diet-induced obese mice published online march 21, 2013; · cd1d-independent nkt cell or an...

Paulsson-Berne, Peter Arner, Göran K. Hansson and Norbert GerdesDaniela Strodthoff, Anna M. Lundberg, Hanna E. Agardh, Daniel F.J. Ketelhuth, Gabrielle

Diet-Induced Obese MiceLack of Invariant Natural Killer T Cells Affects Lipid Metabolism in Adipose Tissue of

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 2013 American Heart Association, Inc. All rights reserved.

Greenville Avenue, Dallas, TX 75231is published by the American Heart Association, 7272Arteriosclerosis, Thrombosis, and Vascular Biology published online March 21, 2013;Arterioscler Thromb Vasc Biol.

http://atvb.ahajournals.org/content/early/2013/03/21/ATVBAHA.112.301105World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://atvb.ahajournals.org/content/suppl/2013/03/21/ATVBAHA.112.301105.DC1.htmlData Supplement (unedited) at:

http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

http://www.lww.com/reprints

Information about reprints can be found online at: Reprints:

document. Question and AnswerPermissions and Rightspage under Services. Further information about this process is available in the

which permission is being requested is located, click Request Permissions in the middle column of the WebCopyright Clearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of theArteriosclerosis, Thrombosis, and Vascular Biologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on April 1, 2013http://atvb.ahajournals.org/Downloaded from

http://atvb.ahajournals.org/content/early/2013/03/21/ATVBAHA.112.301105

http://atvb.ahajournals.org/content/suppl/2013/03/21/ATVBAHA.112.301105.DC1.html

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://atvb.ahajournals.org//subscriptions/

http://atvb.ahajournals.org/

1

Natural killer T (NKT) cells are leukocytes that combine characteristics from both natural killer cells and T lym-

phocytes. In contrast to major histocompatibility complex class I and major histocompatibility complex II–restricted T cells, NKT cells recognize glycolipids presented by CD1d molecules. Two types of NKT cells can be distinguished based on their T-cell receptors. Type I NKT cells express the invariant T-cell receptor chain (Vα14-Vα18 [mouse], Vα24-Vα18 [human]), and they are restricted to recognize antigen presented by the major histocompatibility complex–related CD1d molecule.1 They account for up to 30% of the liver lym-phocytes in mice and are the focus of our study.2 Type II NKT cells on the contrary may have several different Vα chains.1

NKT cells were ascribed a crucial role in numerous pathol-ogies and conditions, such as viral and bacterial infections, autoimmune diseases, atherosclerosis, and cancer; however, little is known about their role in adipose tissue (AT) inflam-mation and metabolic disorders. Obesity is associated with a chronic inflammatory condition in the AT, characterized by

infiltration of macrophages and, as recently discovered, T cells.3–5 Studies of diet-induced obesity (DIO) in mice have demonstrated impairment of insulin sensitivity coinciding with infiltration of T cells into AT in early stages of obesity, when macrophages were not yet present.6 Although T cells might contribute to DIO-driven inflammation, they do not suf-fice to cause insulin resistance in hyperlipidemic, but lean, mice.5 However, several proinflammatory cytokines produced by T cells impact on metabolic pathways in AT, including tumor necrosis factor-α, interleukin-6, and interferon-γ.3,5,7 The liver has a central role in both glucose and lipid metabo-lism. Interestingly, it also harbors the main reservoir of NKT cells in the body, and these cells may exert a pronounced effect on lipid metabolism. In the current study, we examined the impact of invariant NKT (iNKT) cells on glucose and lipid metabolism in liver and AT.

Materials and MethodsMaterials and Methods are available in the online-only Supplement.

© 2013 American Heart Association, Inc.

Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org DOI: 10.1161/ATVBAHA.112.301105

June

130,190

Objective—Obesity promotes a chronic inflammatory condition in adipose tissue (AT). Impairment of insulin sensitivity coincides with infiltration of T cells into AT in early stages of obesity, when macrophages are not yet present. Here, we examine the role of invariant natural killer T (iNKT) cells, a subtype of T cells activated by lipid antigens, on glucose and lipid metabolism in obesity.

Approach and Results—Jα18−/− mice, specifically lacking iNKT cells, and wild-type mice consumed a chow or high-fat diet for 10 weeks. One third of all T lymphocytes in the liver of wild-type mice were iNKT cells, whereas few were detected in AT. Diet-induced obesity increased blood glucose in both genotypes of mice, whereas glucose tolerance test revealed similar kinetics of glucose clearance in Jα18−/− and wild-type mice. Under obese conditions, expression of inflammatory cytokines in AT did not differ between the groups, although the number of T cells and macrophages was lower in Jα18−/− mice. Nonetheless, AT homeostasis in Jα18−/− mice was altered evidenced by lower AT weight, smaller adipocytes, accelerated lipogenesis, increased expression of hormone-sensitive lipase, and accelerated basal lipolysis.

Conclusions—iNKT cells do not affect glucose clearance but rather modulate lipid metabolism in both liver and AT. Only few iNKT cells are found in AT under lean and obese conditions, suggesting that their effects on lipid metabolism are mainly mediated in the liver, their primary host organ. (Arterioscler Thromb Vasc Biol. 2013;33:00-00.)

Key Words: immune system ◼ leukocytes ◼ lipases ◼ NKT cells ◼ obesity

Received on: December 30, 2012; final version accepted on: February 26, 2013.From the Department of Medicine, Center for Molecular Medicine, Karolinska Institute, Karolinska University Hospital, Stockholm, Sweden (D.S.,

A.M.L., H.E.A., D.F.J.K., G.P.-B., G.K.H., N.G.); Department of Medicine, Karolinska University Hospital Huddinge, Karolinska Institute, Stockholm, Sweden (P.A.); Institute for Cardiovascular Prevention (IPEK), Ludwig-Maximilians-University (LMU), Munich, Germany (N.G.); and Department of Medical Biochemistry, Academic Medical Center (AMC), University of Amsterdam, Amsterdam, The Netherlands (N.G.).

*These authors contributed equally to this work.The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.112.301105/-/DC1.Correspondence to Norbert Gerdes, Institute for Cardiovascular Prevention (IPEK), Pettenkoferstrasse 9, Ludwig-Maximilians University Munich,

80336 Munich, Germany. E-mail [email protected]

Lack of Invariant Natural Killer T Cells Affects Lipid Metabolism in Adipose Tissue of Diet-Induced Obese Mice

Daniela Strodthoff,* Anna M. Lundberg,* Hanna E. Agardh, Daniel F.J. Ketelhuth, Gabrielle Paulsson-Berne, Peter Arner, Göran K. Hansson, Norbert Gerdes

Original Article


http://atvb.ahajournals.org/lookup/suppl/10.1161/ATVBAHA.112.301105/-/DC1

mailto:[email protected]


2 Arterioscler Thromb Vasc Biol June 2013

ResultsiNKT Cells Are a Major Lymphocyte Population in Liver, but Not in ATStaining with α-GalCer–loaded CD1d-Ig dimer protein (DimerX) was used in combination with anti-NK1.1 to iden-tify iNKT cells in AT and liver. Flow cytometric analysis of liver leukocytes confirmed specific deficiency of iNKT cells in Jα18−/− mice (<0.2%), whereas this population repre-sented 32% of all CD4+ T cells in the liver of wild-type mice (Figure 1A and 1B). In contrast to liver, AT of wild-type as well as Jα18−/− mice contained very few or no iNKT cells, respectively (Figure 1D and 1E). Jα18−/− mice contained a slightly increased subpopulation of CD4+ lymphocytes that expressed NK1.1 but did not bind DimerX (Figure IC and ID in the online-only Data Supplement), probably representing a CD1d-independent NKT cell or an NKT-cell like T-cell pop-ulation. Additional analyses using counting beads revealed that AT contained <1000 iNKT cells per gram tissue, whereas this number was ≈200-fold higher in the liver (Figure 1C and 1F). Furthermore, real-time polymerase chain reaction analy-sis for the specific Vα14Jα18 T-cell receptor mRNA of liver and AT confirmed that iNKT cells are not present in Jα18−/− mice (Figure IA and IB in the online-only Data Supplement).

Absence of iNKT Cells Does Not Influence Glucose ClearanceTo examine the impact of iNKT cell deficiency on glu-cose metabolism, ipGTT was performed on wild-type and

Jα18−/− mice that consumed high-fat diet (HFD) or chow diet for 10 weeks. Although fasting glucose did not differ between the groups (Figure 2A), Jα18−/− mice had lower body and fat weight, irrespective of the diet (Figure 3A and 3B). ipGTT revealed that lack of iNKT cells does not affect blood glucose clearance after consuming chow diet. DIO impaired blood glucose clearance similarly in Jα18−/− and wild-type mice (Figure 2B and 2C). Similar results were obtained when mice were fasted for only 6 hours before performing ipGTT (Figure IIA and IIB in the online-only Data Supplement). In addition, fasting insulin levels in the circulation were lower in Jα18−/− than in wild-type mice fed with a chow diet but did not differ significantly in mice fed with HFD (Figure 2D). We also measured the kinetics of insulin release after an IP glucose challenge. Interestingly, wild-type mice required much higher insu-lin levels to maintain glycemic control compared with Jα18−/− mice (Figure 2E). Accordingly, calculation of the Homeostasis Model of Assessment–Insulin Resistance confirmed that insulin sensitivity was better preserved in Jα18−/− mice compared with controls (Figure 2F), which also was associated with enhanced expression of insulin-regulated glucose transporters (Glut2 and -4) in liver and AT, respectively (Figure 2G). Plasma leptin concentration was decreased in Jα18−/− mice on HFD as compared with wild-type mice (Figure 2H), whereas circulating levels of adiponectin were unaltered in Jα18−/− compared with wild-type mice (Figure 2I).

100

101

102

103

104

100

101

102

103

104

100

101

102

103

104

100

101

102

103

104

100

101

102

103

104

100

101

102

103

104

100

101

102

103

104

100

101

102

103

104

NK1.1

Dim

erX

NK1.1

Dim

erX

wild-type Jα18-/-

D E

32% 0.19%

NK1.1

Dim

erX

Dim

erX

NK1.1

BA

0.6% 0.17%

F

C

wild-type Jα18-/-0

500

1000

1500

**

HFD

iNK

T c

ells

/g A

dipo

e tis

sue

wild-type Jα18-/-0

50000

100000

150000

200000 ***

HFD

iNK

T c

ells

/g L

iver

Figure 1. Few invariant natural killer T (iNKT) cells are present in adipose tissue (AT) flow cytometric analysis of iNKT cells in liver (A–C) and AT (D–E) from wild-type (A, D) and Jα18−/− mice (B, E) fed high-fat diet (HFD) for 10 to 14 weeks. Representative plots (A, B, D, and E) and absolute numbers (C, F) are shown. Cells in the gate are defined as percentage of single, live CD19−, CD3+, CD4+, DimerX+, NK1.1+ lymphocytes. Absolute number of iNKT cells were analyzed in individual mice. Liver: n=8; AT: n=6 or 8, wild-type and Jα18−/−, respec-tively. Data are presented as mean±SEM. **P<0.01, ***P<0.001



Strodthoff et al iNKT Cells Affect Adipose Tissue Lipid Metabolism 3

iNKT Cell–Deficient Mice Have Less Fat, Smaller Adipocytes, but Similar Lean Body MassThe diminished body weight and AT mass (Figure 3A and 3B) resulted from both lower weight at study start and diminished weight gain over the period of HFD consumption (Figure IIC and IID in the online-only Data Supplement). These data led us to investigate body composition in HFD-fed mice. Indeed, the reduced body fat weight in Jα18−/− mice was confirmed by MRI and attributable to both smaller visceral and subcutaneous fat depots, whereas lean body mass was unchanged (Figure 3C and 3G). This decrease in body fat suggested differences in AT homeostasis in Jα18−/− mice. Indeed, adipocyte volume, calculated from the cell diameter, was decreased in Jα18−/− mice, whereas the number of adipocytes was stable (Figure 3H and 3I). This change in adipocyte size was observed under both dietary conditions but paralleled the reduced AT weight and overall body weight in Jα18−/− mice, which appeared as a result of reduced food intake (Figure 3J). Of note, most parameters (eg, body weight, AT weight, glucose control) were confirmed in female mice (Figure IVA–IVD).

Increased Lipogenesis in iNKT Cell–Deficient Mice Is Counterbalanced by Elevated Lipase Expression and Basal LipolysisLipogenesis, determined by incorporation of radio-labeled glucose into lipids of the adipocytes, was increased in iNKT cell–deficient mice compared with wild-type mice with and without insulin stimulation. Notably, in both strains of mice, lipogenesis was lower after consuming HFD when compared with those fed with chow diet (Figure 4A and 4B). Increased lipogenesis in AT of Jα18−/− mice did not result in elevated plasma triglyceride or cholesterol concentrations (Figure 4C and 4D) on chow diet or HFD. Considering this discrepancy, we analyzed mRNA for lipoprotein lipase (LPL) and hor-mone-sensitive lipase (HSL), the 2 major enzymes involved in degradation and hydrolyzation of triglycerides in AT. Indeed, quantitative polymerase chain reaction revealed increased expression of LPL and HSL in Jα18−/− mice consuming both types of diet, suggesting that the elevated lipogenesis was counterbalanced by enhanced triglyceride breakdown, lead-ing to an overall accelerated triglyceride metabolism in AT of mice lacking iNKT cells (Figure 4E). In accordance with the

0

50

100

150

Chow HFD

Fas

tin

g g

luco

se [

mg

/dl]

0

0.5

1.0

1.5

Chow HFD

**

Insu

lin [

µg

/l]

wild-type Jα18-/-0

4

8

12*

HO

MA

-IR

wild-type Jα18-/-

wild-type Jα18-/-

0

0.5

1.0

1.5

2.0

Glut2 (liver) Glut4 (AT)

** **

Rel

ativ

e m

RN

A le

vels

0

5

10

15

Chow HFD

Ad

ipo

nec

tin

[µ

g/m

l]

A B C

D E F

G H I

wild-typeJα18-/-

0 15 30 45 60 1200

100

200

300

Chow wild-typeChow Jα18-/-

HFD wild-typeHFD Jα18-/-

Time post glucose injection (min)

Glu

cose

[m

g/d

l]

0

10000

20000

30000

Chow HFD

Glu

cose

[A

UC

]

wild-type Jα18-/- 0

1000

2000

3000 *

Insu

lin s

ecre

tio

n [

AU

C]

0

50

100

150

200

Chow HFD

*

Lep

tin

[n

g/m

l]

Figure 2. Absence of invariant natural killer T (iNKT) cells does not influence glucose clearance but improves insulin sensitivity. A, Fasting glucose was measured in wild-type and Jα18−/− mice that consumed either a chow diet or high-fat diet (HFD) for 10 weeks. B, ipGTT was performed on overnight-fasted mice by injecting glucose (1g/kg; IP), and (C) area under the curve (AUC) was calculated. Concentration of glucose and (E) of circulating insulin after glucose stimulation was analyzed in blood drawn from the tail vein. F, Homeostasis Model of Assessment–Insulin Resistance was calculated using fasted glucose and fasted insulin. D, Insulin, (H) leptin, and (I) adiponectin were measured in plasma. G, mRNA expression of Glut2 (liver) and Glut4 (adipose tissue [AT]). Open symbols, wild-type mice; filled symbols, Jα18−/− mice. N=7 to 17 mice per group. Data are presented as mean±SEM, *P<0.05, **P<0.01.




key role of LPL and HSL in lipolysis, we observed signifi-cantly increased lipolysis in Jα18−/− mice on HFD, whereas a similar tendency was seen in the mice on chow diet (Figure

4F). Additional hormonal stimulation of lipolysis did not dif-fer between wild-type and Jα18−/− mice (Figure IIIA and IIIB in the online-only Data Supplement). Similar results were

wild-type Jα18-/- wild-type Jα18-/-0

20

40

60

Chow HFD

*** *****

*

Bo

dy

wei

gh

t [g

]

wild-type Jα18-/- wild-type Jα18-/- 0

0.5

1.0

1.5

*** *

Chow HFD

******

Ad

ipo

se ti

ssu

e [g

]

Subcutaneous fat

Visceral fat

wild-type

Jα18-/-

wild-type Jα18-/-0

10

20

30

Lea

n b

od

y m

ass

[g]

wild-type Jα18-/- wild-type Jα18-/-0

200

400

600

Chow HFD

** *

Ad

ipo

cyte

vo

lum

e [p

L]

wild-type Jα18-/-0

20

40

60

n.s.

Ad

ipo

cyte

cel

l nu

mb

er [x

106 ]

wild-type Jα18-/-0

2

4

6

***

Fo

od

inta

ke [

g/d

ay]

A B

C D E

F G

H I

wild-type Jα18-/-0

5

10

15

To

tal b

od

y fa

t [g

]

wild-type Jα18-/-0

2

4

6*

Vis

cera

l fat

[g

]

J

n.s.

wild-type Jα18-/-0

2

4

6

8*

Su

bcu

tan

eou

s fa

t [g

]*

Figure 3. Less fat, smaller adipocytes, but similar lean body mass, in invariant natural killer T (iNKT) cell–deficient mice. A, Body weight and (B) gonadal adipose tissue (1 fat pad) weight from Jα18−/− and wild-type mice fed either chow diet or high-fat diet (HFD) was mea-sured. C–G, Axial images of abdominal fat deposits were analyzed by MRI (n=8 per genotype). C, Total body fat, (D) visceral, and (E) subcutaneous fat. F, Representative image from wild-type and Jα18−/− mice. G, Lean body mass. H, Adipocyte volume and (I) adipocyte number (n=8 per genotype). J, Food intake (n=11 or 10 for wild-type and Jα18−/− mice, respectively). Data are presented as mean±SEM, *P<0.05, **P<0.01, ***P<0.001.




obtained in female mice (Figure IVE and IVF in the online-only Data Supplement).

Although iNKT cell deficiency influences lipid metabolism in AT, no change in expression of genes involved in triglyc-eride formation (Dgat) or cholesterol synthesis (HMG-CoA reductase) was observed in the liver (Figure 4G and 4H). The latter observation is consistent with the unaltered cho-lesterol concentrations measured in the plasma (Figure 4D). Moreover, mRNA expression of hepatic lipase and MTTP was unaffected in the liver of Jα18−/− mice. As expected, LIGHT mRNA was decreased in Jα18−/− mice in both dietary condi-tions (Figure 4G and 4H), probably because of the absence of LIGHT-expressing iNKT cells in the liver. Finally, levels of free fatty acids increased in mice of both genotypes but did not differ between wild-type mice and those deficient for iNKT cells, suggesting operation of additional mechanism of free fatty acids catabolism (Figure 4I).

Lack of iNKT Cells Does Not Influence Expression of Inflammatory Genes in AT on DIOQuantitative polymerase chain reaction analysis revealed a diminished inflammatory response in the liver of Jα18−/− com-pared with wild-type mice after consuming either chow diet or HFD (Figure VA–VC in the online-only Data Supplement; Figure 5E). Absolute numbers of leukocytes and their major subsets, such as macrophages and T cells, were decreased in the liver of Jα18−/− mice compared with wild-type controls, both when calculating total cells per organ (Figure 5B–5D) or correcting for organ weight (data not shown). In addition, under normal dietary conditions, Jα18−/− mice displayed lower expression in AT of some inflammatory transcripts (eg, CD68 and tumor necrosis factor-α), indicating a role for iNKT cells in immune homeostasis under physiological conditions (Figure VD and VF in the online-only Data Supplement). Although we observed lower numbers of inflammatory cells

Figure 4. Increased lipogenesis in invariant natural killer T (iNKT) cell–deficient mice is counterbalanced by increased lipase expression and enhanced basal lipolysis. Lipogenesis was determined in adipose tissue (AT) of mice consuming (A) chow diet or (B) high-fat diet (HFD) for 10 weeks. For each analysis, 500 mg AT was used; samples were pooled if nec-essary. C, Triglycerides (TG) and (D) cholesterol were mea-sured in plasma from Jα18−/− and wild-type mice fed either chow diet or HFD. E, Transcript of hormone-sensitive lipase (HSL) and lipoprotein lipase (LPL) was analyzed by quantita-tive polymerase chain reaction (qPCR) in AT. F, Basal lipoly-sis was determined by release of glycerol in mice fed with a chow diet or HFD. Transcript of Dgat, HMG-CoA, hepatic lipase, LIGHT, and Mttp was analyzed by qPCR in liver of mice consuming (G) chow diet and (H) HFD. Expression was normalized to Hprt. I, Free fatty acids (FFA) was measured in plasma. Open bars, wild-type; filled bars, Jα18−/− mice. n=7 to 17 mice per group if not indicated otherwise. Data are pre-sented as mean±SEM, *P<0.05, **P<0.01, ***P<0.001.

0

2

4

6

** *

basal + Insulin

Chow

Lip

og

enes

is[n

mo

l glu

cose

/2h

/g T

G]

0

0.5

1.0

1.5

basal + Insulin

* *HFD

Lip

og

enes

is[n

mo

l glu

cose

/2h

/g T

G]

wild-typeJα18-/-A B

0

0.5

1.0

1.5

Chow HFD

Tri

gly

ceri

de

[mM

]

C D

E F

G H

I

0

0.5

1.0

1.5

2.0

HSL LPL

Chow HFD

** ** ** *

HSL LPLHSL

Rel

ativ

e m

RN

A le

vels

Chow HFD

0

1

2

3

4

5**

Chow HFD

Bas

al li

po

lysi

s[G

lyce

rol µ

mo

l/g T

G]

0

1

2

3

4

Chow diet HFD

Ch

ole

ster

ol [

mM

]

0

0.5

1.0

1.5

2.0

Dgat HMGCoA

Hepaticlipase

LIGHT Mttp

*Chow

Rel

ativ

e m

RN

A le

vels

0

0.5

1.0

1.5

2.0

2.5

Dgat HMGCoA

Hepaticlipase

LIGHT Mttp

*HFD

Rel

ativ

e m

RN

A le

vels

wild-type Jα18-/- wild-type Jα18-/- 0

0.2

0.4

0.6

0.8

1.0 ******

FF

A [

mM

]




residing in AT of Jα18−/− mice consuming HFD (Figure 5F–5I), no difference in expression of any of the immune-related transcripts was found in the AT, suggesting that these cells do not directly influence obesity-induced AT inflammation (Figure 5J; Figure VE in the online-only Data Supplement). Of note, total numbers of leukocytes and their subsets were considerably higher in liver than in AT (Figure 5) mirroring the distribution of iNKT cells (Figure 1C and 1F), and sug-gesting that lack of these cells may influence processes in their primary host organ more potently than in AT.

DiscussionThe results of the present study show that iNKT cells modu-late lipid and insulin metabolism without directly affecting plasma glucose clearance. Furthermore, they show that meta-bolically active iNKT cells reside in much greater numbers in the liver rather than in AT. Using a well-defined combination of markers for iNKT cells, we found only few of them in AT, despite repeated analyses of T-cell subpopulations. However, we cannot exclude that these cells, although small in number, may actively modulate adipocyte metabolism. Our data are contrasted by groups reporting large numbers of iNKT cells in AT8–10 but corroborate other studies that failed to detect iNKT cells in AT.11,12 Although we cannot provide a final explanation for this controversy, differences in preparation or experimen-tal protocols are most likely. For instance, subtle differences in the genetic background of the mouse strains under study, as well as microbiological conditions in the different animal facilities, could influence the metabolic and immune state of the animals.13 Furthermore, analytical methods differ with regard to flow cytometric staining procedures and reagents (eg, we used aGalCer-loaded CD1d dimer, whereas others use CD1d tetramers), and the present study is the only one using mRNA analysis to validate conclusions from flow cytometric data. In addition, we found considerable binding of the CD1d dimer to dead cells, which were excluded from our quantita-tive analysis (data not shown). Finally, HFD may change the activation status of the iNKT cells and influence the expres-sion level of their specific T-cell receptors, thus compromising conclusions when using CD1d multimers.

The impairment of glucose clearance on HFD was similar in mice that lack iNKT cells as in wild-type mice. This is in line with a recent published article by Mantell et al11 but con-trasts the findings of Ohmura et al.14 We observed that lack of iNKT cells results in better insulin sensitivity reflected by lower insulin levels and elevated basal expression of insulin-sensitive glucose transporters in liver and AT. However, this increased sensitivity did not result in improved glucose clear-ance, probably as a result of lower insulin level secreted by the mice lacking iNKT cells. We further investigated the cross-talk between glucose and lipid metabolism using the mice devoid of iNKT cells attributable to targeted deletion of the iNKT cell-specific Jα18 T-cell receptor gene. Our study revealed that iNKT cells mainly reside in the liver, and that their absence leads to the following: (1) diminished inflam-matory response in liver; (2) increased lipogenesis, which is counterbalanced by increased HSL and LPL expression and increased lipolysis; and (3) decreased adipocyte volume.

Although a possible malignant role in glucose metabo-lism was recently proposed,14 our data rather suggest that iNKT cells residing in the liver influence AT homeostasis. We observed that mice lacking iNKT cells have lower body weight, implying that AT homeostasis may differ compared with wild-type mice. Indeed, adipocytes from Jα18−/− mice were smaller than those of control mice, whereas their num-ber did not change. In addition, MRI analysis revealed similar lean body mass in the 2 groups, whereas Jα18−/− mice dis-played a reduction of total body fat that was attributable to both the visceral and subcutaneous fat compartment.

Several factors define AT homeostasis, the balance of tri-glyceride synthesis (lipogenesis), and breakdown of triglyc-eride (lipolysis) in adipocytes on one hand and uptake of triglycerides from the circulation on the other. The dynam-ics of adipose lipid turnover seem to play a major role in metabolic disease.15 Indeed, although smaller in volume, adipocytes from Jα18−/− mice have an increased turnover of triglycerides from the circulation indicated by increased expression of LPL in AT. Furthermore, we detected elevated expression of HSL in AT. Of note, LPL and HSL expres-sion and activity can be regulated by different cytokines. Both tumor necrosis factor-α, which was decreased in AT and liver of Jα18−/− mice fed with normal chow diet, and interferon-γ, which was decreased in the liver, are known to inhibit LPL and HSL.16–18 HSL is the key enzyme initiating adipocyte lipolysis,19,20 and we could demonstrate that the increased HSL expression in Jα18−/− mice is accompanied by increased (basal) lipolysis. The increased lipolysis poten-tially promotes release of free fatty acids that is counterbal-anced by improved lipogenesis overall, resulting in plasma triglyceride concentrations that are similar to that of wild-type mice. The discrepancy between an increased lipolysis and unchanged free fatty acids levels in the Jα18−/− mice may also be a consequence of enhanced catabolic processes that use circulating fatty acids and can also result from increased insulin sensitivity.21

There is a strong connection between obesity and fatty liver disease, also termed nonalcoholic steatohepatitis,22 indicating an indirect impact of cytokines/adipokines released in the liver or AT. Indeed, studies observed reduced NKT cell number in leptin-deficient ob/ob mice,23 and adoptive transfer of NKT cells into ob/ob mice positively influenced glucose metabo-lism, resulting in decreased liver fat content.24 Although the mechanisms underlying these findings are not fully under-stood, we studied the connection between liver and AT under obese conditions from another perspective by depleting iNKT cells, which resulted in decreased leptin secretion from AT. Indeed, iNKT cells may directly affect leptin release, but it is also possible that the morphological changes (ie, smaller adi-pocytes and lower fat mass) lead to reduced leptin secretion.

NKT cells are a diverse subpopulation of T cells, classi-fication of which was refined over the past years. They have been described in AT as CD3+NK1.1+ T cells.4 Nowadays, it is widely accepted that the characterization of NKT cells only based on NK1.1 is not sufficient.1 We used the combina-tion of α-GalCer-loaded CD1d dimer (DimerX) and NK1.1 to identify iNKT cells and found only few iNKT cells in AT




of wild-type mice. Of note, a minor DimerX-NK1.1+ T-cell population could be detected in Jα18−/− mice. These cells may represent type II NKT cells, the role of which in AT was recently studied by Satoh et al,12 who corroborate our findings of minor presence and functional involvement of iNKT cells in AT inflammation during DIO. Another explanation is that these DimerX-NK1.1+CD3+CD4+ cells are mucosal-associ-ated NKT cells, so far only detected in the liver, blood, spleen, lymph nodes, and bone marrow.25 However, our study focused on the invariant CD1d-restricted iNKT cells, whereas other subsets were not further investigated.

Furthermore, we could not identify any changes in expres-sion of inflammatory cytokines in AT of obese mice lacking iNKT cells compared with wild-type mice. Corroborating the pivotal role of iNKT cells in the liver, Jα18−/− mice displayed a less inflammatory phenotype compared with the wild-type mice, displayed by reduced CD3, interferon-γ, and LIGHT expression evoked by the removal of a major, liver-resident T-lymphocyte population in these mice. Interestingly, LIGHT was previously shown expressed by hepatic iNKT cells and modulates liver metabolism.26,27 Although in our study, iNKT cell deficiency reduced cellular infiltrate in both liver and AT

wild-type Jα18-/- 0

0.5

1.0

1.5 **

Liv

er w

eig

ht [

g]

wild-type Jα18-/-0

Mac

rop

hag

es[x

106 /to

tal l

iver

]

wild-type Jα18-/-0

*

Leu

kocy

tes

[x 1

06 / to

tal l

iver

]

1

2

3

0.05

0.1

0.15

0CD3

+

*

wild-type

Jα18-/-

0.5

1

1.5

T c

ells

[x10

6 /to

tal l

iver

]

CD4+0

1

2

3

TNFα IL-6 IFNγ

*

Rel

ativ

e m

RN

A le

vels

wild-type Jα18-/-0

**

Leu

kocy

tes

[x10

6 /tota

l AT

]

0.5

1.0

1.5

0CD3+ CD4+

** **

T c

ells

[x10

6 /tota

l AT

]

0.2

0.4

0.6

A CB

D E

F G H

I J

wild-type Jα18-/-0

1

2

3**

AT

wei

gh

t [g

]

wild-type Jα18-/-0

0.2

0.4

0.6

0.8**

Mac

rop

hag

es[x

106 /to

tal A

T]

0

0.5

1.0

1.5

2.0

TNFα IL-6 IFNγ

Rel

ativ

e m

RN

A le

vels

Figure 5. Lack of invariant natural killer T (iNKT) cells reduces leukocyte numbers in liver and adipose tissue (AT) and inflammatory gene expression in the liver. Weight of (A) liver and (F) AT was determined in wild-type and Jα18−/− mice consuming high-fat diet (HFD) for 14 weeks. Total numbers of (B, G) leukocytes, (C, H) macrophages (F4/80+), (D, I) CD3+ and CD4+ T cells in (B–D) liver and (G–I) AT was determined by flow cytometry using counting beads. Transcripts of signature inflammatory cytokines (tumor necrosis factor [TNF]-α, interleukin [IL]-6, and interferon [IFN]-γ) were analyzed by quantitative polymerase chain reaction (qPCR) in (E) liver and (J) AT. Expression was normalized to Hprt. Open bars, wild-type; filled bars, Jα18−/− mice. For qPCR, n=11 and 8 for wild-type and Jα18−/−, respectively. For weight and flow cytometry, n=6 to 8 for both groups. Data are presented as mean±SEM, *P<0.05, **P<0.01, ***P<0.001.




under obese conditions, transcripts of effector cytokines were lower only in the liver, suggesting that the reduced pathol-ogy is mainly mediated in this main iNKT cell reservoir. Nonetheless, these changes in the liver may likely lead to a healthier AT phenotype and overall preserved insulin sensi-tivity. In turn, this enhanced insulin sensitivity in iNKT cell–deficient mice may promote lipogenesis because this process is tightly regulated by insulin.

Our findings were sex-independent because we observed very similar results in female mice. However, male mice, which do not have pronounced hormonal fluctuations and develop more rapid obesity, are most widely used for study-ing DIO.28 Hence, our study focused on the HFD response in male mice.

Notably, our conclusions differ from those recently pre-sented by Ohmura et al,14 who showed reduced numbers of inflammatory cells in AT and improved glucose control in β2 microglobulin-deficient (b2m−/−) mice. The authors attribute these changes to the absence of NKT cells in these mice,14 although β2 microglobulin also functions as a component of the major histocompatibility complex class I complex,29 which is prerequisite for functional CD8+ T cells. Because of the absence of this major T-cell population, a qualified assessment of the contribution of NK and all NKT cells to AT inflammation and metabolic disturbance seems question-able by this study. The important role of CD8+ T cells in AT inflammation and glucose intolerance was initially shown by Nishimura et al30 and was recently confirmed by Mantell et al.11 Using a mouse model that lacks CD1d, but otherwise has a complete T-cell repertoire, they support our data that NKT cells do not impair glucose clearance. Here, we expand the current knowledge by using a highly specific mouse model lacking iNKT cells.

In summary, our data demonstrate a strong metabolic inter-action between liver and AT. We show that depleting iNKT cells in the liver can influence AT homeostasis and lipid metabolism.

AcknowledgmentsWe thank Mikael Karlsson for providing Ja18−/− mice; and Ingrid Törnberg, Anneli Olsson, Anna-Lena Gustafsson, and Melanie Cremer for technical help; and Sahar Nikkhou Aski, Peter Damberg, and Stefan Brene for assistance with the MRI measurements.

Sources of FundingThis work was supported by project grants from the Swedish Research Council (grants 521-2009-4203 [project grant] and 349-2007-8703 [CERIC Linnaeus support]), the Swedish Heart-Lung Foundation, the Foundation for Strategic Research (SSF), Vinnova Foundation, Stockholm County Council, the European Commission (AtheroRemo collaborative project), Loo and Hans Ostermans foundation, Strategic Research Program in Diabetes at Karolinska Institute and the Leducq Network on Atherothrombosis.

DisclosuresNone.

References 1. Godfrey DI, Stankovic S, Baxter AG. Raising the NKT cell family. Nat

Immunol. 2010;11:197–206.

2. Gao B, Jeong WI, Tian Z. Liver: an organ with predominant innate immu-nity. Hepatology. 2008;47:729–736.

3. Hotamisligil GS. Inflammation and metabolic disorders. Nature. 2006;444:860–867.

4. Caspar-Bauguil S, Cousin B, Galinier A, Segafredo C, Nibbelink M, André M, Casteilla L, Pénicaud L. Adipose tissues as an ancestral immune organ: site-specific change in obesity. FEBS Lett. 2005;579:3487–3492.

5. Sultan A, Strodthoff D, Robertson AK, Paulsson-Berne G, Fauconnier J, Parini P, Rydén M, Thierry-Mieg N, Johansson ME, Chibalin AV, Zierath JR, Arner P, Hansson GK. T cell-mediated inflammation in adipose tis-sue does not cause insulin resistance in hyperlipidemic mice. Circ Res. 2009;104:961–968.

6. Kintscher U, Hartge M, Hess K, et al. T-lymphocyte infiltration in visceral adipose tissue: a primary event in adipose tissue inflammation and the development of obesity-mediated insulin resistance. Arterioscler Thromb Vasc Biol. 2008;28:1304–1310.

7. Rocha VZ, Folco EJ, Sukhova G, Shimizu K, Gotsman I, Vernon AH, Libby P. Interferon-gamma, a Th1 cytokine, regulates fat inflammation: a role for adaptive immunity in obesity. Circ Res. 2008;103:467–476.

8. Lynch L, Nowak M, Varghese B, Clark J, Hogan AE, Toxavidis V, Balk SP, O’Shea D, O’Farrelly C, Exley MA. Adipose tissue invariant NKT cells protect against diet-induced obesity and metabolic disorder through regulatory cytokine production. Immunity. 2012;37:574–587.

9. Wu L, Parekh VV, Gabriel CL, Bracy DP, Marks-Shulman PA, Tamboli RA, Kim S, Mendez-Fernandez YV, Besra GS, Lomenick JP, Williams B, Wasserman DH, Van Kaer L. Activation of invariant natural killer T cells by lipid excess promotes tissue inflammation, insulin resistance, and hepatic steatosis in obese mice. Proc Natl Acad Sci USA. 2012;109:E1143–E1152.

10. Schipper HS, Rakhshandehroo M, van de Graaf SF, et al. Natural killer T cells in adipose tissue prevent insulin resistance. J Clin Invest. 2012;122:3343–3354.

11. Mantell BS, Stefanovic-Racic M, Yang X, Dedousis N, Sipula IJ, O’Doherty RM. Mice lacking NKT cells but with a complete complement of CD8+ T-cells are not protected against the metabolic abnormalities of diet-induced obesity. PLoS ONE. 2011;6:e19831.

12. Satoh M, Andoh Y, Clingan CS, Ogura H, Fujii S, Eshima K, Nakayama T, Taniguchi M, Hirata N, Ishimori N, Tsutsui H, Onoé K, Iwabuchi K. Type II NKT cells stimulate diet-induced obesity by mediating adipose tissue inflam-mation, steatohepatitis and insulin resistance. PLoS ONE. 2012;7:e30568.

13. Tremaroli V, Bäckhed F. Functional interactions between the gut micro-biota and host metabolism. Nature. 2012;489:242–249.

14. Ohmura K, Ishimori N, Ohmura Y, Tokuhara S, Nozawa A, Horii S, Andoh Y, Fujii S, Iwabuchi K, Onoé K, Tsutsui H. Natural killer T cells are involved in adipose tissues inflammation and glucose intol-erance in diet-induced obese mice. Arterioscler Thromb Vasc Biol. 2010;30:193–199.

15. Arner P, Bernard S, Salehpour M, Possnert G, Liebl J, Steier P, Buchholz BA, Eriksson M, Arner E, Hauner H, Skurk T, Rydén M, Frayn KN, Spalding KL. Dynamics of human adipose lipid turnover in health and metabolic disease. Nature. 2011;478:110–113.

16. Sumida M, Sekiya K, Okuda H, Tanaka Y, Shiosaka T. Inhibitory effect of tumor necrosis factor on gene expression of hormone sensitive lipase in 3T3-L1 adipocytes. J Biochem. 1990;107:1–2.

17. Morin CL, Schlaepfer IR, Eckel RH. Tumor necrosis factor-alpha elimi-nates binding of NF-Y and an octamer-binding protein to the lipoprotein lipase promoter in 3T3-L1 adipocytes. J Clin Invest. 1995;95:1684–1689.

18. Harris SM, Harvey EJ, Hughes TR, Ramji DP. The interferon-gamma-mediated inhibition of lipoprotein lipase gene transcription in macrophages involves casein kinase 2- and phosphoinositide-3-kinase-mediated regula-tion of transcription factors Sp1 and Sp3. Cell Signal. 2008;20:2296–2301.

19. Kraemer FB, Shen WJ. Hormone-sensitive lipase: control of intracel-lular tri-(di-)acylglycerol and cholesteryl ester hydrolysis. J Lipid Res. 2002;43:1585–1594.

20. Arner P. Human fat cell lipolysis: biochemistry, regulation and clinical role. Best Pract Res Clin Endocrinol Metab. 2005;19:471–482.

21. Carpentier AC, Frisch F, Brassard P, Lavoie F, Bourbonnais A, Cyr D, Giguère R, Baillargeon JP. Mechanism of insulin-stimulated clearance of plasma nonesterified fatty acids in humans. Am J Physiol Endocrinol Metab. 2007;292:E693–E701.

22. el-Hassan AY, Ibrahim EM, al-Mulhim FA, Nabhan AA, Chammas MY. Fatty infiltration of the liver: analysis of prevalence, radiological and clinical features and influence on patient management. Br J Radiol. 1992;65:774–778.

23. Guebre-Xabier M, Yang S, Lin HZ, Schwenk R, Krzych U, Diehl AM. Altered hepatic lymphocyte subpopulations in obesity-related murine fatty




livers: potential mechanism for sensitization to liver damage. Hepatology. 2000;31:633–640.

24. Elinav E, Pappo O, Sklair-Levy M, Margalit M, Shibolet O, Gomori M, Alper R, Thalenfeld B, Engelhardt D, Rabbani E, Ilan Y. Adoptive transfer of regulatory NKT lymphocytes ameliorates non-alcoholic steatohepatitis and glucose intolerance in ob/ob mice and is associated with intrahepatic CD8 trapping. J Pathol. 2006;209:121–128.

25. Wingender G, Kronenberg M. Role of NKT cells in the digestive sys-tem. IV. The role of canonical natural killer T cells in mucosal immu-nity and inflammation. Am J Physiol Gastrointest Liver Physiol. 2008;294:G1–G8.

26. Niemeyer M, Darmoise A, Mollenkopf HJ, Hahnke K, Hurwitz R, Besra GS, Schaible UE, Kaufmann SH. Natural killer T-cell characterization through gene expression profiling: an account of versatility bridging

T helper type 1 (Th1), Th2 and Th17 immune responses. Immunology. 2008;123:45–56.

27. Hansson GK. Medicine. LIGHT hits the liver. Science. 2007;316:206–207. 28. Pettersson US, Waldén TB, Carlsson PO, Jansson L, Phillipson M. Female

mice are protected against high-fat diet induced metabolic syndrome and increase the regulatory T cell population in adipose tissue. PLoS ONE. 2012;7:e46057.

29. Koller BH, Marrack P, Kappler JW, Smithies O. Normal development of mice deficient in beta 2M, MHC class I proteins, and CD8+ T cells. Science. 1990;248:1227–1230.

30. Nishimura S, Manabe I, Nagasaki M, Eto K, Yamashita H, Ohsugi M, Otsu M, Hara K, Ueki K, Sugiura S, Yoshimura K, Kadowaki T, Nagai R. CD8+ effector T cells contribute to macrophage recruitment and adipose tissue inflammation in obesity. Nat Med. 2009;15:914–920.

SignificanceOur study investigated the role of invariant natural killer T (iNKT) cells, a glycolipid-recognizing T-cell subpopulation, in glucose and lipid metabolism using a mouse model, specifically lacking iNKT cells. Contrary to some recent reports, we found that the majority of iNKT cells reside in the liver rather than in adipose tissue. In fact, we believe that the few iNKT cells in visceral fat will not exert a substantial contri-bution to adipose tissue inflammation, which is increasingly recognized as a key pathological process promoting type 2 diabetes mellitus. Nonetheless, rather than affecting glucose control, our additional data indicate that lack of iNKT cells promotes higher lipid turnover in adipose tissue, despite smaller adipocyte size overall associated with a more lean phenotype in diet-induced obese mice. Accordingly, therapies targeting iNKT cells may still hold promise in ameliorating metabolic diseases other than diabetes mellitus.



Supplemental Material to Strodthoff et al.: Lack of Invariant Natural Killer T Cells Affects Lipid

Metabolism in Adipose Tissue of Diet‐Induced Obese Mice Page 1

Supplemental material to Strodthoff et al.:

Lack of Invariant Natural Killer T Cells Affects Lipid Metabolism in Adipose

Tissue of Diet-Induced Obese Mice

Supplementary Figure I: J18-/- mice lack V14J18 T cell receptor-expressing

iNKT cells. Mice were fed HFD for 10 weeks. Real-time PCR analysis of the iNKT

cell-specific V14J18 T cell receptor in (A) liver and (B) AT of wild-type and J18-/-

mice. Analysis of NKT cell sub-populations by flow cytometry in (C) liver (D) AT. Cells

in the gate are defined as percentage of single, live, CD19-, CD3+, CD4+,

lymphocytes. Quantitation of flow cytometry pooled from 4-6 mice (AT 2 pools for

each strain; liver 2 pools (J18-/-) and 3 pools (wild-type)) are shown. Data are

presented as mean±SEM. *P<0.05.





Supplementary Figure II: Absence of iNKT cells does not influence glucose

clearance. (A) ipGTT was performed by injecting glucose (1g/kg; i.p.) in 6h-fasted

mice fed a HFD for 10 weeks (B) AUC was calculated. Glucose concentration was

analyzed in blood drawn from the tail vein. (C) Body weight and (D) body weight gain

was monitored during the time of diet. Open symbol, wild-type mice; filled symbol,

J18-/- mice. N=7-8 mice per genotype. Data are presented as mean±SEM.





Supplementary Figure III: Hormone-stimulated lipolysis is unchanged

Isolated adipocytes were incubated with isoprenaline (Iso), norepinephrine

(NA), and 8-bromo-cyclic AMP (8-Br-cAMP). Lipolysis was determined by release of

glycerol in mice fed a (A) chow diet or (B) HFD and is expressed as fold induction to

basal lipolysis. Open bars, wild-type mice; closed bars, J18-/- mice. Data are

presented as mean±SEM.





Supplementary Figure IV: Absence of iNKT cells does not influence glucose

clearance in female mice. (A) Body weight and (B) gonadal adipose tissue weight of

female mice fed a HFD for 14 weeks. N=6, wild-type and n=5 for J18-/- mice. (C)

ipGTT was performed on overnight-fasted mice by injecting glucose (1g/kg; i.p.) and

(D) AUC was calculated. Glucose concentration was analyzed in blood drawn from





the tail vein. N=7 per genotype (E) Lipogenesis was determined in AT of mice on

HFD by incorporation of radiolabelled glucose into adipocyte lipids under basal and

after insulin stimulation. Lipogenesis is shown in nmol of glucose incorporated in 2

hours/gram lipid. (F) Basal lipolysis was determined by release of glycerol in mice

consuming HFD. N=6, wild-type and n= 3 for J18-/- mice. Open symbol, wild-type

mice; filled symbol, J18-/- mice. Data are presented as mean±SEM. *P<0.05,

**P<0.01, ***P<0.001, n.s.= non significant.





Supplementary Figure V: Lack of iNKT cells reduces inflammatory status in the

liver.

Transcripts of cell type-specific markers were analyzed by RT-PCR in (A, B) liver and

(D,E) AT and transcript of cytokines were analyzed in (C) liver and (F) AT from mice

consuming (A,C,D,F) chow diet or (B,E) HFD. Expression was normalized to Hprt;

open bars, wild-type; filled bars, J18-/- mice. n=8-11 for all. Data are presented as

mean±SEM. *P<0.05, **P<0.01, ***P<0.001.



Material and Methods to Strodthoff et al.: Lack of Invariant Natural Killer T Cells Affects Lipid


Material and Methods to Strodthoff et al.:

Lack of Invariant Natural Killer T Cells Affects Lipid Metabolism in Adipose

Tissue of Diet-Induced Obese Mice

Animal experiments

Five to 8 weeks old J18-/- mice1, which have a non-functional V14J18 T cell

receptor gene and consequently lack iNKT cells, and C57BL/6 wild-type mice were

fed a high-fat diet (HFD; 34.9% fat, Altromin, Lage, Germany) or normal chow diet for

10-14 weeks. Intraperitoneal glucose tolerance test (ipGTT) was performed on

overnight (i.e., 12-14h) fasted mice by injecting glucose (1g/kg body weight; i.p.) and

monitoring glucose concentration (Abbott Scandinavia AB, Solna, Sweden) in blood

drawn from the tail vein. HOMA-IR was calculated as following: fasting glucose

(mg/dl) x fasting insulin (mU/l) / 405. Male mice were used if not indicated otherwise.

All animal experiments were approved by the Stockholm regional board for animal

experimentation.

Tissue processing

Overnight-fasted mice were euthanized and blood was collected in EDTA tubes

before perfusion of the arterial tree with PBS. Tissue was harvested and kept in cold

PBS or was immediately frozen at -80oC (for mRNA analysis) until further processing.

Where applicable, whole body weight, liver and AT weight was determined

immediately. AT weight of one gonadal fat pad was measured.2





Plasma analysis

Blood was centrifuged and plasma collected and aliquoted. Triglyceride and

cholesterol were measured by using an enzymatic colorimetric method (Randox,

Crumlin, UK). NEFA (Wako Chemicals GmbH, Neuss, Germany), glycerol (Sigma-

Aldrich, St. Louis, USA), Insulin (Mercodia, Uppsala, Sweden and Crystal Chem

INC., Illinois, USA), adiponectin (R&D Systems, Minneapolis, USA), and leptin

(Peprotech, Rocky Hill, USA) were measured according to the protocols provided by

the manufacturers.

Flow cytometry

Hepatic mononuclear cells were prepared as described previously.3 Gonadal AT of

the respective group (J18-/-, wild-type) was pooled when indicated and incubated in

Krebs-Ringer buffer (KRB) containing 0.05% collagenase type I and

4% BSA (both Sigma-Aldrich, St. Louis, USA) for 1h at 37oC. The cell suspension

was filtered through a nylon mesh (200µm, Sefar, Heiden, Switzerland) and

centrifuged following erythrocyte lysis (ACK-lysis buffer). Following two washing

steps with PBS the suspension was passed through 100µm cell strainer (BD), and

resuspended in PBS. Cells from this stromal-vascular cell fraction or hepatic

mononuclear cells were incubated with the following antibodies: NK1.1-PE or-FITC,

CD8a-PerCp, CD4-APC-H7, CD3-Pacific Blue, and CD19-APC (all BD, Franklin

Lakes, USA). CD1d dimers (DimerX, BD, Franklin Lakes, USA) were loaded with -

galactosylceramide (GalCer; Larodan Fine Chemicals, Malmö, Sweden) according

to the manufacturer’s instructions and pre-incubated with FITC- or PE-conjugated rat

anti-mouse IgG1 (BD, Franklin Lakes, USA). Dead cells were excluded employing

the Live/Dead® fixable aqua dead cell stain kit (Invitrogen, Carlsbad, USA).

CountBright™ Absolute Counting Beads (Invitrogen) were used to determine





absolute cell number. Samples were fixed in 2% formaldehyde in PBS and analyzed

in a CyanADP flow cytometer (Beckman Coulter, Miami, USA).

Adipocyte cell size and cell number

The diameter of 100 isolated adipocytes per sample was determined by microscopy.

The mean adipocyte volume was calculated according to Hirsch and Gallian and

sagittal diameter was used.4 Five hundred mg AT was used to determine adipocyte

size, lipogenesis, and lipolysis. AT from J18-/- mice was pooled in groups of 2-3

mice to compensate small tissue sample size. Adipocyte number of 8 mice per

genotype was calculated as total fat pad weight in gram divided by the fat cell weight

times 10000.

Lipogenesis

Lipogenesis was analyzed as described in detail elsewhere.5 In brief, isolated fat

cells were incubated for 2 h at 37oC at a concentration of 2% (v/v) in KRB buffer (pH

7.4) containing albumin (20 mg/ml), [3-3H] glucose (5x106 dpm/ml), unlabelled

glucose (1 μM) and human insulin at different concentrations (10-15 -10-6 M).

Incubations were stopped by rapidly chilling to 4°C and adding H2SO4. The

incorporation of radiolabelled glucose into adipocyte lipids was determined in a

scintillator. Lipogenesis was expressed as nmol of glucose incorporated/2 hours/g

TG.

Lipolysis

Glycerol release was used as an index of lipolysis. Adipocytes were incubated for 2

hours at 37°C in a shaking bath in KRB containing 2% BSA, 1 mg/ml glucose;

isoprenaline (Iso) at 10-9-10-5 M, norepinephrine (NA) at 10-12-10-4 M, and 8-bromo-

cyclic AMP (8-Br-cAMP) at 10−3 M. The total lipid content was measured





gravimetrically after extraction with heptane. At the end of the incubation, a cell-free

aliquot of the incubation medium was removed for glycerol determination by a

bioluminescence method. Lipolysis was expressed as the glycerol concentration per

lipid weight of the incubated adipocytes. Hormone induced lipolysis was expressed

as fold induction compared to basal lipolysis.

RNA isolation and quantitative PCR

Total RNA was isolated from liver and gonadal AT using the RNeasy Lipid Mini kit

(Qiagen, Hilden, Germany) and reverse transcription was performed with

Superscript-II (Invitrogen, Carlsbad, USA) and random hexamers. cDNA was

amplified by RT-PCR using primers and probes (obtained as “assays on demand”

from Applied Biosystems, Carlsbad, USA) for CD3, CD68, IFN, TNF, IL-6,

lipoprotein lipase (Lpl), hormone sensitive lipase (Hsl), hepatic lipase, diglyceride

acyltransferase (Dgat), 3-hydroxy-3-methyl-glutaryl-CoA reductase (HMG-CoA),

hepatic lipase, TNF ligand superfamily member 14 (LIGHT), microsomal triglyceride

transfer protein (Mttp) and hypoxanthine-guanine phosphoribosyltransferase (Hprt).

V14J18 TCR was amplified, as also described previously,6 7 using the following

primers; Forward primer: 5'-TGG ATG ACA CTG CCA CCT ACAT-3'; Reverse

primer: 5'-TCC AAA ATG CAG CCT CCC TA-3'; The probe was newly designed:

6FAM-5'-CAG CCT CCC TAA GGC TGA ACC TCT-3'-TAMRA.

All amplification and analyses were performed on an ABI 7900HT Sequence Detector

(Applied Biosystems). Data were calculated as 2-∆∆CT, where ∆∆CT = ∆CT (sample) –

∆CT (calibrator = average CT values of all samples within each group), and ∆CT is the

CT of the target gene after subtraction of the CT value for the housekeeping gene

(Hprt).





Food intake

Food intake was monitored in individually-housed mice over a period of 48h and

displayed as food intake [g] per mouse per day [24h].

Magnetic resonance imaging (MRI)

To determine the distribution of body fat magnetic resonance imaging (MRI) were

performed on isoflurane-anesthetized mice. Animals were placed in a 9.4 T magnet

with a bore size of 31 cm (Varian, Yarnton, UK) in a gradient system with a 12 cm

inner diameter and a maximum gradient strength of 600 mT/m. The core body

temperature and the respiration were monitored during scanning (SA-instruments)

and body temperature was maintained at 37°C during MRI scanning with a

thermostated warm air system (SA-instruments). A volume coil of the millipede

design (Varian.inc) with 40 mm inner diameter and 110 mm RF-window was

employed both for excitation and detection. 40-45 axial slices, from neck to tail, of 1

mm thickness with 0.5 mm gap, with a field-of-view of 40x40 mm2 and a matrix size

of 256x96 were acquired. The spin-echo sequence was employed with an echo time

of 15 ms and one average. Preceding every excitation pulse, a 6 ms five lobe sinc

pulse, achieving a selective 90 flip for water followed by a 2 ms crusher gradient of

11 G/cm was applied to saturate water, while leaving the fat signal unperturbed.

Respiration gating was employed, with ten blocks of data acquired during each

expiration period. A pixel counting-based determination of fat volumes was performed

on two-dimensional image series analyzed using Fiji image analysis. A density factor

of 0.9 g/ml was used to convert fat volumes (in ml) into fat mass (in g). Lean fat mass

was calculated by body weight subtracted by total fat (BW-total fat).

Statistics





Results are expressed as mean ± SEM. Statistically significant differences were

determined by the Mann Whitney test, and p values <0.05 were considered

significant. Area under the curve (AUC) was calculated to compare differences over

time.

Reference

1. Cui J, Shin T, Kawano T, Sato H, Kondo E, Toura I, Kaneko Y, Koseki H, Kanno M, Taniguchi M. Requirement for valpha14 nkt cells in il‐12‐mediated rejection of tumors. Science. 1997;278:1623‐1626

2. Rogers P, Webb GP. Estimation of body fat in normal and obese mice. Br J Nutr. 1980;43:83‐86

3. Watanabe H, Ohtsuka K, Kimura M, Ikarashi Y, Ohmori K, Kusumi A, Ohteki T, Seki S, Abo T. Details of an isolation method for hepatic lymphocytes in mice. J Immunol Methods. 1992;146:145‐154

4. Hirsch J, Gallian E. Methods for the determination of adipose cell size in man and animals. J Lipid Res. 1968;9:110‐119

5. Arner P, Engfeldt P. Fasting‐mediated alteration studies in insulin action on lipolysis and lipogenesis in obese women. Am J Physiol. 1987;253:E193‐201

6. Naumov YN, Bahjat KS, Gausling R, Abraham R, Exley MA, Koezuka Y, Balk SB, Strominger JL, Clare‐Salzer M, Wilson SB. Activation of cd1d‐restricted t cells protects nod mice from developing diabetes by regulating dendritic cell subsets. Proc Natl Acad Sci U S A. 2001;98:13838‐13843

7. Tupin E, Nicoletti A, Elhage R, Rudling M, Ljunggren HG, Hansson GK, Berne GP. Cd1d‐dependent activation of nkt cells aggravates atherosclerosis. J Exp Med. 2004;199:417‐422



diet-induced obese mice published online march 21, 2013; · cd1d-independent nkt cell or an...

Documents