389

Differential impact of adipokines derived from primary adipocytes of

wild-type versus streptozotocin-induced diabetic rats on glucose and
fatty acid metabolism in cardiomyocytes
Rengasamy Palanivel, Vivian Vu, Min Park, Xiangping Fang and Gary Sweeney
Department of Biology, York University, Toronto, Ontario, Canada M3J 1P3
(Correspondence should be addressed to G Sweeney; Email: [email protected])

Abstract
The causal relationship between obesity and cardiovascular production was strongly induced by adipocytes from diabetic
disease is extensively acknowledged; however, the exact rats. Examination of fatty acid uptake revealed that stimulation
mechanisms linking obesity and heart failure remain unclear. only occurred in response to adipokines secreted by wild-type
Here, we investigated the influence of adipokines derived from rat adipocytes. Importantly, oxidation of fatty acids by
primary adipocytes on glucose and fatty acid uptake and cardiomyocytes was decreased by adipokines derived from
metabolism in isolated primary cardiomyocytes. Either co-cul- diabetic rat adipocytes. Analysis of adipokine profiles in diabetic
ture of these cell types or incubation with adipocyte- rat adipocyte-conditioned medium demonstrated the most
conditioned medium significantly increased glucose uptake in significant decreases in adiponectin and leptin with increased IL6
cardiomyocytes. When streptozotocin-induced diabetic rats expression. Taken together, these data suggest that the profile of
were used as a source of adipocytes, there was a lower ability to adipokines secreted by adipocytes from diabetic rats have a
elicit glucose uptake in cardiomyocytes which corresponded deleterious influence on cardiomyocyte metabolism which may
with lower Akt and AMPK phosphorylation. The profile of be of relevance in the pathophysiology of heart failure.
glucose metabolism also differed with oxidation being favored Journal of Endocrinology (2008) 199, 389–397
upon co-culture with wild-type adipocytes whereas lactate

Introduction (Borradaile & Schaffer 2005). A decrease in glucose transport,

glycolysis, and glucose oxidation, together with an increase in
Obesity plays an influential role in dictating the morbidity and fatty acid uptake and oxidation, is typically observed in
mortality associated with cardiomyopathy (Abel et al. 2008). obesity and diabetes (Stanley et al. 2005, Abel et al. 2008).
This has prompted many studies searching for a more complete Regulation of glucose and fatty acid metabolism by adipokines
understanding of the factors and mechanisms involved in cardiac has been extensively demonstrated in liver and skeletal muscle
remodeling in obesity. Consequently, various adipokines have (Badman & Flier 2007). Recent studies have also shown that
emerged as potential players in the pathophysiology of heart adipokines such as leptin (Palanivel et al. 2006) and adiponectin
failure via endocrine effects which impact upon cardiac function (Pineiro et al. 2005, Guo et al. 2007, Li et al. 2007, Palanivel et al.
(Bradham et al. 2002, Hopkins et al. 2007, Karmazyn et al. 2007). 2007) can mediate potent direct effects on cardiomyocyte
These include factors such as leptin, adiponectin, and TNF. It is glucose and fatty acid uptake and metabolism. Nevertheless, our
also of interest to note that many of these factors have now been knowledge of the effects of adipokines on cardiomyocyte
shown to be produced by the heart itself (Purdham et al. 2004, metabolism is derived largely from studies adding recombinant
Pineiro et al. 2005, Guo et al. 2007). proteins to quiescent cells. While this is a significant step in
One of the most significant remodeling events leading to characterizing the effects of each adipokine, it is important to
heart failure is an alteration in cardiomyocyte metabolism appreciate that, in physiological terms, crosstalk between effects
(Lopaschuk et al. 2007). In the healthy heart, under aerobic induced by each adipokine may influence the cellular response to
conditions, the majority of energy required for contractile other adipokines. Therefore, to examine this scenario more
performance is derived from fatty acids while the remainder carefully we established a co-culture system of primary rat
(w30%) is principally obtained via metabolism of glucose adipocytes together with primary neonatal rat cardiomyocytes or
(Stanley et al. 2005, An & Rodrigues 2006). Well-controlled the use of primary adipocyte-conditioned medium, which
fatty acid metabolism is also important to prevent triglyceride allows us to examine the effect of more physiologically relevant
accumulation (McGavock et al. 2006), as this can lead to combinations of adipokines on cardiomyocyte glucose and fatty
lipotoxic effects such as apoptosis or insulin resistance acid metabolism. We proposed that the adipocyte mixture

Journal of Endocrinology (2008) 199, 389–397 DOI: 10.1677/JOE-08-0336

0022–0795/08/0199–389 q 2008 Society for Endocrinology Printed in Great Britain Online version via http://www.endocrinology-journals.org
390 R PALANIVEL AND OTHERS . Effect of adipokines on cardiomyocyte metabolism

derived from primary adipocytes would alter glucose and fatty indirect immunofluorescence staining with an antibody to
acid metabolism in cardiomyocytes, together with associated myosin heavy chain (MF20, a kind gift from Dr J C McDermott,
intracellular signaling events. We also hypothesized that the York University,Toronto, Canada). After incubation at 37 8C in
profile of adipokines secreted by adipocytes derived from wild- humid air with 5% (v/v) CO2 and 95% O2 (v/v) for 24 h, the
type or diabetic rats would differ and may cause distinct metabolic cardiomyocytes were then deprived of serum and incubated for
and signaling effects in cardiomyocytes. another 24 h before treatment.

Experimental induction of diabetes using STZ

Materials and Methods
Male Wistar rats were used at 6–8 weeks (250–300 g), and were
Materials either used directly for adipocyte isolation (wild-type) or
diabetes was induced with STZ prior to adipocyte isolation.
Primaria TM Easy GripTM (surface modified polystyrene, non- Diabetes was induced by i.p. injection of STZ (in 50 mM citrate
pyrogenic) tissue culture dishes and plates were from Becton buffer, pH 4.5) at a dose of 100 mg/kg body weight. Diabetic
Dickinson (Franklin Lakes, NJ, USA). DMEM/F12 medium, animals exhibited slightly decreased body weight and blood
was obtained from Gibco (Grand Island, NY, USA). Gentamicin glucose levels, measured with the OneTouch Ultra Meter
sulfate was obtained from Medlatech Inc. (Hsserndon, VA, glucometer (Lifescan, Burnaby, BC, Canada) upon removal of
USA) and penicillin/streptomycin from Wisent Inc. (St Foy, blood from the tail vein, which were found to increase to
QC, Canada). 5-Aminoimidazole-4-carboxamide-1-b-D-ribo- 31.64G0.642 mmol/l, 6 days following STZ injection.
furanoside (AICAR) was purchased from Toronto Research
Chemicals Inc. (Toronto, ON, Canada). [3H] palmitate,
2-deoxy-D-[3H] glucose, D-[U-14C] glucose, and [1-14C] Isolation of adipocytes from epididymal adipose tissue
palmitate were from Amersham. FATP1, FATP2, and Adipocytes were isolated as described previously (Vu et al. 2007),
FAT/CD36 were obtained from Santa Cruz Biotechnology where epididymal adipose tissue was removed from 6- to
(Santa Cruz, CA, USA). Primary antibodies for phospho- 8-week-old male wild-type and diabetic (6 days post-injection
AMPK (Thr-172), phospho-Akt (Ser-473, Thr-308), ACC of STZ) Wistar rats, always between 1000 and 1200 h to avoid
(Ser-79), and HRP-conjugated anti-rabbit secondary antibody
diurnal variations in adipokine profiles, and chopped with
were purchased from Cell Signaling Technology (Beverly, MA,
scissors into 2 ml Krebs–Ringer–HEPES (KRH) buffer
USA). Enhanced chemiluminescence reagent was purchased
(131.5 mM NaCl, 4.7 mM KCl, 2.5 mM CaCl2, 1.25 mM
from Perkin–Elmer Life Sciences (Boston, MA, USA).
MgSO4, 2.5 mM NaH2PO4, 10.0 mM HEPES), supple-
Streptozotocin (STZ) was purchased from Sigma. All other
mented with 1% BSA. Tissues were digested with collagenase
reagents were of the highest grade available.
type II (1 mg/ml) for 1 h at 37 8C in a 250 r.p.m. shaker. After
1 h of digestion, infranatant was removed and adipocytes were
Isolation and culture of neonatal ventricular myocytes washed with fresh KRH buffer. It is established that using this
The care and use of the animals in the present study were in method of isolation effectively removed macrophages. The
accordance with approved guidelines of the York University number of adipocytes was counted and diluted to 1!106 cells/
Animal Care Committee (Toronto, Canada). Primary cultures ml with 10% FBS DMEM/F12 medium. Our initial
of cardiomyocytes were prepared from the ventricles of 2-to-3- experiments were performed using co-culture of primary rat
day-old Wistar rats by enzymatic digestion by using trypsin as adipocytes and cardiomyocytes; however, to avoid potential
described previously (Palanivel et al. 2007). Briefly, neonatal rats contaminating effects of adipocytes remaining in the wells, we
were put into a glass beaker containing cotton mass wetted with also used adipocyte-conditioned media only. Similar results
ethyl ether. After anesthesia and decapitation, hearts were taken were observed and the latter approach was used in generating the
out immediately and put into ice-cold calcium and bicarbonate data reported in this manuscript.
free Hank’s to HEPES (CBFHH) buffer, and then cut into
pieces. Cells in suspension were collected after several rounds of Preparation of adipocyte-conditioned medium
digestion of heart pieces, for selective enrichment of
cardiomyocytes; cells were then pre-plated into several 100! Adipocytes were stabilized in a 25 cm2 cell suspension flask
20 mm culture dishes and incubated for 1 h. The suspension for 3 h in a humidified atmosphere (95% air and 5% CO2) at
containing unattached cardiomyocytes was then collected and 37 8C. After stabilization, media was removed from adipocytes
seeded at a density of 1!106 cells/ml in culture media (DMEM and changed to serum free DMEM/F12 media to prepare
with 10% fetal bovine serum, 0.1 mM 5-bromodeoxyuridine, conditioned media. Conditioned media was collected from
50 mg/ml gentamicin, 100 U/ml penicillin, and 100 mg/ml incubation of 1!106 adipocytes/ml for 2 h in serum free
streptomycin). BrdU (0.1 mmol/l, Sigma) included in the DMEM/F12 medium and were used to treat primary
culture medium was to prevent proliferation of non-myocytes. cardiomyocytes for different experiments and various times
More than 90% of the cells were myocytes, as evaluated by indicated in the figure legends.

Journal of Endocrinology (2008) 199, 389–397 www.endocrinology-journals.org

 Effect of adipokines on cardiomyocyte metabolism . R PALANIVEL AND OTHERS 391

Glucose and fatty acid uptake measurements in primary Lactate production by cardiomyocytes
cardiomyocytes
Lactate content was determined by the lactate oxidase method
Cardiomyocytes were cultured in 24-well plates and treated using a lactate assay kit (Sigma). Cells were pre-incubated
with adipocyte-conditioned media from control and diabetic with adipocyte-conditioned medium for the time indicated
rats for periods of 0.5, 1, 2, and 3 h:insulin (100 nM) for in the figure, and insulin (100 nM) for 2 h was used as a
20 min was used as positive control for both experiments. positive control, after which the media was collected and used
Consequently, glucose and/or fatty acid transport was assayed for analysis of lactate content and using adipocyte-
for 5 and 1 min, respectively, at room temperature as conditioned media not exposed to cardiomyocytes as control.
described previously (Palanivel et al. 2007). Briefly, the
incubation medium was aspirated, cells were washed with
Immunoblotting analysis of signaling proteins
ice-cold saline, and 200 ml potassium hydroxide (KOH)
(1 mol/l) was added to each well. Aliquots of cell lysates were Cardiomyocytes were seeded on 35!10 mm culture dishes
transferred to scintillation vials for radioactivity counting and and incubated acutely with adipocyte-conditioned medium
the remainder was used for protein assay. Non-specific uptake from wild-type versus STZ-diabetic rat for different time
was determined in the presence of cytochalasin B (10 mmol/l) point (0–10 min) as indicated in the figure legend, and insulin
and was subtracted from all the values. Results are calculated for 10 min as a positive control for phosphorylation study. For
as pmol of glucose or fatty acid uptake/min/mg protein. total protein expression study, the cells were treated with
adipocyte-conditioned medium for 3 h. After appropriate
treatment, cells were washed thrice with ice-cold PBS and
Determination of glucose and fatty acid oxidation in primary
lysates were prepared exactly as described previously
cardiomyocytes
(Palanivel et al. 2007). Prior to loading onto SDS-PAGE
As described previously (Palanivel et al. 2007), glucose and fatty gels, the samples were diluted 1:1 (v/v) with 2!Laemmli
acid oxidation was measured by the production of 14CO2 from sample buffer (62.5 mmol/l Tris–HCl [pH 6.8], 2% [w/v]
14 14
D-[U– C] glucose and [1- C] palmitate respectively. Briefly, SDS, 50 mmol/l dithiothreitol, 0.01% [w/v] bromophenol
cardiomyocytes were seeded in 60!15 mm Petri dishes and blue). Equal amounts of cardiomyocyte proteins (30 mg) were
pretreated with adipocyte-conditioned medium from both resolved by SDS-PAGE (8–12%), and then transferred to
normal and diabetic rats for the time indicated in the figure polyvinylidene difluoride (PVDF) membranes (Bio-Rad).
legend. Cells were then incubated with a medium containing Membranes were probed with phosphorylation-specific
0.15 mCi/ml D-[U–14C] glucose and/or 0.15 mCi/ml [1-14C] antibodies against proteins of interest (phospho Akt (Thr308
palmitate for 2 h. Each Petri dish was sealed with parafilm and Ser473), phospho-AMPK (Thr172) and phospho-ACC
containing a piece of Whatman paper attached to the inside. The (Ser79)) and antibodies against fatty acid transporter proteins
Whatman paper was wetted with 100 ml of phenylethylamine– such as FATP1, FATP2, and CD36. Appropriate HRP-
methanol (1:1) to trap CO2 produced during the incubation conjugated secondary antibodies (anti-rabbit at 1:10 000
period. After 2 h of incubation, 200 ml of H2SO4 (4 mol/l) was dilution,) were used in each case and detected by the
added, followed by further incubation for 1 h at 37 8C. Finally, enhanced chemiluminescence method. Densitometric images
the pieces of Whatman paper were removed and transferred to of data were quantitated using Scion Image software.
scintillation vials for radioactivity counting. Insulin (100 nM)
for glucose oxidation and AICAR (2 mM) for fatty acid
Analysis of adipokines profiles in adipocyte-conditioned medium
oxidation were used as positive controls.
Adipocyte-conditioned medium obtained by culturing epidi-
dymal fat cells (1!106 cells/ml) isolated from wild-type and
Measurement of glycogen synthesis
STZ-diabetic rats were analyzed for adiponectin, leptin, and
Glycogen synthesis was measured by the incorporation of visfatin content by using ELISA kit (ALPCO Diagnostic,
14
D-[U– C] glucose to glycogen as described previously (Palanivel Windham, NH, USA). For quantitative analysis of free fatty
et al. 2007) with a few modifications. Briefly, cardiomyocytes acids in the conditioned medium, we used the HR series
were cultured in 35!10 dishes and pre-incubated with NEFA-HR(2) kit (Wako Pure Chemical Industries Ltd, Osaka,
adipocyte-conditioned medium for various times (0–3 h) as Japan). We used antibodies for resistin (AdipoGen, Seoul, South
indicated in the figure. Consequently, incubation with Korea), TNF, and IL6 (Cedarelane Laboratories, Burlington,
0.15 mCi/ml D-[U–14C] glucose for 2 h and insulin (100 nM Canada) to quantitatively determine changes in expression of
for 2 h) was used as a positive control. The cells were washed three these adipokines by immunoblotting, as described previously
times with cold PBS and lysed in 1 mol/l KOH. To measure with slight modifications. Briefly, adipocyte-conditioned
insulin-stimulated incorporation of glucose into glycogen, cell medium were diluted 1:1 (v/v) with 2! Laemmli sample
lysates were used for overnight glycogen precipitation with buffer (62.5 mmol/l Tris–HCl [pH 6.8], 2% [w/v] SDS,
ethanol. Precipitated glycogen was then dissolved in water 50 mmol/l dithiothreitol, 0.01% [w/v] bromophenol blue).
and transferred to scintillation vials for radioactivity counting. Equal volumes (50 ml) of adipocyte-conditioned medium from

www.endocrinology-journals.org Journal of Endocrinology (2008) 199, 389–397

392 R PALANIVEL AND OTHERS . Effect of adipokines on cardiomyocyte metabolism

wild-type versus STZ-diabetic rats were resolved by SDS-PAGE t-test where appropriate. Differences between the groups
(12–15%), and then transferred to PVDF membranes. were considered statistically significant when P!0.05.
Membranes were probed with specific antibodies against resistin
and TNF at 1:1000 dilution and IL6 (1:500) and appropriate
HRP conjugated secondary antibodies (anti-rabbit or anti-goat Results
at 1:10 000 dilution) were used in each case and detected by the
enhanced chemiluminescence method. We first determined the effect of primary adipocyte-conditioned
medium from wild-type or STZ-diabetic rats on glucose uptake
in primary neonatal cardiomyocytes. We found that incubation
Statistical analysis
with adipocyte-conditioned medium from wild-type rats for
Data are expressed as meansGS.E.M. Statistical analysis was times from 30 min to 3 h significantly stimulated
undertaken using one-way ANOVA or the paired Student’s 2-deoxyglucose transport into cardiomyocytes (Fig. 1A). Similar

Figure 1 Regulation of glucose uptake and phosphorylation of intracellular signaling proteins by adipocyte-conditioned medium. (A) Uptake
of 2-deoxyglucose was measured in response to adipocyte-conditioned medium derived from wild-type or diabetic rats (0 min–3 h). Insulin
(100 nm) for 20 min served as a positive control. Values are expressed as meanGS.E.M. of nZ6; * indicates P!0.05 compared with control
(in the absence of adipocyte-conditioned medium) and # indicates P!0.05 compared with response observed using wild-type adipocytes.
(B) Co-culture effect of adipocyte-conditioned medium (5 and 10 min) on phosphorylation of AMPK (Thr172) and Akt (Thr308 or Ser473).
Representative immunoblots together with quantitative analyses (meanGS.E.M.; nZ4) are shown in all cases.

Journal of Endocrinology (2008) 199, 389–397 www.endocrinology-journals.org

 Effect of adipokines on cardiomyocyte metabolism . R PALANIVEL AND OTHERS 393

data were obtained upon direct co-culture of primary adipocytes of this increase in glucose uptake was much lower than that
with cardiomyocytes (data not shown). When cardiomyocytes observed in response to the adipokine mixture produced by
were treated with adipocyte-conditioned medium produced wild-type rat adipocytes. This observation correlated with the
using adipocytes derived from streptozoticin-induced diabetic fact that phosphorylation of both AMPK (Thr172) and Akt
rats, there was a moderate stimulation of glucose uptake when (Thr308 and Ser473) were stimulated to a greater degree by
compared with the control (Fig. 1A). However, the magnitude wild-type adipocyte-conditioned medium (Fig. 1B).
We then examined various potential routes of glucose
metabolism within the cardiomyocyte. Using adipocytes
derived from wild-type rats we observed that conditioned
media, at times up to 3 h, significantly elevated glucose
oxidation (Fig. 2A), did not alter glycogen synthesis (Fig. 2B),
and caused a slight increase in lactate production by
cardiomyocytes (Fig. 2C). By contrast, there was a striking
increase in lactate production when using adipocytes derived
from diabetic rats (Fig. 2C) and no significant change in
glucose oxidation (Fig. 2A) or glycogen synthesis (Fig. 2B).
We also observed a significant difference in the ability of wild-
type and diabetic rat adipocyte-conditioned media to elicit fatty
acid uptake in cardiomyocytes. At 30 min to 3 h, the former
significantly stimulated fatty acid uptake (Fig. 3A); however, this
response was not seen in response to media prepared using
adipocytes derived from diabetic rats (Fig. 3A). After 3 h of
treatment we did not detect any change in the total expression
levels of CD36, FATP1, and FATP2 (Fig. 3B).
Subsequent examination of palmitate oxidation in cardio-
myocytes suggested that wild-type conditioned media (up to 3 h
treatment) did not dramatically alter oxidation from control,
although a small increase was detected at 2 h (Fig. 4A) and an
apparent increase in ACC phosphorylation was also detected
(Fig. 4B). AICAR was used as positive control in palmitate
oxidation assay (Fig. 4). Interestingly, when we examined
palmitate oxidation in cardiomyocytes after treatment with
diabetic rat adipocyte-conditioned media, we found that the
level of oxidation was even lower that basal at 1 and 2 h (Fig. 4).
Since we have observed differential impact of adipocyte-
conditioned medium obtained from wild-type and STZ-
diabetic rats on cardiomyocyte glucose and fatty acid uptake
and metabolism, we determined changes in the secretion level of
adipokines. This analysis (summarized in Fig. 5 as fold changes
from values found in wild-type adipocyte-conditioned media)
demonstrated that total adiponectin levels decreased most
obviously (0.46G0.10 vs 0.084G0.008 mg/ml in wild-type
and diabetic samples respectively). Leptin and visfatin levels also
decreased significantly (220.34G29.6 vs 77.82G13.09 pg/ml;
2.799G0.1525 vs 1.940G0.057 ng/ml respectively) whereas

Figure 2 Regulation by adipocyte-conditioned medium on glucose

oxidation, glycogen synthesis, and lactate production in primary
cardiomyocytes. (A) The co-culture effect of adipocyte-conditioned
medium from wild-type and/or diabetic rats (30 min–3 h) on basal
14
CO2 production from D-[U-14C] glucose; (B) the incorporation of
14
D-[U- C] glucose into glycogen by adipocyte-conditioned
medium. (C) Lactate production by the cells. Values shown are
expressed as meanGS.E.M. of nZ6 experiments; * indicates P!0.05
with respect to control (in the absence of conditioned medium) and
# indicates P!0.05 compared with response observed using wild-
type adipocytes.

www.endocrinology-journals.org Journal of Endocrinology (2008) 199, 389–397

394 R PALANIVEL AND OTHERS . Effect of adipokines on cardiomyocyte metabolism

Figure 3 Impact of adipocyte-conditioned medium on fatty acid uptake and fatty acid transporter protein expression. (A) The uptake of [3H]
palmitic acid by cardiomyocytes was examined in response to adipocyte-conditioned medium from wild-type and/or diabetic rats (0–3 h).
Values are expressed as meanGS.E.M. of nZ6; * indicates P!0.05 with respect to control (in the absence of adipocyte-conditioned medium)
and # indicates P!0.05 compared with response observed using wild-type adipocytes. (B) Examination of FATP1, FATP2, and CD36 protein
expression in cells treated (3 h) with adipocyte-conditioned medium from wild-type versus diabetic rats. Representative images of four
individual experiments are shown.

there was a modest and non-significant increase in resistin and An & Rodrigues 2006), and have more recently been
TNF expression. IL6 levels increased by 34% and there was no observed in ob/ob or db/db mouse and Zucker rat hearts
significant difference in free fatty acid levels (0.0245G0.0049 vs prior to the onset of hyperglycemia and as early as 10 days
0.0220G0.0072 mM respectively). after the initiation of high-fat feeding (Buchanan et al. 2005,
Golfman et al. 2005, Park et al. 2005, Wang et al. 2005). These
changes were consistently associated with increased myo-
Discussion cardial oxygen consumption and decreased cardiac efficiency.
Severely obese humans also display increased rates of fatty
Changes in substrate metabolism is one of the earliest acid oxidation, increased myocardial oxygen consumption, and
measurable abnormalities in the hearts of both diabetic and reduced cardiac efficiency (Peterson et al. 2004). Therefore,
obese animals and humans (Abel et al. 2008). This precedes obesity and diabetes clearly precipitate heart failure at least
measurable changes in in vivo cardiac function, suggesting an partially via alterations in myocardial substrate metabolism.
important causative role. Changes in myocardial substrate The last decade has seen intense research interest in the role
utilization in diabetes are well established (Stanley et al. 2005, of adipokines as regulators of metabolic homeostasis, with the

Journal of Endocrinology (2008) 199, 389–397 www.endocrinology-journals.org

 Effect of adipokines on cardiomyocyte metabolism . R PALANIVEL AND OTHERS 395

Figure 5 Changes in the adipokine profile level in wild-type versus

STZ-diabetic rat adipocyte-conditioned medium. The level of
adiponectin, leptin, visfatin, resistin, TNF, IL6, and free fatty acid
(FFA) in adipocyte-conditioned medium is summarized here as fold
changes (increase or decrease) in medium prepared using adipocytes
from rats 6 days after injection of STZ compared with values found in
wild-type adipocyte-conditioned media. Absolute values for the
concentration of these adipokines are given where appropriate in the
results section. Values shown are expressed as meanGS.E.M. of nR3
experiments; * indicates P!0.05 when levels in wild-type or STZ-
diabetic rat adipocyte-conditioned media were compared.

demonstrated direct metabolic effects of adipokines on

cardiomyocyte metabolism via treating cells with individual
recombinant forms of each adipokine. However, this
approach inherently neglects the extensive crosstalk between
signaling pathways regulated by various adipokines which
exists in vivo. At the other extreme, in vivo studies are often
confounded by the existence of additional inputs (e.g.,
neuronal). Hence, we recently established the use of a
co-culture system, which allowed direct analysis of effects
induced by more physiologically relevant combinations of
Figure 4 Regulation by adipocyte-conditioned medium on fatty acid
adipokines acting in concert with one another (Vu et al.
oxidation and phosphorylation of acetyl CoA carboxylase in primary 2007). In this study, we utilized a similar approach but with
cardiomyocytes. (A) The impact of adipocyte-conditioned medium primary neonatal rat cardiomyocytes as our target cell.
from wild-type versus diabetic rats (0–3 h) on 14CO2 production from Our results demonstrated that the profile of adipokines
[1-14C] palmitate in cardiomyocytes. AICAR (2 mM, 2 h) was used as
secreted by primary rat adipocytes stimulated glucose uptake in
a positive control. Data are representative of six independent
experiments expressed as meanGS.E.M.; * indicates P!0.05 cardiomyocytes. This correlated with increases in both AMPK
compared with control (no adipocyte-conditioned medium) and (insulin-independent) and Akt (insulin-like) signaling. Obesity
# indicates P!0.05 compared with response observed using and diabetes are known to be associated with decreased
wild-type adipocytes. We examined ACC (Ser79) phosphorylation in circulating adiponectin levels, in particular high molecular
lysates from cells treated with adipocyte-conditioned medium
(0–30 min). (B) A representative image together with quantitative weight (HMW), which correlate closely with various aspects
analysis (meanGS.E.M.) of four individual experiments. of cardiac remodeling and the metabolic syndrome (Liu et al.
2007, Abel et al. 2008). In this study, we used rats treated
majority of studies focusing on liver and skeletal muscle acutely with STZ to induce diabetes as an appropriate model
(Badman & Flier 2007). Given the recent appreciation of the providing a source of primary adipocytes which are known to
important contribution of adipokines to the pathophysiology secrete a reduced HMW adiponectin profile, indeed one
of heart failure (Bradham et al. 2002, Hopkins et al. 2007, which can dictate changes in glucose metabolism in skeletal
Karmazyn et al. 2007), we designed this study to examine muscle (Vu et al. 2007). When we used adipocytes derived from
whether adipokines directly regulate cardiomyocyte glucose these animals as a source of adipokines, the change in glucose
and fatty acid metabolism. Reports from our laboratory transport was significant but lower in magnitude than that
(Palanivel et al. 2006, 2007) and several others (Graveleau et al. elicited by wild-type adipokines. This is reminiscent of our
2005, Pineiro et al. 2005, Guo et al. 2007, Li et al. 2007) have previous studies in skeletal muscle cells and suggests that

www.endocrinology-journals.org Journal of Endocrinology (2008) 199, 389–397

396 R PALANIVEL AND OTHERS . Effect of adipokines on cardiomyocyte metabolism

adiponectin, whose level is decreased in conditioned medium wild-type adipokine mixtures promote a beneficial increase in
prepared using adipocytes from 3-day STZ-treated rats (Vu glucose uptake and oxidation, the increased fatty acid oxidation
et al. 2007) and even further in 6-day STZ-treated rats as shown which we observe here may be detrimental to the heart. While
here, may be the adipokine most likely to mediate these our study model allows analysis of the direct influence of
changes. Our studies using both recombinant adiponectin and adipokine mixtures to be determined without potentially
the globular C-terminal domain of adiponectin (Palanivel et al. confounding effects of other circulating hormones or metab-
2007) and those of others (Pineiro et al. 2005, Guo et al. 2007) olites, as well as centrally mediated neuronal inputs, it is clearly
demonstrate that this adipokine can stimulate glucose uptake in important not to extrapolate our findings too far in terms of
primary neonatal cardiomyocytes. These data suggest that physiological significance, since the latter clearly have a
reduced adiponectin action via reduced secretion from adipose significant physiological role.
tissue or a decrease in expression of its receptors (Guo et al. In summary, these results characterize the direct effects of
2007) may be detrimental in terms of myocardial metabolism physiologically relevant adipokine mixtures on cardiomyo-
in obesity and diabetes. It is also feasible that many other cyte metabolism and highlight the differential effects of
adipokine levels change in the two types of conditioned adipokine profiles derived from wild-type and STZ-induced
medium used in this study and our analysis of adipokine diabetic rat adipocytes. They suggest that having an adequate
profiles have highlighted that the significant decrease in leptin amount of adipose-derived adipokines is beneficial, yet too
and increase in IL6 may play an important role (Ceddia et al. much as is the case in obesity, too little as observed in
2002, Carey & Febbraio 2004). lipoatrophy or a disturbed profile of secreted adipokines, may
One of the most striking observations in our study was the be detrimental to myocardial metabolism and ultimately
enhanced stimulation of lactate production induced by cardiac function.
adipokines derived from diabetic rat adipocytes. This appeared
to result as a switch of metabolism toward lactate production at
the expense of glucose oxidation. The heart uses lactate as an Declaration of interest
energy source and can also produce lactate (An & Rodrigues
The authors declare that there is no conflict of interest that could be perceived
2006) with the physiological consequences of increased lactate as prejudicing the impartiality of the research reported.
production, and decreased ATP production from glucose
oxidation, being detrimental to performance of the heart
(Stanley et al. 2005). Indeed, metabolic abnormalities in STZ- Funding
induced diabetic hearts have been well characterized (Rama-
nadham et al. 1990), rendering these animals more susceptible to Funding for this work was provided by Heart and Stroke Foundation of
Canada. R P and X F also acknowledge Postdoctoral Fellowship and Graduate
heart failure associated with exaggerated left ventricular Studentship support, respectively, from Heart and Stroke Foundation of
remodeling including increased interstitial fibrosis and myocyte Canada, and G S acknowledges support from The Canadian Institutes of
apoptosis (Shiomi et al. 2003). Specifically, cardiac utilization of Health Research (CIHR) via a New Investigator award. V V is supported by a
lactate is reduced to an even greater extent than glucose Graduate Studentship from the Canadian Diabetes Association.
oxidation (Chatham et al. 1999a,b). Our previous study using
recombinant adiponectin showed no change in lactate
production upon acute treatment times as used in this study References
(Palanivel et al. 2007). Hence, the loss of a permissive
contribution of adiponectin in the context of adipokine Abel ED, Litwin SE & Sweeney G 2008 Cardiac remodeling in obesity.
Physiological Reviews 88 389–419.
mixtures or increased levels of another adipokine which directly
An D & Rodrigues B 2006 Role of changes in cardiac metabolism in
stimulates lactate production are potential mechanisms. development of diabetic cardiomyopathy. American Journal of Physiology.
As noted above, efficient uptake and oxidation of fatty acids Heart and Circulatory Physiology 291 H1489–H1506.
are essential in maintaining the bulk of myocardial ATP Badman MK & Flier JS 2007 The adipocyte as an active participant in energy
production; yet, perturbations in fatty acid metabolism to create balance and metabolism. Gastroenterology 132 2103–2115.
Borradaile NM & Schaffer JE 2005 Lipotoxicity in the heart. Current
a discrepancy between fatty acid uptake and oxidation, as often Hypertension Reports 7 412–417.
observed in obese and diabetic individuals or animal models, can Bradham WS, Bozkurt B, Gunasinghe H, Mann D & Spinale FG 2002 Tumor
have additional consequences. For example, accumulation of necrosis factor-alpha and myocardial remodeling in progression of heart
lipotoxic lipid products can mediate many detrimental effects on failure: a current perspective. Cardiovascular Research 53 822–830.
Buchanan J, Mazumder PK, Hu P, Chakrabarti G, Roberts MW, Yun UJ,
the heart, including apoptosis, impairing contractile function,
Cooksey RC, Litwin SE & Abel ED 2005 Reduced cardiac efficiency and
inducing or exacerbating arrhythmias and changes in cell altered substrate metabolism precedes the onset of hyperglycemia and
signaling and membrane function (Schaffer 2003). Our current contractile dysfunction in two mouse models of insulin resistance and
studies demonstrate that the ability of cardiomyocytes to uptake obesity. Endocrinology 146 5341–5349.
palmitate is enhanced by a normal adipokine mixture, which Carey AL & Febbraio MA 2004 Interleukin-6 and insulin sensitivity: friend or
foe? Diabetologia 47 1135–1142.
also elicits a small but significant increase in oxidation. Palmitate Ceddia RB, Koistinen HA, Zierath JR & Sweeney G 2002 Analysis of
uptake and, in particular, oxidation are compromised by paradoxical observations on the association between leptin and insulin
adipokines derived from diabetic rat adipocytes. Thus, although resistance. FASEB Journal 16 1163–1176.

Journal of Endocrinology (2008) 199, 389–397 www.endocrinology-journals.org

 Effect of adipokines on cardiomyocyte metabolism . R PALANIVEL AND OTHERS 397

Chatham JC, Gao ZP, Bonen A & Forder JR 1999a Preferential inhibition of Park SY, Cho YR, Kim HJ, Higashimori T, Danton C, Lee MK, Dey A,
lactate oxidation relative to glucose oxidation in the rat heart following Rothermel B, Kim YB, Kalinowski A et al. 2005 Unraveling the temporal
diabetes. Cardiovascular Research 43 96–106. pattern of diet-induced insulin resistance in individual organs and cardiac
Chatham JC, Gao ZP & Forder JR 1999b Impact of 1 wk of diabetes on the dysfunction in C57BL/6 mice. Diabetes 54 3530–3540.
regulation of myocardial carbohydrate and fatty acid oxidation. American Peterson LR, Herrero P, Schechtman KB, Racette SB, Waggoner AD,
Journal of Physiology 277 E342–E351. Kisrieva-Ware Z, Dence C, Klein S, Marsala J, Meyer T et al. 2004
Golfman LS, Wilson CR, Sharma S, Burgmaier M, Young ME, Guthrie PH, Effect of obesity and insulin resistance on myocardial substrate meta-
Van Arsdall M, Adrogue JV, Brown KK & Taegtmeyer H 2005 Activation of bolism and efficiency in young women. Circulation 109 2191–2196.
PPARgamma enhances myocardial glucose oxidation and improves Pineiro R, Iglesias MJ, Gallego R, Raghay K, Eiras S, Rubio J, Dieguez C,
contractile function in isolated working hearts of ZDF rats. American Gualillo O, Gonzalez-Juanatey JR & Lago F 2005 Adiponectin is
Journal of Physiology. Endocrinology and Metabolism 289 E328–E336. synthesized and secreted by human and murine cardiomyocytes. FEBS
Graveleau C, Zaha VG, Mohajer A, Banerjee RR, Dudley-Rucker N, Letters 579 5163–5169.
Steppan CM, Rajala MW, Scherer PE, Ahima RS, Lazar MA et al. 2005 Purdham DM, Zou MX, Rajapurohitam V & Karmazyn M 2004 Rat heart is
Mouse and human resistins impair glucose transport in primary mouse a site of leptin production and action. American Journal of Physiology. Heart
cardiomyocytes, and oligomerization is required for this biological action. and Circulatory Physiology 287 H2877–H2884.
Journal of Biological Chemistry 280 31679–31685. Ramanadham S, Doroudian A & McNeill JH 1990 Myocardial and metabolic
Guo Z, Xia Z, Yuen VG & McNeill JH 2007 Cardiac expression of abnormalities in streptozotocin-diabetic Wistar and Wistar–Kyoto rats.
adiponectin and its receptors in streptozotocin-induced diabetic rats. Canadian Journal of Cardiology 6 75–82.
Metabolism 56 1363–1371. Schaffer JE 2003 Lipotoxicity: when tissues overeat. Current Opinion in
Hopkins TA, Ouchi N, Shibata R & Walsh K 2007 Adiponectin actions in the Lipidology 14 281–287.
cardiovascular system. Cardiovascular Research 74 11–18. Shiomi T, Tsutsui H, Ikeuchi M, Matsusaka H, Hayashidani S, Suematsu N,
Karmazyn M, Purdham DM, Rajapurohitam V & Zeidan A 2007 Leptin as a Wen J, Kubota T & Takeshita A 2003 Streptozotocin-induced hyperglyce-
cardiac hypertrophic factor: a potential target for therapeutics. Trends in mia exacerbates left ventricular remodeling and failure after experimental
Cardiovascular Medicine 17 206–211. myocardial infarction. Journal of the American College of Cardiology 42 165–172.
Li L, Wu L, Wang C, Liu L & Zhao Y 2007 Adiponectin modulates carnitine Stanley WC, Recchia FA & Lopaschuk GD 2005 Myocardial substrate
palmitoyltransferase-1 through AMPK signaling cascade in rat cardiomyo- metabolism in the normal and failing heart. Physiological Reviews 85
cytes. Regulatory Peptides 139 72–79. 1093–1129.
Liu Y, Retnakaran R, Hanley A, Tungtrongchitr R, Shaw C & Sweeney G 2007 Vu V, Kim W, Fang X, Liu YT, Xu A & Sweeney G 2007 Coculture with
Total and high molecular weight but not trimeric or hexameric forms of primary visceral rat adipocytes from control but not streptozotocin-induced
adiponectin correlate with markers of the metabolic syndrome and liver injury diabetic animals increases glucose uptake in rat skeletal muscle cells: role of
in Thai subjects. Journal of Clinical Endocrinology and Metabolism 92 4313–4318. adiponectin. Endocrinology 148 4411–4419.
Lopaschuk GD, Folmes CD & Stanley WC 2007 Cardiac energy metabolism Wang P, Lloyd SG, Zeng H, Bonen A & Chatham JC 2005 Impact of altered
in obesity. Circulation Research 101 335–347. substrate utilization on cardiac function in isolated hearts from Zucker
McGavock JM, Victor RG, Unger RH & Szczepaniak LS 2006 Adiposity of diabetic fatty rats. American Journal of Physiology. Heart and Circulatory
the heart, revisited. Annals of Internal Medicine 144 517–524. Physiology 288 H2102–H2110.
Palanivel R, Eguchi M, Shuralyova I, Coe I & Sweeney G 2006 Distinct
effects of short- and long-term leptin treatment on glucose and fatty acid
uptake and metabolism in HL-1 cardiomyocytes. Metabolism 55 1067–1075. Received in final form 8 September 2008
Palanivel R, Fang X, Park M, Eguchi M, Pallan S, De Girolamo S, Liu Y,
Wang Y, Xu A & Sweeney G 2007 Globular and full-length forms of
Accepted 10 September 2008
adiponectin mediate specific changes in glucose and fatty acid uptake and Made available online as an Accepted Preprint
metabolism in cardiomyocytes. Cardiovascular Research 75 148–157. 11 September 2008

www.endocrinology-journals.org Journal of Endocrinology (2008) 199, 389–397