db mice

basic research

www.kidney-international.org

Blocking lysophosphatidic acid receptor 1 signaling inhibits diabetic nephropathy in db/db mice Hui Ying Li1,2, Yoon Sin Oh1,3, Ji-Woong Choi4, Ji Yong Jung3,5 and Hee-Sook Jun1,3,4 1 Lee Gil Ya Cancer and Diabetes Institute, Gachon University, Incheon, Korea; 2Department of Internal Medicine, Yanbian University Hospital, Yanji, Jilin Province, China; 3Gachon Medical Research Institute, Gil Hospital, Incheon, Korea; 4College of Pharmacy, Gachon University, Incheon, Korea; and 5Division of Nephrology, Department of Internal Medicine, Gachon University School of Medicine, Incheon, Korea

Lysophosphatidic acid (LPA) is known to regulate various biological responses by binding to LPA receptors. The serum level of LPA is elevated in diabetes, but the involvement of LPA in the development of diabetes and its complications remains unknown. Therefore, we studied LPA signaling in diabetic nephropathy and the molecular mechanisms involved. The expression of autotaxin, an LPA synthesis enzyme, and LPA receptor 1 was significantly increased in both mesangial cells (SV40 MES13) maintained in high-glucose media and the kidney cortex of diabetic db/db mice. Increased urinary albumin excretion, increased glomerular tuft area and volume, and mesangial matrix expansion were observed in db/db mice and reduced by treatment with ki16425, a LPA receptor 1/3 antagonist. Transforming growth factor (TGF)b expression and Smad-2/ 3 phosphorylation were upregulated in SV40 MES13 cells by LPA stimulation or in the kidney cortex of db/db mice, and this was blocked by ki16425 treatment. LPA receptor 1 siRNA treatment inhibited LPA-induced TGFb expression, whereas cells overexpressing LPA receptor 1 showed enhanced LPA-induced TGFb expression. LPA treatment of SV40 MES13 cells increased phosphorylated glycogen synthase kinase (GSK)3b at Ser9 and induced translocation of sterol regulatory element-binding protein (SREBP)1 into the nucleus. Blocking GSK3b phosphorylation inhibited SREBP1 activation and consequently blocked LPA-induced TGFb expression in SV40 MES13 cells. Phosphorylated GSK3b and nuclear SREBP1 accumulation were increased in the kidney cortex of db/db mice and ki16425 treatment blocked these pathways. Thus, LPA receptor 1 signaling increased TGFb expression via GSK3b phosphorylation and SREBP1 activation, contributing to the development of diabetic nephropathy. Kidney International (2017) j.kint.2016.11.010

-, -–-;

http://dx.doi.org/10.1016/

Correspondence: Yoon Sin Oh, Lee Gil Ya Cancer and Diabetes Institute, College of Medicine, Gachon University, 7-45 Songdo-dong, Yeonsu-ku, Incheon, Korea. E-mail: [email protected] or Hee-Sook Jun, Lee Gil Ya Cancer and Diabetes Institute, College of Pharmacy and Gachon Institute of Pharmaceutical Science, 7-45 Songdo-dong, Yeonsu-ku, Incheon, Korea. E-mail: [email protected] Received 19 March 2016; revised 7 November 2016; accepted 10 November 2016 Kidney International (2017) -, -–-

KEYWORDS: diabetic nephropathy; extracellular matrix (ECM); lysophosphatidic acid (LPA); mesangial cell; transforming growth factor–b (TGF-b) Copyright ª 2016, International Society of Nephrology. Published by Elsevier Inc. All rights reserved.

D

iabetic nephropathy is the major microvascular complication of both type 1 and type 2 diabetes.1 High expression of extracellular matrix (ECM) proteins, such as collagen, fibronectin, and laminin in the mesangial cells of glomeruli, leads to thickening of the basement membrane2 and results in glomerulosclerosis, which is a characteristic pathological feature of type 2 diabetes–induced nephropathy.3,4 Transforming growth factor–b (TGF-b) is a key mediator for high expression of ECM proteins5–8; therefore, increased TGF-b expression in mesangial cells promotes ECM accumulation and hypertrophy during progression of diabetic nephropathy. Treatment with reninangiotensin system inhibitors or tight control of glucose levels can prevent diabetic nephropathy, but they are not completely effective in preventing disease progression. Reninangiotensin inhibitors are not tolerated by all patients with diabetic nephropathy1; therefore, early prevention or delaying progression to diabetic nephropathy is necessary. Lysophosphatidic acid (LPA) is a small, ubiquitous phospholipid involved in cellular processes such as proliferation, survival, migration, and suppression of apoptosis.9 LPA is released from activated platelets at sites of injury, and increased LPA production during inflammation mediates proinflammatory effects on several cell types within the kidney.10,11 Moreover, LPA has been demonstrated to be a profibrotic mediator in various organs, such as the liver, lung, kidney, peritoneum, skin, retina, and heart.12–14 LPA mediates its cellular effects via binding to at least 6 G-protein–coupled receptors (LPAR1–LPAR6).15,16 Pharmacological tools specifically targeting LPA receptor subtypes have been developed, including ki16425, which specifically blocks LPAR1 and LPAR3 subtypes,17 and is used for various LPA-related disorders.18,19 It was reported that the plasma level of LPA was increased in high-fat, diet-induced obese mice20 and in the glomeruli of diabetic mice.21 Moreover, the expression of autotaxin (ATX), a hydrolysis enzyme that produces LPA from lysophosphatidyl choline, was significantly increased in the kidney cortex of db/ db mice compared with control mice,22 as well as in the serum of patients with diabetic nephropathy.23 These observations 1

basic research

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

led us to hypothesize that LPA could be involved in the development of glomerulosclerosis and could contribute to the progression of diabetic nephropathy. In the present study, we found that ki16425 reduced the progression of diabetic nephropathy in db/db mice. Moreover, glycogen synthase kinase 3b (GSK3b) (Ser9) phosphorylation and subsequent sterol regulatory element–binding protein (SREBP) 1 activation was involved in TGF-b production induced by LPA in the diabetic condition. RESULTS LPAR1 expression is upregulated in both the renal cortex of diabetic db/db mice and simian virus-transformed mouse mesangial cells maintained in high glucose

It was reported that LPA was elevated in the serum of diabetic animal models,20 and that the mRNA levels of ATX were

a

d

α-SMA

LPAR1

Merge

DAPI

WT-1

LPAR1

Merge

300 250

*

f

* Wild type db/db

200

LPAR1 LPAR2

150

β-actin

100 50

5 mM G

0 ATX LPAR1 LPAR2 LPAR3

5 mM G 25 mM G (mannitol)

G M

ni

M m 5

g Relative density % of con

350

l)

G M G

G M m 5

Relative density % of con

c

an

an

0.0

(m

to ni

M 5

m

db/db

(m

Wild type

0.5

m

0.0

1.0

to

1.0

1.5

25

2.0

G

β-actin

3.0

m

LPAR3

4.0

2.0

5

LPAR2

LPAR2 mRNA level (relative ratio)

LPAR1 mRNA level (relative ratio)

ATX

LPAR1

* 5.0

G

db

ild

ty

ty pe db /d b

ild

e

l)

0.4 0.0 pe

0.4 0.0

M

0.8

m

1.2 0.8

25

1.2

/d b

1.6

W

ild W

DAPI

LPAR3

1.6

/d b

ty

pe

ty pe db /d b

ild W

b

*

6.0 5.0 4.0 3.0 2.0 1.0 0.0

db

*

2.5 2.0 1.5 1.0 0.5 0.0

LPAR2

LPAR1

W

ATX mRNA level (relative ratio)

significantly increased in the adipose tissue of patients with diabetes.24 We first examined the expression of ATX, the enzyme required for LPA production, in the kidney cortex of db/db mice. mRNA (Figure 1a) and protein (Figure 1b and c) levels of ATX were significantly increased in db/db mice compared with wild-type mice, which was consistent with a previously reported study.22 mRNA expression of lipid phosphate phosphatases, which are enzymes associated with extracellular LPA degradation, were not different between the 2 groups (Supplementary Figure S1). Next, we examined the expression of LPAR subtypes, and found that LPAR1, LPAR2, and LPAR3 were expressed, but other receptor subtypes (e.g., LPAR4 and LPAR 5) were not detected in the kidney cortex of both db/db and wild-type mice (data not shown). The mRNA (Figure 1a) and protein (Figure 1b and c) levels of LPAR1 were significantly increased in db/db mice compared with

500

* 5 mM G

400

5 mM G (mannitol)

300

25 mM G

200 100 0 LPAR1

LPAR2

Figure 1 | mRNA and protein levels of lysophosphatidic acid receptor 1 (LPAR1) are upregulated in both the renal cortex of db/db mice and simian virus-transformed mouse mesangial (SV40 MES13) cells maintained in high glucose. mRNAs and proteins were isolated from the renal cortex of 16-week-old wild-type and db/db mice. (a) mRNA levels of autotaxin (ATX), LPAR1, LPAR2, and LPAR3 were measured by quantitative real-time polymerase chain reaction (qRT PCR) (n ¼ 3–4). (b) Protein levels of ATX, LPAR1, LPAR2, and LPAR3 were measured by Western blot. (c) The results were quantified and b-actin was used as a loading control (n ¼ 6–8). Representative pictures of (d) immunofluorescence staining for colocalization of a–smooth muscle actin (a-SMA) or Wilms’ tumor 1 (WT-1) with LPAR1, which were examined in the kidney sections of 16-week-old db/db mice. Bars ¼ 20 mm (n ¼ 6). (e,f) mRNA and protein levels of LPAR1 and LPAR2 in SV40 MES13 cells maintained in low glucose (5 mM), low glucose (5 mM) þ mannitol (19.44 mM), and high glucose (25 mM) were examined by qRT PCR and Western blot. (g) The results were quantified, and b-actin was used as a loading control (n ¼ 3 independent experiments). *P < 0.05. Data represent the mean  SEM. DAPI, 40 ,6-diamidino-2-phenylindole. 2

Kidney International (2017) -, -–-

basic research

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

wild-type mice, but the levels of LPAR2 and LPAR3 were not different between the 2 groups. To investigate which cell types expressed LPAR1 within the kidney cortex, we performed immunofluorescence staining of kidney sections with antibodies against LPAR1, a–smooth muscle actin (a-SMA), which is a marker for injured mesangial cells, and Wilms’ tumor 1 (WT-1), which is a marker for podocytes.25,26 We found that a-SMA–positive cells were increased in the glomeruli of db/db mice compared with wild-type mice (Supplementary Figure S2A), and most a-SMA–positive cells also expressed LPAR1. In contrast, few WT-1–positive podocytes expressed LPAR1 (Figure 1d). When we examined the expression of SMA, a marker for normal mesangial cells,27 we found SMA-positive cells in the glomeruli in both wild-type and db/db mice, and the cells also immunoreacted with the LPAR1 antibody (Supplementary Figure S2B). Because LPAR1 was mainly expressed in mesangial cells, we checked the expression level of LPAR mRNA and protein in a mouse renal mesangial cell line— simian virus-transformed mouse mesangial cells (SV40 MES13). As shown in Figure 1e to g, LPAR1 and LPAR2 were expressed in SV40 MES13 cells, but not LPAR3 (data not shown). Moreover, both mRNA and protein levels of LPAR1 were significantly increased in high glucose (25 mM) compared with cells incubated in low glucose (5 mM) or glucose (5 mM) þ mannitol (19.44 mM) as an osmotic control. In contrast, there was no difference in the LPAR2 mRNA or protein levels between cells incubated in low or high glucose concentrations (Figure 1e–g). Treatment with ki16425 ameliorates albuminuria without affecting glucose homeostasis in db/db mice

Because we found that the expression levels of LPAR1 were significantly increased in the renal cortex of db/db mice, we investigated whether treatment with ki16425 affected the progression of diabetic nephropathy. Eight-week-old db/db mice were injected daily with ki16425 or vehicle for 8 weeks, and various clinical and pathological parameters were

analyzed. Serum creatinine, urinary albumin, and the albumin/creatinine ratio were increased in db/db mice compared with wild-type mice, and these levels were significantly decreased by ki16425 treatment (Table 1), which is consistent with a reduction in kidney injury. Body weights, kidney weights, cholesterol, and triglycerides were not changed in ki16425-treated mice compared with vehicle-treated db/db mice (Table 1). Similarly, ki16425 treatment did not affect glycosylated hemoglobin (HbA1c), nonfasting blood glucose (Table 1), or glucose tolerance and insulin tolerance (Supplementary Figure S3) in db/db mice. Treatment with ki16425 ameliorates mesangial matrix accumulation in glomeruli of db/db mice

Because glomerular matrix accumulation is a hallmark of diabetic nephropathy, we measured glomerular histopathological changes of kidney tissues in wild-type, db/db, and ki16425-treated db/db mice by hematoxylin-eosin staining, periodic acid-Schiff staining, and fibronectin immunohistochemistry (Figure 2a). We found that the glomerular tuft area, glomerular volume, and mesangial matrix index were significantly increased in db/db mice, which reflects the degree of ECM deposition (Figure 2b–d). The increases in glomerular tuft area, glomerular volume, and mesangial matrix index in db/db mice were significantly ameliorated by ki16425 treatment (Figure 2b–d). Blockade of LPAR1 signaling reduces the expression of TGF-b and fibronectin both in the renal cortex of db/db mice and in SV40 MES13 cells

TGF-b is a profibrotic factor involved in mesangial matrix accumulation and glomerulosclerosis in diabetic nephropathy.2,5 Therefore, we investigated whether TGF-b expression is altered in the kidney cortex of db/db mice. mRNA and protein levels of TGF-b were significantly increased in db/db mice compared with wild-type mice (Figure 3a–c). Fibronectin mRNA (Supplementary Figure S4A) and phosphorylated Smad2/3 (Figure 3d), which are major downstream signaling

Table 1 | Biochemical and physical characteristics of wild-type, db/db, and ki16425-treated db/db mice at the end of the 8-week experimental period Characteristic Body weight (g) Kidney weight (g) HbA1c (%) Non-fasting blood glucose (mg/dl) Food intake (g/d) Water intake (ml/d) Cholesterol (mg/dl) Triglyceride (mg/dl) Serum creatinine (mg/dl) 24-h albuminuria (mg) ACR (mg/mg)

Wild type 23.3 0.34 4.5 142.2 3.22 3.94 93.18 75.03 0.32 0.07 303.82

          

0.3 0.007 0.1 12.6 0.03 0.16 2.72 6.61 0.01 0.01 26.16

db/db 45.8 0.43 11.1 591.4 6.15 17.94 118.19 138.66 0.40 1.71 1229.07

          

0.5a 0.02a 0.6a 16.8a 0.12a 0.12a 5.11a 7.32a 0.004a 0.26a 130.04a

db/db D ki16425 43.2 0.43 10.2 519.4 6.04 16.67 113.36 131.26 0.37 1.10 780.46

          

1.4a 0.02a 0.3a 15.2a 0.38a 0.47a 3.32a 6.83a 0.01a,b 0.17a,b 97.13a,b

ACR, albumin/creatinine ratio; HbA1c, glycosylated hemoglobin. Data are mean  SEM. a P < 0.05 versus wild-type mice; bP < 0.05 versus db/db, (n ¼ 9 to 10).

Kidney International (2017) -, -–-

3

basic research

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

a

Wild type

db/db

db/db+ki16425

H&E

PAS

Fibronectin

*

*

150 100 50

5 42

b

16

/d

ki

db

b+

/d db

ty

pe

0 ild

Mesangial matrix index (%)

200

W

25

b

64

/d

i1

db

pe W

ild

ty

42

b

16

/d

ki

db db

/d

b+

ty ild W

*

50 0

5

0

*

250 200 150 100

+k

50

d

db /d b

100

pe

Glomerular tuft area (% of control)

*

*

200 150

Glomerular volume (% of control)

c

b

Figure 2 | Treatment with ki16425 ameliorates mesangial matrix accumulation in cortex of db/db mice. Eight-week-old wild-type and db/ db mice were injected with vehicle or ki16425 (10 mg/kg i.p.) for 8 weeks, and mice were killed. (a) Representative staining pictures of hematoxylin and eosin (H & E), periodic acid–Schiff (PAS), and fibronectin staining of kidney tissues in mice. Bars ¼ 20 mm. (b) Glomerular tuft area, (c) glomerular volume, and (d) the mesangial matrix index (percentage) were quantified from PAS staining using ImageJ Software (30 glomeruli per mouse, n ¼ 6). *P < 0.05. Data represent mean  SEM.

molecules of TGF-b,28 were also increased in db/db mice. In ki16425-treated db/db mice, TGF-b (mRNA and protein), fibronectin mRNA and phosphorylated Smad2/3 were significantly decreased compared with db/db mice (Figure 3a–d and Supplementary Figure S4A). In SV40 MES13 cells, the mRNA level of TGF-b was significantly increased at 3 hours after LPA treatment (Figure 3e), and protein levels were significantly increased at 3, 6, and 12 hours after LPA stimulation compared with untreated cells (Figure 3f and g). Fibronectin mRNA levels were also increased by LPA treatment in SV40 MES13 cells (Supplementary Figure S4B and C), similar to the findings in db/db mice. LPA-induced expression of TGF-b (Figure 3h–j) and fibronectin (Supplementary Figure S4D) was significantly reduced by 10 mM of ki16425 treatment in SV40 MES13 cells. Similarly, treatment of primary human mesangial cells (NHMC; Lonza, Walkersville, MD) with LPA (10 mM) increased TGF-b expression, which was reduced by ki16425 treatment (Figure 3k). When we inhibited TGF-b expression by siRNA treatment, upregulation of fibronectin expression in LPA-treated cells was inhibited (Figure 3l and m). To confirm that LPAR1 is involved in the induction of TGF-b expression, LPAR1 expression was transiently downregulated by siRNA treatment in SV40 MES13 cells, and TGF-b expression was analyzed. LPAR1 expression was efficiently reduced after 48 hours of LPAR1 siRNA 4

transfection (Figure 4a), and LPA-enhanced TGF-b expression was inhibited in LPAR1 siRNA-treated cells compared with scrambled siRNA-treated cells (Figure 4b). In other experiments, LPAR1 expression in SV40 MES13 cells was upregulated by transfection with an LPAR1-overexpressing vector (Figure 4c and d). In these cells, TGF-b expression by LPA treatment was increased in LPAR1-overexpressing SV40 MES13 cells compared with LPA-treated, control vectortransfected cells (Figure 4e). Treatment with H2L5186303, an LPAR2 antagonist, did not affect LPAinduced TGF-b or fibronectin mRNA expression (Supplementary Figure S4E and F). GSK3b phosphorylation, SREBP1 activation, and the PI3K pathway are involved in LPA-induced TGF-b transcription in SV40 MES13 cells

It is known that GSK3b is involved in the development of fibrosis, and GSK3b (Ser9) phosphorylation is increased in the kidney cortex of db/db and streptozotocin (STZ)-induced diabetic mice.29,30 In addition, mature-SREBP1 is one of the factors that regulates TGF-b transcription.31,32 Therefore, we checked whether LPA activated these signaling pathways in SV40 MES13 cells. SV40 MES13 cells were incubated with 10 mM LPA for 5 minutes, 1 hour, or 3 hours, and the phosphorylation of GSK3b at Ser9 and nuclear SREBP1 expression were examined by Western blot analysis. We Kidney International (2017) -, -–-

basic research

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

a

b

d P-Smad2/3

*

β-actin

200 0

f

db db /d b+ /db ki 16 42 5

db db /d b+ /db ki 16 42 5

W ild

ty pe

0.0

e

*

400

ty pe

0.5

*

600

W ild

c

1.0

db/db+ki16425

db/db

Wild type

1.5

con

2.0

LPA

1.5 1.0 0.5 0.0 0h

1h

3h

TGF-β

TGF-β β-actin

+

+ +

TGF-β β-actin

600

sicon siTGF-β

Fibronectin

0

β-actin LP A

m

200

LP A

400

5+

42

LPA

sicon siTGF-β sicon siTGF-β + + -

ki

- + 12 h

16

16

- + 6h

5 LP A 5+ LP A

n co

ki

- + 3h

+ -

l

*

*

n+

0.5 0.0

100 0 LPA

1.0

800

-

42

200

Ki16425 LPA

16

*

1.5

5

*

+ +

j

2.0

ki

300

*

+

-

co

400

*

n

g

*

2.5

42

12 h

h

co

6h

3h

+

16

-

+

-

ki

+

-

TGF-β / β-actin (relative density % of con)

+

β-actin

Ki16425 LPA

42

-

LPA

TGF-β mRNA level (relative ratio)

β-actin

6 h 12 h

k

i

TGF-β

TGF-β / β-actin (relative density % of con)

*

2.5

TGF-β mRNA levels (relative ratio)

*

TGF-β / β-actin (relative density % of con)

TGF-β mRNA levels (relative ratio)

TGF-β 2.0

db/db+ki16425

db/db

Wild type

Figure 3 | Treatment with ki16425 reduces the expression of transforming growth factor (TGF-b) in both the renal cortex of db/db mice and lysophosphatidic acid (LPA)-treated simian virus-transformed mouse mesangial (SV40 MES13) cells. Mice were treated as described in Figure 2. (a) mRNA expression of TGF-b in the renal cortex was determined by quantitative real-time polymerase chain reaction (qRT PCR) (n ¼ 9 to 10). (b) Representative Western blots of TGF-b protein expression. (c) The results were quantified, and b-actin was used as a loading control (n ¼ 9 to 10). (d) Representative pictures of immunofluorescence staining for phospho-Smad2/3 protein expression. Bars ¼ 20 mm (n ¼ 6). (eLg) SV40 MES13 cells were treated with 10 mM of LPA for the indicated times. (e) TGF-b mRNA levels were analyzed by qRT PCR. (f) TGF-b protein levels were analyzed by Western blot. (g) The results were quantified, and b-actin was used as loading control (n ¼ 3 independent experiments). (hLj) SV40 MES13 cells were treated with LPA (10 mM) with or without ki16425 (10 mM) for 3 hours. (h) TGF-b mRNA levels were analyzed by qRT PCR. (i) Protein levels were analyzed by Western blot. (j) The results were quantified, and b-actin was used as a loading control. (k) Normal human mesangial cells were treated with LPA (10 mM) with or without ki16425 (10 mM) for 1.5 hours. TGF-b protein levels were analyzed by Western blot (n ¼ 3 independent experiments). (l) TGF-b siRNA and scrambled siRNA (sicon) were transfected into SV40 MES13 cells, and TGF-b protein levels were analyzed by Western blot (n ¼ 3 independent experiments). (m) LPA-induced fibronectin protein expression was examined by Western blot analysis at 12 hours after treatment in cells with TGF-b siRNA or sicon (n ¼ 3 independent experiments). *P < 0.05. Data represent mean  SEM.

found that phospho–GSK3b (Ser9) was increased at 5 minutes and peaked at 1 hour after LPA stimulation (Figure 5a). The phospho-GSK3b (Ser9)/GSK3b ratio was significantly increased after 1 hour of LPA treatment (Figure 5b). The nuclear-translocated SREBP1 protein level was also significantly increased at 1 hour after LPA treatment (Figure 5c and d). When we blocked GSK3b (Ser9) phosphorylation using a 4-amino-5-(4-chlorophenyl)-7-(t-butyl) pyrazolo[3,4-D]pyrimidine inhibitor,30 phosphorylation of GSK3b at Ser9 was reduced, and subsequent LPA-induced SREBP1 translocation into the nucleus was significantly reduced (Figure 6a–c). Overexpression of the hemagglutinin Kidney International (2017) -, -–-

(HA)-tagged, constitutively active mutant of GSK3b (HAGSK3b-S9A)29,33 markedly reduced nuclear SREBP1 accumulation induced by LPA stimulation (Figure 6d–f). It was reported that nuclear SREBP1 is degraded by the ubiquitin-proteasome system, which is regulated by activated GSK3b.34 Therefore, we examined the changes of LPAinduced nuclear SREBP1 accumulation after MG132 (proteasome inhibitor) treatment in HA-GSK3b-S9A-transfected cells. After MG132 treatment, a large amount of SREBP accumulated in the nucleus, indicating that GSK3b (Ser9) phosphorylation was involved in LPA-induced SREBP accumulation (Figure 6e and f). 5

basic research

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

a

b

LPAR1

TGF-β β-actin

β-actin sicon

siLPAR1 LPA

LPAR1

c

*

mRNA levels (relative ratio)

1000 500

sicon siLPAR1 -

d

e

LPAR1

TGF-β

β-actin 5 4 3 2 1 0

β-actin vector LPAR1 LPA

vector

sicon siLPAR1 + +

vector LPAR1 vector LPAR1 + +

LPAR1

Figure 4 | Knockdown of lysophosphatidic acid receptor 1 (LPAR1) decreases LPA-induced transforming growth factorLb (TGF-b) expression and overexpression of LPAR1 upregulates LPA-induced TGF-b expression in simian virus-transformed mouse mesangial (SV40 MES13) cells. (a) LPAR1 siRNA or scrambled siRNA (sicon) was transfected into SV40 MES13 cells, and LPAR1 protein levels were analyzed by Western blot (n ¼ 3 independent experiments). (b) LPA-induced TGF-b protein expression was examined by Western blot analysis at 6 hours after LPA treatment in cells treated with LPAR1 siRNA or sicon (n ¼ 3 independent experiments). (c) pEGFP-LPAR1 (LPAR1) or an empty vector (vector) was transfected into SV40 MES13 cells, and LPAR1 mRNA levels were analyzed by quantitative real-time polymerase chain reaction (n ¼ 3 independent experiments). (d) Protein level of LPAR1 was measured by Western blot (n ¼ 3 independent experiments). (e) LPA-induced TGF-b was examined at 6 hours after LPA treatment in cells overexpressing LPAR1 and analyzed by Western blot. *P < 0.05. Data represent mean  SEM.

To investigate whether mature SREBP1 directly regulates TGF-b expression, we treated SV40 MES13 cells with the SREBP1 activation inhibitor, fatostatin (20 mM), in the presence or absence of LPA stimulation. We found that fatostatin decreased mature SREBP1 in the presence or absence of LPA (Figure 6g and h) and also reduced LPAinduced TGF-b mRNA (Figure 6i) and protein (Figure 6j and k) levels in SV40 MES13 cells. To investigate whether the PI3K pathway is involved in LPA-induced TGF-b expression, SV40 MES13 cells were pretreated with the PI3K inhibitor, LY294002 (20 mM), for 30

Treatment with ki16425 inhibits GSK3b (Ser9) phosphorylation and mature-SREBP1 nuclear accumulation in the renal cortex of db/db mice

To determine whether the GSK3b-SREBP1 pathway is involved in diabetic nephropathy, we checked protein levels in db/db mice with or without ki16425 treatment. Western blot analysis showed that phospho-GSK3b (Ser9) and

c

a P-GSK3β Cell lysate

GSK3β β-actin LPA

- + 5 min

- + 1h

-

+ 3h

b

m-SREBP1 Lamin B LPA - + 5 min

Nuclear - + 1h

-

+ 3h

d 250

*

200 150 100 50 0 LPA - + - + - + 5 min 1 h 3h

m-SREBP1 / Lamin B (relative density % of con)

P-GSK3β / GSK3β (relative density % of con)

minutes and then treated with LPA. We found that LY294002 inhibited the expression of the LPA-induced TGF-b protein (Supplementary Figure S5).

600

*

500 400 300 200 100 0 LPA

- + - + - + 5 min 1 h 3h

Figure 5 | Lysophosphatidic acid (LPA) phosphorylates glycogen synthase kinase 3b (GSK3b) at Ser9 and induces nuclear mature SREBP1 (m-SREBP1) accumulation in simian virus-transformed mouse mesangial (SV40 MES13) cells. (a) GSK3b, phospho-GSK3b (Ser9), and (c) m-SREBP1 and Lamin B expression after 10 mM of LPA treatment was analyzed by Western blot. (b,d) The results were quantified, and b-actin and Lamin B were used as loading controls (n ¼ 3 independent experiments). *P < 0.05. Data represent mean  SEM. 6

Kidney International (2017) -, -–-

basic research

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

co

n pp 2 pp L P 2+ A LP A

m-SREBP1 / Lamin B (relative density % of con)

+ + +

+ +

st

*

300 200 100

in

in at

st to

co

n+

LP A +L PA

0

fa

fa

to

fa

*

500 400

n

+ +

+ + -

+

at

+

k

st

+ -

*

*

3.0 2.5 2.0 1.5 1.0 0.5 0.0

to

-

+ -

-

co

Lamin B fatostatin LPA

fatostatin LPA

to

Nuclear

m-SREBP1

β-actin

0

i

g

+ + -

+ -

TGF-β

50

fa

+ + +

vector + HA-GSK3β-S9A LPA MG132

j

to

+ + -

0

fa

+ + -

SREBP1 Lamin B

50

100

to

+ -

100

150

fa

vector + HA-GSK3β-S9A LPA MG132

*

vector HA-GSKβ-S9A

e

in LP at A in +L PA

*

800 350 300 250 200 150 100 50 0

150

TGF-β / β-actin (relative density % of con)

*

1000

d HA β-actin

*

*

200

st

1200

*

200

in LP at A in +L PA

m-SREBP1 / Lamin B (relative density % of con)

h 1400

TGF-β mRNA level (relative ratio)

f

0

n

+

*

250

at

-

30

at

+

60

co

-

90

n

PP2

120

co

+

c

*

st

+

SREBP1 / Lamin B (relative density % of con)

-

*

co n pp 2 pp LP 2+ A LP A

m-SREBP1 -

P-GSK3β / GSK3β (relative density % of con)

GSK3β

150

Nuclear

Cell lysate Nuclear

P-GSK3β

Lamin B LPA

b

st

a

Figure 6 | Glycogen synthase kinase 3b (GSK3b) (Ser9) phosphorylation and SREBP1 activation are involved in lysophosphatidic acid (LPA)-induced transforming growth factorLb (TGF-b) expression in simian virus-transformed mouse mesangial (SV40 MES13) cells. (aLc) SV40 MES13 cells were pretreated with PP2 (20 mM) for 1 hour, then treated with LPA (10 mM) for 1 hour. The total cell lysate and nuclear fraction were extracted. (a) GSK3b, phospho-GSK3b (Ser9), and mature (m)-SREBP1 levels were analyzed by Western blot. (b,c) The results were quantified and Lamin B was used as a loading control (n ¼ 3 independent experiments). (d) HA-GSK3b-S9A (constitutively active GSK3b) and empty vector (vector) were transfected into SV40 MES13 cells, and HA protein levels were analyzed by Western blot (n ¼ 3 independent experiments). (e) After transfection with HA-GSK3b-S9A or control vector in SV40 MES13 cells, cells were treated LPA (10 mM) for 1 hour with or without MG132 (10 mM). The nuclear fraction was extracted, and m-SREBP1 levels were analyzed by Western blot. (f) The results were quantified, and lamin B was used as a loading control (n ¼ 3 independent experiments). (g) Cells were treated with LPA (10 mM) for 1 hour with or without preincubation with fatostatin (20 mM) for 4 hours. The nuclear fraction was extracted, and m-SREBP1 levels were analyzed by Western blot. (h) The results were quantified, and Lamin B was used as a loading control (n ¼ 3 independent experiments). (i,j) Cells were treated with LPA (10 mM) for 3 hours (mRNA) and 6 hours (protein) with or without preincubation with fatostatin (20 mM) for 4 hours; the mRNA and protein levels of TGF-b were analyzed by quantitative real-time polymerase chain reaction and Western blot. (k) The results were quantified, and b-actin was used as a loading control (n ¼ 3 independent experiments). *P < 0.05. Data represent mean  SEM. PP2, kinase inhibitor (4-amino-5-[4chlorophenyl]-7-(dimethylethyl)pyrazolo[3,4-d]pyrimidine).

nuclear-translocated SREBP1 were dramatically increased in the kidney cortex of db/db mice, and these changes were reduced after ki16425 treatment (Figure 7a and c). As shown in Figure 7b and d, quantitative data also showed that phospho-GSK3b (Ser9)/GSK3b and mature-SREBP1/Lamin B ratios were significantly increased in db/db mice compared with wild-type mice, but these were inhibited by ki16425 treatment. DISCUSSION

The development of diabetic nephropathy is a complex process: proteases, growth factors, cytokines, and chemokines Kidney International (2017) -, -–-

are released from renal cells and interact with each other to promote neovascularization and fibrosis.35,36 Dysfunction and morphological changes in glomeruli, such as hypertrophy, increased glomerular Ig uptake, and mesangial matrix expansion in early stages, are associated with the progress of diabetic nephropathy.4 Serum LPA levels are increased in high-fat, diet-induced diabetic mice,20 and the expression of ATX, an enzyme for LPA synthesis, is increased in the kidneys of diabetic db/db mice.22 In this study, we demonstrated the involvement of LPA in the development of glomerular injury in diabetic nephropathy. Both in vitro and in vivo data suggested that LPA 7

basic research

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

c

a P-GSK3β

m-SREBP1

GSK3β

Lamin B

Wild type db/db db/db+ki16425

β-actin

200

ki 16 42 5

db /d b

db /d b+

W ild

ty pe

0

300 200 100 0

ki 16 42 5

400

*

*

400

db /d b

600

d

db /d b+

*

ty pe

*

800

W ild

b

db/db db/db+ki16425

m-SREBP1 / Lamin B (relative density % of con)

Wild type

P-GSK3β / GSK3β (relative density % of con)

Nuclear

Figure 7 | Treatment with ki16425 inhibits glycogen synthase kinase 3b (GSK3b) (Ser9) phosphorylation and mature (m)-SREBP1 nuclear accumulation in the renal cortex of db/db mice. Mice were treated as described in Figure 2. (a) Representative blots of GSK3b, phospho-GSK3b (Ser9) of whole renal cortical tissue, and (c) m-SREBP1 in nuclear fractions of renal cortical tissue, which were analyzed by Western blot. (b,d) The results were quantified, and b-actin and Lamin B were used as loading controls (n ¼ 9 to 10). *P < 0.05. Data represent mean  SEM.

was able to upregulate TGF-b and fibronectin expression in mesangial cells at the mRNA and protein levels. This was associated with upregulation of mesangial matrix accumulation in the kidney cortex of db/db mice. We found that the expression of ATX and LPAR1 in the renal cortex of db/db mice was increased compared with wildtype mice, and that ki16425 treatment reduced glomerular injury, which suggested that LPAR1 signaling was involved in the pathogenesis of diabetic nephropathy. It was previously reported that the expression of LPAR1 was increased in white blood cells of patients with essential hypertension,37 which is a major factor for kidney injury during diabetes progression.1,3,4 These results indicated that LPAR1 expression was correlated with kidney injury. Both glomerular and tubular injury develops during the diabetic nephropathy process; glomerulopathy occurs in the early stages, whereas tubulointerstitial fibrosis is present in the advanced stages.38,39 In 16-week-old db/db diabetic mice, Breyer et al. reported that glomerular hypertrophy and ECM accumulation were consistently seen, but there was no evidence of tubulointerstitial fibrosis4; these results were consistent with our histopathological analysis. Ki16425 is a mixed antagonist for LPAR1 and LPAR3 with close affinity for the two subtypes. It is known that LPAR3 is not as responsive as LPAR1 to LPA stimulation,40 and we found that LPAR1 was predominantly expressed in the kidney cortex of db/db mice and high glucose–treated mesangial cells. Our LPAR1 knockdown and overexpression experiments in SV40 MES13 cells suggested that the fibrotic activity of LPA in glomeruli was likely mediated mainly through LPAR1, although an effect of LPAR3 could not completely be ruled out. Similarly, genetic deletion or pharmacological antagonism of LPAR1 showed antifibrotic activity in the skin and lungs.41–43 8

Rancoule et al. reported that LPA exerted a deleterious effect on glucose tolerance, and ki16425 treatment showed an antihyperglycemic effect via enhanced islet number and insulin secretion in mice fed a high-fat diet.20 In our study, treatment of db/db mice with ki16425 ameliorated albuminuria and mesangial matrix accumulation, but had no impact on glucose homeostasis. We found that LPAR1 expression was more abundant in the kidney cortex than in islets of db/db mice (unpublished data), thus the islets might be less sensitive to LPA than glomeruli. In addition, db/db mice with a C57BLKS/J background, which we used in our study, are known to show more severe b-cell dysfunction than other type 2 diabetic animal models.44 These factors might have affected the different effects of ki16425 on glucose homeostasis. We observed that ki16425 treatment ameliorated glomerular hypertrophy and albuminuria, but did not affect kidney weight. Similarly, other studies in db/db mice and chronic kidney disease models showed lack of changes in kidney weight despite inhibition of glomerular hypertrophy.45,46 Because kidney hypertrophy is caused not only by glomerular changes, but also by proximal tubular cell changes,35 ki16425 treatment might affect glomerular hypertrophy, but not kidney weight. Because the expression of TGF-b, a profibrogenic cytokine, is increased in kidney injury,5 we investigated whether LPA increases TGF-b expression. Treatment with 10 mM of LPA47 or LPAR1 overexpression increased the expression of TGF-b, and ki16425 treatment or knockdown of LPAR1 expression inhibited this increase in mouse mesangial cells, which was also confirmed in human mesangial cells. These results showed that LPAR1 signaling is clearly involved in LPA-induced TGF-b expression. In addition, exogenous TGF-b augmented ATX expression in SV40 MES cells Kidney International (2017) -, -–-

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

(Supplementary Figure S6A and B), which indicated that activation of LPAR1 enhanced TGF-b expression, which, in turn, created a vicious cycle that accelerated fibrosis in the glomeruli of db/db mice. A high level of SREBP1 was observed in the kidney cortex of STZ-induced diabetic mice,48 and overexpression of glomerular-specific, SREBP1-induced diabetic nephropathy in mice.49 In addition, SREBP1 activation is known to be involved in glucose- or angiotensin II-induced TGF-b upregulation and glomerular fibrosis.31,32 We also found that LPA increased activation of SREBP1 and subsequently induced TGF-b upregulation in mesangial cells, which likely contributed to the pathogenesis of diabetic nephropathy. SREBP1 was also found in other glomerular cell types, including podocytes and endothelial cells50,51; however, it is not known whether LPA activates SREBP1 in these cells. The well-known signaling pathway for SREBP1 activation is that of the SREBP cleavage-activating protein (SCAP)–assisted transport of SREBP1 to the Golgi for cleavage by site-1 and -2 proteases to produce the mature transcription factor.52 Our study used the SCAP inhibitor fatostatin, which indicated that the SCAP/ site-1/site-2 protease pathway is involved in SREBP1 activation by LPA. We found that LPA leads to phosphorylation of GSK3b (Ser9), which allowed SREBP1 activation, but nonsterolmediated activation of SREBP1 was also reported.53 We observed that renal cortex homogenates of db/db mice exhibited increased Ser9 phosphorylation of GSK3b, which inhibited its activity.54 Moreover, in view of SREBP1 activation by LPA treatment, it was blocked by constitutively active HA-GSK3b-S9A or kinase inhibitor (PP2) treatment, which suggested that phosphorylation of GSK3b at Ser9 was involved in LPAR1-dependent SREBP1 activation. GSK3b is active in the normal state of the tissue and normally functions as a repressor of protein synthesis and transcription. Therefore, phosphorylation at Ser9 and inactivation of GSK3b might lead to increased protein synthesis such as TGF-b. Other studies also reported that diabetic mice (e.g., STZinduced diabetic or db/db mice) showed increased phosphorylation of renal cortical GSK3b at Ser9, which was correlated with renal hypertrophy and increased fibronectin protein content.29,30 Because we found that LPA-induced TGF-b expression was inhibited by PI3K inhibitor treatment, it was possible that increased LPA under diabetic conditions was bound to LPAR1, which activated phosphorylation of PI3K/AKT, leading to the phosphorylation of GSK3b at Ser9.55 The nuclear form of SREBP1 is known to be rapidly degraded by the ubiquitin-proteasome system, which is also regulated by GSK3b.34 We found that HA-GSK3b-S9A– transfected cells treated with a proteasome inhibitor accumulated a large amount of SREBP1 in the nucleus. These results indicated that LPA-induced GSK3b phosphorylation at Ser9 might also increase the stability of nuclear SREBP1 via inhibition of SREBP1 degradation and SCAP activity. Our study also showed the phosphorylation of Erk, JNK, and p38 Kidney International (2017) -, -–-

basic research

in LPA-treated SV40 MES13 cells (Supplementary Figure S7), which indicated that LPA might also may contribute to mitogen-activated protein kinase signaling-induced TGF-b expression. These results were consistent with previous studies that showed TGF-b expression was dependent on mitogen-activated protein kinase in high glucose conditions or diabetes.35,56 In summary, our study showed that upregulation of LPA under diabetic conditions activates LPAR1 in mesangial cells, which resulted in GSK3b (Ser9) phosphorylation and SREBP1 activation and subsequent induction of TGF-b expression (Figure 8). These pathways contributed to glomerular injury in db/db mice and might provide a target for therapeutic strategies to prevent the development of diabetic nephropathy. METHODS Animals Eight-week-old male diabetic db/db (BKS.Cg-leprdb/leprdb) mice of C57BLKS/J background were obtained from Jackson Laboratories (Bar Harbor, ME).3,4 Age-matched nondiabetic wild-type (BKS.Cgleprþ/leprþ) mice were used as the control group. Vehicle or ki16425 (Biorbyt Ltd., Cambridge, UK) was injected daily at a dosage of 10 mg/kg i.p. for 8 weeks. All of the experiments using mice were approved by the Institutional Animal Care and Use Committee at Gachon University. Biochemical parameters in blood and urine Mice were placed in individual mice metabolic cages for 24 hours. HbA1c was determined from red blood cell lysates by a DCA Systems HbA1c Reagent Kit (Siemens, Tarrytown, New York). Biochemical parameters were measured using a biochemistry analyzer (Beckman, Miami, Florida). The albumin/creatinine ratio was calculated as urinary albumin (milligrams per deciliter)/urinary creatinine (grams per deciliter). Nonfasting blood glucose from the tail vein blood was measured using a glucose analyzer (One Touch Ultra, Lifescan Johnson & Johnson, Milpitas, CA). Renal histological assessment and immunostaining Peroxidase staining was performed with 3, 3-DAB as a chromogen (Liquid DABþ Substrate Chromogen System; Dako, Carpinteria, California). For immunofluorescence, Texas red- or fluorescein isothiocyanate–conjugated secondary antibodies were used, and the slides were observed under a confocal microscope (LSM 700, Carl Zeiss Inc, Oberkochen, West Germany). Cell culture SV40 MES13 cells were maintained in Dulbecco’s modified Eagle’s medium, containing 5% fetal bovine serum, and 1% penicillinstreptomycin. To investigate the effect of LPA (Avanti POLAR LIPIDS, Alabaster, Alabama), the cells were plated and pretreated with 0.1% fatty acid-free bovine serum albumin (BSA) for 12 to 16 hours and then treated with LPA or various inhibitors. Primary human renal mesangial cells were purchased from Lonza, and cultured in MsGM medium (Lonza, Walkersville, MD) containing 5% fetal bovine serum and 0.1% gentamicin sulfate. Studies were performed using the cells in between the fourth and tenth passages. 9

basic research

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

LPA

LPA

LPA

LPAR1

Renal mesangial cell

Gi PI3K/AKT GSK3β

Ser9 GSK3β

SREBP1 Reduce degradation

Increase enzyme activity for translocation and processing (SCAP, S1P, S2P)

SREBP1 (Mature Form)

TGF-β

nucleus

Glomerular injury Diabetic nephropathy Figure 8 | Schematic diagram showing the pathway involved in lysophosphatidic acid (LPA)–induced transforming growth factor–b (TGF-b) synthesis and glomerular injury, which is mediated by phospho–glycogen synthase kinase 3b (GSK3b) (Ser9) and SREBP1 activation. SCAP, SREBP cleavage-activating protein.

Transfection The pEGFP-LPAR1 plasmid DNA was obtained from Origene Technologies Inc. (Rockville, MD). SV40 MES13 cells were transfected with empty vector, pEGFP-LPAR1 or HA-GSK3b-S9A (alanine 9 mutant constitutively active GSK3b, Addgene, Cambridge, MA) plasmid DNA using the Lipofectamine 2000 reagent (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions. For small-interfering RNA transfection, SV40 MES13 cells were seeded into 6-well plates and were grown until 60% confluent. The cells were transiently transfected with 10 pM of TGF-b siRNA (Bioneer, Daejeon, Korea), 40 pM of LPAR1 siRNA, or scrambled siRNA (Santa Cruz Biotechnology, Santa Cruz, California) using Lipofectamine RNAi MAX (Invitrogen) reagent in accordance with the manufacturer’s protocol. After 36 hours, the medium was replaced with serum-free medium containing 0.1% fatty acid-free BSA for 12 to 16 hours, and the cells were subsequently stimulated with LPA (10 mM) for various times.

Western blot analysis To extract nuclear proteins, extraction buffers (A: 10 mM N-2-hydroxyethylpiperazine-N0 -2-ethanesulfonic acid [HEPES], 10 mM potassium chloride, 0.1 mM ethylenediamine tetraacetic acid [EDTA], 0.1 mM ethyleneglycol-bis-(b-aminoethylether)N,N,N0 ,N0 -tetraacetic acid [EGTA], and 1 mM dithiothreitol [DTT]; B: 20 mM HEPES, 400 mM sodium chloride, 1 mM EDTA, 1 mM EGTA, and 1 mM DTT) were used. Signals were detected by using an enhanced chemiluminescence detection system (Millipore, Watford, UK). The western blots were quantified using ImageJ software (National Institutes of Health, Bethesda, MD).

Quantitative real-time polymerase chain reaction Total RNA was prepared and cDNA was synthesized as described previously.57 The relative mRNA transcript levels were calculated according to the 2-DCTmethod, and the cyclophilin internal control. Primers used for quantitative real-time polymerase chain reaction (qRT-PCR) are listed in Supplementary Table S1.

DISCLOSURE

10

Statistical analyses The results were expressed as mean  SEM. Differences among more than 2 groups were analyzed by 1-way analysis of variance followed by Tukey’s post hoc multiple comparison tests. An unpaired 2-tailed t-test was used for analysis of the 2 groups. Significance was declared if P values were <0.05.

All the authors declared no competing interests. ACKNOWLEDGMENTS

This study was supported by grants from the National Research Foundation of Korea (NRF) grant funded by the Korea government Kidney International (2017) -, -–-

basic research

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

(MSIP) (No. 2016R1A2B2013347) and the Ministry of Health & Welfare, Republic of Korea (grant number: HI14C1135).

SUPPLEMENTARY MATERIAL Figure S1. Expression level of lipid phosphate phosphatases (LPPs) is not changed in the kidney cortex of db/db mice compared with wildtype mice. mRNAs were isolated from the renal cortex of 16-week-old wild-type and db/db mice, and mRNA levels of (A) LPP1, (B) LPP2, and (C) LPP3 were measured by q-real-time polymerase chain reaction (q-RT PCR) (n ¼ 3). Data are expressed as the ratio relative to the levels found in wild-type mice. Figure S2. LPAR1 expression is increased in a–smooth muscle actin (a-SMA)positive cells of db/db mice. Representative pictures of immunofluorescence staining for colocalization of (A) a-SMA or (B) SMA with LPAR1, which were examined in the kidney sections of 16week-old wild-type mice and db/db mice. Bars ¼ 20 mm (n ¼ 6). Figure S3. Treatment with ki16425 does not affect the glucose tolerance and insulin sensitivity of db/db mice. Eight-week-old wildtype and db/db mice were injected i.p. with vehicle or ki16425 (10 mg/kg) for 8 weeks. (A) For glucose tolerance tests, mice fasted overnight and were then injected with glucose (2 g/kg i.p.). A glucometer (Onetouch; Lifescan, Milpitas, CA) was used to measure glucose in the tail vein blood at 0, 30, 60, 120, and 150 minutes after injection. (B) For insulin tolerance tests, mice were injected with insulin (2 U/kg, i.p.; Green Cross, Yongin, Korea), and tail vein blood glucose was measured at 0, 30, 60, and 90 minutes after injection. Figure S4. Treatment with ki16425, but not H2L5186303, reduces the expression of fibronectin in the renal cortex of db/db mice or lysophosphatidic acid (LPA)-treated simian virus-transformed mouse mesangial (SV40 MES13) cells. Eight-week-old wild-type and db/db mice were injected with vehicle or ki16425 (10 mg/kg i.p.) for 8 weeks. (A) The mRNA level of fibronectin in the renal cortex was determined by quantitative real-time polymerase chain reaction (qRT PCR) (n ¼ 9 to 10). (B) SV40 MES 13 cells were treated with various concentrations of LPA for 24 hours, and the mRNA level of fibronectin was analyzed by qRT PCR (n ¼ 3 independent experiments). (C) SV40 MES 13 cells were treated with LPA (10 mM) for the indicated times, and mRNA level of fibronectin was analyzed by qRT PCR (n ¼ 3 independent experiments). (D) Cells were treated with LPA (10 mM) without (con) or with ki16425 (10 mM) for 3 hours, and fibronectin mRNA levels were analyzed by qRT PCR (n ¼ 3 independent experiments). (E and F) Cells were treated with LPA (10 mM) with or without various concentrations of H2L5186303 for 3 hours, and (E) transforming growth factor-b (TGF-b) and (F) fibronectin mRNA levels were analyzed by qRT PCR (n ¼ 3 independent experiments). *P < 0.05. Data represent mean  SEM. Figure S5. AKT/PI3K pathway is involved in lysophosphatidic acid (LPA)-induced transforming growth factor-b (TGF-b) expression in simian virus-transformed mouse mesangial (SV40 MES13 cells). SV40 MES13 cells were treated with LPA (10 mM) for 6 hours with or without preincubation with LY294002 (20 mM) for 30 minutes. TGF-b protein expression was analyzed by Western blot (n ¼ 3 independent experiments, representative blot shown). Figure S6. Exogenous transforming growth factor-b (TGF-b) increases the protein expression level of autotaxin (ATX) in simian virustransformed mouse mesangial (SV40 MES) cells. (A) SV40 MES13 cells were treated with 5 ng/ml of TGF-b for 12 hours and ATX protein expression was analyzed by Western blot. (B) The results were quantified and b-actin was used as loading control (n ¼ 3 independent experiments). Figure S7. Lysophosphatidic acid (LPA) activates the mitogenactivated protein kinase (MAPK) pathway in simian virus-transformed mouse mesangial (SV40 MES) cells. SV40 MES13 cells were treated Kidney International (2017) -, -–-

with 10 mM of LPA for the indicated times. Phospho-ERK, phosphoJNK, phospho-p38, and b-actin expression were analyzed by Western blot (n ¼ 3 independent experiments; representative blot shown). Table S1. Primers (mouse) for quantitative real-time polymerase chain reaction Supplementary material is linked to the online version of the paper at www.kidney-international.org. REFERENCES 1. Gonzalez Suarez ML, Thomas DB, Barisoni L, et al. Diabetic nephropathy: is it time yet for routine kidney biopsy? World J Diabetes. 2013;4:245–255. 2. Asbun J, Villarreal FJ. The pathogenesis of myocardial fibrosis in the setting of diabetic cardiomyopathy. J Am Coll Cardiol. 2006;47:693–700. 3. Sharma K, McCue P, Dunn SR. Diabetic kidney disease in the db/db mouse. Am J Physiol Renal Physiol. 2003;284:F1138–F1144. 4. Breyer MD, Bottinger E, Brosius FC, 3rd, et al. Mouse models of diabetic nephropathy. J Am Soc Nephrol. 2005;16:27–45. 5. Schnaper HW, Hayashida T, Hubchak SC, et al. TGF-beta signal transduction and mesangial cell fibrogenesis. Am J Physiol Renal Physiol. 2003;284:F243–F252. 6. Ziyadeh FN, Han DC, Cohen JA, et al. Glycated albumin stimulates fibronectin gene expression in glomerular mesangial cells: involvement of the transforming growth factor-beta system. Kidney Int. 1998;53: 631–638. 7. Ziyadeh FN, Sharma K, Ericksen M, et al. Stimulation of collagen gene expression and protein synthesis in murine mesangial cells by high glucose is mediated by autocrine activation of transforming growth factor-beta. J Clin Invest. 1994;93:536–542. 8. Sharma K, Jin Y, Guo J, et al. Neutralization of TGF-beta by anti-TGF-beta antibody attenuates kidney hypertrophy and the enhanced extracellular matrix gene expression in STZ-induced diabetic mice. Diabetes. 1996;45: 522–530. 9. Tigyi G. Aiming drug discovery at lysophosphatidic acid targets. Br J Pharmacol. 2010;161:241–270. 10. Wada A, Tojo H, Sugiura T, et al. Group II phospholipase A2 as an autocrine growth factor mediating interleukin-1 action on mesangial cells. Biochim Biophys Acta. 1997;1345:99–108. 11. Pfeilschifter J, Schalkwijk C, Briner VA, et al. Cytokine-stimulated secretion of group II phospholipase A2 by rat mesangial cells. Its contribution to arachidonic acid release and prostaglandin synthesis by cultured rat glomerular cells. J Clin Invest. 1993;92:2516–2523. 12. Ikeda H, Yatomi Y, Yanase M, et al. Effects of lysophosphatidic acid on proliferation of stellate cells and hepatocytes in culture. Biochem Biophys Res Commun. 1998;248:436–440. 13. Tager AM, LaCamera P, Shea BS, et al. The lysophosphatidic acid receptor LPA1 links pulmonary fibrosis to lung injury by mediating fibroblast recruitment and vascular leak. Nature Med. 2008;14:45–54. 14. Pradere JP, Klein J, Gres S, et al. LPA1 receptor activation promotes renal interstitial fibrosis. J Am Soc Nephrol. 2007;18:3110–3118. 15. Choi JW, Herr DR, Noguchi K, et al. LPA receptors: subtypes and biological actions. Ann Rev Pharmacol Toxicol. 2010;50:157–186. 16. Lin ME, Herr DR, Chun J. Lysophosphatidic acid (LPA) receptors: signaling properties and disease relevance. Prostaglandins Other Lipid Mediat. 2010;91:130–138. 17. Ohta H, Sato K, Murata N, et al. Ki16425, a subtype-selective antagonist for EDG-family lysophosphatidic acid receptors. Mol. Pharmacol. 2003;64: 994–1005. 18. Zhao J, Wei J, Weathington N, et al. Lysophosphatidic acid receptor 1 antagonist ki16425 blunts abdominal and systemic inflammation in a mouse model of peritoneal sepsis. Translational Res. 2015;166:80–88. 19. David M, Sahay D, Mege F, et al. Identification of heparin-binding EGFlike growth factor (HB-EGF) as a biomarker for lysophosphatidic acid receptor type 1 (LPA1) activation in human breast and prostate cancers. PloS One. 2014;9:e97771. 20. Rancoule C, Attane C, Gres S, et al. Lysophosphatidic acid impairs glucose homeostasis and inhibits insulin secretion in high-fat diet obese mice. Diabetologia. 2013;56:1394–1402. 21. Grove KJ, Voziyan PA, Spraggins JM, et al. Diabetic nephropathy induces alterations in the glomerular and tubule lipid profiles. J Lipid Res. 2014;55:1375–1385. 22. Ferry G, Tellier E, Try A, et al. Autotaxin is released from adipocytes, catalyzes lysophosphatidic acid synthesis, and activates preadipocyte

11

basic research

23.

24.

25. 26.

27.

28.

29.

30.

31.

32.

33.

34.

35. 36. 37. 38.

39.

40.

12

proliferation. Up-regulated expression with adipocyte differentiation and obesity. J Biolog Chem. 2003;278:18162–18169. Shimizu M, Furuichi K, Toyama T, et al. Serum autotaxin levels are associated with proteinuria and kidney lesions in Japanese type 2 diabetic patients with biopsy-proven diabetic nephropathy. Intern Med. 2016;55:215–221. Rancoule C, Dusaulcy R, Treguer K, et al. Depot-specific regulation of autotaxin with obesity in human adipose tissue. J Physiol Biochem. 2012;68:635–644. Vaughan MR, Quaggin SE. How do mesangial and endothelial cells form the glomerular tuft? J Am Soc Nephrol. 2008;19:24–33. Sarrab RM, Lennon R, Ni L, et al. Establishment of conditionally immortalized human glomerular mesangial cells in culture, with unique migratory properties. Am J Physiol Renal Physiol. 2011;301:F1131–F1138. Rops AL, van der Vlag J, Jacobs CW, et al. Isolation and characterization of conditionally immortalized mouse glomerular endothelial cell lines. Kidney Int. 2004;66:2193–2201. Heldin CH, Miyazono K, ten Dijke P. TGF-beta signalling from cell membrane to nucleus through SMAD proteins. Nature. 1997;390: 465–471. Mariappan MM, Shetty M, Sataranatarajan K, et al. Glycogen synthase kinase 3beta is a novel regulator of high glucose- and high insulininduced extracellular matrix protein synthesis in renal proximal tubular epithelial cells. J Biol Chem. 2008;283:30566–30575. Mariappan MM, Prasad S, D’Silva K, et al. Activation of glycogen synthase kinase 3beta ameliorates diabetes-induced kidney injury. J Biol Chem. 2014;289:35363–35375. Wang TN, Chen X, Li R, et al. SREBP-1 Mediates angiotensin II-induced TGF-beta1 upregulation and glomerular fibrosis. J Am Soc Nephrol. 2015;26:1839–1854. Uttarwar L, Gao B, Ingram AJ, et al. SREBP-1 activation by glucose mediates TGF-beta upregulation in mesangial cells. Am J Physiol Renal Physiol. 2012;302:F329–F341. Stambolic V, Woodgett JR. Mitogen inactivation of glycogen synthase kinase-3 beta in intact cells via serine 9 phosphorylation. Biochem J. 1994;303(Pt 3):701–704. Sundqvist A, Bengoechea-Alonso MT, Ye X, et al. Control of lipid metabolism by phosphorylation-dependent degradation of the SREBP family of transcription factors by SCF(Fbw7). Cell Metab. 2005;1:379–391. Wolf G, Ziyadeh FN. Molecular mechanisms of diabetic renal hypertrophy. Kidney Int. 1999;56:393–405. Soetikno V, Arozal W, Louisa M, et al. New insight into the molecular drug target of diabetic nephropathy. Int J Endocrinol. 2014;2014:968681. Xu K, Ma L, Li Y, et al. Genetic and functional evidence supports LPAR1 as a susceptibility gene for hypertension. Hypertension. 2015;66:641–646. Fioretto P, Caramori ML, Mauer M. The kidney in diabetes: dynamic pathways of injury and repair. The Camillo Golgi Lecture 2007. Diabetologia. 2008;51:1347–1355. Bohle A, Wehrmann M, Bogenschutz O, et al. The pathogenesis of chronic renal failure in diabetic nephropathy. Investigation of 488 cases of diabetic glomerulosclerosis. Pathol Res Pract. 1991;187:251–259. Gardell SE, Dubin AE, Chun J. Emerging medicinal roles for lysophospholipid signaling. Trends Mol Med. 2006;12:65–75.

HY Li et al.: The role of LPAR1 signaling in diabetic nephropathy

41. Castelino FV, Goldberg H, Dellaripa PF. The impact of rheumatological evaluation in the management of patients with interstitial lung disease. Rheumatol (Oxford). 2011;50:489–493. 42. Swaney JS, Chapman C, Correa LD, et al. Pharmacokinetic and pharmacodynamic characterization of an oral lysophosphatidic acid type 1 receptor-selective antagonist. J Pharmacol Exper Ther. 2011;336:693–700. 43. Qian Y, Hamilton M, Sidduri A, et al. Discovery of highly selective and orally active lysophosphatidic acid receptor-1 antagonists with potent activity on human lung fibroblasts. J Med. Chem. 2012;55:7920–7939. 44. Leiter EH, Lee CH. Mouse models and the genetics of diabetes: is there evidence for genetic overlap between type 1 and type 2 diabetes? Diabetes. 2005;54(Suppl 2):S151–S158. 45. Shibuya K, Kanasaki K, Isono M, et al. N-acetyl-seryl-aspartyl-lysyl-proline prevents renal insufficiency and mesangial matrix expansion in diabetic db/db mice. Diabetes. 2005;54:838–845. 46. Tsuprykov O, Ando R, Reichetzeder C, et al. The dipeptidyl peptidase inhibitor linagliptin and the angiotensin II receptor blocker telmisartan show renal benefit by different pathways in rats with 5/6 nephrectomy. Kidney Int. 2016;89:1049–1061. 47. Rancoule C, Viaud M, Gres S, et al. Pro-fibrotic activity of lysophosphatidic acid in adipose tissue: in vivo and in vitro evidence. Biochim Biophys Acta. 2014;1841:88–96. 48. Sun L, Halaihel N, Zhang W, et al. Role of sterol regulatory elementbinding protein 1 in regulation of renal lipid metabolism and glomerulosclerosis in diabetes mellitus. J Biol Chem. 2002;277: 18919–18927. 49. Ishigaki N, Yamamoto T, Shimizu Y, et al. Involvement of glomerular SREBP-1c in diabetic nephropathy. Biochem Biophys Res Commun. 2007;364:502–508. 50. Liu Y, Chen BP, Lu M, et al. Shear stress activation of SREBP1 in endothelial cells is mediated by integrins. Arterioscl Thromb Vasc Biol. 2002;22:76–81. 51. Jiang T, Wang XX, Scherzer P, et al. Farnesoid X receptor modulates renal lipid metabolism, fibrosis, and diabetic nephropathy. Diabetes. 2007;56: 2485–2493. 52. Bengoechea-Alonso MT, Ericsson J. SREBP in signal transduction: cholesterol metabolism and beyond. Curr Opin Cell Biol. 2007;19: 215–222. 53. Kim YM, Song I, Seo YH, et al. Glycogen synthase kinase 3 inactivation induces cell senescence through sterol regulatory element binding protein 1-mediated lipogenesis in Chang cells. Endocrinol Metab (Seoul). 2013;28:297–308. 54. Leger B, Cartoni R, Praz M, et al. Akt signalling through GSK-3beta, mTOR and Foxo1 is involved in human skeletal muscle hypertrophy and atrophy. J Physiol. 2006;576:923–933. 55. Badri L, Lama VN. Lysophosphatidic acid induces migration of human lung-resident mesenchymal stem cells through the beta-catenin pathway. Stem Cells. 2012;30:2010–2019. 56. Ni WJ, Tang LQ, Wei W. Research progress in signalling pathway in diabetic nephropathy. Diabetes/Metab Res Rev. 2015;31:221–233. 57. Li HY, Oh YS, Lee YJ, et al. Amelioration of high fat diet-induced glucose intolerance by blockade of Smad4 in pancreatic beta-cells. Exper Clin Endocrinol Diabetes. 2015;123:221–226.

Kidney International (2017) -, -–-