Logo of ajpendoPublished ArticleArchivesSubscriptionsSubmissionsContact UsAJP - Endocrinology and MetabolismAmerican Physiological Society
Am J Physiol Endocrinol Metab. Sep 2011; 301(3): E517–E526.
Published online Jun 14, 2011. doi:  10.1152/ajpendo.00241.2011
PMCID: PMC3174531

Loss of PDGF-B activity increases hepatic vascular permeability and enhances insulin sensitivity


Hepatic vasculature is not thought to pose a permeability barrier for diffusion of macromolecules from the bloodstream to hepatocytes. In contrast, in extrahepatic tissues, the microvasculature is critically important for insulin action, because transport of insulin across the endothelial cell layer is rate limiting for insulin-stimulated glucose disposal. However, very little is known concerning the role in this process of pericytes, the mural cells lining the basolateral membrane of endothelial cells. PDGF-B is a growth factor involved in the recruitment and function of pericytes. We studied insulin action in mice expressing PDGF-B lacking the proteoglycan binding domain, producing a protein with a partial loss of function (PDGF-Bret/ret). Insulin action was assessed through measurements of insulin signaling and insulin and glucose tolerance tests. PDGF-B deficiency enhanced hepatic vascular transendothelial transport. One outcome of this change was an increase in hepatic insulin signaling. This correlated with enhanced whole body glucose homeostasis and increased insulin clearance from the circulation during an insulin tolerance test. In obese mice, PDGF-B deficiency was associated with an 80% reduction in fasting insulin and drastically reduced insulin secretion. These mice did not have significantly higher glucose levels, reflecting a dramatic increase in insulin action. Our findings show that, despite already having a high permeability, hepatic transendothelial transport can be further enhanced. To the best of our knowledge, this is the first study to connect PDGF-B-induced changes in hepatic sinusoidal transport to changes in insulin action, demonstrating a link between PDGF-B signaling and insulin sensitivity.

Keywords: platelet-derived growth factor B, pericytes, diabetes

platelet-derived growth factor b (PDGF-B) and its receptor PDGF receptor-β (PDGFRβ) mediate the recruitment and proliferation of pericytes in numerous vascular beds (23). Pericytes are mural cells associated with arterioles, venules, capillaries, and sinusoids. They function in numerous pathways, including vasoconstriction, angiogenesis, endothelial cell maturation, and vascular barrier function (23).

Vascular pathology is associated with susceptibility to diabetes and plays an important role in diabetic complications such as retinopathy, nephropathy, neuropathy, and atherosclerosis (22, 44). The islet microvasculature is critical for efficient insulin delivery to the bloodstream. Mice with reduced islet vasculature deliver less secreted insulin from islets to the bloodstream in vivo and have reduced glucose tolerance (14, 32). The microvasculature is also important for insulin action. In muscle, the process of delivering insulin from the bloodstream across the endothelial cell layer into the interstitium appears to be a rate-limiting step for insulin action and may contribute to insulin resistance (6, 18). Uptake of FITC-insulin by rat skeletal muscle endothelial cells occurs within 10 min of intravenous injection, but the FITC-insulin remains concentrated in endothelial cells for up to 60 min after injection (50). The process of transendothelial insulin transport requires the expression of caveolin-1, suggesting a caveloae-mediated mechanism (51). In contrast, there appears to be no time delay between the appearance of insulin in lymph and activation of the insulin receptor in myocytes (39). Accordingly, after insulin is directly injected into the muscle interstitium, there is a strong correlation between the interstitial insulin concentration and local glucose uptake (17). Thus, once insulin enters the interstitial space, receptor binding, signaling, and stimulation of glucose uptake are essentially instantaneous.

The liver displays a unique sinusoidal structure, characterized by gaps between endothelial cells, large fenestrae, and lack of a well-formed basement membrane. Fenestrae occupy 10% of the sinusoidal surface and are organized into clusters, termed “sieve plates.” They connect the sinusoidal lumen directly to the space of Disse, which lines the basolateral surface of the endothelium and interacts with hepatocyte cell membranes and microvilli (4). These fenestrated discontinuous hepatic sinusoids are highly permeable and allow particles smaller than 100 nm to freely diffuse into the space of Disse (34). Therefore, unlike muscle, the liver vasculature is considered not to be rate limiting for nutrient transport under normal circumstances.

However, hepatic nutrient transport out of the bloodstream and diffusion through the space of Disse is hindered during certain pathological conditions. “Capillarization” is the process by which hepatic sinusoids undergo morphological changes that include sinusoidal defenestration, the development of basement membrane, and excessive distortion of the space of Disse due to extracellular matrix (ECM) deposition (34). These ultrastructural changes result from the activation of liver pericytes, also termed hepatic stellate cells (HSCs) (45), and PDGF-B signaling can induce HSC activation (20). Capillarization-like phenotypes have been noted in fibrosis (34), aging (34), and type 1 diabetes (8, 9, 30, 33). Defenestration and perisinusoidal fibrosis are also present in nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis, the two most common manifestations of hepatic insulin resistance (22). Capillarization can impede the delivery of nutrients and small molecules to hepatocytes (15, 34), although hepatic insulin delivery has not been examined under these conditions.

Few studies have examined the effects of PDGF-B signaling and pericytes in insulin action. PDGF-B signaling can induce glucose transporter 4 translocation in both cultured adipocytes (52) and mouse skeletal muscle (55), but only under conditions of PDGFRβ overexpression. The effect of PDGF-B on insulin action in hepatocytes has not been examined, although the low endogenous expression of PDGFRβ in these cells suggests that PDGF-B would have little direct effect on hepatocyte function (46) but rather through modulation of pericytes and the vasculature.

One pathway potentially modulated by PDGF-B is the maintenance of vascular barrier function. Mice with reduced pericyte coverage of the vasculature display widespread and often lethal vascular hemorrhage and edema (27, 28, 35, 36, 47). Loss of pericyte retention on blood vessels in tumors leads to vascular leakage, with an increased potential for tumor cell metastasis (1, 53). Pericytes also play a critical role in the development and maintenance of the blood-brain barrier, and pericyte deficiency both increases the permeability of the blood-brain barrier (5, 7, 21) and reduces brain capillary perfusion (7). Diabetic retinopathy is characterized by excessive vascular permeability in the retina and is caused by continual pericyte loss over time (25). Finally, high doses of Gleevec, a pharmacological inhibitor of PDGF-B signaling, lead to increased risk of edema and interstitial leakage of fluid (31). These data led us to hypothesize that reduced PDGF-B signaling may increase vascular permeability and thereby improve insulin delivery to its target cells.

To address this hypothesis, we studied PDGF-B retention-deficient (PDGF-Bret/ret) mice. PDGF-B is secreted as a homodimer from endothelial cells, where it is retained on the cell surface by a COOH-terminal heparan-sulfate proteoglycan-binding retention motif (40). Deletion of this motif in the PDGF-Bret/ret mice leads to decreased retention of PDGF-B and a reduction in PDGF-B signaling (40, 43). PDGF-Bret/ret mice display a loss of pericyte density, pericyte detachment, and abnormal capillary morphology in the developing brain, kidney, and retina as well as the postnatal kidney and retina (37).

In the present study, we demonstrate that impaired PDGF-B signaling enhances vascular transendothelial transport in the liver. The resulting elevations in hepatic insulin signaling lead to improved whole body glucose clearance. To metabolically challenge these animals, we bred the Leptinob (Lepob) mutation (56) into the PDGF-Bret/ret background to generate obese PDGF-B retention-deficient mice. Despite an 88% reduction in plasma insulin, the mice had normal glucose levels, reflecting a drastic improvement in insulin action. These phenotypes appear to result from elevated insulin transport and signaling rather than a loss of β-cell secretory capacity. To the best of our knowledge, this represents the first study that links reduced PDGF-B signaling to increased insulin action and establishes that hepatic vasculature can be made even more permeable.



C57BL/6J (B6) Lepob/+ animals were purchased from The Jackson Laboratory (Bar Harbor, ME) and bred at the University of Wisconsin-Madison. PDGF-Bret/ret Lepob/ob mice were created by crossing B6 Lep+/+ PDGF-Bret/ret (37) and B6 Lepob/+ PDGF-B+/+ mice. For all studies, either wild-type or PDGF-Bret/+ littermates (indistinguishable from wild type) were used as controls. Either Lep+/+ or Lepob/+ mice were used in studies involving lean mice. Mice were housed in an environmentally controlled facility (12:12-h light-dark cycles) with free access to food (Purina) and water. All animal procedures were approved by the University of Wisconsin Animal Care and Use Committee.

FITC-dextran leakage.

Mice were anesthetized with a 250 mg/kg dose of Avertin. The right jugular was cannulated and injected with 200 ml of 2 mg/ml lysine-fixable FITC-dextran [10 kDa molecular weight (MW); Invitrogen]. The FITC-dextran circulated for 30 min. Frozen 10-μm sections were blocked with 5% donkey serum and stained overnight at +4°C with 1:50 rat anti-platelet endothelial cell adhesion molecule-1 (PECAM-1) antibody (BD Pharmingen) to mark the vasculature. Sections were stained with 1:100 Cy3-conjugated donkey anti-rat secondary antibody (Jackson ImmunoResearch). Confocal images were analyzed using Metamorph software in a blinded manner, and FITC-dextran associated with the vasculature was quantified as a signal that colocalized with PECAM-1. Extravascular FITC-dextran intensity was quantified as a signal that did not colocalize with PECAM-1.

Pericyte coverage immunohistochemistry.

Two frozen 10-μm sections/animal were blocked with 5% donkey serum containing 1% Triton X-100 and stained overnight at +4°C with 1:50 rat anti-PECAM-1 antibody (as described in FITC-dextran leakage) and 1:100 rabbit anti-desmin antibody (Abcam) to mark HSCs. Sections were stained with anti-rat (above) or 1:100 FITC-conjugated goat anti-rabbit (Jackson ImmunoResearch) secondary antibodies. We imaged three random images per section, totaling six images per animal. Pericyte coverage was calculated as the desmin area divided by the PECAM-1 area per animal as determined by Metamorph.

Transmission electron microscopy of liver sinusoids.

Sections were cut and contrasted with 8% uranyl acetate in 50% ethanol for 7 min at room temperature, followed by Reynolds lead citrate for 5 min at room temperature. Images of hepatic sinusoids were taken at ×8,800 magnification on a Philips CM120 transmission electron microscope. Sinusoids were scored in a blinded manner as “intact” (fenestrated with a well-formed space of Disse) or “disrupted” (defenestrated with a disorganized, electron-dense space of Disse). Results are presented as intact or disrupted sinusoids as a percentage of total images analyzed per animal.

Insulin signaling Western blots.

Mice were given a 5-min portal vein injection of 5 U of insulin to measure insulin receptor-β (IRβ) signaling or were injected with 0.1 U/kg insulin via jugular vein for 15 min to measure Akt signaling. Livers were lysed on ice for 30 min in 50 mM HEPES buffer (pH 7.4) containing 150 mM NaCl, 2 mM EDTA, 1% Nonidet P-40, and 10% glycerol and supplemented with a protease and phosphatase inhibitor cocktail (Roche). Supernatants were collected via centrifugation at 15,000 g for 20 min, and protein concentration was measured by BCA assay.

Total liver homogenates (200 mg for IRβ, 50 mg for Akt) were subjected to SDS-PAGE, transferred to nitrocellulose membrane, and blotted with a mouse anti-phosphotyrosine antibody (1:1,000; Millipore) for IRβ or a mouse anti-phospho-Akt (Ser473) antibody (1:1,000; Cell Signaling Technology) overnight. To detect total protein, blots were stripped and probed overnight with a rabbit anti-IRβ antibody (1:1,000; Santa Cruz Biotechnology) or a rabbit anti-Akt antibody (1:1,000; Cell Signaling Technology).

2-[14C]deoxyglucose uptake.

Nonfasted animals were injected via right jugular cannulation with 0.1 U/kg insulin, 2 g/kg glucose, and 300 μCi/kg 2-[1-14C]deoxy-d-glucose (PerkinElmer). Mice were bled 15 min after tracer injection, and organs were snap-frozen.

Plasma was precipitated with 0.3 N Ba(OH)2 and 0.3 N Zn(SO)4 and centrifuged for 5 min at high speed. Glucose-specific activity (disintegrations·min−1·mg−1) was calculated by dividing the supernatant radioactivity (disintegrations·min−1·ml−1) by the plasma glucose concentration (mg/ml).

Tissues (50 mg) were homogenized in 0.5% perchloric acid and centrifuged for 20 min at 2,000 g. Radioactivity was counted from the initial supernatant (supernatant 1) and the supernatant following Ba(OH)2 and Zn(SO)4 precipitation (supernatant 2). 2-[14C]deoxyglucose radioactivity (disintegrations·min−1·mg−1) was calculated as the difference between supernatants 1 and 2 divided by the sample weight. This value, divided by the glucose-specific activity, generated the amount of 2-[14C]deoxyglucose phosphate per tissue (mg 14C/mg tissue).

Insulin and glucose tolerance tests and glucose-stimulated insulin secretion from isolated islets.

Insulin tolerance tests were performed on nonfasted animals. Blood samples were drawn before insulin injection (Novolin R; Novo Nordisk) and at 5, 15, 30, 60, and 120 min after intraperitoneal injection. Lep+/+ mice were given 0.1 U/kg body wt insulin in sterile saline, whereas Lepob mice were given 20 U/kg body wt. Human plasma insulin was measured using a radioimmunoassay (Millipore).

Glucose tolerance tests were performed on mice fasted overnight. Glucose (2 mg/g body wt in sterile saline) was administered intraperitoneally. Blood samples were drawn before glucose injection and at 15, 30, 60, 120, 240, and 360 min after injection. Plasma C-peptide levels were measured using a radioimmunoassay (Linco-Millipore).

In vitro insulin secretion assays were performed as described previously (42).

Statistical analysis.

Fasting plasma insulin values did not have a normal distribution and were thus analyzed using a Mann-Whitney test. Area under the curve was determined for insulin and glucose tolerance tests and analyzed using unpaired Student t-tests. Curves were not corrected for baseline in these analyses. Individual time points during the insulin and glucose tolerance tests were analyzed by unpaired Student t-tests without adjusting for multiple comparisons; during the insulin tolerance tests, raw (not normalized to time 0) glucose values were used for this analysis. All other studies were analyzed using unpaired Student t-tests.


PDGF-Bret/ret livers display increased vascular permeability.

To study vascular permeability in liver, gastrocnemius muscle, and heart, we carried out intrajugular injection of 10 kDa FITC-dextran. The most striking difference was found in liver, where after only 30 min of circulation, we observed an approximately threefold increase in FITC-dextran accumulation in PDGF-Bret/ret relative to PDGF-Bret/+ control mice (Fig. 1, A–C). Of the accumulated FITC-dextran, 38% remained associated with the vasculature in control liver compared with only 14% in PDGF-Bret/ret liver, suggesting more rapid diffusion of dextran out of the vascular space in PDGF-deficient mice. We did not observe leakage in skeletal muscle but found increased leakage in PDGF-Bret/ret heart after only 24 h of circulation.

Fig. 1.
Enhanced vascular leakage resulting from reduced sinusoidal disruption in platelet-derived growth factor B (PDGF-B)ret/ret liver. Eight-week-old male Leptinob (Lepob/+) PDGF-Bret/+ control (A) and PDGF-Bret/ret mice (B) were infused with FITC-dextran ...

PDGF-B is a known mitogen and chemotactic growth factor for liver pericytes, also known as hepatic stellate cells (HSCs) or Ito cells (45). In addition, PDGF-B induces pericyte recruitment and attachment to endothelial cells in other vascular beds (23). Pericyte coverage is believed to play an important role in the maintenance of vascular barrier function (44). Therefore, loss of pericyte coverage induced by the PDGF-Bret/ret mutation may explain the difference in vascular permeability noted in the PDGF-Bret/ret liver.

To quantify hepatic pericyte coverage, we performed immunohistological analysis of wild-type and PDGF-Bret/ret liver vasculature and staining for sinusoidal endothelial cells (PECAM-1) and hepatic stellate cells (desmin). Despite the loss of pericytes in numerous other tissues, we did not observe a loss of HSC coverage in PDGF-Bret/ret liver (Fig. 1, D–F). This implies that PDGF-B deficiency increases hepatic vascular permeability independently of its effects on liver pericyte density.

PDGF-B signaling also contributes to hepatic stellate cell activation (20). This process is characterized by sinusoidal capillarization, a process by which liver sinusoids become defenestrated and display fibrotic alterations to the space of Disse (22). Capillarization can prevent the delivery of macromolecules to hepatocytes in models of liver fibrosis, injury, obesity, and aging (34, 38). Therefore, we analyzed PDGF-Bwt/wt and PDGF-Bret/ret sinusoidal ultrastructure by transmission electron microscopy. The majority of PDGF-Bret/ret sinusoids (80%) were intact, as marked by a highly fenestrated endothelium and well-formed space of Disse with numerous hepatocyte microvilli (Fig. 1, H and I). In contrast, ~60% of PDGF-Bwt/wt sinusoids displayed some form of disruption, including defenestration and a disordered electron-dense space of Disse (Fig. 1, G and I). Together, these data suggest that PDGF-B deficiency prevents capillarization-like alterations to the hepatic sinusoidal ultrastructure, thereby enhancing transendothelial transport in the liver.

PDGF-Bret/ret mice display enhanced hepatic insulin signaling and whole body glucose uptake.

PDGF-Bret/ret mice have elevated hepatic sinusoidal transport that appears to result from a reduction in capillarization-like changes to the sinusoidal architecture. Given that the molecular weight of the FITC-dextran we used was 10 kDa and that of insulin was ~6 kDa, we hypothesized that PDGF-B-deficient sinusoids might be more permissive for insulin delivery to hepatocytes. This would lead to greater downstream insulin signaling. Therefore, we examined phosphorylation of the IRβ and Akt at 5 and 15 min, respectively, after intraportal injection of insulin. Phosphorylation of both IRβ and Akt was increased in the livers of PDGF-Bret/ret mice (3-fold increase over wild type for IRβ, 5-fold for Akt; Fig. 2, A and B).

Fig. 2.
Increased insulin signaling and glucose clearance in PDGF-Bret/ret liver. A: representative Western blots of insulin receptor-β (IRβ) and Akt phosphorylation. Phosphotyrosine and total IRβ were measured in 8-wk-old Lepob/+ male ...

To evaluate the downstream effects of insulin stimulation, we quantified the mRNA abundance of insulin-regulated genes. Insulin signaling induces the expression of glucokinase, and we observed a twofold elevation in glucokinase mRNA in PDGF-Bret/ret liver (Fig. 2C). Insulin signaling also suppresses the expression of glucose-6-phosphatase, and accordingly, we observed an 80% reduction in glucose-6-phosphatase mRNA in PDGF-Bret/ret liver (Fig. 2C). These results support our hypothesis that elevated hepatic vascular permeability increases insulin signaling as a consequence of PDGF-B deficiency.

We next asked whether this increased hepatic insulin signaling could affect hepatic glucose uptake. In hepatocytes, glucose transport is not determined by the abundance of glucose transporters but is largely determined by the capacity of hepatocytes to phosphorylate glucose, i.e., the abundance of glucokinase. Mice were injected intravenously with insulin (0.1 U/kg), glucose, and a 2-[14C]deoxyglucose tracer. After 15 min of circulation, 14C radioactivity was measured in the liver. Hepatic glucose uptake was increased twofold in PDGF-Bret/ret mice compared with PDGF-Bwt/wt mice (Fig. 2D). This measurement correlates with increased whole body insulin-stimulated glucose clearance in PDGF-Bret/ret mice following an intraperitoneal insulin injection (Fig. 2E). Taken together, these results indicate that increased hepatic insulin signaling present during PDGF-B deficiency contributes to whole body insulin sensitivity and increased glucose clearance.

PDGF-Bret/ret partially restores insulin sensitivity to Lepob mice.

To study the role of PDGF-B signaling in insulin and glucose homeostasis under conditions of metabolic stress evoked by obesity, we created obese PDGF retention-deficient mice by breeding the Lepob mutation into the PDGF-Bret/ret mice (Table 1). Lepob mice normally have severe insulin resistance and hyperinsulinemia (typical fasting insulin levels are elevated ~8-fold). PDGF-Bret/ret Lepob mice displayed improved peripheral insulin sensitivity. When challenged by insulin, PDGF-Bret/ret Lepob mice cleared ~30% more glucose than control Lepob mice (Fig. 3A). PDGF-Bret/ret Lepob mice cleared ~65% more injected insulin than control Lepob mice (Fig. 3B).

Table 1.
PDGF-Bret/ret Lepob fasting plasma phenotypes
Fig. 3.
Enhanced whole body glucose and insulin clearance in PDGF-Bret/ret Lepob mice. Plasma glucose (%time 0; A) and human insulin concentration in plasma (B) during insulin tolerance tests on 8-wk-old male control Lepob (n = 8) and PDGF-Bret/ret Lepob mice ...

PDGF-Bret/ret Lepob mice display severely reduced plasma insulin without elevated hyperglycemia.

The improved insulin sensitivity of the PDGF-Bret/ret Lepob mice would predict an amelioration of the severe obesity-induced hyperinsulinemia normally seen in Lepob mice. Indeed, the fasting plasma insulin levels of the PDGF-Bret/ret Lepob mice were reduced by 88% at 14 wk of age (Fig. 4A). Despite a severe reduction in fasting plasma insulin and a severe reduction in the insulin secretory response to a glucose challenge (Fig. 4C), fasting glucose (Fig. 4B) and postchallenge glucose (Fig. 4D) were at the normal range of Lepob. Together, these data show that the loss of PDGF-B function greatly enhances insulin action in vivo, enabling severely obese mice to avoid hyperglycemia despite a very low insulin level.

Fig. 4.
Severely reduced plasma insulin without a change in glucose tolerance in PDGF-Bret/ret Lepob mice. A and B: plasma insulin (A) and plasma glucose (B) from male control Lepob and PDGF-Bret/ret Lepob mice measured after a 4-h fast. C and D: in vivo insulin ...

PDGF-Bret/ret Lepob mice display intact β-cell secretory capacity.

A surprising result was the essentially flat insulin response of the PDGF-Bret/ret Lepob mice after a glucose challenge (Fig. 4C). This is a consequence of very rapid insulin clearance (Fig. 3B), but it may also reflect reduced insulin secretion. To assess β-cell secretion in vivo, we quantified C-peptide levels. After a glucose challenge, the PDGF-Bret/ret Lepob C-peptide area under the curve was reduced by 36% relative to Lepob controls (Fig. 5A).

Fig. 5.
Intact β-cell secretory capacity in PDGF-Bret/ret Lepob mice. A: C-peptide secretion during glucose tolerance tests on 8-wk-old male control Lepob (n = 6) and PDGF-Bret/ret Lepob mice (n = 4). B: plasma C-peptide/insulin ratios during a glucose ...

The C-peptide/insulin ratio can be used as a surrogate measurement of insulin clearance because C-peptide is released in equimolar amounts to insulin but is cleared from circulation at approximately one-tenth the rate of insulin (48). The C-peptide/insulin ratio was significantly higher over time in PDGF-Bret/ret Lepob mice (Fig. 5B), consistent with the reduced area under the curve of plasma insulin during the insulin tolerance curve (Fig. 3B).

To directly assess β-cell insulin secretory function, we measured glucose-stimulated insulin secretion from isolated islets in vitro. Both insulin secretion in response to high glucose and total cellular insulin content were reduced significantly in islets from PDGF-Bret/ret Lepob mice (Fig. 5C). However, when normalized for insulin content, PDGF-Bret/ret Lepob fractional insulin release was not significantly different from control Lepob islets (Fig. 5D), indicating that the secretory capacity of PDGF-Bret/ret islets is intact and that the reduced insulin secretion is due to a large reduction in the pool of available insulin in the β-cells. The β-cell area (Fig. 5E), islet size (data not shown), and β-cell architecture (data not shown) were not significantly different between PDGF-Bwt/wt Lepob and PDGF-Bret/ret Lepob mice, suggesting that the reduced insulin content results from diminished insulin synthesis in the β-cells rather than a loss of β-cells.


In the present study, we demonstrate that PDGF-B deficiency prevents hepatic capillarization and thereby increases hepatic vascular permeability. This results in increased hepatic insulin signaling, suggesting improved insulin delivery to hepatocytes. In obese mice, this enhanced hepatic insulin signaling partially ameliorates the insulin resistance and hyperinsulinemia normally associated with obesity (Fig. 6) and is associated with a far lower glucose level than one would normally expect in obese mice with low insulin levels.

Fig. 6.
Reduced PDGF-B signaling improves hepatic vascular delivery of insulin and enhances whole body insulin and glucose homeostasis. In C57BL/6J mice, normal PDGF-B retention and signaling leads to ultrastructural changes to the hepatic sinusoids, including ...

Unlike muscle, where transendothelial transport is known to be rate limiting for insulin delivery (6, 18), the liver is considered to be freely permeable to most solutes, and vascular delivery to hepatocytes has not been believed to limit hormone action (34). Although defenestration and alterations to the perisinusoidal space can impede the delivery of macromolecules to hepatocytes (34, 38), most studies have focused on cases of advanced cirrhosis and fibrosis. We show that hepatic sinusoidal transendothelial transport can be improved even in animals lacking overt liver fibrosis and that such changes affect whole body insulin and glucose homeostasis.

We were surprised to observe equal HSC coverage of sinusoidal vessels in PDGF-Bwt/wt and PDGF-Bret/ret liver. However, in PDGF-B−/− embryos, which do not survive to birth, HSCs develop normally in the liver (28), suggesting that another factor can induce HSC recruitment in the face of reduced hepatic PDGF-B signaling. Multiple growth factors that function in signaling and recruitment of pericytes have been identified (23), and the identity of the pericyte recruitment factor in the liver remains to be elucidated. Our data imply that PDGF-B controls hepatic sinusoidal function independently of its role in pericyte recruitment in the liver.

One such potential function for PDGF-B involves HSC activation, the process by which normally quiescent HSCs begin synthesizing ECM (45). HSC activation and the resulting ECM production are classically attributed to transforming growth factor-β (TGFβ) signaling (11). However, hepatic PDGF-B overexpression has more recently been shown to trigger HSC activation and fibrosis independently of TGFβ (20). In our study, ~60% of PDGF-Bwt/wt sinusoids were noticeably altered. These disruptions resembled the sinusoidal changes reported for HSC activation, including loss of endothelial fenestration and accumulation of electron-dense ECM-like material in the space of Disse (45). In contrast, hepatic sinusoidal architecture was noticeably improved in PDGF-Bret/ret mice. The results of our study provide further evidence that PDGF-B plays an active role in sinusoidal remodeling.

Hepatic steatosis/NAFLD is the most common manifestation of liver insulin resistance, and B6 mice are predisposed to the development of fatty liver, even in the lean state (10). Ultrastructural changes to the hepatic sinusoids are common in NAFLD (3), and this may explain why reduced PDGF-B signaling actually improves hepatic sinusoidal structure in the B6 background. Indeed, black and tan, tufted mice are resistant to the development of fatty liver, and they display reduced sinusoidal alterations compared with B6 mice (data not shown). These studies suggest that sinusoidal dysfunction plays a role in the development of hepatic insulin resistance.

Our data illustrate that changes in liver transendothelial transport due to loss of PDGF-B retention result in elevated whole body insulin and glucose clearance. This is a unique mechanism for enhancement of insulin sensitivity, because insulin resistance is classically thought to occur through defects in the intracellular insulin-signaling cascade downstream of the insulin receptor. Our findings demonstrate a novel role for PDGF-B in hepatic insulin signaling and maintenance of euglycemia and emphasize that hepatic delivery of insulin is an important step in insulin action.

One of the major observations in the present study is that the loss of PDGF-B retention causes a profound increase in insulin sensitivity in PDGF-Bret/ret Lepob mice. This improvement in insulin sensitivity prevents the expected obesity-induced insulin resistance and hyperinsulinemia of PDGF-Bret/ret Lepob mice. The mice maintain the same abnormal fasting glucose values as the Lepob controls and similar glucose tolerance despite drastically reduced fasting insulin levels and reduced glucose-stimulated insulin secretion. The β-cell secretory machinery remains functional upon PDGF-B deficiency, as demonstrated by normal in vitro fractional insulin secretion from isolated islets. The observed decrease in insulin secretion is most likely due to reduced insulin demand. Improvements in insulin sensitivity are often accompanied by corresponding decreases in insulin secretion and β-cell insulin content (16).

The present study provides new insights into the effect of imatinib mesylate (Gleevec) on diabetes. Imatinib mesylate inhibits several receptor tyrosine kinases, including PDGFRβ (13). Imatinib mesylate lowers fasting blood glucose levels in diabetic patients treated for chronic myeloid leukemia (2, 12, 13, 49). Treatment of db/db mice with imatinib mesylate improves insulin sensitivity and decreases in vivo insulin secretion in response to a glucose challenge (26). It is possible that the observed increase in insulin sensitivity may involve PDGFRβ inhibition and subsequent effects on hepatic sinusoidal structure. Sinusoidal fibrosis is reduced upon imatinib mesylate treatment (54). This suggests that further investigation into the effects of imatinib mesylate and other PDGFRβ inhibitors on HSC function might uncover new avenues, leading to the treatment of insulin resistance and type 2 diabetes.

Our results also provide a basis for further study into the mechanisms underlying the action of Sorcs1, a PDGF-B-binding protein (29). Several studies have suggested that Sorcs1 expression modulates diabetes in both rodents and humans (19, 24, 41), although its mechanism of action is not clear. The present study demonstrates that modulation of PDGF-B levels can affect insulin action, and Sorcs1 binding to PDGF-B could potentially affect PDGF-B availability, clearance, or function.

In conclusion, this study shows that loss of PDGF-B enhances hepatic insulin action, resulting in improved glucose tolerance. This enhanced glycemic control reduces the need for excessive insulin secretion, alleviating the hyperinsulinemic state normally seen in obese animals. Most importantly, our work reveals a novel role for PDGF-B in hepatic insulin delivery and supports the concept that insulin access to its target tissues is a vital part of insulin sensitivity.


This study was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-66369 and DK-58037 (A. D. Attie), National Institutes of Health Training Grant T-32-GN-07215 (S. M. Raines and O. C. Richards), and a Wisconsin Alumni Research Foundation fellowship (O. C. Richards).


D. Attie is a member of Pfizer's Cardiovascular Medicine TASAP Committee.


We thank Elaine Raines (University of Washington) for suggesting the collaboration between our laboratory and that of Christer Betsholtz. We are grateful to Kristen Bolstad, Nicole Spencer, Matthew Flowers, Harini Sampath (University of Wisconsin-Madison), and Sudha Biddinger (Harvard Medical School) for expert technical assistance and advice. We thank Laura Vanderploeg for graphic design expertise. We are grateful to Susanne Clee (University of British Columbia) for critical review of the manuscript and helpful discussion. We thank the University of Wisconsin Medical School Electron Microscope Facility for assistance with transmission electron microscopy studies.


1. Abramsson A, Lindblom P, Betsholtz C. Endothelial and nonendothelial sources of PDGF-B regulate pericyte recruitment and influence vascular pattern formation in tumors. J Clin Invest 112: 1142–1151, 2003. [PMC free article] [PubMed]
2. Agostino N, Chinchilli VM, Lynch CJ, Koszyk-Szewczyk A, Gingrich R, Sivik J, Drabick JJ. Effect of the tyrosine kinase inhibitors (sunitinib, sorafenib, dasatinib, and imatinib) on blood glucose levels in diabetic and nondiabetic patients in general clinical practice. J Oncol Pharm Pract. In press. [PubMed]
3. Ahishali E, Demir K, Ahishali B, Akyuz F, Pinarbasi B, Poturoglu S, Ibrisim D, Gulluoglu M, Ozdil S, Besisik F, Kaymakoglu S, Boztas G, Cakaloglu Y, Mungan Z, Canberk Y, Okten A. Electron microscopic findings in non-alcoholic fatty liver disease: is there a difference between hepatosteatosis and steatohepatitis? J Gastroenterol Hepatol 25: 619–626, 2010. [PubMed]
4. Aird WC. Phenotypic heterogeneity of the endothelium: II. Representative vascular beds. Circ Res 100: 174–190, 2007. [PubMed]
5. Armulik A, Genové G, Mäe M, Nisancioglu MH, Wallgard E, Niaudet C, He L, Norlin J, Lindblom P, Strittmatter K, Johansson BR, Betsholtz C. Pericytes regulate the blood-brain barrier. Nature 468: 557–561, 2010. [PubMed]
6. Barrett EJ, Eggleston EM, Inyard AC, Wang H, Li G, Chai W, Liu Z. The vascular actions of insulin control its delivery to muscle and regulate the rate-limiting step in skeletal muscle insulin action. Diabetologia 52: 752–764, 2009. [PMC free article] [PubMed]
7. Bell RD, Winkler EA, Sagare AP, Singh I, LaRue B, Deane R, Zlokovic BV. Pericytes control key neurovascular functions and neuronal phenotype in the adult brain and during brain aging. Neuron 68: 409–427, 2010. [PMC free article] [PubMed]
8. Bernuau D, Guillot R, Durand AM, Raoux N, Gabreau T, Passa P, Feldmann G. Ultrastructural aspects of the liver perisinusoidal space in diabetic patients with and without microangiopathy. Diabetes 31: 1061–1067, 1982. [PubMed]
9. Bernuau D, Guillot R, Durand-Schneider AM, Poussier P, Moreau A, Feldmann G. Liver perisinusoidal fibrosis in BB rats with or without overt diabetes. Am J Pathol 120: 38–45, 1985. [PMC free article] [PubMed]
10. Biddinger SB, Almind K, Miyazaki M, Kokkotou E, Ntambi JM, Kahn CR. Effects of diet and genetic background on sterol regulatory element-binding protein-1c, stearoyl-CoA desaturase 1, and the development of the metabolic syndrome. Diabetes 54: 1314–1323, 2005. [PubMed]
11. Bissell DM, Roulot D, George J. Transforming growth factor beta and the liver. Hepatology 34: 859–867, 2001. [PubMed]
12. Breccia M, Muscaritoli M, Alimena G. Reduction of glycosylated hemoglobin with stable insulin levels in a diabetic patient with chronic myeloid leukemia responsive to imatinib. Haematologica 90, Suppl: ECR21, 2005. [PubMed]
13. Breccia M, Muscaritoli M, Aversa Z, Mandelli F, Alimena G. Imatinib mesylate may improve fasting blood glucose in diabetic Ph+ chronic myelogenous leukemia patients responsive to treatment. J Clin Oncol 22: 4653–4655, 2004. [PubMed]
14. Brissova M, Shostak A, Shiota M, Wiebe PO, Poffenberger G, Kantz J, Chen Z, Carr C, Jerome WG, Chen J, Baldwin HS, Nicholson W, Bader DM, Jetton T, Gannon M, Powers AC. Pancreatic islet production of vascular endothelial growth factor-a is essential for islet vascularization, revascularization, and function. Diabetes 55: 2974–2985, 2006. [PubMed]
15. Brock RW, Dorman RB. Obesity, insulin resistance and hepatic perfusion. Microcirculation 14: 339–347, 2007. [PubMed]
16. Cederroth CR, Vinciguerra M, Gjinovci A, Kühne F, Klein M, Cederroth M, Caille D, Suter M, Neumann D, James RW, Doerge DR, Wallimann T, Meda P, Foti M, Rohner-Jeanrenaud F, Vassalli JD, Nef S. Dietary phytoestrogens activate AMP-activated protein kinase with improvement in lipid and glucose metabolism. Diabetes 57: 1176–1185, 2008. [PubMed]
17. Chiu JD, Richey JM, Harrison LN, Zuniga E, Kolka CM, Kirkman E, Ellmerer M, Bergman RN. Direct administration of insulin into skeletal muscle reveals that the transport of insulin across the capillary endothelium limits the time course of insulin to activate glucose disposal. Diabetes 57: 828–835, 2008. [PubMed]
18. Clark MG. Impaired microvascular perfusion: a consequence of vascular dysfunction and a potential cause of insulin resistance in muscle. Am J Physiol Endocrinol Metab 295: E732–E750, 2008. [PMC free article] [PubMed]
19. Clee SM, Yandell BS, Schueler KM, Rabaglia ME, Richards OC, Raines SM, Kabara EA, Klass DM, Mui ET, Stapleton DS, Gray-Keller MP, Young MB, Stoehr JP, Lan H, Boronenkov I, Raess PW, Flowers MT, Attie AD. Positional cloning of Sorcs1, a type 2 diabetes quantitative trait locus. Nat Genet 38: 688–693, 2006. [PubMed]
20. Czochra P, Klopcic B, Meyer E, Herkel J, Garcia-Lazaro JF, Thieringer F, Schirmacher P, Biesterfeld S, Galle PR, Lohse AW, Kanzler S. Liver fibrosis induced by hepatic overexpression of PDGF-B in transgenic mice. J Hepatol 45: 419–428, 2006. [PubMed]
21. Daneman R, Zhou L, Kebede AA, Barres BA. Pericytes are required for blood-brain barrier integrity during embryogenesis. Nature 468: 562–566, 2010. [PMC free article] [PubMed]
22. Farrell GC, Teoh NC, McCuskey RS. Hepatic microcirculation in fatty liver disease. Anat Rec (Hoboken) 291: 684–692, 2008. [PubMed]
23. Gaengel K, Genove G, Armulik A, Betsholtz C. Endothelial-mural cell signaling in vascular development and angiogenesis. Arterioscler Thromb Vasc Biol 29: 630–638, 2009. [PubMed]
24. Goodarzi MO, Lehman DM, Taylor KD, Guo X, Cui J, Quinones MJ, Clee SM, Yandell BS, Blangero J, Hsueh WA, Attie AD, Stern MP, Rotter JI. SORCS1: a novel human type 2 diabetes susceptibility gene suggested by the mouse. Diabetes 56: 1922–1929, 2007. [PubMed]
25. Hammes HP. Pericytes and the pathogenesis of diabetic retinopathy. Horm Metab Res 37, Suppl 1: 39–43, 2005. [PubMed]
26. Han MS, Chung KW, Cheon HG, Rhee SD, Yoon CH, Lee MK, Kim KW, Lee MS. Imatinib mesylate reduces endoplasmic reticulum stress and induces remission of diabetes in db/db mice. Diabetes 58: 329–336, 2009. [PMC free article] [PubMed]
27. Hellstrom M, Gerhardt H, Kalen M, Li X, Eriksson U, Wolburg H, Betsholtz C. Lack of pericytes leads to endothelial hyperplasia and abnormal vascular morphogenesis. J Cell Biol 153: 543–553, 2001. [PMC free article] [PubMed]
28. Hellstrom M, Kalen M, Lindahl P, Abramsson A, Betsholtz C. Role of PDGF-B and PDGFR-beta in recruitment of vascular smooth muscle cells and pericytes during embryonic blood vessel formation in the mouse. Development 126: 3047–3055, 1999. [PubMed]
29. Hermey G, Sjøgaard SS, Petersen CM, Nykjaer A, Gliemann J. Tumour necrosis factor alpha-converting enzyme mediates ectodomain shedding of Vps10p-domain receptor family members. Biochem J 395: 285–293, 2006. [PMC free article] [PubMed]
30. Jamieson HA, Cogger VC, Twigg SM, McLennan SV, Warren A, Cheluvappa R, Hilmer SN, Fraser R, de Cabo R, Le Couteur DG. Alterations in liver sinusoidal endothelium in a baboon model of type 1 diabetes. Diabetologia 50: 1969–1976, 2007. [PubMed]
31. Kantarjian H, Pasquini R, Hamerschlak N, Rousselot P, Holowiecki J, Jootar S, Robak T, Khoroshko N, Masszi T, Skotnicki A, Hellmann A, Zaritsky A, Golenkov A, Radich J, Hughes T, Countouriotis A, Shah N. Dasatinib or high-dose imatinib for chronic-phase chronic myeloid leukemia after failure of first-line imatinib: a randomized phase 2 trial. Blood 109: 5143–5150, 2007. [PubMed]
32. Lammert E, Gu G, McLaughlin M, Brown D, Brekken R, Murtaugh LC, Gerber HP, Ferrara N, Melton DA. Role of VEGF-A in vascularization of pancreatic islets. Curr Biol 13: 1070–1074, 2003. [PubMed]
33. Latry P, Bioulac-Sage P, Echinard E, Gin H, Boussarie L, Grimaud JA, Balabaud C. Perisinusoidal fibrosis and basement membrane-like material in the livers of diabetic patients. Hum Pathol 18: 775–780, 1987. [PubMed]
34. Le Couteur DG, Fraser R, Hilmer S, Rivory LP, McLean AJ. The hepatic sinusoid in aging and cirrhosis: effects on hepatic substrate disposition and drug clearance. Clin Pharmacokinet 44: 187–200, 2005. [PubMed]
35. Leveen P, Pekny M, Gebre-Medhin S, Swolin B, Larsson E, Betsholtz C. Mice deficient for PDGF B show renal, cardiovascular, and hematological abnormalities. Genes Dev 8: 1875–1887, 1994. [PubMed]
36. Lindahl P, Johansson BR, Leveen P, Betsholtz C. Pericyte loss and microaneurysm formation in PDGF-B-deficient mice. Science 277: 242–245, 1997. [PubMed]
37. Lindblom P, Gerhardt H, Liebner S, Abramsson A, Enge M, Hellstrom M, Backstrom G, Fredriksson S, Landegren U, Nystrom HC, Bergstrom G, Dejana E, Ostman A, Lindahl P, Betsholtz C. Endothelial PDGF-B retention is required for proper investment of pericytes in the microvessel wall. Genes Dev 17: 1835–1840, 2003. [PMC free article] [PubMed]
38. McCuskey RS, Ito Y, Robertson GR, McCuskey MK, Perry M, Farrell GC. Hepatic microvascular dysfunction during evolution of dietary steatohepatitis in mice. Hepatology 40: 386–393, 2004. [PubMed]
39. Miles PD, Levisetti M, Reichart D, Khoursheed M, Moossa AR, Olefsky JM. Kinetics of insulin action in vivo. Identification of rate-limiting steps. Diabetes 44: 947–953, 1995. [PubMed]
40. Ostman A, Andersson M, Betsholtz C, Westermark B, Heldin CH. Identification of a cell retention signal in the B-chain of platelet-derived growth factor and in the long splice version of the A-chain. Cell Regul 2: 503–512, 1991. [PMC free article] [PubMed]
41. Paterson AD, Waggott D, Boright AP, Hosseini SM, Shen E, Sylvestre MP, Wong I, Bharaj B, Cleary PA, Lachin JM, Below JE, Nicolae D, Cox NJ, Canty AJ, Sun L, Bull SB. A genome-wide association study identifies a novel major locus for glycemic control in type 1 diabetes, as measured by both A1C and glucose. Diabetes 59: 539–549, 2010. [PMC free article] [PubMed]
42. Rabaglia ME, Gray-Keller MP, Frey BL, Shortreed MR, Smith LM, Attie AD. α-Ketoisocaproate-induced hypersecretion of insulin by islets from diabetes-susceptible mice. Am J Physiol Endocrinol Metab 289: E218–E224, 2005. [PubMed]
43. Raines EW, Ross R. Compartmentalization of PDGF on extracellular binding sites dependent on exon-6-encoded sequences. J Cell Biol 116: 533–543, 1992. [PMC free article] [PubMed]
44. Richards OC, Raines SM, Attie AD. The role of blood vessels, endothelial cells, and vascular pericytes in insulin secretion and peripheral insulin action. Endocr Rev 31: 343–363, 2010. [PMC free article] [PubMed]
45. Sato M, Suzuki S, Senoo H. Hepatic stellate cells: unique characteristics in cell biology and phenotype. Cell Struct Funct 28: 105–112, 2003. [PubMed]
46. Schoemaker MH, Rots MG, Beljaars L, Ypma AY, Jansen PL, Poelstra K, Moshage H, Haisma HJ. PDGF-receptor beta-targeted adenovirus redirects gene transfer from hepatocytes to activated stellate cells. Mol Pharm 5: 399–406, 2008. [PubMed]
47. Soriano P. Abnormal kidney development and hematological disorders in PDGF beta-receptor mutant mice. Genes Dev 8: 1888–1896, 1994. [PubMed]
48. Steiner DF. The proinsulin C-peptide—a multirole model. Exp Diabesity Res 5: 7–14, 2004. [PMC free article] [PubMed]
49. Veneri D, Franchini M, Bonora E. Imatinib and regression of type 2 diabetes. N Engl J Med 352: 1049–1050, 2005. [PubMed]
50. Wang H, Liu Z, Li G, Barrett EJ. The vascular endothelial cell mediates insulin transport into skeletal muscle. Am J Physiol Endocrinol Metab 291: E323–E332, 2006. [PubMed]
51. Wang H, Wang AX, Barrett EJ. Caveolin-1 is required for vascular endothelial insulin uptake. Am J Physiol Endocrinol Metab 300: E134–E144, 2011. [PMC free article] [PubMed]
52. Whiteman EL, Chen JJ, Birnbaum MJ. Platelet-derived growth factor (PDGF) stimulates glucose transport in 3T3-L1 adipocytes overexpressing PDGF receptor by a pathway independent of insulin receptor substrates. Endocrinology 144: 3811–3820, 2003. [PubMed]
53. Xian X, Hakansson J, Stahlberg A, Lindblom P, Betsholtz C, Gerhardt H, Semb H. Pericytes limit tumor cell metastasis. J Clin Invest 116: 642–651, 2006. [PMC free article] [PubMed]
54. Yoshiji H, Noguchi R, Kuriyama S, Ikenaka Y, Yoshii J, Yanase K, Namisaki T, Kitade M, Masaki T, Fukui H. Imatinib mesylate (STI-571) attenuates liver fibrosis development in rats. Am J Physiol Gastrointest Liver Physiol 288: G907–G913, 2005. [PubMed]
55. Yuasa T, Kakuhata R, Kishi K, Obata T, Shinohara Y, Bando Y, Izumi K, Kajiura F, Matsumoto M, Ebina Y. Platelet-derived growth factor stimulates glucose transport in skeletal muscles of transgenic mice specifically expressing platelet-derived growth factor receptor in the muscle, but it does not affect blood glucose levels. Diabetes 53: 2776–2786, 2004. [PubMed]
56. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. Positional cloning of the mouse obese gene and its human homologue. Nature 372: 425–432, 1994. [PubMed]

Articles from American Journal of Physiology - Endocrinology and Metabolism are provided here courtesy of American Physiological Society

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...