly and rapidly after BS
such as RYGB [17]
[18]. Moreover, bile acids have recently been
recognized as pleiotropic hormones affecting white adipose tissue function and
adipokine secretion [23]
[24]. Thus, we investigated a broad panel of
human bile acid species in murine adipocytes for their potential to modulate CAMP
gene expression. We could demonstrate that certain bile acids in a non-toxic dose
range are able to upregulate CAMP expression. This observation might be explained by
the fact that adipocytes synthesize one of the main bile acid-responsive
transcription factor, FXRalpha (farnesoid-X-receptor-alpha). Tauromurocholic acid
(TMCA) induced CAMP expression in the present study. It is important to remember
that 3T3-L1 adipocytes represent a murine cell line and that TMCA is one of the main
bile acids found in mice. TMCA differs from primary bile acids found in humans such
as cholic acid (CA) which had no effect in our experiments. The secondary and
conjugated bile acids deoxycholic acid (DCA) and taurodeoxycholic acid (TDCA) were
also without any effects on CAMP expression. Moreover, the results are supported by
our recent study on white adipocytes providing evidence of a functional bile acid
signaling pathway [24] in adipocytes. Thus,
bile acids might represent future and therapeutic modulators of CAMP in the context
of infection and inflammation similar to certain bile acids such as ursodeoxycholic
acid and obeticholic acid that have been used for the treatment of inflammatory
liver diseases.
Since CAMP represents an anti-microbial peptide, acute infections might upregulate
CAMP serum levels. However, patients were investigated by a physician at every time
point of the study and we excluded patients with signs or symptoms of a clinical
infection. Moreover (as already published earlier [8]), CRP concentrations were evaluated and shown not to differ between
bariatric patients and patients under diet.
Incretin hormones such as GLP-1 have also been reported to increase regularly after
BS [19]
[20]. Since we observed a decline of elevated CAMP levels between V3 and
V12 after BS, we tested GLP-1 for its ability to suppress CAMP expression in
adipocytes in vitro. This hypothesis could be verified by a large number of cell
culture experiments in vitro. However, it is a limitation of the present study that
bile acid species and incretin hormones could not be measured in the two cohorts.
Future studies have to address this question by using tandem mass spectrometry
(LC-MS/MS) in order to correlate the bile acid metabolome [25] in obesity and after BS with CAMP levels.
Moreover, recent data already proved [25] that
the human bile acid metabolome is rapidly responsive to an oral lipid uptake. Thus,
the modulation of bile acid metabolome (and also incretins) during LCD should be
investigated systematically by LC-MS/MS.
The gender-specific effect showing reduced serum CAMP levels and reduced subcutaneous
adipose tissue gene expression levels in females is interesting but remains unclear.
Since sex hormones did not influence CAMP expression in adipocytes in vitro under
serum-free conditions (and in the absence of incretins or bile acids), other
systemic or multifactorial mechanisms seem to play a role. Moreover, adipose tissue
is not the only potential cellular source of circulating CAMP. This is highlighted
by the fact that neither subcutaneous nor visceral adipose tissue CAMP expression
was correlated to the respective serum CAMP concentrations in 156 patients. Thus,
other cellular sources of CAMP seem to exist. Based on this observation, CAMP cannot
longer be regarded as a typical adipokine (such as adiponectin) with an
adipocyte-specific expression and secretion profile. This has to be borne in mind
when interpreting serum CAMP levels in future studies. The present study
investigated CAMP mRNA expression in human total subcutaneous and visceral adipose
tissue. The limitation is given by the fact that other cell types possibly
expressing CAMP are present in total adipose tissue (stroma vascular cell fraction,
immune cells). Since a reliable and well established human adipocyte cell line
(without any syndromal or tumorigenic background) is lacking, future studies are
necessary to collect human adipose tissue for the preparation of (pre-adipocyte)
precursor cells and subsequent hormonal differentiation into mature adipocytes.
LPS as a potent TLR-4 ligand upregulates CAMP expression in neutrophils [26] and the effects of MALP-2 on CAMP
expression in pre-adipocytes and in mature adipocytes was documented in two recent
studies [6]
[7]. Based on this, CAMP levels seem to be correlated with inflammatory
response syndromes. This is important to remember because CAMP serum levels
and/or CAMP expression in adipose tissue were shown to correlate positively
with unfavorable metabolic factors (BMI, HbA1c, triglycerides, resistin)
and negatively with favorable factors (HDL cholesterol, adiponectin) in the present
study. To the best of our knowledge, the present study provides the largest cohort
of obese patients that has been investigated in order to quantify circulating CAMP
before and after weight loss. Thus, the present data might be helpful for the future
establishment of reference values.
Conclusions
There exist gender-specific and AT compartment-specific effects on the regulation of
CAMP gene expression. Weight loss induced by BS (but not by LCD) upregulated CAMP
serum levels suggesting the involvement of weight loss-independent mechanisms on
CAMP regulation such as bile acids, incretins, and metabolic factors. CAMP might
represent an adipokine at the interface between metabolism and innate immune
response. The presented serum concentrations in two large cohorts might be useful to
establish reference values for human CAMP.
