DOI 10.
1515/jpem-2012-0407      J Pediatr Endocr Met 2013; 26(9-10): 995–998
Short communication
Ximena Lopez*, Allison B. Goldfine, William L. Holland, Ruth Gordillo and Philipp E. Scherer
Plasma ceramides are elevated in female children
and adolescents with type 2 diabetes
Abstract: Accumulation of ceramides within tissues
induces insulin resistance. Moreover, adiponectin exerts its
beneficial metabolic effects at least partially through cera-
mide catabolism. We hypothesized that specific plasma
ceramide subspecies are elevated in obese children and
adolescents with type 2 diabetes (T2D), and that they
inversely correlate with adiponectin and measures of insu-
lin sensitivity. This was a cross-sectional study. Participants
included 14 obese female subjects with T2D, ages 10–17, and
14 lean healthy controls of the same age and gender. Fast-
ing plasma ceramide subspecies were measured by quan-
titative tandem mass spectrometry. Subjects with T2D had
higher concentrations of C22:0 and C20:0 ceramides, with
a 2-fold increase in C18:0 ceramide and C24:1 dihydrocera-
mide (p < 0.05). C22:0, C20:0 and C18:0 ceramide correlated
with decreased adiponectin concentrations, increased
HOMA-IR, BMI Z-score, triglyceride and fasting blood glu-
cose concentrations (p < 0.05). Plasma levels of C18:0, C20:0
and C22:0 ceramide, as well as C24:1 dihydroceramide, were
elevated in obese female children and adolescents with
T2D. This may be a reflection of tissue insulin resistance
and could be a result of low adiponectin levels.
Keywords: adolescents; ceramides; children; insulin
resistance; obesity; type 2 diabetes.
*Corresponding author: Ximena Lopez, University of Texas
Southwestern, Pediatrics, 5323 Harry Hines Blvd., Dallas, TX 75390-
9063, USA, Phone: +1-214-648-3501, Fax: +1-213-456-2940,
E-mail: ximena.lopez@utsouthwestern.edu
Allison B. Goldfine: Harvard Medical School and Joslin Diabetes
Center, Boston, MA, USA
William L. Holland, Ruth Gordillo and Philipp E. Scherer:
University of Texas Southwestern, Touchstone Diabetes Center,
Dallas, TX, USA
Introduction
A growing body of literature implicates accumulation of
tissue ceramides in the development of insulin resistance
atherosclerosis, β-cell dysfunction and apoptosis (1). In
humans, muscle ceramide content has been associated
with insulin resistance, and is nearly doubled in obese
insulin-resistant individuals compared to lean insulin-
sensitive subjects (2).
Ceramides can be generated by de novo biosynthesis.
The rate limiting biosynthetic enzyme, serine palmitoyl
transferase, is stimulated by the presence of long-chain
saturated fatty acids, but not by unsaturated fatty acids
(3). Ceramides induce insulin resistance in part through
activation of protein phosphatase 2A and protein kinase
C, leading to inhibition of Akt/PKB (4). Ceramides are also
tightly linked with inflammation, as ceramide synthesis is
essential for toll-like receptor-4 dependent insulin resist-
ance (3). The dependence of ceramide synthesis on satu-
rated fats makes these sphingolipids attractive candidate
metabolites, linking lipid oversupply to the antagonism of
insulin signaling.
Moreover, adiponectin, a hormone with insulin-sen-
sitizing, anti-inflammatory and anti-apoptotic functions,
exerts its beneficial metabolic effects through the stimula-
tion of a ceramidase that enhances ceramide catabolism
and formation of its anti-apoptotic metabolite, sphingo-
sine-1-phosphate (S1P) (5). Plasma ceramides have the
potential to be a novel biomarker of insulin resistance
and inflammation, although the relative flux of ceramides
between tissues and plasma is not well established. The
objective of this study is to compare plasma ceramide
subspecies in children and adolescents with obesity and
T2D vs. healthy controls, and to determine if these cor-
relate with adiponectin and other measures of insulin
sensitivity.
Methods
The Institutional Review Board of the University of Texas Southwest-
ern approved the study. Written informed consent and assent were
obtained from all parents and subjects before any testing procedures.
This is a cross-sectional study. Fourteen female subjects aged 10–
17 years with obesity (BMI  ≥  95th percentile) and previous diagnosis
of T2D, and 14 healthy lean female subjects (BMI < 85th percentile) of
the same age group, were invited to participate in the study. A blood
Brought to you by | Linköping University Library
Authenticated
Download Date | 10/26/17 9:58 PM
996      Lopez et al.: Elevated plasma ceramides in children

specimen was obtained from all subjects after an 8–12  h overnight

fast. Eight subjects from the control group had well controlled hypo-
thyroidism (normal TSH and free T4 levels at the time of evaluation).
In the T2D group, six subjects were being treated with metformin as
the only treatment, four with metformin and insulin, one with insu-
lin alone, and three were not receiving any diabetes treatment. T2D
subjects remained on their diabetes treatment at the time of the study
evaluation.
Quantitative analyses of plasma ceramide subspecies were
performed via liquid chromatography-tandem mass spectrometry
(LC-MS/MS) (5). Total adiponectin levels were measured with a com-
mercial ELISA kit (Millipore/Linco Research, St. Charles, MO, USA).
Plasma free fatty acids (FFA) concentrations were determined by
colorimetric assay (Wako Pure Chemical Industries, Osaka, Japan).
Plasma glucose concentrations were determined using the VITROS
GLU slide method (VITROS, Buckinghamshire, UK). Insulin concen-
trations were determined with human insulin ELISA (Millipore/Mer-
ck KGaA, Darmstadt, Germany). Triglycerides were determined using
the VITROS TRIG slide method (VITROS).
Subject characteristics are presented as mean ± SD and results as
mean ± SE. Two tailed Student’s unpaired t-test and χ2 were used for
comparisons of continuous and categorical variables, respectively.
Univariate linear regression was performed for correlation analysis.
All statistical analyses were performed using Stat View version 5.0.1
(SAS Institute, Cary, NC, USA).
Results
The demographic, clinical and metabolic characteristics
are shown in Table 1. The average HbA1c level in the T2D
group was 8.8% ± 1.8%.
Concentrations of ceramide subspecies in plasma of
the T2D subjects and controls are shown in Figure 1. C18:0,
C20:0 and C22:0 were elevated in the T2D group [49.5 ± 4.1
vs. 22.6 ± 2.0 ng/mL in controls (p = 0.000005), 74.3 ± 6.9 vs.
54.3 ± 5.4 ng/mL in controls (p = 0.03), and 373.8 ± 31.8 vs.
286.7 ± 25.45 ng/mL in controls (p = 0.04), respectively].
Also C24:1 dihydroceramide, a precursor of ceramide
formation, was increased in the T2D group (34.4 ± 4.0 vs.
16.8 ± 2.9 ng/mL in controls, p < 0.002). Plasma S1P levels
were not different between the two groups (p = 0.06).
However, C24:1 glucosylceramide, derived from the glu-
cosylation of C24:1 ceramide, decreased in the T2D group
(248.3 ± 22.3 vs. 339.0 ± 35.1 ng/mL in controls, p = 0.04).
There were no differences between groups in con-
centrations of total ceramides (p = 0.9), as well as the
ceramides subspecies C16:0, C24:0, C24:1, in C24:0 dihy-
droceramide, and in ceramide derivatives C16:0-1P, C16:0
lac-ceramide, C16:0 and C18:0 glucosylceramides, sphin-
ganine-1P, and in the sphingomyelins SM18:0, SM24:0 and
SM24:1.
Adiponectin inversely correlated with C18:0 (r = –0.55,
p < 0.03), C20:0 (r = –0.50, p < 0.007) and C22:0 (r = –0.47,
p = 0.01) (Figure 1). In contrast, adiponectin had a positive
correlation with C24:1 glucosylceramide (r = 0.47, p = 0.01).
HOMA-IR correlated with C18:0 (r = 0.67, p = 0.0004), C20:0
(r = 0.64, p = 0.0009) and C22:0 (r = 0.57, p = 0.004). Fasting
plasma glucose correlated with C18:0 (r = 0.46, p = 0.01),
C20:0 (r = 0.44, p < 0.02) and C22:0 (r = 0.42, p < 0.03).

Table 1 Subject characteristics.

Healthy controls Type 2 Diabetes subjects p-Value

n 14 14 –
Age, years 13.7 ± 4.1 14.3 ± 1.8 0.44
Gender, male/female 0/14 0/14 NS
Ethnicity, H/NH/U 8/6/0 7/6/1 NS
Race, W, AA, O, U 11/1/1/1 8/6/0/0  < 0.05
BMI, kg/m2 19.5 ± 2.1 37.4 ± 10.4  < 0.001
BMI z-score 0.007 ± 0.553 2.309 ± 0.279  < 0.001
Waist Circumference, cm 66.6 ± 10.9 81.0 ± 3.0  < 0.001
Tanner stage, I-V 3.7 ± 1.3 4.7 ± 0.5 0.02
FPG, mmol/L, mg/dL 5.0 ± 0.0, 90.1 ± 4.8 10.0 ± 4.1, 180 ± 73.3  < 0.001
FPI, pmol/L, uU/mL 28.4 ± 4.7, 4.1 ± 0.7 97.2 ± 60.6, 14.0 ± 8.7  < 0.001
HOMA-IRa 0.9 ± 0.2 4.7 ± 1.4 0.003

NEFA, mmol/L or mEq/L 0.45 ± 0.17 0.51 ± 0.17 NS

TG, mmol/L, mg/dL 0.9 ± 0.3, 80.4 ± 26.9 2.3 ± 1.1, 204.8 ± 95.7  < 0.001
Adiponectin, µg/mL 12.1 ± 3.8 6.3 ± 3.4  < 0.001

Data are mean ± SD. H, Hispanic; NH, non-Hispanic; U, unknown; W, White; AA, African American; O, Other; FPG, fasting plasma glucose;
FPI, fasting plasma insulin; NEFA, non-esterified fatty acids; TG, triglycerides. aHOMA-IR was not calculated in the subjects receiving insulin
therapy with (n = 4) or without metformin (n = 1).

Brought to you by | Linköping University Library

Authenticated
Download Date | 10/26/17 9:58 PM
 Lopez et al.: Elevated plasma ceramides in children      997

A
*
400
B
80
T2D
300
Controls 70 p=0.002
60

C18:0, ng/mL
ng/mL

200 50

40

100 30
*
*** 20
**
10
0 0 5 10 15 20
C22:0 C20:0 C18:0 C24:1DH
Adiponectin, µg/mL

C 140 D 700
120
600
p=0.01
100 p=0.006
C20:0, ng/mL

500
80 C22:0, ng/mL
400
60

40 300

20 200

0 100
0 5 10 15 20 0 5 10 15 20
Adiponectin, µg/mL Adiponectin, µg/mL

Figure 1 (A) Elevated plasma ceramides in type 2 diabetes (T2D) vs. control subjects. Correlation of plasma adiponectin with plasma C18:0
(B), C20:0 (C), and C22:0 (D) ceramides. (•) T2D, (○) controls. Data are mean ± SE. *p < 0.05. **p < 0.002. ***p < 0.00001.

Fasting FFA levels did not correlate with plasma cera-
mides, but plasma triglycerides showed a strong correla-
tion with C18:0 (r = 0.76, p < 0.0001), C20:0 (r = 0.74, p = 0.004)
and C22:0 (r = 0.64, p = 0.0002). BMI Z-score correlated with
C18:0 (r = 0.69, p < 0.0001), C20:0 (r = 0.39, p = 0.04) and
C22:0 (r = 0.39, p < 0.04), while it showed a negative cor-
relation with C24:1 glucosylceramide (r = –0.45, p = 0.01).
S1P did not correlate with adiponectin, HOMA-IR, fasting
plasma glucose, FFA, triglyceride levels or BMI Z-score.
Discussion
The present study shows for the first time that plasma con-
centrations of specific ceramide subspecies are elevated in
the circulation of obese female children and adolescents
with T2D. Moreover, these correlate inversely with adiponec-
tin, and directly with HOMA-IR, fasting plasma glucose,
triglyceride levels and BMI-Z score. This is consistent with
adult studies (6), which also show the greatest increase in
the C18:0 subspecies, although there was a larger difference
in our younger population (118% vs. 32%). In muscle, the
greatest increase in adult obese subjects has been observed
in C16:0 and C20:0 (2), and while our subjects had elevated
levels of C20:0, no differences were seen in C16:0 levels.
Dihydroceramides are precursors of ceramide formation
and are elevated in adipose tissue of obese humans (7).
Consistently, we found that plasma dihydroceramides are
elevated in adolescents with obesity and T2D.
Only one other study has evaluated the role of cera-
mides in a younger age group (8). Majumdar and Mas-
trandrea showed a correlation of plasma ceramides with
insulin resistance in overweight adolescents. Differently
from our study, only 20% of the overweight subjects had
the metabolic syndrome, and no increase of plasma cera-
mides was found. Therefore, the severity of insulin resist-
ance/metabolic syndrome might be a better predictor of
ceramide concentrations compared to obesity alone.
The mechanisms responsible for increased plasma
ceramide concentrations were not directly evaluated.
However, we found plasma ceramides inversely corre-
late with adiponectin. This observation is consistent with
animal data showing adiponectin improves insulin sensi-
tivity by reducing tissue concentrations of ceramides (5).
Also, triglycerides were strongly associated with ceramide
levels. In rodents, a high fat diet or infusion of lipids high
in saturated fat induces an increase in ceramide concen-
trations in plasma and other tissues (9, 10). Therefore,

Brought to you by | Linköping University Library

Authenticated
Download Date | 10/26/17 9:58 PM
998      Lopez et al.: Elevated plasma ceramides in children
one could speculate that increased saturated fat intake
could play a role in differences observed in ceramide
levels between groups. Plasma levels of C22:0 and C20:0
ceramide, as well as of C18:0 and C24:1 dihydroceramide
are elevated in female children and adolescents with T2D.
This may be a reflection of tissue insulin resistance and
could be a result of low adiponectin levels.
Unexpectedly, glucosylceramide levels were decreased
in the T2D group. The importance of this finding is not clear
as inhibition of glycosphingolipid synthesis in rodents
improves glycemic control and insulin sensitivity (11).
Only female subjects were included in this study as
adiponectin levels decrease significantly during puberty
in males (12), and adiponectin may decrease tissue cera-
mide levels (5). Therefore, our findings may not apply to
male children and adolescents. The control group had a
slightly lower Tanner stage, but we did not observe an
association between ceramide levels and Tanner stage
(data not shown). Subjects with T2D were on heterogene-
ous diabetes therapies. We recognize the potential effects
References
1. Holland WL, Summers SA. Sphingolipids, insulin resistance, and
metabolic disease: new insights from in vivo manipulation of
sphingolipid metabolism. Endocr Rev 2008;29:381–402.
2. Adams JM 2nd, Pratipanawatr T, Berria R, Wang E,
DeFronzo RA, et al. Ceramide content is increased in skeletal
muscle from obese insulin-resistant humans. Diabetes
2004;53:25–31.
3. Holland WL, Bikman BT, Wang LP, Yuguang G, Sargent KM,
et al. Lipid-induced insulin resistance mediated by the
proinflammatory receptor TLR4 requires saturated fatty
acid-induced ceramide biosynthesis in mice. J Clin Invest
2011;121:1858–70.
4. Schmitz-Peiffer C, Craig DL, Biden TJ. Ceramide generation is
sufficient to account for the inhibition of the insulin-stimulated
PKB pathway in C2C12 skeletal muscle cells pretreated with
palmitate. J Biol Chem 1999;274:24202–10.
5. Holland WL, Miller RA, Wang ZV, Sun K, Barth BM, et al.
Receptor-mediated activation of ceramidase activity initiates
the pleiotropic actions of adiponectin. Nat Med 2011;17:
55–63.
6. Haus JM, Kashyap SR, Kasumov T, Zhang R, Kelly KR, et al. Plasma
ceramides are elevated in obese subjects with type 2 diabetes
that different treatments may have on ceramide concen-
trations. We are focusing at this stage only on plasma
sphingolipids primarily due to the easy availability of
samples. Fundamental limitations of such an approach
include: firstly, we do not know how levels of individual
subspecies of ceramides correlate with levels of these
metabolites in tissues. Secondly, we do not know which
tissue is the main driving force influencing the plasma
concentrations of a given subspecies. As these sphingoli­
pids are unlikely to exert a direct effect on insulin sensiti­
vity while in plasma, we have approached this issue from
the perspective that these altered plasma sphingolipid
levels serve as powerful biomarkers for various aspects
of the metabolic syndrome. As such, the correlations
seen are very strong, indicating that the measurement of
these species may serve as metabolic biomarkers in the
future.
Received December 17, 2012; accepted February 18, 2013; previously
published online April 24, 2013
and correlate with the severity of insulin resistance. Diabetes
2009;58:337–43.
7. Błachnio-Zabielska AU, Pułka M, Baranowski M, Nikołajuk A,
Zabielski P, et al. Ceramide metabolism is affected by
obesity and diabetes in human adipose tissue. J Cell Physiol
2012;227:550–7.
8. Majumdar I, Mastrandrea LD. Serum sphingolipids and
inflammatory mediators in adolescents at risk for metabolic
syndrome. Endocrine 2012;41:442–9.
9. Holland WL, Brozinick JT, Wang LP, Hawkins ED, Sargent
KM, et al. Inhibition of ceramide synthesis ameliorates
glucocorticoid-, saturated-fat-, and obesity-induced insulin
resistance. Cell Metab 2007;5:167–79.
10. Boini KM, Zhang C, Xia M, Poklis JL, Li PL. Role of sphingolipid
mediator ceramide in obesity and renal injury in mice fed a
high-fat diet. J Pharmacol Exp Ther 2010;334:839–46.
11. Langeveld M, Aerts JM. Glycosphingolipids and insulin
resistance. Prog Lipid Res 2009;48:196–205.
12. Böttner A, Kratzsch J, Müller G, Kapellen TM, Blüher S, et al.
Gender differences of adiponectin levels develop during the
progression of puberty and are related to serum androgen
levels. J Clin Endocrinol Metab 2004;89:4053–61.
Brought to you by | Linköping University Library
Authenticated
Download Date | 10/26/17 9:58 PM