the role of blood vessels, endothelial cells, and vascular ... · 2/17/2010 · the role of blood...

Endocr. Rev. 2010 31:343-363 originally published online Feb 17, 2010; , doi: 10.1210/er.2009-0035

Oliver C. Richards, Summer M. Raines and Alan D. Attie

Secretion and Peripheral Insulin ActionThe Role of Blood Vessels, Endothelial Cells, and Vascular Pericytes in Insulin

Society please go to: http://edrv.endojournals.org//subscriptions/ or any of the other journals published by The EndocrineEndocrine ReviewsTo subscribe to

Copyright © The Endocrine Society. All rights reserved. Print ISSN: 0021-972X. Online

The Role of Blood Vessels, Endothelial Cells, andVascular Pericytes in Insulin Secretion and PeripheralInsulin Action

Oliver C. Richards, Summer M. Raines, and Alan D. Attie

Department of Biochemistry (O.C.R., S.M.R., A.D.A.) and Graduate Program in Cellular and Molecular Biology (O.C.R.,A.D.A.), University of Wisconsin-Madison, Madison, Wisconsin 53706

The pathogenesis of type 2 diabetes is intimately intertwined with the vasculature. Insulin must efficientlyenter the bloodstream from pancreatic �-cells, circulate throughout the body, and efficiently exit thebloodstream to reach target tissues and mediate its effects. Defects in the vasculature of pancreatic isletscan lead to diabetic phenotypes. Similarly, insulin resistance is accompanied by defects in the vasculatureof skeletal muscle, which ultimately reduce the ability of insulin and nutrients to reach myocytes. Anunderappreciated participant in these processes is the vascular pericyte. Pericytes, the smooth muscle-likecells lining the outsides of blood vessels throughout the body, have not been directly implicated in insulinsecretion or peripheral insulin delivery. Here, we review the role of the vasculature in insulin secretion, isletfunction, and peripheral insulin delivery, and highlight a potential role for the vascular pericyte in theseprocesses. (Endocrine Reviews 31: 343–363, 2010)

I. IntroductionII. Endothelial Cells and the Heterogeneity of Vascular

BedsIII. Islet Vasculature and Insulin Secretion

A. Introduction to islet vasculatureB. The importance of the vasculature for pancreas

developmentC. Proper vascularization is also required for mature

islet functionD. The role of islet revascularization during islet

transplantationIV. Peripheral Vasculature and Insulin Delivery

A. Introduction to peripheral vasculature and insulindelivery

B. Transendothelial transport of insulinC. Effects of insulin on blood flowD. Insulin-induced capillary recruitmentE. Molecular mechanism of capillary recruitmentF. Insulin resistance and muscle vasculatureG. Exercise-induced vascular changes

V. Vascular Pericytes: More Than Inert Contractile CellsA. Introduction to pericytesB. Platelet-derived growth factor-B: a key mediator of

pericyte function

C. Diabetic complications: a key role for pericytesD. Are pericytes multipotent progenitor cells?E. Pericytes in normal islet functionF. Pericytes in islet tumorsG. A role for PDGF-B signaling in glucose uptake?H. Inhibition of PDGFR� and diabetes therapyI. A role for pericytes in insulin-induced hemodynamic

changesVI. Summary/Conclusions

I. Introduction

Type 2 diabetes is a growing world epidemic (1, 2).There appear to be two key steps in the development

of type 2 diabetes: 1) the development of insulin resistance;and 2) �-cell decompensation. Although both of these pro-cesses are beginning to be understood at the molecularlevel, much remains to be elucidated. An important recentdevelopment is the discovery of the role that blood vesselsplay in the pathogenesis of these two conditions. The focusof this review is an investigation of the role that blood

ISSN Print 0021-972X ISSN Online 1945-7197Printed in U.S.A.Copyright © 2010 by The Endocrine Societydoi: 10.1210/er.2009-0035 Received September 1, 2009. Accepted December 17, 2009.First Published Online February 17, 2010

Abbreviations: Ang-1, Angiopoeitin-1; CEU, contrast-enhanced ultrasound; eNOS, endo-thelial NO synthase; FITC, fluorescein isothiocyanate; HIF-1�, hypoxia-inducible factor-1�;IRS, insulin receptor substrate; MSC, mesenchymal stem cell; 1-MX, 1-methylxanthine;NG-2, neuron-glial 2; NO, nitric oxide; PDGF, platelet-derived growth factor; PDGFR, PDGFreceptor; Pdx-1, pancreatic and duodenal homeobox 1; PI3K, phosphatidylinositol-3-ki-nase; PKC, protein kinase C; PKG, protein kinase G; RIP, rat insulin promoter; �-SMA,�-smooth muscle actin; Vegf-A, vascular endothelial growth factor-A; vSMC, vascularsmooth muscle cell; ZF, Zucker fatty (rats).

R E V I E W

Endocrine Reviews, June 2010, 31(3):343–363 edrv.endojournals.org 343

vessels and their constituent endothelial cells, vascularsmooth muscle cells (vSMCs), and pericytes play in �-cellfunctionand thedevelopmentof insulin resistance. Severalexcellent reviews have described several of these topics (3,4), but this review will have a broader focus including boththe role of blood vessels in islet development and functionand introducing the pericyte as a novel mediator of theseeffects.

II. Endothelial Cells and the Heterogeneity ofVascular Beds

Blood vessels in the vascular beds of different tissues ex-hibit large structural variability, especially in the numberof fenestrae and caveolae (5–8). Fenestrae are the approx-imately 100-nm pores covered by a permeable diaphragmresulting from the fusion of apical and basolateral plasmamembranes. Caveolae are the 60- to 80-nm plasma mem-brane pits thought to be involved in endocytosis and trans-cytosis. For example, the highly permeable liver endothe-lium is termed “discontinuous” and contains larger thannormal fenestrae that lack diaphragms (5). Liver endothe-lium also has many intercellular gaps that allow for easyaccess of blood-borne molecules to hepatocytes (5). Incontrast, the nonfenestrated, caveolae-free endotheliumof the brain vasculature contains many tight junctions andhas very low permeability (5). This helps to form the blood-brain barrier, which regulates the entry of blood-borne mol-ecules into the brain and preserves ionic homeostasis (9).The permeability characteristics of pancreatic islet andmuscle vasculature lie somewhere between these two ex-tremes. Islet vasculature is relatively permeable, and al-though it does not have gaps between endothelial cells, theendothelial cells are highly fenestrated to allow for facilenutrient sampling from blood, enabling islets to respondquickly to fluctuations in blood glucose and adjust insulinsecretion as needed (10, 11). In contrast, both cardiac andskeletal muscle vasculature are relatively impermeable toblood-borne macromolecules (5, 6, 12).

Endothelial cells form the inner lining of blood vessels.They are critically involved in many physiological func-tions, including control of vasomotor tone, blood cell traf-ficking, hemostatic balance, permeability, proliferation,survival, and immunity (5). However, endothelial cells arenot the only component of blood vessels. Pericytes andvSMCs line the outer walls of blood vessels and play crit-ical roles in blood vessel and endothelial cell function.Interestingly, despite the recent interest in the role of bloodvessels in insulin secretion and insulin action, a role forpericytes in these processes has not been well described.

III. Islet Vasculature and Insulin Secretion

A. Introduction to islet vasculatureA complete understanding of the role of the vasculature

in islet function requires an understanding of islet vasculararchitecture. Pancreatic islets are highly vascularized by adense network of capillaries. The islet capillary network isapproximately five times denser in islets than in the sur-rounding exocrine tissue (11). Although pancreatic isletsaccount for approximately 1% of total pancreatic mass,they receive approximately 7–10% of total pancreaticblood flow (13). Traditionally, blood has been thought toflow from one to three arterioles into the intraislet capil-lary network and empty into venules in the islet periphery(14). However, real-time fluorescent tracing of blood flowpatterns in mouse islets demonstrates that in the majorityof cases, islets exhibit an inner-to-outer flow patternwhere capillaries perfuse �-cells before other islet celltypes and larger vessels exhibit an efferent rather thanafferent flow pattern (15). �-Cells are intimately associ-ated with islet endothelial cells and are usually no morethan one cell distant from the bloodstream (16). Isletcapillaries are highly fenestrated (10, 11); in fact, theycontain about 10 times more fenestrae than the sur-rounding exocrine tissue (8, 11). This presumably al-lows for delivery of nutrients and growth factors intoislets and enables efficient sampling of the blood to al-low for rapid glucose-sensing.

Islet blood flow can be regulated by a number of met-abolic and nonmetabolic factors. For instance, glucose al-most doubles islet blood flow (17). Insulin decreases isletblood flow, likely due to resultant hypoglycemia, ratherthan a direct effect of hyperinsulinemia (18). ATP signalsthrough islet A1 adenosine receptors to increase isletblood flow (19). This is in agreement with the effects ofglucose on islet blood flow and suggests that increased isletcell metabolism promotes increased islet blood flow.Other nutrients and hormones have indirect effects on isletblood flow by interacting with vasodilators and vasocon-strictors (20, 21). Finally, the central nervous systemimpacts islet perfusion; brain stem-dead donors have im-paired islet hormone secretion (22). These studies dem-onstrate that islet blood flow is carefully controlled byvarious nutrients and growth factors and is exquisitelyintertwined with the metabolic state of the organism.

B. The importance of the vasculature forpancreas development

A close relationship between islet endocrine cells andendothelial cells has been described, from early islet de-velopment through mature islet function. In the develop-ing embryo, the pancreatic endoderm is located immedi-ately beside the dorsal aorta, and the presence of the dorsal

344 Richards et al. The Vasculature and Insulin Action Endocrine Reviews, June 2010, 31(3):343–363

aorta is required for pancreatic differentiation and insulinexpression (23). Overexpression of vascular endothelialgrowth factor-A (Vegf-a) in the developing pancreas undercontrol of the pancreatic and duodenal homeobox 1(PDX1) promoter leads to a hypervascularized pancreaswith a 3-fold increase in islet number and islet area as wellas ectopic insulin expression in the developing gut, sug-gesting that the vasculature is not only critical for initia-tion of pancreas development but also remains importantfor islet growth throughout maturity (23). The transcrip-tion factor PTF1A was identified as a crucial signal that isinduced by signals from aortic endothelial cells andpromotes the budding of the PDX1-positive pancreaticendoderm as well as insulin and glucagon expression (24).

Subsequent to the early requirement for pancreaticbudding, endothelial cells, both directly and indirectlythrough sphingosine-1-phosphate, promote growth andbudding of the dorsal pancreatic endoderm by inductionof mesenchymal cell proliferation (25, 26). This close re-lationship between islet endothelial cells and endocrinecells continues during the period of postnatal islet expan-sion. During the pronounced expansion of islet endocrinecells 1 wk after birth, there is an even more pronouncedexpansion of islet endothelial cells, leading to a markedincrease in intraislet vascular density (27). Overexpressionof VEGF-A by a tetracycline-inducible pancreas-specificbitransgenic system demonstrates increased islet endothe-lial cell formation. However, this actually causes reducedpostnatal �-cell mass, suggesting that islet angiogenic sig-nals must be maintained within narrow limits (28). Thesedata indicate that endothelial cues are critically importantfor initiation of pancreatic budding and that endothelialcells remain vitally important for pancreas and isletgrowth throughout later embryonic and early postnataldevelopment.

C. Proper vascularization is also required for matureislet function

The close relationship between islet endocrine cells andendothelial cells is important not only for pancreas devel-opment, but also for mature islet function. Knockout ofVEGF-A byPdx1-promoter-driven Cre recombinase leadsto severely reduced islet vascular density, resulting in glu-cose intolerance (29). Interestingly, these mice display lossof endothelial fenestrae and increased numbers of endo-thelial caveolae, suggesting that caveolae attempt to com-pensate for the loss of secretory capacity that is broughtabout by the “tightening” of the remaining blood vessels(29). It is somewhat surprising that these mice do not dis-play more severe glucose homeostasis defects given thesevere reduction of islet vasculature. Additionally, no lossof islet cell area was noted despite the evidence linking

endothelial cell cues to islet proliferation, as describedabove in Section III. B.

Several studies directly examined the role of VEGF-Ain adult �-cell function. Because Pdx1 is expressed athigh levels throughout the developing pancreas and atlower levels in adult �-cells, these studies used rat insulinpromoter (RIP)-driven CRE to conditionally knock outVEGF-A specifically in �-cells. Similar to the PDX1-CREknockout mice, the RIP-Cre VEGF-A knockouts dis-played impaired glucose tolerance and defective in vivoinsulin secretion (30, 31). A role for VEGF-A in the adultislet has also been directly tested: tamoxifen-induciblePDX-1-CRE VEGF-A knockouts show glucose intoler-ance and reduced vessel density, although the effects ofVEGF-A loss are less severe than when VEGF-A is absentfrom the beginning of pancreatic development (32). Thesestudies suggest that loss of islet endothelial cells can neg-atively affect insulin secretion and contribute to glucoseintolerance.

Not surprisingly, loss of islet endothelial cells has amore severe effect in obese animals. RIP-CRE-mediateddeletion of VEGF-A in high-fat diet-fed animals leads tomore severe glucose intolerance and reduction in insulinsecretion than in lean knockouts (33). Interestingly, thesehigh-fat diet-fed mice are able to normally expand their�-cell mass, suggesting that normal islet endothelial celldensity is dispensable for compensatory islet hyperplasiain response to obesity. Two potential explanations for thepreserved ability of these mice to expand their �-cell masscould be residual VEGF-A due to inefficiency of the Crepromoter, which can vary with the distance between loxPsites (34), or the presence of other VEGF isoforms in islets(35). Because Cre is usually not expressed in 100% of itstarget cells, the remaining cells could produce enoughVEGF-A to explain the continued presence of some isletendothelial cells. These remaining endothelial cells couldbe sufficient to promote �-cell mass expansion.

In addition to VEGF-A, several other islet proteins havebeen demonstrated to play a role in adult islet vascular-ization and function. Fyn-related kinase, when overex-pressed under control of the RIP, causes reduced in vivoinsulin secretion and mild glucose intolerance, which ap-pears to be due to reduced islet blood flow and abnormalcapillary morphology (36). Similarly, �-cell or pancreas-specific deletion of von Hippel-Lindau factor, the proteinthat controls the degradation of hypoxia-induciblefactor-1� (HIF-1�), leads to glucose intolerance and im-paired insulin secretion, which is mediated through anincrease in Hif-1� expression (37). This study demon-strates that a direct effect of vessel loss (i.e., increasedhypoxia) is able to mediate the same effect as actual vesselloss. This might seem somewhat paradoxical, because


HIF-1� directly up-regulates Vegf-a expression and thuswould be expected to increase islet vascularization (38).However, HIF-1� also directly alters the expression ofgenes involved in �-cell function in these mice, reducingthe capacity of �-cells to mediate glucose uptake, glucosemetabolism, and insulin secretion (37). Conversely, micein which the antiangiogenic factor thrombospondin-1 isknocked out have enlarged hypervascular islets (39).Taken together, these data suggest that decreased islet vas-cularization or blood flow can have a deleterious effect onislet insulin secretion and whole-body glucose tolerance.Conversely, increased islet vascularization has the oppo-site effect.

In addition to the data gleaned from various knockoutmice, several animal models of diabetes suggest a closerelationship between islet vascularization and insulin se-cretion. For example, in Zucker fatty (ZF) and Zuckerdiabetic fatty rats, islets become more vascularized duringthe obesity-induced expansion of islet mass (40). Interest-ingly, as diabetes and loss of �-cell mass ensue in theZucker diabetic fatty rats, islet vasculature decreases (40).A similar association between loss of islet capillary densityand progression of diabetes is observed in the Otsuka-Long-Evans-Tokushima fatty rat model (41). These dataprovide further evidence that islet mass and the amount ofislet vasculature are critically linked. Similarly, islet bloodflow and blood pressure are elevated in nonobese diabeticGK rats, obese Zucker rats, obese Wistar rats, and GK-Wistar F1 hybrid rats (42–44). Islet blood flow is similarlyincreased in 1-month-old ob/ob vs. lean B6 mice during aperiod of hyperglycemia, hyperinsulinemia, and �-cell ex-pansion (45). Islet blood flow was normalized in 6- to7-month-old ob/ob mice, suggesting that increased isletblood flow is important during expansion of �-cell mass inresponse to hyperglycemia. A number of islet capillarychanges were noted in adult db/db mice, including loss ofislet capillaries, increased capillary diameter, and pericytehypertrophy (46). Together, these data suggest that anincrease in islet vascularization and blood flow accompa-nies compensatory �-cell mass expansion in response tohyperglycemia. When �-cells are no longer able to expandand decompensation occurs, loss of islet vasculature alsooccurs. Interestingly, insulin-deficient mice have more is-let capillaries and bigger islets (47), which could indicatethat insulin inhibits islet vascularization. Alternatively,this effect may be a compensatory response whereby isletvasculature and islet size increase in response to insuffi-cient levels of insulin secretion.

Islet endothelial cells have recently been shown to playa critical role in producing islet basement membrane.�-Cells appear incapable of forming their own basementmembrane (48), and deletion of VEGF-A under the Pdx1

promoter results in a loss of islet, but not acinar tissuebasement membrane (49). Vascular endothelium-pro-duced laminin-411 and laminin-511 appear to be criticalfor insulin gene expression and �-cell proliferation in a�1-integrin-dependent manner (49). Similarly, treatmentof islets with endothelium-conditioned culture mediumincreases glucose-stimulated insulin secretion and islet in-sulin content, an effect that is blocked by addition of aneutralizing antibody to the �1-chain of laminin (50). Pu-rified islet endothelial cells can also stimulate �-cell pro-liferation through secretion of hepatocyte growth factor(51). These results might shed some light on the describedlink between islet hyperplasia and increased islet vascu-larization. Hyperglycemia could trigger VEGF-A secre-tion from islet endocrine cells. This in turn stimulatesvessel growth, basement membrane production, and he-patocyte growth factor production and secretion, whichultimately leads to increased insulin production and �-cellproliferation. Some caution is necessary in applying thesefindings to human islets, however, because it was recentlydemonstrated that blood vessels in human islets are sur-rounded by a unique double basement membrane (52).

D. The role of islet revascularization duringislet transplantation

One area in which the process of islet vascularization isthought to be especially important is during islet trans-plantation. The success of islet transplantation can criti-cally hinge on the ability of transplanted islets to establishfunctional vasculature (53, 54). In addition to immuno-rejection of newly transplanted islets, islet survival hasproved a major challenge to the success of this procedure(55). Isolation of islets for transplantation damages isletendothelial cells and obviously involves the severing ofnative vasculature (56–58). Thus, newly transplanted is-lets must reestablish a functional vascular network, a pro-cess that is believed to involve angiogenesis and possiblyvasculogenesis (57). Not surprisingly, newly transplantedislets are less vascularized immediately after transplant,and even after several weeks they remain less vascularizedand maintain a lower oxygen tension than native pancre-atic islets (59, 60). Islet death and apoptosis are commonproblems after transplant, with islet cell apoptosis increas-ing and �-cell mass decreasing 1–3 d after transplant (61,62). Given the important connection between islet endo-crine cells and islet endothelial cells described in SectionIII. C, it is not hard to imagine that lack of a functionalvasculature might play a direct role in this increased isletdeath. Additionally, the resultant hypoxia and/or isch-emia could also strongly contribute to increased islet ap-optosis. In support of this, Hif-1� expression is increasedin newly transplanted islets, and suppression of Hif-1� intransplanted islets reduces �-cell death (63). Once again,


this might seem paradoxical, due to the ability of HIF-1�

to increase Vegf-a expression (38), but HIF-1� also stim-ulates many other responses, including hypoxia-inducedgrowth arrest and apoptosis, both of which would nega-tively impact the survival of transplanted islets (64).

Endothelial cells from the transplant recipient wereoriginally thought to be the major contributor to post-transplantation vessel formation, but recent evidence sug-gests that donor endothelial cells also play an importantrole in islet revascularization. By using lacZ- or GFP-tagged donor endothelial cells, it has been demonstratedthat after transplantation, islet vasculature is chimeric andis composed of host and donor endothelial cells, both ofwhich contribute to functional blood vessels (65–67).These data suggest that factors that stimulate angiogenesisin isolated islets during culture and transplantation mighthave a positive impact on the success of the transplanta-tion process.

Considerable effort is being devoted to discover theeffects of proangiogenic factors on the success of revas-cularization and survival of newly transplanted islets. Forexample, Vegf-a overexpression in transplanted mouse is-lets has been shown to cause increased vascularization ofand increased blood flow to newly transplanted islets, re-sulting in increased islet insulin content, improved recip-ient glucose tolerance, and increased �-cell survival (68–70). Similarly, overexpression of the proangiogenic growthfactor angiopoeitin-1 (Ang-1), which is normally pro-duced by islets (30), in transplanted islets leads to in-creased glucose tolerance, glucose-stimulated insulinsecretion, islet vascular density, and islet survival (71).Interestingly, Ang-1 can stimulate pericyte migration toendothelial cells, so these studies may suggest a role forpericytes in newly transplanted islets and revasculariza-tion. Pericytes are well-described to influence endothelialcell maturation and proliferation, so proper pericyte cov-erage could be important in this regard. Conversely,blockade of the angiogenesis inhibitor thrombospondin-1in transplanted islets leads to increased vascularizationand improved glucose-stimulated insulin secretion (72).Prolactin overexpression in newly transplanted islets hassimilar beneficial effects on islet vascularization and func-tionality (73). Finally, coculture of human endothelialcells and mesenchymal stem cells (MSCs) with isolatedislets increases islet angiogenesis and suggests a potentialmethod for increasing vascularization of newly trans-planted islets (74). Each of these studies demonstrates theutility of promoting efficient islet vascularization aftertransplantation and demonstrates beneficial effects on�-cell survival and function. Therefore, the elucidation ofadditional factors that are critical for islet vascularizationmay contribute greatly to the treatment of diabetes.

One fascinating new model for studying the revas-cularization of newly transplanted islets was recentlydescribed. Isolated islets were transplanted onto the ret-ina of nude mice, and this procedure was able to effi-ciently normalize streptozotocin-induced hyperglycemia(75). Incontrast toothersitesof transplant, retinal islet trans-plant allows for real-time in vivo monitoring of islet revas-cularization and could provide a useful model system forfuture studies of islet vascularization during transplantation.

IV. Peripheral Vasculature and Insulin Delivery

A. Introduction to peripheral vasculature andinsulin delivery

Insulin signaling in skeletal muscle is critical in glucosedisposal, accounting for almost 90% of whole-body glu-cose disposal in humans (76). This is compared with ad-ipose tissue, which is estimated to account for less than 1%of whole-body glucose disposal in humans and rodents(77, 78). The role of the muscle vasculature is beginning tobe appreciated as a factor that influences the myocyte’sresponse to insulin. Loss of skeletal muscle capillary den-sity is observed in both insulin resistance and in type 2diabetes, and insulin action is positively correlated withcapillary density (79). In cultured myocytes, insulin acti-vates insulin receptor and insulin receptor substrate (IRS)proteins and stimulates glucose uptake in a matter of min-utes (80–82). This suggests that when insulin is present atthe myocyte cell surface, it is able to act almost instanta-neously. Although it is tempting to assume that insulin canact as swiftly in vivo, results from several groups suggestthat this is not the case. Measurements of insulin in lymph,which is derived from interstitial fluid, suggest that lymphinsulin concentrations are reduced for up to 3 h during aninsulin infusion compared with plasma insulin levels (83).Additionally, muscle glucose utilization correlates muchmore strongly with lymph than with plasma insulin levels(83). These data suggest that there is a delay between theappearance of insulin in the bloodstream and its appear-ance in the muscle interstitium. As is the case in culturedmyocytes, insulin acts very quickly upon appearance in theinterstitium. In support of this latter statement, direct in-jection of insulin into the muscle interstitium triggers mus-cle glucose uptake within minutes and circumvents thedelay associated with iv insulin delivery (84).

Recently, the delivery of insulin to myocytes has comeunder increased scrutiny for many of the reasons men-tioned above. Wang et al. (85) demonstrated that after 10min of fluorescein isothiocyanate (FITC)-insulin infusion,the majority of FITC-insulin in muscle is localized withinendothelial cells, suggesting that insulin is rapidly trans-ported from the bloodstream into endothelial cells. Inter-


estingly, after 1 h, most FITC-insulin is still concentratedin muscle endothelial cells, suggesting that transport fromendothelial cells to the muscle interstitium is a slow pro-cess (85). Several studies suggest that insulin is only de-graded at low levels by endothelial cells and that musclelymph flow is relatively slow compared with other tissues,suggesting that any clearance of insulin from the muscleinterstitium is likely due to uptake by myocytes (86–88).These results suggest that the transendothelial transport ofinsulin is a rate-limiting step in muscle insulin action. Insupport of this conclusion, direct injection of insulin intomuscle lymph causes very rapid insulin action, suggestingthat direct injection sidesteps the rate-limiting step (84).Interestingly, although the interstitial passage of insulindoes not appear to be a rate-limiting step under normalconditions, injection of insulin into lymph of ob/ob miceshows a delayed action as compared with lean mice, sug-gesting that the movement of insulin within the muscleinterstitium might be impaired under conditions of insulinresistance (89).

The heterogeneity of the body’s vascular beds providesfurther support for a muscle-specific delay in insulin ac-tion. As mentioned above, the liver vasculature is highlypermeable (5), and in fact, insulin-stimulated inhibition ofhepatic glucose output occurs much more quickly thaninsulin-stimulated muscle glucose disposal (90). Althoughthe more rapid effect of insulin on liver has not been di-rectly linked to increased delivery of insulin, the higherpermeability of liver vs. muscle blood vessels suggests thatthis is a likely cause. Interestingly, inhibition of hepaticglucose output occurs at even low levels of insulin infu-sion, suggesting that the highly permeable liver vascula-ture does not impede the transport of insulin as is the casein muscle (91). In support of this, the concentration oflymph insulin required to stimulate muscle glucose uptakeis similar to the plasma insulin concentration required toinhibit hepatic glucose output (3, 91).

B. Transendothelial transport of insulinThe mechanism for insulin transport across the endo-

thelium is not clear. Several in vitro studies suggest that theendothelial uptake of insulin is mediated by the insulinreceptor (86, 92, 93). However, the study of a transendo-thelial transport process in vitro is complicated by the useof cultured cells or the severing of native blood vessels.Therefore, the in vivo setting seems to be a more reliableplace to study the vasculature. Indeed, in vivo results sug-gest that the transport of insulin across the endotheliumdoes not appear to be saturable, even at pharmacologicalconcentrations of insulin, suggesting that it is not a recep-tor-mediated process (94, 95). At a minimum, these dataseem to suggest that at least at high levels of insulin, trans-port can occur by a non-insulin receptor-mediated trans-

cellular or paracellular process. In some respects, a para-cellular transport process would be somewhat surprisinggiven the tightness of muscle endothelium (5, 6, 12).

Recent research has focused on possible mechanismsfor insulin transport across endothelial cells. In vitro stud-ies in cultured endothelial cells support an insulin recep-tor-mediated pathway for insulin uptake because block-ade of several insulin-signaling pathways inhibits thisprocess (96). Confocal imaging studies suggest colocal-ization of FITC-insulin with the insulin receptor in muscleendothelial cells (85). These proteins also colocalize withcaveolin-1, a protein involved in the formation of caveo-lae. Interestingly, caveolae are also increased in �-cell-spe-cific VEGF-A knockouts, which display a dramatic loss ofislet endothelial fenestrae, as mentioned in Section III. C.In this situation, caveolae are thought to mediate the trans-endothelial transport of insulin from �-cells to the blood-stream to compensate for the loss of endothelial cellfenestrae, so there is some precedent for caveolae-medi-ated transport of insulin across endothelial cells. Althoughthe cellular colocalization of labeled insulin, insulin re-ceptor, and caveloae is intriguing, further studies will benecessary to provide a conclusive link between transen-dothelial insulin transport in muscle and caveolae. Inter-estingly, transendothelial transport of insulin has beendirectly demonstrated by using 125I-labeled insulin inheart muscle (97).

C. Effects of insulin on blood flowInsulin has been demonstrated to cause a number of

direct effects on the vasculature. Although somewhat con-troversial, insulin appears to increase total blood flowwithin skeletal muscle. Baron (98) was the first to theorizethat the ability of insulin to increase limb/muscle bloodflow might be a critical part of its delivery and its abilityto stimulate glucose uptake. Although this has been sup-ported by a number of studies in both normal and obese orinsulin-resistant situations (99–111), there are also anumber of studies that failed to show an insulin-inducedincrease in blood flow, especially at physiological levels ofinsulin (112–119). This has led to some disagreement inthe field as to whether insulin can increase blood flow atphysiological levels (120, 121). There are a number oftheories to explain why an insulin-stimulated increase inblood flow is not always observed. Barrett et al. (3) arguethat pharmacological doses of insulin clearly act to in-crease limb blood flow, but the necessity for large doses ofinsulin and the long durations required to observe an effectcast doubt on the normal physiological relevance of thiseffect. Clark (4) and Barrett et al. (3) both suggest thatalthough there is not always an observed increase in totalblood flow, insulin can influence which vessels in muscleare perfused without changing total blood flow. Specifi-


cally, this switch involves the ability of insulin to increasevasomotion, the relaxationof terminal arterioles, and shiftblood flow from a nonnutritive pathway, i.e., one that haslittle or no exchange with the interstitial fluid surroundingthe muscle fibers, to a nutritive pathway, i.e., one whereblood flows through capillaries that allow for nutrientexchange with the muscle interstitium that is underper-fused in the basal state (4). This insulin-induced shift pro-vides a greater surface area for nutrient exchange and maybe the primary vascular effect caused by insulin (4).

D. Insulin-induced capillary recruitmentThe ability of insulin to stimulate relaxation of terminal

arteries and promote flow through the nutritive pathwayis termed capillary or microvascular recruitment and, incontrast to insulin-induced effects on total blood flow,seems to be gaining general acceptance (3, 4). In normalmuscle tissue, only about two thirds of the vasculature isnormally perfused with erythrocytes at any given time(122), although other studies have demonstrated by liveimaging that nearly all capillaries in the spinotrapeziusmuscle and diaphragm are perfused at rest (123). Poten-tially, differences in flow rates and the extent of perfusionbetween different muscle capillary beds could explain thisapparent inconsistency. By shifting blood flow from non-nutritive vessels to nutritive capillaries, insulin is able toincrease the available surface area for nutrient exchangeand theoretically increase the surface area for its own de-livery and that of glucose to myocytes (3) (Fig. 1, A and B;and a color version in Supplemental Fig. 1, published assupplemental data on The Endocrine Society’s JournalsOnline web site at http://edrv.endojournals.org).

Two notable experimental advances greatly aided theelucidation of insulin’s ability to promote capillary re-cruitment: 1-methylxanthine (1-MX) metabolism (124),and contrast-enhanced ultrasound (CEU) (125). In short,1-MX is a substrate for the endothelial cell enzyme xan-thine oxidase, which is found only in smaller arterioles andcapillaries and catalyzes the conversion of 1-MX to1-methylurate (124). Disappearance of 1-MX from bloodsuggests an increase in available endothelial cell area thatallows for 1-MX entry (124). Insulin administration sig-nificantly increases 1-MX metabolism, suggesting a dila-tion-mediated increase in endothelial cell area (124). CEUuses gas-filled microbubbles as a contrast agent and a sur-rogate for erythrocytes to measure blood flow and hasdemonstrated insulin-induced increases of muscle capil-lary volume without changes in total blood flow (125).Importantly, both 1-MX metabolism and CEU have dem-onstrated the ability of physiological levels of insulin toinduce capillary recruitment (124–127). Insulin-inducedcapillary recruitment has also been investigated in skin,which has the advantage of being much easier to access

and monitor (3, 4, 128). These results are generally con-sistent with those in muscle and have been reviewed indetail elsewhere (3, 4).

E. Molecular mechanism of capillary recruitmentInsulin stimulates its vasodilatory actions associated

with capillary recruitment through up-regulation of en-dothelial nitric oxide synthase (eNOS) in muscle endothe-lial cells (129–132) (Fig. 2A, and a color version in Sup-plemental Fig. 2). eNOS production is induced by insulinreceptor activation of the phosphatidylinositol-3-kinase(PI3K) signaling cascade, which ultimately increaseseNOS expression and subsequent nitric oxide (NO) pro-duction for vasodilation (129, 131–133). Somewhat par-adoxically, insulin can also stimulate vasoconstriction byactivating endothelin via the ERK 1/2 pathway (131, 134)(Fig. 2B). Therefore, it seems that in the normal case, vesselrelaxation and constriction are tightly controlled by an

A Normal Myocytes

Myocytes

D Pericyte Loss Myocytes

Myocytes

C InsulinResistance Myocytes

Myocytes

B Insulin Induced Vasodilation

Myocytes

Myocytes

FIG. 1. Transendothelial transport of insulin and glucose to muscleinterstitium. A, In the normal state, insulin and glucose must travelfrom the blood stream across endothelial cells and potentially pericytesto reach the muscle interstitium and activate insulin signaling inmyocytes. The basal flow rate is depicted by the horizontal arrow. B,Insulin induces vasodilation of muscle vasculature, which is thought toincrease nutritive blood flow, which increases transport of insulin andglucose to the muscle interstitium. C, In insulin-resistant situations,insulin exerts only vasoconstriction signals, which decreases nutritiveflow, leading to reduced glucose and insulin transendothelialtransport. D, Loss of pericytes is thought to increase muscle vesselpermeability and transendothelial transport. A color version of thisfigure is available as Supplemental Figure 1.


organism’s metabolic state, and this suggests that pertur-bations to this balance might affect delivery of nutrientsand insulin itself to myocytes. Indeed, vessels from obeseZF rats display reduced levels of eNOS protein comparedwith lean rats, and insulin treatment stimulates only va-soconstriction, which presumably contributes to the de-velopment of insulin resistance (135). Similarly, blockadeof endothelin-1 receptors in obese humans results in sig-nificant vasodilation, but a similar effect is not seen in leanhumans (136). A shift toward insulin-induced vasocon-striction is also observed by pharmacologically inhibitingeNOS with NO inhibitors, confirming the NO indepen-dence of this vasoconstriction (134, 135). As might bepredicted, eNOS knockout mice display insulin resistance(137). These data suggest that under conditions of insulinresistance, the balance of insulin action shifts toward va-soconstriction. This further exacerbates insulin resistanceby reducing the access of insulin and nutrients to myocytesby decreasing nutritive flow and available capillary sur-face area.

F. Insulin resistance and muscle vasculatureAs is the case in pancreatic islets, muscle vascular de-

fects worsen with insulin resistance and type 2 diabetes.

Numerous studies demonstrate that the insulin-inducedincrease in capillary recruitment is blunted in cases of in-sulin resistance and obesity (Fig. 1C). For example, insu-lin-resistant ZF rats show reduced capillary recruitment(138). Similarly, insulin resistance induced by variousagents such as TNF-�, intralipid/heparin, glucosamine, or�-methylserotonin also reduces capillary recruitment(139–142). Importantly, the extent of muscle capillaryrecruitment is positively correlated with glucose uptake(140, 141), suggesting a possible causal relationship be-tween these two processes. Human type 2 diabetic patientsdemonstrate a reduced muscle capillary permeability-sur-face area in response to hyperinsulinemia, suggesting adefect in insulin-induced capillary recruitment (143). Alikely mechanism for these observations, as mentioned inSection IV. E, is the insulin resistance-induced decrease ofeNOS activation, which leads to reduced vasodilation.These data suggest that muscle vascular dysfunction ac-companies insulin resistance.

Whether insulin resistance causes vascular dysfunctionor vice versa is not entirely clear. Evidence suggests thatmuscle vascular defects are among the earliest phenotypesobserved as insulin resistance progresses and that dimin-

Insulin

eNOS

GuanylylCyclase

G-Protein

Myosin

PI3K

Akt

[cGMP]

[Ca2+]

PKG

PLCPKCIP3

DAG

↑

↓

MyosinLight-Chain

Phosphatase

KATPChannels

NitricOxide

MyosinKinase

MyosinKinase

InsulinResistance

InsulinResistance

Myosin

Vasorelaxation Vasoconstriction

[Ca2+]↑

KATPChannels

EndothelialCell

A B

Pericyte/SmoothMuscleCell

IRS-1

ERK1/2

Endothelin-1

Endothelin-1

EndothelialCell

Pericyte/SmoothMuscleCell

IRIR

Myosin

Myosin P

ETA /ETB

P

P

IRS-1/2P

Insulin

MyosinLight-Chain

Phosphatase

NitricOxide

NitricOxide

P

P

P

FIG. 2. Insulin-induced vasodilation and vasoconstriction signaling in endothelial cells and pericytes. A, Insulin signaling in endothelial cells resultsin increased NO production and secretion, which stimulates the dephosphorylation of myosin in pericytes and vSMCs and results in vasorelaxation.B, Conversely, insulin also stimulates endothelin-1 production, which stimulates myosin phosphorylation and results in vasoconstriction. DAG,Diacylglycerol; IR, insulin receptor; IP3, insositol tris-phosphate; P, phosphate; PLC, phospholipase C, ETA/ETB, endothelin receptor.


ished NO production and increased vasoconstriction pre-cede the development of type 2 diabetes (144). Due to theimportance of insulin signaling in capillary recruitment, itwould not be surprising if reduced eNOS activation wasmediated by the same mechanism as desensitization ofmyocyte insulin signaling. Insulin resistance in myocyteshas been demonstrated to occur downstream of insulinreceptor binding, likely at the level of the IRS proteins(145, 146). One can imagine that the same desensitizationof insulin signaling could blunt the PI3K pathway inendothelial cells and subsequent eNOS activation, thusleading to reduced vascular dilation. The subsequent re-duction in available capillary surface area would likelyreduce transendothelial insulin transport to myocytes andfurther blunt myocyte insulin signaling, continuing a vi-cious cycle. As the potential involvement of the insulinreceptor in transendothelial transport is further eluci-dated, defects in this process might also become importantfor the development of insulin resistance.

Several animal models are informative regarding theeffects of insulin on muscle vasculature. Prominent amongthese models are the vascular endothelial cell insulin re-ceptor knockout (VENIRKO) mice, which somewhat sur-prisingly, are not insulin resistant (147). An initial reactionto this result might be that insulin-mediated vasodilationis not a critical process for muscle glucose disposal. Thissuggests that insulin delivery to the periphery is possiblewithout the endothelial cell insulin receptor under normalcircumstances. However, both eNOS and endothelin-1mRNA levels are reduced in the VENIRKO mice (147),suggesting that the lack of insulin resistance in these micemay not be surprising because the signals for both vaso-constriction and vasodilation are reduced, resulting in nonet changes in vasomotion. When considering these re-sults, it is important to note that in lean mice, creatinginsulin resistance has historically been somewhat compli-cated because even muscle-specific knockout of the insulinreceptor (MIRKO mice), one of the most critical proteinsinvolved in muscle insulin signaling, results in normal glu-cose tolerance and insulin sensitivity (148). Finally, theinsulin-like growth factor-1 receptor, which is found athigh levels in muscle endothelial cells (149), could alsocompensate for loss of the insulin receptor. Additionally,as mentioned in Section IV. E, eNOS knockout mice arealso insulin resistant (137). No studies linking knockout ofendothelin-1 to amelioration of insulin resistance havebeen reported, but this result might be predicted.

G. Exercise-induced vascular changesExercise has been demonstrated to promote many of

the same changes in the muscle vasculature as insulin(150). Exercise efficiently increases muscle capillary re-cruitment, as well as total muscle blood flow (151). Inter-

estingly, the mechanism for insulin- and exercise-inducedincreases in muscle perfusion appear distinct because theexercise mechanism is NO-independent (152). Exercise-induced vascular changes are not blunted by insulin resis-tance or obesity (153). These data suggest that even incases of prolonged disease progression, exercise might beable to remedy some of the effects of insulin resistance onthe muscle vasculature.

V. Vascular Pericytes: More Than InertContractile Cells

A. Introduction to pericytesAlthough pericytes were first described almost 150 yr

ago, they are still not entirely understood. Charles Rougetwas the first to identify pericytes in 1873 when he de-scribed a population of perivascular cells that he regardedas contractile elements (154). In 1923, Zimmermannnamed these cells “Rouget cells” after their discoverer, or“pericytes” because of their location in proximity to en-dothelial cells (peri, around; cyte, cell) (155). Together,pericytes and the related vSMCs make up a class of cells,termed mural cells, which provide support to blood vesselsof all sizes (156). By convention, the mural cells associatedwith larger vessels such as arteries and veins are calledvSMCs, whereas those associated with smaller vessels,such as capillaries, arterioles, and venules, are termed peri-cytes (157). vSMCs are highly contractile, typically ex-press high levels of �-smooth muscle actin (�-SMA), andcan form multiple concentric layers around blood vessels(156). Additionally, vSMCs have their own basementmembrane, which is rich in elastin and fibrillar collagenand is separate from the vascular basement membrane(156, 158). Pericytes, on the other hand, typically form asingle discontinuous layer around smaller vessels (157).They are intimately associated with endothelial cells andreside in a shared basement membrane that is produced byboth pericytes and endothelial cells (159). Pericytes typi-cally express high levels of cell-surface chondroitin sulfateproteoglycan neuron-glial 2 (NG-2), which is commonlyused as a histological pericyte marker (155).

Pericyte coverage and morphology can vary between dif-ferentvascularbeds.Pericytes incertaintissueshaveacquiredspecialized names, such as Ito cells in liver (named after theirdiscoverer, Toshio Ito) and mesangial cells in the kidney glo-merulus (155, 160). The highest level of pericyte coverage isseen in the central nervous system, where pericytes arethought toplayan important role in the formationandmain-tenance of the blood-brain barrier (161, 162).

Although pericytes were originally thought to be solelyinvolved in contractile processes, they have recently beenshown to have many additional functions. Pericytes, like


vSMCs, can indeed mediate vessel contraction and influ-ence vascular diameter and capillary blood flow (163).Loss of pericytes leads to excessive endothelial membranefolding and luminal cytoplasmic protrusions, which likelyimpacts vessel perfusion and possibly nutrient exchange(164). Pericytes can communicate with endothelial cellsthrough a variety of signaling pathways, and they promoteendothelial cell differentiation and maturation (165). Peri-cytes are also thought to limit endothelial cell prolifera-tion; thus, loss of pericytes can lead to endothelial cellhyperplasia (164). Pericytes also play a critical role in an-giogenesis, because they are typically found near the tipsof sprouting vessels and where they are thought to guidethe sprouting processes by expressing VEGF (166–170).

B. Platelet-derived growth factor-B: a key mediator ofpericyte function

Platelet-derived growth factor B (PDGF-B) and its re-ceptor, PDGF receptor-� (PDGFR�), are critically in-volved in pericyte recruitment and proliferation (157).PDGF-B belongs to the PDGF family of proteins, whichshares structural homology with the VEGF protein family(171, 172). In addition to PDGF-B, there are three othermammalian PDGF family members, PDGF-A, PDGF-C,and PDGF-D (172). PDGF was originally identified as aconstituent of whole blood that was absent in cell-freeplasma (173–175) and was later purified from humanplatelets (176–179). The PDGF family members functionas homo- and heterodimers (180). PDGF-B is normallyexpressed in vascular endothelial cells, megakaryocytes,and neurons, whereas PDGF-A and PDGF-C are highlyexpressed in epithelial cells, muscle, and neuronal progen-itor cells (172).

There are two PDGF receptors, PDGFR� and PDGFR�.Ligand binding causes homo- or heterodimerization ofPDGFRs, which activates their cytoplasmic tyrosine ki-nase domains and results in autophosphorylation of sev-eral tyrosine residues in their cytoplasmic tails (181). Thephospho-tyrosine residues create docking sites for a vari-ety of adaptor proteins, leading to activation of a widevariety of signaling pathways, including the RAS-MAPK,JNK/SAPK, PI3K/AKT, and protein kinase C (PKC) path-ways (182, 183). Interestingly, PDGFR signaling throughdifferent pathways appears to be additive rather thanproducing distinct outcomes. In a tour de force study,Tallquist et al. (184) created an allelic series of tyrosine tophenylalanine mutations, which resulted in blunted butqualitatively similar developmental effects. Ligand occu-pancy promotes internalization and subsequent lysosomaldegradation of PDGFR complexes, limiting the durationof signaling (185–187).

PDGFR� binds both PDGF-A and PDGF-B with highaffinity, whereas PDGFR� binds only PDGF-B with high

affinity (180). However, the receptor complex thought tomediate in vivo PDGF-B signaling is the PDGFR� ho-modimer (157, 158, 172). In addition, outside of humanplatelets, the expression patterns of PDGF-A and PDGF-Bare generally nonoverlapping, suggesting that the PDGF-ABheterodimer, which can bind to both PDGFRs, might havelimited in vivo significance (180, 183).

PDGF-A and PDGF-B are the best described PDGF li-gands. Although PDGF-A is involved in a variety of di-verse developmental and organogenesis processes (172),the major function of PDGF-B is its role in stimulating themigration of pericytes and vSMCs to growing blood ves-sels (172). PDGF-B is secreted as a homodimer from en-dothelial cells, where it is retained on the cell surface by aC-terminal heparan-sulfate proteoglycan-binding reten-tion motif (188, 189). PDGFR� on the surface of pericytesand vSMCs binds to PDGF-B, thus recruiting mural cellsto the endothelial cell wall. Loss of the retention motifleads to decreased retention of PDGF-B on the cell surface(188–190). Lindblom et al. (191) have generated andcharacterized mice in which the PDGF-B retention motifhas been deleted by targeted mutagenesis, causing a re-duction in PDGF-B signaling. These mice display a loss ofpericyte density, pericyte detachment, and abnormal cap-illary morphology in the developing brain, kidney, andretina, as well as the postnatal kidney and retina (191).PDGF-B and PDGFR� knockout mice display similar, butmore severe defects: a drastic loss of pericytes, widespreadvascular leakage, general heart defects, and perinatal le-thality (164, 192–195).

Although PDGF-B and PDGFR� are critically importantfor pericyte recruitment, the presence of pericytes in someorgans of PDGF-B/PDGFR� knockouts suggests that otherfactors are also important for pericyte biology. Notably,TGF-� is an important mediator of vSMC/pericyte differen-tiation (196, 197). Additionally, Ang-1 and its receptor, en-dothelium-specific receptor tyrosine kinase-2, are importantfor signaling from pericytes/vSMCs to endothelial cells andplay a role in pericyte recruitment (157).

Although the importance of PDGF-B signaling andpericyte coverage in the microvasculature has beenwell described, little has been reported regarding anyinfluence pericytes might have on insulin secretionand peripheral insulin action. Pericytes are found atrelatively high densities in skeletal and cardiac muscle(198, 199) and in pancreatic islets (200). PDGF-B sig-naling can promote �-cell proliferation in vitro, butonly under conditions of PDGFR� ectopic overexpres-sion; endogenous PDGFR� expression in �-cells is quitelow (201). Therefore, PDGF-B does not directly signal to�-cells.


C. Diabetic complications: a key role for pericytesPericytes play a central role in diabetic complica-

tions. Loss of pericytes is one of the first observablechanges in diabetic retinopathy and is ultimately fol-lowed by increased vascular permeability (202). Inter-estingly, Geraldes et al. (203) recently demonstrated thatglucose increases the expression of PKC-�, which results indown-regulation of PDGFR�/AKT survival signals andpericyte death. The vasculature also plays a clear role inthe progression of nephropathy (204). Knockout ofPDGF-B/PDGFR� or loss of PDGF-B retention results indefects in the pericyte-like mesangial cells, leading to de-fective glomerulogenesis, glomerulosclerosis (191, 194,195), and proteinurea (191), a hallmark of diabeticnephropathy. Diabetic neuropathy is characterized by re-duced blood flow, capillary basement membrane thicken-ing, and pericyte degeneration (205). Tilton et al. (206)used transmission electron microscopy to study pericytemorphology and density in a variety of skeletal musclesfrom nondiabetic and diabetic humans. Similar to what isobserved in diabetic retinopathy, they noted an increaseddegeneration of pericytes in the type 2 diabetic muscles.Pericyte changes are therefore associated with diabetesand are found in most of the microvascular complicationsassociated with diabetes.

D. Are pericytes multipotent progenitor cells?One exciting new field of pericytes involves their po-

tential role as MSC-like progenitor cells. MSCs, alsoknown as multipotent mesenchymal stromal cells, are un-differentiated, self-renewable cells that are present in bonemarrow and mesenchymal tissues (207). Interest in MSCsin relation to diabetes intensified when it was reported thattransplanted bone marrow MSCs initiated pancreas re-generation and improved diabetes in mice and humans(208–210). However, other studies did not demonstrateany evidence for transdifferentiation of bone marrow cellsinto �-cells (211, 212), so this process could be dependenton the stage of diabetes progression and/or the isolationand transplantation techniques. Transplantation of puri-fied MSCs was likewise able to normalize hyperglycemiaand promote islet growth in streptozotocin-treated mice(213). Crisan et al. (214) recently demonstrated that peri-cytes isolated from human mesenchymal tissues, includingskeletal muscle, pancreas, and adipose tissue, were able toserve as multilineage progenitor cells reminiscent ofMSCs. Specifically, this study demonstrated that purifiedpericytes from any of these organs, when cultured in atissue-specific growth medium, could differentiate intomyocytes, adipocytes, osteocytes, and chondrocytes(214). Although pericytes isolated from pancreas wereused as progenitor cells, the authors did not report on theability of isolated pericytes to differentiate into �-cells or

any other pancreatic cell type. This presents the intriguingquestion: can pericytes serve as �-cell progenitors? In vivolineage-tracing experiments using a tetracycline-inducibleCRE recombinase under the control of the adipogenic per-oxisome proliferator-activated receptor-� promoter to in-delibly mark cells with �-galactosidase demonstrate thatadipocyte progenitors are perivascular cells that expressseveral pericyte markers, including NG-2, PDGFR�, and�-SMA (215). One can imagine using a similar system witha �-cell-specific CRE recombinase such as Pdx1-CRE todetermine whether islet pericytes can analogously serve asprogenitors to �-cells.

E. Pericytes in normal islet functionThe role of pericytes in normal islet function is not

completely understood, but islet pericyte changes associ-ated with a number of pathological conditions have beendescribed. In obese animals, islet pericytes become morehypertrophied and assume vSMC characteristics (46). No-tably, pericytes take on more of a smooth muscle cell-likeappearance, and it has been speculated that this is poten-tially in response to increased islet blood pressure (46). Weand others have similarly observed an increase in �-SMAand NG-2 staining densities in islets from ob/ob mice,consistent with obesity-induced pericyte hypertrophy(Ref. 216 and Supplemental Fig. 3). Similarly, hyperten-sive Ren2 rat islets have increased pericyte proliferation,migration, hypertrophy, and �-SMA staining (217). Inrats overexpressing human islet amyloid polypeptide,there is a reduction in both �-cell mass and islet capillarydensity, along with increased pericyte apoptosis and loss(218). Finally, in a rat model of type 2 diabetes, the matrixbetween islets and the surrounding exocrine tissue widens,due to an increase in pericytes and inflammatory cells inthis region (219). The authors of these studies noted evi-dence of pericyte differentiation into stellate cells, but theuse of �-SMA as a stellate-cell marker seems somewhatquestionable due to its well-described role as a vSMCmarker (155).

F. Pericytes in islet tumorsThe role of pericytes in islet-cell tumors has been in-

tensively investigated (220–227). PDGFRs are expressedin tumor pericytes, and treatment with a drug that selec-tively inhibits PDGFR� and PDGFR� blocks furthergrowth of end-stage tumors by causing pericyte detach-ment and disrupting tumor vasculature (220). Pericyte lossinduced by treatment with imatinib mesylate improves theefficacy of metronomic chemotherapy by rendering endo-thelial cells more sensitive to the actions of the cytotoxicdrugs (224). Treatment with imatinib mesylate, metro-nomic chemotherapy, and a selective VEGFR inhibitor


elicits regression of solid islet tumors and increases mediansurvival (224). Some of the effects of PDGFR� inhibitionin islet tumors are mediated by elimination of PDGFR�-positive perivascular progenitor cells, which can differen-tiate into mature pericytes (226). These data suggest thatlack of pericytes improves drug delivery to tumor cells andmight be beneficial for treatment.

Although the absence of tumor pericytes might be ben-eficial for drug delivery, it is also associated with increasedtumor metastasis. �-Cell tumors that are deficient for neu-ral cell adhesion molecule have leaky blood vessels withdetached pericytes, which correlates with an increased in-cidence of metastasis (225). In agreement with this,PDGF-B retention-deficient mice with �-cell tumors ex-hibit increased metastasis, demonstrating a direct causallink between defective pericyte recruitment and increasedmetastasis (225). These findings directly translate to hu-mans, because decreased �-SMA-positive pericyte cover-age of tumor vessels correlates with increased metastasisand results in a poorer prognosis (228). Thus, althoughinhibition of PDGFR is an attractive option for improvingtumor treatment, the benefits must be weighed against theincreased metastatic potential associated with reduced tu-mor pericyte coverage.

G. A role for PDGF-B signaling in glucose uptake?PDGFR� has been investigated in relation to insulin

signaling because it can activate several of the same sig-naling pathways as the insulin receptor. Specifically, theability of PDGFR� to activate AKT/PI3K signaling (183)has been investigated in relation to glucose uptake intomyocytes and adipocytes. PDGF-B signaling can induceglucose transporter type 4 translocation in both culturedadipocytes (229) and mouse skeletal muscle (230).However, due to minimal endogenous expression of thereceptor, this is only possible under conditions involvingoverexpression of PDGFR�. These actions of PDGF-B aremediated independently of IRS-1 activation (229, 230).Interestingly, exogenous overexpression of PDGFR� inskeletal muscle has been used to demonstrate that defectsin the insulin signaling pathway independent of IRS-1 canlead to insulin resistance (231). These data suggest thatany effect of PDGF-B on insulin action is unlikely due toa direct effect of PDGF-B on parenchymal cells.

Although PDGF-B does not normally signal directly tomyocytes, we have recently demonstrated that loss ofPDGF-B activity does impact peripheral insulin sensitivity(200). In ob/ob mice, loss of PDGF-B retention causesdecreased in vivo insulin secretion without a change inglucose tolerance (200). These mice have defective peri-cyte coverage in peripheral tissues involved in insulin ac-tion (200). Loss of pericytes can lead to increased vascularleakage (164, 191–195), and indeed these mice display

increased vascular permeability, especially in heart (200).Ultimately, this leads to increased transendothelial trans-port of insulin and increased whole-body insulin sensitiv-ity (200). These data demonstrate an important and novelrole for pericytes and PDGF-B in delivery of insulin toperipheral tissues.

H. Inhibition of PDGFR� and diabetes therapyOne potentially interesting link between PDGF-B ac-

tivity and diabetes concerns the effect of imatinib mesylate(Gleevec). Imatinib mesylate inhibits several receptor ty-rosine kinases, including c-abl, c-kit, and PDGFR� (232).Several studies have reported that imatinib mesylate low-ers fasting blood glucose levels in diabetic patients treatedfor chronic myeloid leukemia (233–235). Another studyreported no effect of imatinib mesylate treatment on glu-cose levels (236). However, Han et al. (237) recently dem-onstrated that treatment of db/db mice with imatinibmesylate drastically improves peripheral insulin sensitiv-ity. The authors observed improved insulin sensitivity anddecreased in vivo insulin secretion in response to a glucosechallenge, but they attributed these effects to an amelio-ration of c-abl-induced endoplasmic reticulum stress inliver and adipose tissues. However, it is also possible thatthe observed increase in insulin sensitivity may also in-volve PDGFR� inhibition. In accordance with this, treat-ment of mice with imatinib mesylate or a soluble form ofPDGFR� both prevents and reverses type 1 diabetes,whereas treatment with a c-kit inhibitor had little effect(238). None of these studies investigated the effect of ima-tinib mesylate treatment on pericyte coverage of islets orperipheral tissues, which could provide a potential mech-anism for the improvement of diabetes. In support of thispossibility, tumor-bearing mice treated with imatinib me-sylate demonstrate a decrease in pericyte coverage andincreased vessel leakiness, suggesting that a similar im-provement of vascular permeability in peripheral tissuescould improve insulin or nutrient delivery and possiblyexplain the effects of imatinib mesylate on diabetes (239).This suggests that further investigation into the effects ofimatinib mesylate and other PDGFR� inhibitors on dia-betes treatment and pericyte coverage may be informative.

I. A role for pericytes in insulin-inducedhemodynamic changes

In addition to the role for pericytes in insulin deliveryto myocytes, pericytes and vSMCs also play an importantrole in insulin-induced capillary recruitment. NO producedby endothelial-expressed eNOS diffuses into vSMCs whereit binds to and activates the heme moiety of guanylyl cy-clase (240, 241) (Fig. 2A). This results in an increase of thelocal concentration of cGMP, which increases cGMP-de-pendent protein kinase G (PKG) signaling (240, 241). This


results in activation of myosin light-chain phosphataseand the opening of KATP channels (240). NO can alsodirectly nitrosylate KATP channels, resulting in hyperpo-larization of the vSMC plasma membrane and inhibitionof calcium entry (240). Each of these NO-mediated effectsresults in vSMC relaxation and concomitant vasodilation.Conversely, endothelin-1 signals through its G protein-coupled receptors on the surface of vSMCs, ETA and ETB,

to activate PKC signaling that results in increased calciumlevels and vascular contraction (242–244) (Fig. 2B). In-terestingly, PKC signaling can be activated by lipidby-products like diacylglycerol and long-chain acyl-coenzyme A, which are increased in obesity (145) and canfurther promote vasoconstriction and exacerbate insulinresistance. Endothelin-1 and NO mediate their vasocon-strictionandvasodilationeffectsonpericytes through sim-ilar mechanisms (163).

Modulation of vSMC or pericyte coverage and its ef-fects on insulin-induced vascular constriction and dilationhave not been directly studied, although one can imaginethat loss of vSMCs or pericytes could reduce the ability ofvessels to respond to insulin-induced changes in vasculartone. Additionally, loss of vSMCs/pericytes might lead togeneral increases in vessel dilation and muscle perfusion,resulting in increased capillary surface area and insulin/nutrient transport to muscle interstitium. Conversely,vSMC/pericyte hyperplasia might result in increased basalvascular contraction. Notably, vSMCs isolated from ZFrats showed reduced PKG activation by NO and cGMP,suggesting that insulin resistance also impedes this stepin insulin-induced vasodilation (245). This defect wasthought to be due to increased levels of superoxide anionsbecause it was rescued by antioxidant treatment (245). Inkeeping with this, NO can be consumed by reactive oxy-gen species, resulting in the production of peroxynitrite(246). This suggests that increased metabolism and in-creased generation of metabolic by-products like reactiveoxygen species could begin to explain a mechanism fordefective eNOS/NO signaling in insulin-resistant animalsand humans.

PDGF-B could play a direct role in controlling vasculardilation and constriction. Like insulin, PDGF-AB and to alesser extent PDGF-A and PDGF-B increase eNOS expres-sion (247). This suggests that, in addition to decreasingpericyte coverage, defects in PDGF signaling could di-rectly contribute to the development of insulin resistanceby reducing eNOS levels and increasing vasconstriction.

VI. Summary/Conclusions

Type 2 diabetes is a growing worldwide epidemic. Theinvolvement of the vasculature in the processes of islet

development, insulin secretion, and peripheral insulinaction is undeniable. Although most recent research hasfocused on endothelial cells, the vascular pericyte is anintriguing candidate to play a role in these processes. Fromtheir involvement in diabetic complications to their func-tion as mediators of insulin-induced vasodilation and va-soconstriction, the available information suggests a rolefor pericytes in the development of insulin resistance andtype 2 diabetes. Future research will be critical to elucidatethe role of the vascular pericyte in these processes.

Acknowledgments

We are grateful to Robin Davies and Laura Vanderploeg for expertassistance in preparing Figs. 1 and 2. We thank William Dove, Jon Odorico,Anath Shalev, and Xin Sun for critical review of the manuscript andhelpful discussion.

Address all correspondence and requests for reprints to: Alan Attie,University of Wisconsin-Madison, 433 Babcock Drive, Madison, Wis-consin 53706. E-mail: [email protected].

This work was supported by National Institute of Diabetes andDigestive and Kidney Diseases Grants DK66369 and DK58037, Na-tional Institutes of Health Training Grant T32GN07215 (to O.C.R.and S.M.R.), and a Wisconsin Alumni Research Foundation fellow-ship (to O.C.R.).

Disclosure Summary: The authors have nothing to disclose.

References

1. Mokdad AH, Bowman BA, Ford ES, Vinicor F, Marks JS,Koplan JP 2001 The continuing epidemics of obesity anddiabetes in the United States. JAMA 286:1195–1200

2. Diamond J 2003 The double puzzle of diabetes. Nature423:599–602

3. Barrett EJ, Eggleston EM, Inyard AC, Wang H, Li G, ChaiW, Liu Z 2009 The vascular actions of insulin control itsdelivery to muscle and regulate the rate-limiting step inskeletal muscle insulin action. Diabetologia 52:752–764

4. Clark MG 2008 Impaired microvascular perfusion: a con-sequence of vascular dysfunction and a potential cause ofinsulin resistance in muscle. Am J Physiol EndocrinolMetab 295:E732–E750

5. Aird WC 2007 Phenotypic heterogeneity of the endothe-lium. I. Structure, function, and mechanisms. Circ Res 100:158–173

6. Aird WC 2007 Phenotypic heterogeneity of the endo-thelium. II. Representative vascular beds. Circ Res 100:174 –190

7. Parton RG, Simons K 2007 The multiple faces of caveolae.Nat Rev Mol Cell Biol 8:185–194

8. Zanone MM, Favaro E, Camussi G 2008 From endothelialto �-cells: insights into pancreatic islet microendothelium.Curr Diabetes Rev 4:1–9

9. Stamatovic SM, Keep RF, Andjelkovic AV 2008 Brain en-dothelial cell-cell junctions: how to “open” the blood brainbarrier. Curr Neuropharmacol 6:179–192

10. Bearer EL, Orci L 1985 Endothelial fenestral diaphragms:a quick-freeze, deep-etch study. J Cell Biol 100:418–428


11. Henderson JR, Moss MC 1985 A morphometric study ofthe endocrine and exocrine capillaries of the pancreas. Q JExp Physiol 70:347–356

12. Simionescu M, Gafencu A, Antohe F 2002 Transcytosis ofplasma macromolecules in endothelial cells: a cell biolog-ical survey. Microsc Res Tech 57:269–288

13. Jansson L, Carlsson PO 2002 Graft vascular functionafter transplantation of pancreatic islets. Diabetologia45:749 –763

14. Bonner-Weir S, Orci L 1982 New perspectives on the mi-crovasculature of the islets of Langerhans in the rat. Dia-betes 31:883–889

15. Nyman LR, Wells KS, Head WS, McCaughey M, Ford E,Brissova M, Piston DW, Powers AC 2008 Real-time, mul-tidimensional in vivo imaging used to investigate bloodflow in mouse pancreatic islets. J Clin Invest 118:3790–3797

16. Olsson R, Carlsson PO 2006 The pancreatic islet endothe-lial cell: emerging roles in islet function and disease. IntJ Biochem Cell Biol 38:710–714

17. Jansson L, Hellerstrom C 1983 Stimulation by glucose ofthe blood flow to the pancreatic islets of the rat. Diabeto-logia 25:45–50

18. Jansson L, Andersson A, Bodin B, Kallskog O 2007 Pan-creatic islet blood flow during euglycaemic, hyperinsuli-naemic clamp in anaesthetized rats. Acta Physiol (Oxf)189:319–324

19. Carlsson PO, Olsson R, Kallskog O, Bodin B, AnderssonA, Jansson L 2002 Glucose-induced islet blood flow in-crease in rats: interaction between nervous and metabolicmediators. Am J Physiol Endocrinol Metab 283:E457–E464

20. Ballian N, Brunicardi FC 2007 Islet vasculature as a reg-ulator of endocrine pancreas function. World J Surg 31:705–714

21. Carlsson PO, Iwase M, Jansson L 1999 Stimulation of in-testinal glucoreceptors in rats increases pancreatic isletblood flow through vagal mechanisms. Am J Physiol 276:R233–R236

22. Brunicardi FC, Dyen Y, Brostrom L, Kleinman R, ColonnaJ, Gelabert H, Gingerich R 2000 The circulating hormonalmilieu of the endocrine pancreas in healthy individuals,organ donors, and the isolated perfused human pancreas.Pancreas 21:203–211

23. Lammert E, Cleaver O, Melton D 2001 Induction of pan-creatic differentiation by signals from blood vessels. Sci-ence 294:564–567

24. Yoshitomi H, Zaret KS 2004 Endothelial cell interac-tions initiate dorsal pancreas development by selectivelyinducing the transcription factor Ptf1a. Development131:807– 817

25. Edsbagge J, Johansson JK, Esni F, Luo Y, Radice GL, SembH 2005 Vascular function and sphingosine-1-phosphateregulate development of the dorsal pancreatic mesen-chyme. Development 132:1085–1092

26. Jacquemin P, Yoshitomi H, Kashima Y, Rousseau GG,Lemaigre FP, Zaret KS 2006 An endothelial-mesenchymalrelay pathway regulates early phases of pancreas develop-ment. Dev Biol 290:189–199

27. Johansson M, Andersson A, Carlsson PO, Jansson L 2006Perinatal development of the pancreatic islet microvascu-lature in rats. J Anat 208:191–196

28. Cai Q, Brissova M, Shostak A, Powers AC 2009 Increasedexpression of VEGF-A in �-cells increases endothelial cellsbut impairs islet morphogenesis and postnatal �-cellgrowth. Diabetes 58(S1):A56 (Abstract)

29. Lammert E, Gu G, McLaughlin M, Brown D, Brekken R,Murtaugh LC, Gerber HP, Ferrara N, Melton DA 2003Role of VEGF-A in vascularization of pancreatic islets.Curr Biol 13:1070–1074

30. Brissova M, Shostak A, Shiota M, Wiebe PO, PoffenbergerG, Kantz J, Chen Z, Carr C, Jerome WG, Chen J, BaldwinHS, Nicholson W, Bader DM, Jetton T, Gannon M, PowersAC 2006 Pancreatic islet production of vascular endothe-lial growth factor-a is essential for islet vascularization,revascularization, and function. Diabetes 55:2974–2985

31. Iwashita N, Uchida T, Choi JB, Azuma K, Ogihara T,Ferrara N, Gerber H, Kawamori R, Inoue M, Watada H2007 Impaired insulin secretion in vivo but enhanced in-sulin secretion from isolated islets in pancreatic � cell-spe-cific vascular endothelial growth factor-A knock-out mice.Diabetologia 50:380–389

32. Reinert RB, Brissova M, Kantz J, Powers AC 2009 Islet-derived vascular endothelial growth factor-A (VEGF-A) isimportant for maintenance of islet vasculature and func-tion in adult mice. Diabetes 58(S1):A56 (Abstract)

33. Toyofuku Y, Uchida T, Nakayama S, Hirose T, KawamoriR, Fujitani Y, Inoue M, Watada H 2009 Normal islet vas-cularization is dispensable for expansion of �-cell mass inresponse to high-fat diet induced insulin resistance. Bio-chem Biophys Res Commun 383:303–307

34. Branda CS, Dymecki SM 2004 Talking about a revolution:the impact of site-specific recombinases on genetic analysesin mice. Dev Cell 6:7–28

35. Inoue M, Hager JH, Ferrara N, Gerber HP, Hanahan D2002 VEGF-A has a critical, nonredundant role in angio-genic switching and pancreatic �-cell carcinogenesis. Can-cer Cell 1:193–202

36. Anneren C, Welsh M, Jansson L 2007 Glucose intoleranceand reduced islet blood flow in transgenic mice expressingthe FRK tyrosine kinase under the control of the rat insulinpromoter. Am J Physiol Endocrinol Metab 292:E1183–E1190

37. Cantley J, Selman C, Shukla D, Abramov AY, ForstreuterF, Esteban MA, Claret M, Lingard SJ, Clements M, HartenSK, Asare-Anane H, Batterham RL, Herrera PL, PersaudSJ, Duchen MR, Maxwell PH, Withers DJ 2009 Deletionof the von Hippel-Lindau gene in pancreatic � cells impairsglucose homeostasis in mice. J Clin Invest 119:125–135

38. Forsythe JA, Jiang BH, Iyer NV, Agani F, Leung SW, KoosRD, Semenza GL 1996 Activation of vascular endothelialgrowth factor gene transcription by hypoxia-inducible fac-tor 1. Mol Cell Biol 16:4604–4613

39. Crawford SE, Stellmach V, Murphy-Ullrich JE, RibeiroSM, Lawler J, Hynes RO, Boivin GP, Bouck N 1998Thrombospondin-1 is a major activator of TGF-�1 in vivo.Cell 93:1159–1170

40. Li X, Zhang L, Meshinchi S, Dias-Leme C, Raffin D,Johnson JD, Treutelaar MK, Burant CF 2006 Islet micro-vasculature in islet hyperplasia and failure in a model oftype 2 diabetes. Diabetes 55:2965–2973

41. Mizuno A, Noma Y, Kuwajima M, Murakami T, Zhu M,Shima K 1999 Changes in islet capillary angioarchitecturecoincide with impaired B-cell function but not with insulin


resistance in male Otsuka-Long-Evans-Tokushima fattyrats: dimorphism of the diabetic phenotype at an advancedage. Metabolism 48:477–483

42. Carlsson PO, Jansson L, Ostenson CG, Kallskog O 1997Islet capillary blood pressure increase mediated by hyper-glycemia in NIDDM GK rats. Diabetes 46:947–952

43. Atef N, Ktorza A, Picon L, Penicaud L 1992 Increased isletblood flow in obese rats: role of the autonomic nervoussystem. Am J Physiol 262:E736–E740

44. Svensson AM, Abdel-Halim SM, Efendic S, Jansson L,Ostenson CG 1994 Pancreatic and islet blood flow in F1-hybrids of the non-insulin-dependent diabetic GK-Wistarrat. Eur J Endocrinol 130:612–616

45. Carlsson PO, Andersson A, Jansson L 1998 Influence ofage, hyperglycemia, leptin, and NPY on islet blood flow inobese-hyperglycemic mice. Am J Physiol 275:E594–E601

46. Nakamura M, Kitamura H, Konishi S, Nishimura M, OnoJ, Ina K, Shimada T, Takaki R 1995 The endocrine pan-creas of spontaneously diabetic db/db mice: microangiopa-thy as revealed by transmission electron microscopy. Dia-betes Res Clin Pract 30:89–100

47. Duvillie B, Currie C, Chrones T, Bucchini D, Jami J, JoshiRL, Hill DJ 2002 Increased islet cell proliferation, de-creased apoptosis, and greater vascularization leading to�-cell hyperplasia in mutant mice lacking insulin. Endo-crinology 143:1530–1537

48. Jiang FX, Naselli G, Harrison LC 2002 Distinct distribu-tion of laminin and its integrin receptors in the pancreas.J Histochem Cytochem 50:1625–1632

49. Nikolova G, Jabs N, Konstantinova I, Domogatskaya A,Tryggvason K, Sorokin L, Fassler R, Gu G, Gerber HP,Ferrara N, Melton DA, Lammert E 2006 The vascularbasement membrane: a niche for insulin gene expressionand � cell proliferation. Dev Cell 10:397–405

50. Johansson A, Lau J, Sandberg M, Borg LA, MagnussonPU, Carlsson PO 2009 Endothelial cell signalling supportspancreatic � cell function in the rat. Diabetologia 52:2385–2394

51. Johansson M, Mattsson G, Andersson A, Jansson L,Carlsson PO 2006 Islet endothelial cells and pancreatic�-cell proliferation: studies in vitro and during pregnancyin adult rats. Endocrinology 147:2315–2324

52. Virtanen I, Banerjee M, Palgi J, Korsgren O, Lukinius A,Thornell LE, Kikkawa Y, Sekiguchi K, Hukkanen M,Konttinen YT, Otonkoski T 2008 Blood vessels of humanislets of Langerhans are surrounded by a double basementmembrane. Diabetologia 51:1181–1191

53. Shapiro AM, Lakey JR, Ryan EA, Korbutt GS, Toth E,Warnock GL, Kneteman NM, Rajotte RV 2000 Islet trans-plantation in seven patients with type 1 diabetes mellitususing a glucocorticoid-free immunosuppressive regimen.N Engl J Med 343:230–238

54. Shapiro AM, Nanji SA, Lakey JR 2003 Clinical islettransplant: current and future directions towards toler-ance. Immunol Rev 196:219 –236

55. Merani S, Shapiro AM 2006 Current status of pancreaticislet transplantation. Clin Sci (Lond) 110:611–625

56. Lukinius A, Jansson L, Korsgren O 1995 Ultrastructuralevidence for blood microvessels devoid of an endothelialcell lining in transplanted pancreatic islets. Am J Pathol146:429–435

57. Brissova M, Powers AC 2008 Revascularization of trans-planted islets: can it be improved? Diabetes 57:2269–2271

58. Parr EL, Bowen KM, Lafferty KJ 1980 Cellular changes incultured mouse thyroid glands and islets of Langerhans.Transplantation 30:135–141

59. Carlsson PO, Palm F, Andersson A, Liss P 2001 Mark-edly decreased oxygen tension in transplanted rat pan-creatic islets irrespective of the implantation site. Diabetes50:489–495

60. Mattsson G, Jansson L, Carlsson PO 2002 Decreased vas-cular density in mouse pancreatic islets after transplanta-tion. Diabetes 51:1362–1366

61. Davalli AM, Ogawa Y, Scaglia L, Wu YJ, Hollister J,Bonner-Weir S, Weir GC 1995 Function, mass, and repli-cation of porcine and rat islets transplanted into diabeticnude mice. Diabetes 44:104–111

62. Davalli AM, Scaglia L, Zangen DH, Hollister J, Bonner-Weir S, Weir GC 1996 Vulnerability of islets in the imme-diate posttransplantation period. Dynamic changes instructure and function. Diabetes 45:1161–1167

63. Miao G, Ostrowski RP, Mace J, Hough J, Hopper A,Peverini R, Chinnock R, Zhang J, Hathout E 2006 Dy-namic production of hypoxia-inducible factor-1� in earlytransplanted islets. Am J Transplant 6:2636–2643

64. Carmeliet P, Dor Y, Herbert JM, Fukumura D, BrusselmansK, Dewerchin M, Neeman M, Bono F, Abramovitch R,Maxwell P, Koch CJ, Ratcliffe P, Moons L, Jain RK, CollenD, Keshert E, Keshet E 1998 Role of HIF-1� in hypoxia-mediated apoptosis, cell proliferation and tumour angio-genesis. Nature 394:485–490

65. Linn T, Schneider K, Hammes HP, Preissner KT,Brandhorst H, Morgenstern E, Kiefer F, Bretzel RG 2003Angiogenic capacity of endothelial cells in islets of Lang-erhans. FASEB J 17:881–883

66. Brissova M, Fowler M, Wiebe P, Shostak A, Shiota M,Radhika A, Lin PC, Gannon M, Powers AC 2004 Intraisletendothelial cells contribute to revascularization of trans-planted pancreatic islets. Diabetes 53:1318–1325

67. Nyqvist D, Kohler M, Wahlstedt H, Berggren PO 2005Donor islet endothelial cells participate in formation offunctional vessels within pancreatic islet grafts. Diabetes54:2287–2293

68. Zhang N, Richter A, Suriawinata J, Harbaran S, AltomonteJ, Cong L, Zhang H, Song K, Meseck M, Bromberg J, DongH 2004 Elevated vascular endothelial growth factor pro-duction in islets improves islet graft vascularization.Diabetes 53:963–970

69. Narang AS, Cheng K, Henry J, Zhang C, Sabek O, FragaD, Kotb M, Gaber AO, Mahato RI 2004 Vascular endo-thelial growth factor gene delivery for revascularization intransplanted human islets. Pharm Res 21:15–25

70. Lai Y, Schneider D, Kidszun A, Hauck-Schmalenberger I,Breier G, Brandhorst D, Brandhorst H, Iken M, BrendelMD, Bretzel RG, Linn T 2005 Vascular endothelial growthfactor increases functional �-cell mass by improvement ofangiogenesis of isolated human and murine pancreatic is-lets. Transplantation 79:1530–1536

71. Su D, Zhang N, He J, Qu S, Slusher S, Bottino R, BerteraS, Bromberg J, Dong HH 2007 Angiopoietin-1 productionin islets improves islet engraftment and protects islets fromcytokine-induced apoptosis. Diabetes 56:2274–2283

72. Olerud J, Johansson M, Lawler J, Welsh N, Carlsson PO


2008 Improved vascular engraftment and graft functionafter inhibition of the angiostatic factor thrombospondin-1in mouse pancreatic islets. Diabetes 57:1870–1877

73. Johansson M, Olerud J, Jansson L, Carlsson PO 2009 Pro-lactin treatment improves engraftment and function oftransplanted pancreatic islets. Endocrinology 150:1646–1653

74. Johansson U, Rasmusson I, Niclou SP, Forslund N,Gustavsson L, Nilsson B, Korsgren O, Magnusson PU2008 Formation of composite endothelial cell-mesenchy-mal stem cell islets: a novel approach to promote islet re-vascularization. Diabetes 57:2393–2401

75. Speier S, Nyqvist D, Cabrera O, Yu J, Molano RD, PileggiA, Moede T, Kohler M, Wilbertz J, Leibiger B, Ricordi C,Leibiger IB, Caicedo A, Berggren PO 2008 Noninvasive invivo imaging of pancreatic islet cell biology. Nat Med 14:574–578

76. DeFronzo RA, Gunnarsson R, Bjorkman O, Olsson M,Wahren J 1985 Effects of insulin on peripheral andsplanchnic glucose metabolism in noninsulin-dependent(type II) diabetes mellitus. J Clin Invest 76:149–155

77. Bjorntorp P, Berchtold P, Holm J, Larsson B 1971 Theglucose uptake of human adipose tissue in obesity. EurJ Clin Invest 1:480–485

78. Bjorntorp P, Krotkiewski M, Larsson B, Somlo-Szucs Z1970 Effects of feeding states on lipid radioactivity in liver,muscle and adipose tissue after injection of labelled glucosein the rat. Acta Physiol Scand 80:29–38

79. Lillioja S, Young AA, Culter CL, Ivy JL, Abbott WG,Zawadzki JK, Yki-Jarvinen H, Christin L, Secomb TW,Bogardus C 1987 Skeletal muscle capillary density andfiber type are possible determinants of in vivo insulin re-sistance in man. J Clin Invest 80:415–424

80. Ogihara T, Shin BC, Anai M, Katagiri H, Inukai K, FunakiM, Fukushima Y, Ishihara H, Takata K, Kikuchi M,Yazaki Y, Oka Y, Asano T 1997 Insulin receptor substrate(IRS)-2 is dephosphorylated more rapidly than IRS-1 via itsassociation with phosphatidylinositol 3-kinase in skeletalmuscle cells. J Biol Chem 272:12868–12873

81. Sarabia V, Lam L, Burdett E, Leiter LA, Klip A 1992 Glu-cose transport in human skeletal muscle cells in culture.Stimulation by insulin and metformin. J Clin Invest 90:1386–1395

82. Sarabia V, Ramlal T, Klip A 1990 Glucose uptake in hu-man and animal muscle cells in culture. Biochem Cell Biol68:536–542

83. Yang YJ, Hope ID, Ader M, Bergman RN 1989 Insulintransport across capillaries is rate limiting for insulin actionin dogs. J Clin Invest 84:1620–1628

84. Chiu JD, Richey JM, Harrison LN, Zuniga E, Kolka CM,Kirkman E, Ellmerer M, Bergman RN 2008 Direct admin-istration of insulin into skeletal muscle reveals that thetransport of insulin across the capillary endothelium limitsthe time course of insulin to activate glucose disposal. Di-abetes 57:828–835

85. Wang H, Liu Z, Li G, Barrett EJ 2006 The vascular endo-thelial cell mediates insulin transport into skeletal muscle.Am J Physiol Endocrinol Metab 291:E323–E332

86. Dernovsek KD, Bar RS 1985 Processing of cell-bound in-sulin by capillary and macrovascular endothelial cells inculture. Am J Physiol 248:E244–E251

87. Jialal I, King GL, Buchwald S, Kahn CR, Crettaz M 1984

Processing of insulin by bovine endothelial cells in culture.Internalization without degradation. Diabetes 33:794–800

88. Renkin EM, Wiig H 1994 Limits to steady-state lymphflow rates derived from plasma-to-tissue uptake measure-ments. Microvasc Res 47:318–328

89. Kolka CM, Harrison LN, Lottati M, Kirkman EL,Bergman RN 2009 Diet-induced obesity reduces insulinaccess to skeletal muscle causing insulin resistance. Diabe-tes 58 (S1):A68 (Abstract)

90. Miles PD, Levisetti M, Reichart D, Khoursheed M, MoossaAR, Olefsky JM 1995 Kinetics of insulin action in vivo.Identification of rate-limiting steps. Diabetes 44:947–953

91. Yang YJ, Hope I, Ader M, Poulin RA, Bergman RN 1992Dose-response relationship between lymph insulin and glu-cose uptake reveals enhanced insulin sensitivity of periph-eral tissues. Diabetes 41:241–253

92. King GL, Johnson SM 1985 Receptor-mediated transportof insulin across endothelial cells. Science 227:1583–1586

93. Bar RS, Siddle K, Dolash S, Boes M, Dake B 1988 Actionsof insulin and insulin like growth factors I and II in culturedmicrovessel endothelial cells from bovine adipose tissue.Metabolism 37:714–720

94. Steil GM, Ader M, Moore DM, Rebrin K, Bergman RN1996 Transendothelial insulin transport is not saturable invivo. No evidence for a receptor-mediated process. J ClinInvest 97:1497–1503

95. Hamilton-Wessler M, Ader M, Dea MK, Moore D,Loftager M, Markussen J, Bergman RN 2002 Mode oftranscapillary transport of insulin and insulin analogNN304 in dog hindlimb: evidence for passive diffusion.Diabetes 51:574–582

96. Wang H, Wang AX, Liu Z, Barrett EJ 2008 Insulin sig-naling stimulates insulin transport by bovine aortic endo-thelial cells. Diabetes 57:540–547

97. Bar RS, Boes M, Sandra A 1988 Vascular transport ofinsulin to rat cardiac muscle. Central role of the capillaryendothelium. J Clin Invest 81:1225–1233

98. Baron AD 1994 Hemodynamic actions of insulin. Am JPhysiol 267:E187–E202

99. Baron AD, Laakso M, Brechtel G, Edelman SV 1991Mechanism of insulin resistance in insulin-dependent dia-betes mellitus: a major role for reduced skeletal muscleblood flow. J Clin Endocrinol Metab 73:637–643

100. Laakso M, Edelman SV, Brechtel G, Baron AD 1990 De-creased effect of insulin to stimulate skeletal muscle bloodflow in obese man. A novel mechanism for insulin resis-tance. J Clin Invest 85:1844–1852

101. Laakso M, Edelman SV, Brechtel G, Baron AD 1992 Im-paired insulin-mediated skeletal muscle blood flow in pa-tients with NIDDM. Diabetes 41:1076–1083

102. Vollenweider P, Tappy L, Randin D, Schneiter P, Jequier E,Nicod P, Scherrer U 1993 Differential effects of hyperin-sulinemia and carbohydrate metabolism on sympatheticnerve activity and muscle blood flow in humans. J ClinInvest 92:147–154

103. Raitakari M, Knuuti MJ, Ruotsalainen U, Laine H, MakeaP, Teras M, Sipila H, Niskanen T, Raitakari OT, Iida H1995 Insulin increases blood volume in human skeletalmuscle: studies using [15O]CO and positron emission to-mography. Am J Physiol 269:E1000–E1005

104. Utriainen T, Nuutila P, Takala T, Vicini P, RuotsalainenU, Ronnemaa T, Tolvanen T, Raitakari M, Haaparanta


M, Kirvela O, Cobelli C, Yki-Jarvinen H 1997 Intact in-sulin stimulation of skeletal muscle blood flow, its heter-ogeneity and redistribution, but not of glucose uptake innon-insulin-dependent diabetes mellitus. J Clin Invest 100:777–785

105. Raitakari M, Nuutila P, Knuuti J, Raitakari OT, Laine H,Ruotsalainen U, Kirvela O, Takala TO, Iida H, Yki-JarvinenH 1997 Effects of insulin on blood flow and volume inskeletal muscle of patients with IDDM: studies using[15O]H2O, [15O]CO, and positron emission tomogra-phy. Diabetes 46:2017–2021

106. Tack CJ, Ong MK, Lutterman JA, Smits P 1998 Insulin-induced vasodilatation and endothelial function in obesity/insulin resistance. Effects of troglitazone. Diabetologia 41:569–576

107. James DE, Burleigh KM, Storlien LH, Bennett SP, KraegenEW 1986 Heterogeneity of insulin action in muscle: influ-ence of blood flow. Am J Physiol 251:E422–E430

108. Liang C, Doherty JU, Faillace R, Maekawa K, Arnold S,Gavras H, Hood Jr WB 1982 Insulin infusion in consciousdogs. Effects on systemic and coronary hemodynamics,regional blood flows, and plasma catecholamines. J ClinInvest 69:1321–1336

109. Fisher BM, Gillen G, Dargie HJ, Inglis GC, Frier BM 1987The effects of insulin-induced hypoglycaemia on cardio-vascular function in normal man: studies using radionu-clide ventriculography. Diabetologia 30:841–845

110. Creager MA, Liang CS, Coffman JD 1985 �-Adrenergic-mediated vasodilator response to insulin in the humanforearm. J Pharmacol Exp Ther 235:709–714

111. Richter EA, Mikines KJ, Galbo H, Kiens B 1989 Effect ofexercise on insulin action in human skeletal muscle. J ApplPhysiol 66:876–885

112. DeFronzo RA, Jacot E, Jequier E, Maeder E, Wahren J,Felber JP 1981 The effect of insulin on the disposal ofintravenous glucose. Results from indirect calorimetry andhepatic and femoral venous catheterization. Diabetes 30:1000–1007

113. DeFronzo RA, Ferrannini E, Hendler R, Felig P, Wahren J1983 Regulation of splanchnic and peripheral glucose up-take by insulin and hyperglycemia in man. Diabetes 32:35–45

114. Jackson RA, Hamling JB, Blix PM, Sim BM, Hawa MI,Jaspan JB, Belin J, Nabarro JD 1986 The influence ofgraded hyperglycemia with and without physiologicalhyperinsulinemia on forearm glucose uptake and othermetabolic responses in man. J Clin Endocrinol Metab63:594 – 604

115. Jackson RA, Roshania RD, Hawa MI, Sim BM, DiSilvio L1986 Impact of glucose ingestion on hepatic and peripheralglucose metabolism in man: an analysis based on simulta-neous use of the forearm and double isotope techniques.J Clin Endocrinol Metab 63:541–549

116. Yki-Jarvinen H, Young AA, Lamkin C, Foley JE 1987 Ki-netics of glucose disposal in whole body and across theforearm in man. J Clin Invest 79:1713–1719

117. Taddei S, Virdis A, Mattei P, Natali A, Ferrannini E,Salvetti A 1995 Effect of insulin on acetylcholine-inducedvasodilation in normotensive subjects and patients withessential hypertension. Circulation 92:2911–2918

118. Bonadonna RC, Saccomani MP, Del Prato S, Bonora E,DeFronzo RA, Cobelli C 1998 Role of tissue-specific blood

flow and tissue recruitment in insulin-mediated glucoseuptake of human skeletal muscle. Circulation 98:234–241

119. Natali A, Buzzigoli G, Taddei S, Santoro D, Cerri M,Pedrinelli R, Ferrannini E 1990 Effects of insulin on he-modynamics and metabolism in human forearm. Diabetes39:490–500

120. Yki-Jarvinen H, Utriainen T 1998 Insulin-induced vaso-dilatation: physiology or pharmacology? Diabetologia 41:369–379

121. Steinberg HO, Baron AD 1999 Insulin-mediated vasodi-lation: why one’s physiology could be the other’s pharma-cology. Diabetologia 42:493–495

122. Honig CR, Odoroff CL, Frierson JL 1982 Active and pas-sive capillary control in red muscle at rest and in exercise.Am J Physiol 243:H196–H206

123. Poole DC, Brown MD, Hudlicka O 2008 Counterpoint:There is not capillary recruitment in active skeletal muscleduring exercise. J Appl Physiol 104:891–893; discussion893–894

124. Rattigan S, Clark MG, Barrett EJ 1997 Hemodynamic ac-tions of insulin in rat skeletal muscle: evidence for capillaryrecruitment. Diabetes 46:1381–1388

125. Coggins M, Lindner J, Rattigan S, Jahn L, Fasy E, Kaul S,Barrett E 2001 Physiologic hyperinsulinemia enhances hu-man skeletal muscle perfusion by capillary recruitment.Diabetes 50:2682–2690

126. Vincent MA, Barrett EJ, Lindner JR, Clark MG, RattiganS 2003 Inhibiting NOS blocks microvascular recruitmentand blunts muscle glucose uptake in response to insulin.Am J Physiol Endocrinol Metab 285:E123–E129

127. Zhang L, Vincent MA, Richards SM, Clerk LH, RattiganS, Clark MG, Barrett EJ 2004 Insulin sensitivity of musclecapillary recruitment in vivo. Diabetes 53:447–453

128. Serne EH, IJzerman RG, Gans RO, Nijveldt R, De Vries G,Evertz R, Donker AJ, Stehouwer CD 2002 Direct evidencefor insulin-induced capillary recruitment in skin of healthysubjects during physiological hyperinsulinemia. Diabetes51:1515–1522

129. Zeng G, Quon MJ 1996 Insulin-stimulated productionof nitric oxide is inhibited by wortmannin. Direct mea-surement in vascular endothelial cells. J Clin Invest 98:894 – 898

130. Montagnani M, Chen H, Barr VA, Quon MJ 2001 Insulin-stimulated activation of eNOS is independent of Ca2� butrequires phosphorylation by Akt at Ser(1179). J Biol Chem276:30392–30398

131. Eringa EC, Stehouwer CD, Merlijn T, Westerhof N,Sipkema P 2002 Physiological concentrations of insulininduce endothelin-mediated vasoconstriction during inhi-bition of NOS or PI3-kinase in skeletal muscle arterioles.Cardiovasc Res 56:464–471

132. Kubota T, Kubota N, Kozono H, Takahashi T, Itoh S,Ueki K, Kadowaki T 2008 Insulin signaling in endothelialcells participates in the regulation of skeletal muscle insulinsensitivity. Diabetes 57(S1):A369 (Abstract)

133. Eringa EC, Stehouwer CD, Walburg K, Clark AD, vanNieuw Amerongen GP, Westerhof N, Sipkema P 2006Physiological concentrations of insulin induce endothelin-dependent vasoconstriction of skeletal muscle resistancearteries in the presence of tumor necrosis factor-� depen-dence on c-Jun N-terminal kinase. Arterioscler ThrombVasc Biol 26:274–280


134. Eringa EC, Stehouwer CD, van Nieuw Amerongen GP,Ouwehand L, Westerhof N, Sipkema P 2004 Vasocon-strictor effects of insulin in skeletal muscle arterioles aremediated by ERK1/2 activation in endothelium. Am JPhysiol Heart Circ Physiol 287:H2043–H2048

135. Eringa EC, Stehouwer CD, Roos MH, Westerhof N,Sipkema P 2007 Selective resistance to vasoactive effects ofinsulin in muscle resistance arteries of obese Zucker (fa/fa)rats. Am J Physiol Endocrinol Metab 293:E1134–E1139

136. Cardillo C, Campia U, Iantorno M, Panza JA 2004 En-hanced vascular activity of endogenous endothelin-1 inobese hypertensive patients. Hypertension 43:36–40

137. Shankar RR, Wu Y, Shen HQ, Zhu JS, Baron AD 2000Mice with gene disruption of both endothelial and neuro-nal nitric oxide synthase exhibit insulin resistance. Diabe-tes 49:684–687

138. Wallis MG, Wheatley CM, Rattigan S, Barrett EJ, ClarkAD, Clark MG 2002 Insulin-mediated hemodynamicchanges are impaired in muscle of Zucker obese rats.Diabetes 51:3492–3498

139. Clerk LH, Rattigan S, Clark MG 2002 Lipid infusion im-pairs physiologic insulin-mediated capillary recruitmentand muscle glucose uptake in vivo. Diabetes 51:1138-1145

140. Rattigan S, Clark MG, Barrett EJ 1999 Acute vasocon-striction-induced insulin resistance in rat muscle in vivo.Diabetes 48:564–569

141. Youd JM, Rattigan S, Clark MG 2000 Acute impairmentof insulin-mediated capillary recruitment and glucose up-take in rat skeletal muscle in vivo by TNF-alpha. Diabetes49:1904–1909

142. Wallis MG, Smith ME, Kolka CM, Zhang L, Richards SM,Rattigan S, Clark MG 2005 Acute glucosamine-inducedinsulin resistance in muscle in vivo is associated with im-paired capillary recruitment. Diabetologia 48:2131–2139

143. Gudbjornsdottir S, Sjostrand M, Strindberg L, Lonnroth P2005 Decreased muscle capillary permeability surface areain type 2 diabetic subjects. J Clin Endocrinol Metab 90:1078–1082

144. Lesniewski LA, Donato AJ, Behnke BJ, Woodman CR,Laughlin MH, Ray CA, Delp MD 2008 Decreased NOsignaling leads to enhanced vasoconstrictor responsivenessin skeletal muscle arterioles of the ZDF rat prior to overtdiabetes and hypertension. Am J Physiol Heart Circ Physiol294:H1840–H1850

145. Savage DB, Petersen KF, Shulman GI 2007 Disorderedlipid metabolism and the pathogenesis of insulin resis-tance. Physiol Rev 87:507–520

146. Yu C, Chen Y, Cline GW, Zhang D, Zong H, Wang Y,Bergeron R, Kim JK, Cushman SW, Cooney GJ, AtchesonB, White MF, Kraegen EW, Shulman GI 2002 Mechanismby which fatty acids inhibit insulin activation of insulinreceptor substrate-1 (IRS-1)-associated phosphatidylino-sitol 3-kinase activity in muscle. J Biol Chem 277:50230–50236

147. Vicent D, Ilany J, Kondo T, Naruse K, Fisher SJ, KisanukiYY, Bursell S, Yanagisawa M, King GL, Kahn CR 2003The role of endothelial insulin signaling in the regulationof vascular tone and insulin resistance. J Clin Invest 111:1373–1380

148. Bruning JC, Michael MD, Winnay JN, Hayashi T, HorschD, Accili D, Goodyear LJ, Kahn CR 1998 A muscle-specific

insulin receptor knockout exhibits features of the meta-bolic syndrome of NIDDM without altering glucose tol-erance. Mol Cell 2:559–569

149. Chisalita SI, Arnqvist HJ 2004 Insulin-like growth factor Ireceptors are more abundant than insulin receptors in hu-man micro- and macrovascular endothelial cells. Am JPhysiol Endocrinol Metab 286:E896–E901

150. Clifford PS 2007 Skeletal muscle vasodilatation at the on-set of exercise. J Physiol 583:825–833

151. Rattigan S, Wheatley C, Richards SM, Barrett EJ, ClarkMG 2005 Exercise and insulin-mediated capillary recruit-ment in muscle. Exerc Sport Sci Rev 33:43–48

152. Ross RM, Wadley GD, Clark MG, Rattigan S, McConellGK 2007 Local nitric oxide synthase inhibition reducesskeletal muscle glucose uptake but not capillary blood flowduring in situ muscle contraction in rats. Diabetes56:2885–2892

153. Wheatley CM, Rattigan S, Richards SM, Barrett EJ, ClarkMG 2004 Skeletal muscle contraction stimulates capillaryrecruitment and glucose uptake in insulin-resistant obeseZucker rats. Am J Physiol Endocrinol Metab 287:E804–E809

154. Krueger M, Bechmann I 2010 CNS pericytes: concepts,misconceptions, and a way out. Glia 58:1–10

155. Bergers G, Song S 2005 The role of pericytes in blood-vesselformation and maintenance. Neuro Oncol 7:452–464

156. Gerhardt H, Semb H 2008 Pericytes: gatekeepers in tu-mour cell metastasis? J Mol Med 86:135–144

157. Gaengel K, Genove G, Armulik A, Betsholtz C 2009Endothelial-mural cell signaling in vascular development andangiogenesis. Arterioscler Thromb Vasc Biol 29:630–638

158. Armulik A, Abramsson A, Betsholtz C 2005 Endothelial/pericyte interactions. Circ Res 97:512–523

159. Mandarino LJ, Sundarraj N, Finlayson J, Hassell HR 1993Regulation of fibronectin and laminin synthesis by retinalcapillary endothelial cells and pericytes in vitro. Exp EyeRes 57:609–621

160. Suematsu M, Aiso S 2001 Professor Toshio Ito: a clair-voyant in pericyte biology. Keio J Med 50:66–71

161. Balabanov R, Dore-Duffy P 1998 Role of the CNS micro-vascular pericyte in the blood-brain barrier. J Neurosci Res53:637–644

162. Kunz J, Krause D, Gehrmann J, Dermietzel R 1995Changes in the expression pattern of blood-brain barrier-associated pericytic aminopeptidase N (pAP N) in thecourse of acute experimental autoimmune encephalomy-elitis. J Neuroimmunol 59:41–55

163. Rucker HK, Wynder HJ, Thomas WE 2000 Cellular mech-anisms of CNS pericytes. Brain Res Bull 51:363–369

164. Hellstrom M, Gerhardt H, Kalen M, Li X, Eriksson U,Wolburg H, Betsholtz C 2001 Lack of pericytes leads toendothelial hyperplasia and abnormal vascular morpho-genesis. J Cell Biol 153:543–553

165. Gerhardt H, Betsholtz C 2003 Endothelial-pericyte inter-actions in angiogenesis. Cell Tissue Res 314:15–23

166. Ozerdem U, Grako KA, Dahlin-Huppe K, Monosov E,Stallcup WB 2001 NG2 proteoglycan is expressed exclu-sively by mural cells during vascular morphogenesis. DevDyn 222:218–227

167. Ozerdem U, Stallcup WB 2003 Early contribution of peri-cytes to angiogenic sprouting and tube formation. Angio-genesis 6:241–249


168. Gerhardt H, Golding M, Fruttiger M, Ruhrberg C,Lundkvist A, Abramsson A, Jeltsch M, Mitchell C, AlitaloK, Shima D, Betsholtz C 2003 VEGF guides angiogenicsprouting utilizing endothelial tip cell filopodia. J Cell Biol161:1163–1177

169. Redmer DA, Doraiswamy V, Bortnem BJ, Fisher K,Jablonka-Shariff A, Grazul-Bilska AT, Reynolds LP 2001Evidence for a role of capillary pericytes in vascular growthof the developing ovine corpus luteum. Biol Reprod 65:879–889

170. Evensen L, Micklem DR, Blois A, Berge SV, Aarsaether N,Littlewood-Evans A, Wood J, Lorens JB 2009 Mural cellassociated VEGF is required for organotypic vessel forma-tion. PLoS One 4:e5798

171. Reigstad LJ, Sande HM, Fluge Ø, Bruland O, Muga A,Varhaug JE, Martinez A, Lillehaug JR 2003 Platelet-derived growth factor (PDGF)-C, a PDGF family memberwith a vascular endothelial growth factor-like structure.J Biol Chem 278:17114–17120

172. Andrae J, Gallini R, Betsholtz C 2008 Role of platelet-derived growth factors in physiology and medicine. GenesDev 22:1276–1312

173. Kohler N, Lipton A 1974 Platelets as a source of fibroblastgrowth-promoting activity. Exp Cell Res 87:297–301

174. Ross R, Glomset J, Kariya B, Harker L 1974 A platelet-dependent serum factor that stimulates the proliferation ofarterial smooth muscle cells in vitro. Proc Natl Acad SciUSA 71:1207–1210

175. Westermark B, Wasteson A 1976 A platelet factor stimu-lating human normal glial cells. Exp Cell Res 98:170–174

176. Antoniades HN, Scher CD, Stiles CD 1979 Purification ofhuman platelet-derived growth factor. Proc Natl Acad SciUSA 76:1809–1813

177. Deuel TF, Huang JS, Proffitt RT, Baenziger JU, Chang D,Kennedy BB 1981 Human platelet-derived growth factor.Purification and resolution into two active protein frac-tions. J Biol Chem 256:8896–8899

178. Heldin CH, Westermark B, Wasteson A 1979 Platelet-de-rived growth factor: purification and partial characteriza-tion. Proc Natl Acad Sci USA 76:3722–3726

179. Raines EW, Ross R 1982 Platelet-derived growth factor. I.High yield purification and evidence for multiple forms.J Biol Chem 257:5154–5160

180. Heldin CH, Westermark B 1999 Mechanism of action andin vivo role of platelet-derived growth factor. Physiol Rev79:1283–1316

181. Kelly JD, Haldeman BA, Grant FJ, Murray MJ, Seifert RA,Bowen-Pope DF, Cooper JA, Kazlauskas A 1991 Platelet-derived growth factor (PDGF) stimulates PDGF receptorsubunit dimerization and intersubunit trans-phosphoryla-tion. J Biol Chem 266:8987–8992

182. Kazlauskas A, Cooper JA 1989 Autophosphorylation ofthe PDGF receptor in the kinase insert region regulatesinteractions with cell proteins. Cell 58:1121–1133

183. Hoch RV, Soriano P 2003 Roles of PDGF in animal de-velopment. Development 130:4769–4784

184. Tallquist MD, French WJ, Soriano P 2003 Additive effectsof PDGF receptor � signaling pathways in vascular smoothmuscle cell development. PLoS Biol 1:E52

185. Heldin CH, Wasteson A, Westermark B 1982 Interactionof platelet-derived growth factor with its fibroblast recep-

tor. Demonstration of ligand degradation and receptormodulation. J Biol Chem 257:4216–4221

186. Sorkin A, Westermark B, Heldin CH, Claesson-Welsh L1991 Effect of receptor kinase inactivation on the rate ofinternalization and degradation of PDGF and the PDGF�-receptor. J Cell Biol 112:469–478

187. Mori S, Kanaki H, Tanaka K, Morisaki N, Saito Y 1995Ligand-activated platelet-derived growth factor �-recep-tor is degraded through proteasome-dependent proteolyticpathway. Biochem Biophys Res Commun 217:224–229

188. LaRochelle WJ, May-Siroff M, Robbins KC, Aaronson SA1991 A novel mechanism regulating growth factor asso-ciation with the cell surface: identification of a PDGF re-tention domain. Genes Dev 5:1191–1199

189. Ostman A, Andersson M, Betsholtz C, Westermark B,Heldin CH 1991 Identification of a cell retention signal inthe B-chain of platelet-derived growth factor and in thelong splice version of the A-chain. Cell Regul 2:503–512

190. Raines EW, Ross R 1992 Compartmentalization of PDGFon extracellular binding sites dependent on exon-6-en-coded sequences. J Cell Biol 116:533–543

191. Lindblom P, Gerhardt H, Liebner S, Abramsson A, EngeM, Hellstrom M, Backstrom G, Fredriksson S, LandegrenU, Nystrom HC, Bergstrom G, Dejana E, Ostman A,Lindahl P, Betsholtz C 2003 Endothelial PDGF-B retentionis required for proper investment of pericytes in the mi-crovessel wall. Genes Dev 17:1835–1840

192. Lindahl P, Johansson BR, Leveen P, Betsholtz C 1997 Peri-cyte loss and microaneurysm formation in PDGF-B-defi-cient mice. Science 277:242–245

193. Hellstrom M, Kalen M, Lindahl P, Abramsson A, BetsholtzC 1999 Role of PDGF-B and PDGFR-� in recruitment ofvascular smooth muscle cells and pericytes during embry-onic blood vessel formation in the mouse. Development126:3047–3055

194. Leveen P, Pekny M, Gebre-Medhin S, Swolin B, Larsson E,Betsholtz C 1994 Mice deficient for PDGF B show renal,cardiovascular, and hematological abnormalities. GenesDev 8:1875–1887

195. Soriano P 1994 Abnormal kidney development and hema-tological disorders in PDGF �-receptor mutant mice.Genes Dev 8:1888–1896

196. Hirschi KK, Rohovsky SA, D’Amore PA 1998 PDGF,TGF-�, and heterotypic cell-cell interactions mediate en-dothelial cell-induced recruitment of 10T1/2 cells and theirdifferentiation to a smooth muscle fate. J Cell Biol 141:805–814

197. Hirschi KK, Rohovsky SA, Beck LH, Smith SR, D’AmorePA 1999 Endothelial cells modulate the proliferation ofmural cell precursors via platelet-derived growth factor-BBand heterotypic cell contact. Circ Res 84:298–305

198. Tilton RG, Kilo C, Williamson JR 1979 Pericyte-endothe-lial relationships in cardiac and skeletal muscle capillaries.Microvasc Res 18:325–335

199. Egginton S, Hudlicka O, Brown MD, Graciotti L, GranataAL 1996 In vivo pericyte-endothelial cell interaction dur-ing angiogenesis in adult cardiac and skeletal muscle. Mi-crovasc Res 51:213–228

200. Raines S, Richards O, Scheuler K, Attie A 2009 Decreasein PDGF-B signalling reduces in vivo insulin secretion.Diabetes 58(S1):A376 (Abstract)

201. Welsh M, Claesson-Welsh L, Hallberg A, Welsh N,


Betsholtz C, Arkhammar P, Nilsson T, Heldin CH,Berggren PO 1990 Coexpression of the platelet-derivedgrowth factor (PDGF) B chain and the PDGF � receptor inisolated pancreatic islet cells stimulates DNA synthesis.Proc Natl Acad Sci USA 87:5807–5811

202. Hammes HP 2005 Pericytes and the pathogenesis of dia-betic retinopathy. Horm Metab Res 37(Suppl 1):39–43

203. Geraldes P, Hiraoka-Yamamoto J, Matsumoto M,Clermont A, Leitges M, Marette A, Aiello LP, Kern TS,King GL 2009 Activation of PKC-� and SHP-1 by hyper-glycemia causes vascular cell apoptosis and diabetic reti-nopathy. Nat Med 15:1298–1306

204. Wolf G, Chen S, Ziyadeh FN 2005 From the periphery ofthe glomerular capillary wall toward the center of disease:podocyte injury comes of age in diabetic nephropathy.Diabetes 54:1626 –1634

205. Siemionow M, Demir Y 2004 Diabetic neuropathy: patho-genesis and treatment. J Reconstr Microsurg 20:241–252

206. Tilton RG, Hoffmann PL, Kilo C, Williamson JR 1981Pericyte degeneration and basement membrane thickeningin skeletal muscle capillaries of human diabetics. Diabetes30:326–334

207. Caplan AI 1994 The mesengenic process. Clin Plast Surg21:429–435

208. Hess D, Li L, Martin M, Sakano S, Hill D, Strutt B, ThyssenS, Gray DA, Bhatia M 2003 Bone marrow-derived stemcells initiate pancreatic regeneration. Nat Biotechnol 21:763–770

209. Ianus A, Holz GG, Theise ND, Hussain MA 2003 In vivoderivation of glucose-competent pancreatic endocrine cellsfrom bone marrow without evidence of cell fusion. J ClinInvest 111:843–850

210. Voltarelli JC, Couri CE, Stracieri AB, Oliveira MC,Moraes DA, Pieroni F, Coutinho M, Malmegrim KC,Foss-Freitas MC, Simoes BP, Foss MC, Squiers E, Burt RK2007 Autologous nonmyeloablative hematopoietic stemcell transplantation in newly diagnosed type 1 diabetesmellitus. JAMA 297:1568–1576

211. Choi JB, Uchino H, Azuma K, Iwashita N, Tanaka Y,Mochizuki H, Migita M, Shimada T, Kawamori R,Watada H 2003 Little evidence of transdifferentiation ofbone marrow-derived cells into pancreatic � cells. Diabe-tologia 46:1366–1374

212. Lechner A, Yang YG, Blacken RA, Wang L, Nolan AL,Habener JF 2004 No evidence for significant transdiffer-entiation of bone marrow into pancreatic �-cells in vivo.Diabetes 53:616–623

213. Ezquer FE, Ezquer ME, Parrau DB, Carpio D, Yanez AJ,Conget PA 2008 Systemic administration of multipotentmesenchymal stromal cells reverts hyperglycemia andprevents nephropathy in type 1 diabetic mice. Biol BloodMarrow Transplant 14:631– 640

214. Crisan M, Yap S, Casteilla L, Chen CW, Corselli M, ParkTS, Andriolo G, Sun B, Zheng B, Zhang L, Norotte C,Teng PN, Traas J, Schugar R, Deasy BM, Badylak S,Buhring HJ, Giacobino JP, Lazzari L, Huard J, Peault B2008 A perivascular origin for mesenchymal stem cells inmultiple human organs. Cell Stem Cell 3:301–313

215. Tang W, Zeve D, Suh JM, Bosnakovski D, Kyba M,Hammer RE, Tallquist MD, Graff JM 2008 White fat pro-genitor cells reside in the adipose vasculature. Science 322:583–586

216. Dai C, Brissova M, Nyman LR, Shiota M, Powers AC 2009Islet vasculature changes in response to insulin resistance.Diabetes 58(S1):A57 (Abstract)

217. Hayden MR, Karuparthi PR, Habibi J, Wasekar C, LastraG, Manrique C, Stas S, Sowers JR 2007 Ultrastructuralislet study of early fibrosis in the Ren2 rat model of hy-pertension. Emerging role of the islet pancreatic pericyte-stellate cell. JOP 8:725–738

218. Hayden MR, Karuparthi PR, Habibi J, Lastra G, Patel K,Wasekar C, Manrique CM, Ozerdem U, Stas S, Sowers JR2008 Ultrastructure of islet microcirculation, pericytes andthe islet exocrine interface in the HIP rat model of diabetes.Exp Biol Med (Maywood) 233:1109–1123

219. Hayden MR, Patel K, Habibi J, Gupta D, Tekwani SS,Whaley-Connell A, Sowers JR 2008 Attenuation of endo-crine-exocrine pancreatic communication in type 2 diabe-tes: pancreatic extracellular matrix ultrastructural abnor-malities. J Cardiometab Syndr 3:234–243

220. Bergers G, Song S, Meyer-Morse N, Bergsland E, HanahanD 2003 Benefits of targeting both pericytes and endothelialcells in the tumor vasculature with kinase inhibitors. J ClinInvest 111:1287–1295

221. Baluk P, Morikawa S, Haskell A, Mancuso M, McDonaldDM 2003 Abnormalities of basement membrane on bloodvessels and endothelial sprouts in tumors. Am J Pathol163:1801–1815

222. Joyce JA, Laakkonen P, Bernasconi M, Bergers G,Ruoslahti E, Hanahan D 2003 Stage-specific vascularmarkers revealed by phage display in a mouse model ofpancreatic islet tumorigenesis. Cancer Cell 4:393–403

223. Berger M, Bergers G, Arnold B, Hammerling GJ, Ganss R2005 Regulator of G-protein signaling-5 induction in peri-cytes coincides with active vessel remodeling during neo-vascularization. Blood 105:1094–1101

224. Pietras K, Hanahan D 2005 A multitargeted, metronomic,and maximum-tolerated dose “chemo-switch” regimen isantiangiogenic, producing objective responses and sur-vival benefit in a mouse model of cancer. J Clin Oncol23:939–952

225. Xian X, Håkansson J, Ståhlberg A, Lindblom P, BetsholtzC, Gerhardt H, Semb H 2006 Pericytes limit tumor cellmetastasis. J Clin Invest 116:642–651

226. Song S, Ewald AJ, Stallcup W, Werb Z, Bergers G 2005PDGFR�� perivascular progenitor cells in tumours regu-late pericyte differentiation and vascular survival. Nat CellBiol 7:870–879

227. Sennino B, Falcon BL, McCauley D, Le T, McCauley T,Kurz JC, Haskell A, Epstein DM, McDonald DM 2007Sequential loss of tumor vessel pericytes and endothelialcells after inhibition of platelet-derived growth factor B byselective aptamer AX102. Cancer Res 67:7358–7367

228. Yonenaga Y, Mori A, Onodera H, Yasuda S, Oe H,Fujimoto A, Tachibana T, Imamura M 2005 Absence ofsmooth muscle actin-positive pericyte coverage of tumor ves-sels correlates with hematogenous metastasis and prognosisof colorectal cancer patients. Oncology 69:159–166

229. Whiteman EL, Chen JJ, Birnbaum MJ 2003 Platelet-de-rived growth factor (PDGF) stimulates glucose transport in3T3–L1 adipocytes overexpressing PDGF receptor by apathway independent of insulin receptor substrates. En-docrinology 144:3811–3820

230. Yuasa T, Kakuhata R, Kishi K, Obata T, Shinohara Y,


Bando Y, Izumi K, Kajiura F, Matsumoto M, Ebina Y 2004Platelet-derived growth factor stimulates glucose transportin skeletal muscles of transgenic mice specifically express-ing platelet-derived growth factor receptor in the muscle,but it does not affect blood glucose levels. Diabetes 53:2776–2786

231. Hoehn KL, Hohnen-Behrens C, Cederberg A, Wu LE,Turner N, Yuasa T, Ebina Y, James DE 2008 IRS1-inde-pendent defects define major nodes of insulin resistance.Cell Metab 7:421–433

232. Druker BJ 2002 STI571 (Gleevec) as a paradigm for cancertherapy. Trends Mol Med 8:S14–S18

233. Veneri D, Franchini M, Bonora E 2005 Imatinib and re-gression of type 2 diabetes. N Engl J Med 352:1049–1050

234. Breccia M, Muscaritoli M, Aversa Z, Mandelli F, AlimenaG 2004 Imatinib mesylate may improve fasting blood glu-cose in diabetic Ph� chronic myelogenous leukemia pa-tients responsive to treatment. J Clin Oncol 22:4653–4655

235. Breccia M, Muscaritoli M, Alimena G 2005 Reduction ofglycosylated hemoglobin with stable insulin levels in a di-abetic patient with chronic myeloid leukemia responsive toimatinib. Haematologica 90 Suppl:ECR21

236. Dingli D, Wolf RC, Vella A 2007 Imatinib and type 2diabetes. Endocr Pract 13:126–130

237. Han MS, Chung KW, Cheon HG, Rhee SD, Yoon CH, LeeMK, Kim KW, Lee MS 2009 Imatinib mesylate reducesendoplasmic reticulum stress and induces remission of di-abetes in db/db mice. Diabetes 58:329–336

238. Louvet C, Szot GL, Lang J, Lee MR, Martinier N, BollagG, Zhu S, Weiss A, Bluestone JA 2008 Tyrosine kinaseinhibitors reverse type 1 diabetes in nonobese diabeticmice. Proc Natl Acad Sci USA 105:18895–18900

239. Kano MR, Komuta Y, Iwata C, Oka M, Shirai YT,Morishita Y, Ouchi Y, Kataoka K, Miyazono K 2009Comparison of the effects of the kinase inhibitors imatinib,sorafenib, and transforming growth factor-� receptor in-hibitor on extravasation of nanoparticles from neovascu-lature. Cancer Sci 100:173–180

240. Cersosimo E, DeFronzo RA 2006 Insulin resistance andendothelial dysfunction: the road map to cardiovasculardiseases. Diabetes Metab Res Rev 22:423–436

241. Murad F 2008 Nitric oxide and cyclic guanosine mono-phosphate signaling in the eye. Can J Ophthalmol 43:291–294

242. BrunnerF,Bras-SilvaC,CerdeiraAS,Leite-MoreiraAF2006Cardiovascular endothelins: essential regulators of cardio-vascular homeostasis. Pharmacol Ther 111:508–531

243. Agapitov AV, Haynes WG 2002 Role of endothelin in car-diovascular disease. J Renin Angiotensin Aldosterone Syst3:1–15

244. Schubert R, Lidington D, Bolz SS 2008 The emerging roleof Ca2� sensitivity regulation in promoting myogenic va-soconstriction. Cardiovasc Res 77:8–18

245. Russo I, Del Mese P, Doronzo G, Mattiello L, Viretto M,Bosia A, Anfossi G, Trovati M2008 Resistance to the nitricoxide/cyclic guanosine 5�-monophosphate/protein kinaseG pathway in vascular smooth muscle cells from the obeseZucker rat, a classical animal model of insulin resistance:role of oxidative stress. Endocrinology 149:1480–1489

246. Pacher P, Beckman JS, Liaudet L 2007 Nitric oxide and per-oxynitrite in health and disease. Physiol Rev 87:315–424

247. Guillot PV, Guan J, Liu L, Kuivenhoven JA, RosenbergRD, Sessa WC, Aird WC 1999 A vascular bed-specificpathway. J Clin Invest 103:799–805


the role of blood vessels, endothelial cells, and vascular ... · 2/17/2010 · the role of blood...

Documents