Increased Expression of Vascular Endothelial Growth Factor in Kidney Leads to Progressive Impairment of Glomerular Functions
Enqi Liu*,,
Masatoshi Morimoto*,
Shuji Kitajima*,
Tomonari Koike,
Ying Yu,
Hideo Shiiki,
Michio Nagata||,
Teruo Watanabe* and
Jianglin Fan
* Analytical Research Center for Experimental Sciences, Saga University, Saga, Japan; Laboratory Animal Center, Xi'an Jiaotong University, Xi'an, China; Department of Molecular Pathology, Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Yamanashi, Japan; Department of Nephrology, Uda City Hospital, Nara, Japan; and || Department of Molecular Pathology, Institute of Basic Medical Sciences, University of Tsukuba, Tsukuba, Japan
Correspondence: Dr. Jianglin Fan, Department of Molecular Pathology, Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, 1110 Shimokato, Chuo-City, Yamanashi 409-3898, Japan. Phone: +81-55-273-9520; Fax: +81-55-273-9520; E-mail: fan_molpatho{at}yahoo.co.jp
Received for publication January 24, 2006.
Accepted for publication April 17, 2007.
Vascular endothelial growth factor (VEGF) is an important mediatorin maintaining normal kidney functions. In addition, severallines of evidence suggest that upregulation of VEGF in glomerulimay be associated with or cause renal dysfunction such as diabeticnephropathy. For elucidation of the pathologic consequencesof high levels of VEGF in glomeruli, transgenic (Tg) rabbitsthat express human VEGF165 isoform in both kidney and liverunder the control of the human -1-antitrypsin promoter weregenerated and characterized. With the use of heterozygous Tgrabbits and their littermates aged 8 to 55 wk, renal functionsand structures were investigated. Compared with control rabbits,Tg rabbits exhibited progressive proteinuria with increasedGFR at the early stage and decreased GFR at the later stage.Histologic examinations revealed that Tg rabbit kidneys werecharacterized by considerable glomerular hypertrophy as a resultof increased proliferation of both glomerular capillaries andmesangial cells accompanied by prominent podocyte hypertrophy.With increasing age starting from 20 wk, Tg rabbit kidneys showedprominent formation of microaneurysms and capillary proliferationat the vascular pole area. At a later stage (55 wk), many glomerulishowed sclerosis and tuft collapse with the formation of glomerularcysts on a background of tubular atrophy and interstitial fibrosis.This study provides the first evidence that increased expressionof VEGF in glomeruli directly causes the glomerular hypertrophythat is associated with proteinuria, suggesting that VEGF exertsmultiple effects on the glomerular pathophysiologic processes.
Vascular endothelial growth factor (VEGF) is an important regulatorof angiogenesis. VEGF stimulates endothelial cell proliferationand differentiation, increases vascular permeability, mediatesendothelium-dependent vasodilation, and supports vascular survivalby preventing endothelial apoptosis. In general, VEGF exertsits physiologic functions via binding to specific cellular surfacereceptors, VEGFR-1 through -3 and neuropilin 1 to 2 (see reviewby Ferrara1).
In the kidney, VEGF and its receptors are widely expressed indifferent types of cells. VEGF mRNA is expressed most prominentlyin glomerular podocytes, distal tubules, and collecting ductsand to a lesser extent in some proximal tubules,2,3 whereasVEGF receptors are mainly expressed by endothelial cells ofglomerular and peritubular capillaries.4 VEGF exerts a numberof physiologic and pathophysiologic effects on the kidneys.It is widely known that VEGF is essential for normal nephrogenesis,particularly glomerulogenesis,5 and homozygous deletion of VEGFgenes in glomeruli resulted in perinatal lethality of knockoutmice.6 Ample data suggest that increased VEGF may be involvedin the pathogenesis of glomerulopathies such as diabetic nephropathyand FSGS.3 VEGF and its receptors are upregulated in the diabeticstates of experimental animals.7,8 Eremina et al.6 showed thatpodocyte-specific overexpression of the VEGF164 isoform in miceresulted in renal failure as a result of a collapsing glomerulopathy.Several reports showed that there may be a close relationshipbetween plasma levels of VEGF and proteinuria.9 The data collectedthus far suggest that constitutive expression of VEGF in theglomerular epithelial cells essentially maintains normal glomerularfunctions and provides a filtration barrier, whereas aberrantincreased expression of VEGF may potentially be detrimentalin terms of renal functions, although this hypothesis has notbeen verified.
Although previous studies so far have established a link betweenincreased VEGF and glomerular injury and proteinuria, it remainsunknown whether increased VEGF is causative of or simply a consequenceof these pathologic processes. In fact, there is still controversyregarding the beneficial and deleterious effects of VEGF onthe kidney because some studies even suggested that VEGF mayprotect against renal injury.10 However, the precise elucidationof the direct pathophysiologic effects of VEGF in the previousstudies may have been complicated by the presence of other factors,such as advanced glycation end products, plasminogen, and highlevels of plasma glucose. It is also not clear whether glomerularVEGF can directly affect the filtration of the glomerulus orplays a central role in the pathogenesis of glomerulopathiesand proteinuria.
One of the valid and powerful methods to answer this "cause-and-effect"question may rely on the application of transgenic (Tg) animalmodels in which VEGF is overexpressed in glomeruli while excludingother interfering factors such as high levels of blood glucose.To achieve this goal, our laboratory generated Tg rabbits tostudy the relationship between VEGF and glomerulopathies. Therabbit was selected for this undertaking because of its usefulnessin the study of vascular biology and because the large sizeof rabbits compared with rodents may also facilitate many biochemicalanalyses of the plasma and surgical intervention.11 In thisstudy, we characterized VEGF Tg rabbits and found that overexpressionof VEGF in the glomerular cells has profound effects on thekidney pathologies.
Characterization of VEGF Expression in Tg Rabbits
Analysis using real-time reverse transcriptase–PCR revealedthat the renal expression of human VEGF (hVEGF) was 4.5% (arbitraryunits) of the hepatic expression (Figure 1A) and the total VEGFmRNA expression (sum of human transgene and rabbit endogenousgene) in the whole kidneys of Tg rabbits was slightly increasedcompared with that of control rabbits (19% increase over thecontrol kidneys on average, arbitrary units). To determine whetherhVEGF was expressed in glomeruli, we isolated glomeruli fromtubules by a sieving method and found that the expression ofhVEGF indeed occurred in the glomeruli of Tg rabbits (Figure 1B).Tg rabbit glomeruli were hypertrophic compared with controlglomeruli (Figures 3 and 4). Immunohistochemical staining showedthat the hypertrophic podocytes in the hypertrophic glomeruliof Tg rabbits exhibited discernible VEGF staining, whereas thetubules of both Tg and control rabbits expressed VEGF (Figure 1C).
Figure 1. Detection of human vascular endothelial growth factor (hVEGF) expression in transgenic (Tg) rabbit (at 20 wk) kidney and liver (A) and glomeruli (B) by real-time reverse transcriptase–PCR as described in the Materials and Methods section. (C) Paraffin sections were immunohistochemically stained with hVEGF antibody. Compared with the controls (left), Tg rabbit glomeruli were larger in size (middle). VEGF staining was clearly present in hypertrophic podocytes of Tg rabbits (right). Arrowheads are VEGF-positive podocytes of Tg rabbits.
Figure 3. Micrographs of glomeruli of control and Tg rabbits at 12 wk. Paraffin sections from rabbits were stained with either periodic acid-Schiff (PAS; top) or periodic acid-methenamine silver (PAM; bottom). Compared with the controls, the glomeruli of Tg rabbits showed increased proliferation of endothelial cells and mesangial cells and increased mesangial matrix contents.
Figure 4. Morphometric analysis of glomerular area and cell number. PAM-stained sections were examined for glomerular area (x103/µm2) and total cell number per glomerulus (A and B). The number of endothelial cells and mesangial cells were further calculated on the basis of the morphology on PAS-stained sections (C and D). PAM-stained area was measured and expressed as mesangial matrix content (%; E). Data are means ± SEM; n = approximately 5 to 6 rabbits for each group. *P < 0.05, **P < 0.01 versus control.
To determine whether a certain amount of hVEGF may be presentin the circulation of Tg rabbits, we measured the plasma VEGFusing two different ELISA with human VEGF as a positive control.As shown in Table 1, Biotrak ELISA kits revealed that the plasmalevels of VEGF were 1.2-fold higher in Tg rabbits than thatin control rabbits, although not statistically significant (P= 0.272). CytELISA (hVEGF specific) kits along with Westernblotting analysis (data not shown), however, did not detectthe presence of human VEGF in the plasma of Tg rabbits. Thesedata suggest that the transgenic hVEGF proteins were indeedexpressed locally or associated with extracellular matrix (ECM),and little, if any, were released into the circulation. Comparedwith control rabbits, Tg rabbits had slightly lower body weightwith relatively enlarged kidneys (Table 2).
Renal Functional Changes in Tg Rabbits
To evaluate the effects of increased expression of VEGF on renalfunctions, we first examined whether Tg rabbits exhibited proteinuria.Proteinuria was not detected in Tg rabbits until 12 wk and graduallybecame prominent thereafter with increasing age and reacheda plateau at 28 wk (Figure 2A). To quantify the contents ofurinary proteins, we analyzed the 24-h urinary proteins thatwere collected from rabbits at various ages when proteinuriawas markedly increased. As shown in Table 3, the levels of urinaryproteins in Tg rabbits were 12.7-, 34-, and 33-fold higher thanthose in control rabbits at 23, 28, and 42 wk. SDS-PAGE analysisdemonstrated that the major urinary protein was albumin on thebasis of its molecular size of approximately 68 kD (Figure 2A,insert). In addition, GFR of Tg rabbits at 23 wk was increasedby 10% but progressively declined by 45 and 35% at 30 and 42wk, respectively, compared with that in control rabbits (Figure 2B),suggesting that upregulation of VEGF impairs the glomerularbarrier functions of Tg rabbits. Biochemical analysis of theplasma showed that Tg rabbits had hyperlipidemia and lower levelof albumin at 30 wk, whereas their glucose, blood urea nitrogen,creatinine, and total proteins were not significantly differentfrom control rabbits (Table 4). To exclude the possibility thathepatic dysfunctions of Tg rabbits may be responsible for loweralbumin levels in plasma, we measured the plasma glutamic-oxaloacetictransaminase (GOT) and glutamic pyruvic transaminase (GPT) andfound that there was no significant difference between controland Tg rabbits at 30 wk (GPT 32.2 ± 10.9 IU/L in controlversus 44.3 ± 15.9 IU/L in Tg; GOT 40.5 ± 9.1IU/L in control versus 46 ± 13.3 IU/L in Tg).
Figure 2. Measurements of urinary proteins (A) and GFR (B). Urinary protein levels were determined in rabbits at various ages as described in the Concise Methods section. Urine samples were also analyzed using SDS-PAGE (A, insert). GFR was measured at the ages of 23, 30, and 42 wk and expressed as percentage of the control. Data are means ± SEM; n = approximately 5 to 6 rabbits for each group. *P < 0.05, **P < 0.01 versus control.
Table 4. Comparison of plasma parameters of VEGF Tg rabbits at various agesa
Histopathologic Examinations
Because Tg rabbits exhibited apparent renal dysfunction as describedabove, we next examined the pathologic changes of the kidneysof Tg rabbits at various ages. For the convenience of descriptionsof glomerular changes, we arbitrarily divided the renal lesionsof Tg rabbits into (1) early stage (preproteinuria stage, before8 wk), (2) intermediate stage (12, 20, and 35 wk), and (3) laterstage (55 wk) according to the appearance and extent of proteinuriaand severity of renal pathology.
At first glance, the renal histologic morphology of Tg rabbitsstarting from 8 to 20 wk was characterized by progressive glomerularhypertrophy with increased cell numbers per glomerulus (Figure 3).The glomerular size in Tg rabbits was increased by 1.17-, 1.76-,and 2.4-fold at 8, 12, and 20 wk compared with that in controlrabbits (Figure 4A). The glomerular hypertrophy in Tg rabbitswas caused by an increase of the number of cells, includingboth endothelial cells and mesangial cells on the basis of themorphologic features and immunohistochemical staining with CD31mAb (Figures 4, B through D, and 5).
Figure 5. Immunohistochemical staining using CD31 mAb. Rabbits at 35 wk were used. Compared with normal glomerulus (left), Tg glomerular CD31-positive endothelial cells are increased in number (middle and right).
It should be noted that in the early stage, even though proteinuriawas still not detected, the number of endothelial cells in Tgglomeruli was significantly increased. In the intermediate stage,proteinuria started to became obvious starting from 12 wk; however,mesangial cells along with endothelial cells were concomitantlyincreased in Tg rabbits compared with those of control rabbits(Figure 4, C and D). Furthermore, the mesangial matrix contentswere significantly increased at 20 wk in Tg rabbits comparedwith the controls (Figures 3 and 4E).
Another important histologic feature of the glomeruli of Tgrabbits was the appearance of pronounced hypertrophy of podocytes(Figure 6A), whereas the number of podocytes (Wilms’ tumor-1[WT-1]-positive cells) per glomerulus was slightly increased(average podocyte number per glomerulus 7.7 ± 1.8 incontrol versus 9.6 ± 3.0 in Tg at 20 wk; P < 0.001;Figure 6B). Hypertrophic podocytes were larger in size thanthose of the controls (three-fold increase on average; P <0001) and were characterized by a prominent large single nucleusor occasionally two nuclei (Figure 6A; see also transmissionelectron microscopy [TEM] results).
Figure 6. Demonstration of podocytes in the glomerulus. The podocytes of Tg rabbits are increased in size and contained one or two large nuclei, shown by PAM staining (A). Arrowheads are podocytes of control and Tg rabbits. The podocytes can be stained by mAb against Wilms’ tumor 1 (WT-1) as shown in B.
TEM examinations (Figure 7A) confirmed that podocyte hypertrophywas the major feature of Tg glomeruli, as shown by light microscopy.In addition, compared with normal rabbit podocytes, Tg rabbitpodocytes showed loss of slit pores, and their foot processesappeared fused together and effaced (Figure 7B), which may bethe major structural abnormalities responsible for proteinuriain Tg rabbits. Whereas mesangial cells appear large under TEM,endothelial cells exhibited several abnormal alterations thatwere not present in normal rabbit. Endothelial cells were oftenloosely attached on the ECM as a result of subendothelial edema,and some showed apoptotic changes (Figure 7, A and C).
Figure 7. Transmission electron microscopic observation of glomerular cells. (A) Representative pictures of glomerular cells from control and Tg rabbits at 25 wk. Compared with control rabbit glomeruli (left), capillary loops of Tg rabbit are dilated and endothelial cells (EC) show subendothelial edema (arrow, right). Podocytes (PC) of Tg rabbits are clearly large in size, and their foot processes are not clear compared with that of control rabbits. Mesangial cells (MC) are also large in Tg rabbits. (B) Compared with the control glomerulus in which podocyte foot processes are arranged regularly perpendicular to the capillary loop (CL) basement membrane (top left), Tg podocyte foot processes appear irregular or fused together (top right) or effaced along the surface of the capillary basement membrane (bottom). (C) Endothelial cells of Tg glomerulus show marked subendothelial edema (left, arrows), subendothelial vacuolization (top right, arrow), and apoptotic changes (bottom right).
From 20 wk, the proliferation of endothelial cells and mesangialcells reached the peak value of the experimental periods (Figure 4,C and D), and the hypertrophic glomeruli of Tg rabbits wereuniformly characterized by two remarkable pathologic findings:The formation of so-called microaneurysms12 and marked endothelialcell proliferation in the vascular pole area (Figure 8). Microaneurysmswere formed by the dilated capillaries and were considered tobe caused by mesangiolysis.12 They were either in a multiple"glandular" or in a lobular pattern or combined into a largedilated vessel (Figure 8A). In the vascular pole area, endothelialcells proliferated and aggregated or formed a vascular channel(Figure 8B). The proliferation index (5-bromo-2-deoxyuridine[BrdU] incorporation) of glomerular cells was significantlyhigher in Tg rabbits than that in control rabbits (3.1 ±0.4% in Tg versus 0.9 ± 0.3% in control rabbits; P <0.01; Figure 9).
Figure 8. Representative micrographs of glomeruli from Tg rabbits at 35 wk. Paraffin sections were stained with PAM, and glomeruli with the formation of microaneurysms (A) or endothelial cell proliferation in the vascular pole area (B) were seen.
Figure 9. Representative pictures showing the proliferating cells in glomerulus. 5-Bromo-2-deoxyuridine (BrdU)-positive cells were stained as described in the Concise Methods section. Tg rabbit (at 39 wk) glomerulus contains more BrdU-positive cells than does control glomerulus.
In the later stage (55 wk), most glomeruli in Tg rabbits showedmore severe and more complicated lesions, including evidenceof the healing process, after glomerular injury. In additionto the lesions (proliferation of capillary and mesangial cellsand microaneurysms) described, the majority of glomeruli showedeither global or segmental sclerosis or tuft collapse (partiallyor completely) or glomerular cysts (Figure 10, A and B). Insclerotic glomeruli, some cells around the Bowman's capsulewere positively stained with mAb against -smooth muscle actin(Figure 10, C through E). Apart from the lesions seen in glomeruli,tubular atrophy accompanied by interstitial fibrosis was frequentlyfound throughout in Tg rabbit kidneys (data not shown). Therewas no macrophage infiltration in the glomeruli of either Tgor control rabbits, as confirmed by immunohistochemical staining(data not shown).
Figure 10. Representative micrographs of glomeruli from Tg rabbits at 55 wk. Paraffin sections were stained with PAS (A), PAM (B and C), Azan (D), and -smooth muscle actin (E). (A) Glomeruli showed either global or partial tuft collapse accompanied by cystic formation. (B) At higher magnification, glomeruli showed obvious tuft collapse with sclerosis and Bowman's capsule fibrosis. A glomerulus with severe global sclerosis is shown (C through E), and increased fibrosis (D) and -smooth muscle actin–positive cells were present (E).
In this study, we characterized Tg rabbits that expressed thehVEGF165 isoform in the kidney in an attempt to evaluate theeffect of increased VEGF on kidney pathophysiology. Human VEGFtransgene expression was under the control of -1-antitrypsinpromoter, so the transgene was mainly expressed in the liverand, to a lesser extent, in the kidney.13 Although this promoteris not kidney specific, we found that Tg rabbits showed a mildincrease of VEGF expression in the glomeruli (approximately19% increase of mRNA expression over that in control rabbitkidneys). hVEGF proteins were not detected in the plasma ofTg rabbits by ELISA and Western blotting. This result was initiallysurprising but may suggest that either hVEGF was associatedwith plasma inhibitors or the levels of plasma hVEGF were toolow to be detected or hVEGF may be bound to the ECM. Despitethis, our studies suggest that increased VEGF expression inkidney and liver has dramatic and multiple biologic effectson both glomerular cells and ECM production.
VEGF and Glomerular Hypertrophy
Compared with control rabbit glomeruli, the glomeruli of Tgrabbits were markedly hypertrophic, and the hypertrophy wasassociated with increased capillaries and mesangial cell proliferation.It is unlikely that these lesions were mediated by immune dysfunctions,because IgG staining was negative (data not shown). It seemsthat increased VEGF expression initially induces proliferationof endothelial cells and later induces that of mesangial cells,even though both types of cells express VEGF receptors.4,14In this respect, Ostendorf et al.15 showed that administrationof VEGF165 antagonist led to a reduction of glomerular endothelialcell proliferation and an increase in endothelial cell deathin nephritis, whereas proteinuria was not improved. Of note,podocytes were markedly hypertrophic (up to three-fold increasein size over the control) in Tg rabbits, accompanied by theloss of podocyte processes as observed by TEM. The mechanismby which VEGF induces podocyte hypertrophy is unknown but maypossibly be mediated in an autocrine manner because podocytesnot only produce VEGF but also express VEGF receptors.16,17Although this hypothesis remains to be verified, podocytes mayhave the potential to bind to the VEGF that they secrete themselves.Consistent with the previous finding that VEGF can also stimulatepodocytes to proliferate and increase their survival in vitro,18we found that hypertrophic podocytes in Tg rabbits were increasedin number and had two or more nuclei, suggesting that VEGF mayenhance podocyte growth in vivo.
The pathologic changes that were observed in Tg rabbits at laterstages were also interesting. For example, glomerular sclerosisas defined by the periodic acid-methenamine silver (PAM)-stainedarea was markedly increased with age and accompanied by severeproteinuria. Such features of Tg rabbit glomeruli may mimicmany pathologic characteristics of renal diseases, such as FSGSin humans,19 suggesting that the upregulation of VEGF may bea potentially causative factor for the progression of FSGS.Two mechanisms may be operative in terms of VEGF-induced glomerulosclerosis.First, it has been reported that glomerular hypertrophy candirectly cause segmental glomerulosclerosis via podocyte damage.20,21Second, VEGF derived from podocytes may stimulate mesangialcells to proliferate and synthesize ECM.22 In support of thisspeculation, both mesangial cells and ECM were markedly increasedin the glomeruli of Tg rabbits after 20 wk. Of note, massiveformation of glomerular microaneurysms was another feature ofthe Tg rabbits after 20 wk. Glomerular microaneurysms can beinduced in experimental mesangioproliferative anti–Thy-1nephritis.12 Although the pathologic significance of such lesionsis unknown, mesangial cell damage (mesangiolysis) is proposedto be responsible for the formation of microaneurysms. If thisis really operative, it remains to be determined whether VEGFeither directly or indirectly induces mesangiolysis in future.
VEGF and Proteinuria
Podocytes play an important role in maintaining the glomerularfiltration barrier because podocytes contribute structurallyby the provision of the slit diaphragm and glomerular basementmembrane.23 Glomerular VEGF derived from podocytes is also involvedin the maintenance of the glomerular endothelium (includingmaintaining fenestration) and/or selective permeability to macromolecules.15,23Therefore, it is reasonable to speculate that hypertrophic podocytesthat are observed in Tg rabbits may lose their normal physiologicfunction of maintaining the macromolecular selectivity of theglomerular filtration barrier as one of the mechanisms thatare responsible for the prominent proteinuria that is observedin Tg rabbits. In Tg rabbits, the proteinuria started to becomeobvious from 12 wk, suggesting that the deleterious effect ofVEGF was progressive and required a certain time rather thanstarting from embryogenesis. Previous reports have shown thatincreased plasma levels of VEGF may be associated with proteinuriain patients with diabetes,9 and the administration of VEGF antibodiesameliorates proteinuria in diabetic nephropathy.24,25 Our resultsare in support of the notion that VEGF is an important mediatorof glomerular functions and that the chronic upregulation ofVEGF in podocytes leads to prominent and progressive proteinuria.Increased VEGF in the glomeruli may result in proteinuria throughtwo possible different yet closely overlapping or related mechanisms.In the early stage, in addition to the abnormalities of hypertrophicpodocytes discussed, high levels of VEGF derived from podocytescan strongly bind to capillary endothelial cells through specificVEGF receptors, which may result in increased permeability orglomerular hyperfiltration by altering capillary fenestrationor basement membrane components26 or indirectly through theinduction of nitric oxide and prostacyclin.27 In our study,we found that many endothelial cells in glomeruli showed subendothelialedema (Figure 7C). It has been reported that VEGF stimulatesincreased synthesis of collagenase by endothelial cells, whichwould result in the proteolytic disruption of the basement membrane.22In the later stage, however, there were severe secondary glomerularinjuries, including microaneurysms, tuft collapse, and glomerularsclerosis along with tubular injuries, all of which may participatein the enhancement of proteinuria. Therefore, the mechanismsthat are responsible for proteinuria in Tg rabbits may be differentin the early stage (mainly primary effect directly mediatedby VEGF) from those in the later stage (both VEGF-induced effectsand/or secondary effects on glomerular injury). Consistent withthis notion, our study showed that in Tg rabbits, there wasa increase of GFR at 23 wk, but later on, GFR was progressivelydecreased, possibly reflecting the severe glomerulosclerosis.
Although Eremina et al.6 reported that podocyte-specific overexpressionof the VEGF164 led to a striking collapsing glomerulopathy inmice, our Tg rabbits apparently differed from Tg mice. First,unlike Tg mice, which showed kidney abnormalities and albuminuriaas early as 5 d of age, Tg rabbits showed normal kidney developmentuntil 8 wk and showed proteinuria after 12 wk. This discrepancybetween Tg mice and rabbits is probably due to the species differencesand gene expression (levels and specificity) or developmentalexpression (from capillary loop stage of mice) versus laterexpression (in the rabbit model). Tg rabbits showed a mild increaseof VEGF expression (1.2-fold increase over control) comparedwith Tg mice (30-fold increase over control), and transgeneexpression in Tg rabbits may not be podocyte-specific becausesome VEGF may be derived from tubuli or liver. In addition,Tg rabbits exhibited dynamic and diverse lesions in the kidneys,including glomerular hypertrophy (at the early stage), prominentglomerulosclerosis, microaneurysms, and vascular pole sclerosis(from the intermediate to later stages), whereas collapsingglomerulopathy was observed only at the age of 55 wk. Importantly,VEGF-induced glomerulopathies have several implications regardingthe pathologic roles of VEGF in human kidney diseases such asdiabetic nephropathies and FSGS, in which VEGF is upregulated7,8and have been proposed to play a critical role in the pathogenesisof proteinuria.9 High concentrations of ambient glucose inducethe expression of VEGF in both podocytes and mesangial cells28–30
Our study demonstrated that local overexpression of the hVEGF165gene in the podocytes of Tg rabbits resulted in glomerular hypertrophyand increased proliferation of both endothelial cells and mesangialcells and glomerular sclerosis associated with progressive proteinuria.Podocytes were markedly hypertrophic and showed effacement oftheir foot processes. These data support the hypothesis thatVEGF is involved in the maintenance of the glomerular barrierand in the pathogenesis of proteinuria. It is interesting thatTg rabbits exhibited many pathologic features similar to thosefound in human FSGS. These results suggest that local upregulationof VEGF may play diverse roles in glomerulopathies and proteinuria.Inhibition of VEGF in the glomeruli may become a potential targetfor the treatment of kidney diseases such as diabetic nephropathiesand FSGS.
Tg Rabbits Expressing hVEGF
Tg rabbits that express hVEGF165 cDNA under the control of thehuman -1-antitrypsin promoter31 were generated in our laboratory.32A previous study showed that the -1-antitrypsin promoter containedan element that was able to direct the expression of the transgenein the kidney in addition to the liver.13 Three founders wereobtained, but one founder died before breeding. Autopsy examinationrevealed that this founder had hepatic and kidney lesions. Theblood levels of hVEGF of all three founders were below the detection).For this study, heterozygous male Tg rabbits and control littermatesfrom V58 line were analyzed at the ages of 8 to 55 wk. All rabbitswere kept individually in a room where the temperature and humiditywere maintained at 24 ± 2°C and 55 ± 15%,respectively. The rabbits were submitted to a regular 12-h light/darkcycle and given water and a standard chow diet (CBR-1; CLEA,Tokyo, Japan) ad libitum. All experimental protocols were approvedby the Saga University Animal Experimentation Committee andperformed under the Saga University Guidelines for Animal Experimentation.For monitoring the health status of Tg rabbits, blood was collectedfrom rabbits after overnight fasting (16 h), and the plasmaalbumin, blood urea nitrogen, creatinine, glucose, total protein,GOT, and GPT were determined using a compact VetScan unit (Abaxia,Union City, CA) in a Point-of-care blood analyzer (Abaxia, Sunnyvale,CA).
Reverse Transcriptase–PCR Analysis
Total RNA from different tissues was rapidly isolated usingTrizol reagent (Invitrogen, Tokyo, Japan) and reverse-transcribedinto cDNA using Qiagen reverse transcription reagents (Tokyo,Japan), and real-time reverse transcriptase–PCR (DNA EngineOpticon; MJ Research, Tokyo, Japan) was performed using DyNAmoSYBR Green qPCR kits (Finnzymes, Espoo, Finland) according tothe manufacturer's instructions.33 We measured the human-specificVEGF and rabbit endogenous VEGF using specific primers as shownin Table 5. In addition, we separated glomeruli from tubulesusing a sieving method34 to examine the hVEGF mRNA expressionin glomeruli.
Table 5. Primers for real-time reverse transcriptase–PCRa
Urinary Protein Examination
Urine was collected in the morning and examined for urobilinogen,blood, bilirubin, ketones, glucose, protein, pH, leukocytes,specific gravity, and ascorbic acid using Protest 10II strips(Wako Pure Chemical Industries, Osaka, Japan). The total amountsof urinary proteins excreted during 24 h were measured fromrabbits at 23, 28, and 42 wk using protein assay kits (Bio-RadLaboratories, Hercules, CA). In addition, urinary proteins werefractionated by 10% SDS-PAGE.
Evaluation of Renal Functions
The GFR was indirectly evaluated by measurement of the inulinclearance rate in the blood. Briefly, inulin (Wako) was dissolvedin saline and infused into rabbits (which had fasted for 16h) at a dosage of 100 mg/kg body wt at a constant rate over30 s through the auricular vein. Blood samples were collectedthrough the intermediate auricular artery at 3, 10, 20, 40,80, 120, 180, and 240 min. The concentration of serum inulinwas determined by the indole-3-acetic acid (Sigma, St. Louis,MO) colorimetric assay as described previously.35 GFR was estimatedby measurement of serum clearance of inulin according to theformula derived from Brown36: GFR (ml/min per kg body wt) =D/AUC/body wt. D is the amount of injected inulin, and AUC (mgx min/ml) is the total area under the serum inulin disappearancecurve. The AUC was calculated according to the trapezium ruledescribed previously.37
Histologic Examinations and Immunohistochemistry
For the morphologic analysis of kidneys, Tg and littermate controlrabbits were killed at various ages and the kidneys were removedand fixed in a 10% neutral buffered formalin solution. The middlepart of the left and right kidneys was embedded in paraffin,and serial sections were stained with hematoxylin and eosin,periodic acid-Schiff, Azan, and PAM, which were used for morphometricanalysis (see next section). In addition, paraffin sectionswere immunohistochemically stained using mAb against rabbitmacrophages (RAM11; 1:200; Dako, Carpinteria, CA); -smooth muscleactin (1:400; Dako), and hVEGF (ab3; 1:100; Neomarkers, Fremont,CA), which showed cross-reaction with rabbit VEGF. In addition,we performed immunohistochemical staining using two cellularmarkers for human kidney, including endothelial cell markersCD31 and podocyte markers WT-1.38 We found that two CD31 andWT-1 could cross-react with rabbit kidneys. Immunohistochemistrywas performed with the Dako En-vision system following the manufacturer'sinstructions.
Morphometric Analysis
For the quantitative analysis of glomerular changes includingglomerular size, cellular number, matrix expansion, or fibrosis,PAM-stained sections were examined under a light microscopeand 40 constitutive images for each section were recorded usingan Olympus DP70 digital camera (Tokyo, Japan). Glomerular sizeand mesangial matrix area were manually traced and measuredusing the Lumina Vision image analysis system (Mitani, Tokyo,Japan). Mesangial matrix area was defined as the PAM-stainedarea (%) within the tuft area as described by Inada et al.39We calculated the cellular numbers using periodic acid-Schiff–stainedsections on the basis of the cellular morphology and structureunder light microscopy. All analyses were performed by two independentresearchers blindly. For each assay, at least 40 consecutiveintact glomeruli from each section were evaluated.
Proliferation Index in Glomerulus
To evaluate whether hypercellularity in the glomeruli of Tgrabbits was caused by increased proliferation, we examined theproliferating cells using the thymidine analogue BrdU labelingtechnique.40 The proliferation index of glomeruli of Tg rabbitswas compared with that of control rabbits at 40 wk. BrdU (50mg/kg) was subcutaneously injected at 1, 8, 16, and 24 h beforethe removal of the kidney. BrdU-positive cells were reactedwith a mouse mAb against BrdU (1:100; Dako, Copenhagen, Denmark),followed by goat anti-mouse IgG antibodies conjugated to peroxidase,and detected with diacylaminobentidine. The number of BrdU-positivenuclei were counted from approximately 50 to 60 consecutiveglomeruli per rabbit in a blinded manner. The BrdU labelingindex was expressed as BrdU incorporation or percentage of totalglomerular cells.
TEM
A small piece of kidney from rabbits at the age of 25 wk wasfixed in 2.5% glutaraldehyde and postfixed in 1% osmium tetroxideusing a standard protocol. Ultrathin sections were doubly stainedwith uranyl acetate and lead citrate and were observed underan JEM-100CX electron microscope (JEOL Ltd, Tokyo, Japan) asdescribed previously.32
Analysis of Plasma VEGF
To determine whether Tg rabbits had high levels of VEGF in plasma,we used two ELISA kits that had different specificity. CytELISAkits (CytImmune Sciences, Rockville, MD) can detect both humanand rabbit endogenous VEGF, whereas Biotrak ELISA kits (AmershamPharmacia Biotech, Buckinghamshire, UK) were specific for hVEGF(without cross-reaction with rabbit VEGF). To ensure the precisedetermination of hVEGF in Tg rabbit plasma, we used human recombinantVEGF (R&D Systems, Minneapolis, MN) as a positive controlin all ELISA assays. Because VEGF may bind to plasma proteins,which might interfere with ELISA results, we also collectedthe plasma from a normal rabbit that was administered an intravenousinjection of 400 ng/kg human recombinant VEGF as a positivecontrol. The same samples were also fractionated by 10% SDS-PAGEand analyzed by Western blotting.41
Statistical Analyses
Results are expressed as means ± SEM. Statistical analyseswere performed using either the t test when the F value wasequal or Welch t test when the F value was not equal. P <0.05 was considered statistically significant.
This work was supported in part by grants-in-aid for scientificresearch from the Ministry of Education, Culture, Sports andTechnology, Japan (19790226 and 19390099 to J.F.); a researchgrant from AstraZeneca, the Takeda Science Foundation; JapanHeart Foundation; and the Naito Foundation for the Promotionof Science.
We thank F. Muto, H. Ideguchi, and S. Nakahara for technicalassistance.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Ferrara N: Vascular endothelial growth factor: Basic science and clinical progress.
Endocr Rev 25
: 581
–611, 2004[Abstract/Free Full Text]
Simon M, Rockl W, Hornig C, Grone EF, Theis H, Weich HA, Fuchs E, Yayon A, Grone HJ: Receptors of vascular endothelial growth factor/vascular permeability factor (VEGF/VPF) in fetal and adult human kidney: Localization and [125I]VEGF binding sites.
J Am Soc Nephrol 9
: 1032
–1044, 1998[Abstract]
Schrijvers BF, Flyvbjerg A, De Vriese AS: The role of vascular endothelial growth factor (VEGF) in renal pathophysiology.
Kidney Int 65
: 2003
–2017, 2004[CrossRef][Medline]
Robert B, Zhao X, Abrahamson DR: Coexpression of neuropilin-1, Flk1, and VEGF(164) in developing and mature mouse kidney glomeruli.
Am J Physiol Renal Physiol 279
: F275
–F282, 2000[Abstract/Free Full Text]
Kitamoto Y, Tokunaga H, Tomita K: Vascular endothelial growth factor is an essential molecule for mouse kidney development: Glomerulogenesis and nephrogenesis.
J Clin Invest 99
: 2351
–2357, 1997[Medline]
Eremina V, Sood M, Haigh J, Nagy A, Lajoie G, Ferrara N, Gerber HP, Kikkawa Y, Miner JH, Quaggin SE: Glomerular-specific alterations of VEGF-A expression lead to distinct congenital and acquired renal diseases.
J Clin Invest 111
: 707
–716, 2003[CrossRef][Medline]
Cooper ME, Vranes D, Youssef S, Stacker SA, Cox AJ, Rizkalla B, Casley DJ, Bach LA, Kelly DJ, Gilbert RE: Increased renal expression of vascular endothelial growth factor (VEGF) and its receptor VEGFR-2 in experimental diabetes.
Diabetes 48
: 2229
–2239, 1999[Abstract]
Tsuchida K, Makita Z, Yamagishi S, Atsumi T, Miyoshi H, Obara S, Ishida M, Ishikawa S, Yasumura K, Koike T: Suppression of transforming growth factor beta and vascular endothelial growth factor in diabetic nephropathy in rats by a novel advanced glycation end product inhibitor, OPB-9195.
Diabetologia 42
: 579
–588, 1999[CrossRef][Medline]
Kim NH, Oh JH, Seo JA, Lee KW, Kim SG, Choi KM, Baik SH, Choi DS, Kang YS, Han SY, Han KH, Ji YH, Cha DR: Vascular endothelial growth factor (VEGF) and soluble VEGF receptor FLT-1 in diabetic nephropathy.
Kidney Int 67
: 167
–177, 2005[CrossRef][Medline]
Miyamoto K, Kitamoto Y, Tokunaga H, Takeya M, Ezaki T, Imamura T, Tomita K: Protective effect of vascular endothelial growth factor/vascular permeability factor 165 and 121 on glomerular endothelial cell injury in the rat.
Lab Invest 84
: 1126
–1136, 2004[CrossRef][Medline]
Fan J, Watanabe T: Transgenic rabbits as therapeutic protein bioreactors and human disease models.
Pharmacol Ther 99
: 261
–282, 2003[CrossRef][Medline]
Kriz W, Hahnel B, Hosser H, Ostendorf T, Gaertner S, Kranzlin B, Gretz N, Shimizu F, Floege J: Pathways to recovery and loss of nephrons in anti-Thy-1 nephritis.
J Am Soc Nephrol 14
: 1904
–1926, 2003[Abstract/Free Full Text]
Ruther U, Tripodi M, Cortese R, Wagner EF: The human alpha-1-antitrypsin gene is efficiently expressed from two tissue-specific promotors in transgenic mice.
Nucleic Acids Res 15
: 7519
–7529, 1987[Abstract/Free Full Text]
Thomas S, Vanuystel J, Gruden G, Rodriguez V, Burt D, Gnudi L, Hartley B, Viberti G: Vascular endothelial growth factor receptors in human mesangium in vitro and in glomerular disease.
J Am Soc Nephrol 11
: 1236
–1243, 2000[Abstract/Free Full Text]
Ostendorf T, Kunter U, Eitner F, Loos A, Regele H, Kerjaschki D, Henninger DD, Janjic N, Floege J: VEGF(165) mediates glomerular endothelial repair.
J Clin Invest 104
: 913
–923, 1999[Medline]
Harper SJ, Xing CY, Whittle C, Parry R, Gillatt D, Peat D, Mathieson PW: Expression of neuropilin-1 by human glomerular epithelial cells in vitro and in vivo.
Clin Sci (Lond) 101
: 439
–446, 2001[Medline]
Chen S, Kasama Y, Lee JS, Jim B, Marin M, Ziyadeh FN: Podocyte-derived vascular endothelial growth factor mediates the stimulation of alpha3(IV) collagen production by transforming growth factor-beta1 in mouse podocytes.
Diabetes 53
: 2939
–2949, 2004[Abstract/Free Full Text]
Foster RR, Hole R, Anderson K, Satchell SC, Coward RJ, Mathieson PW, Gillatt DA, Saleem MA, Bates DO, Harper SJ: Functional evidence that vascular endothelial growth factor may act as an autocrine factor on human podocytes.
Am J Physiol Renal Physiol 284
: F1263
–F1273, 2003[Abstract/Free Full Text]
Magil AB: Ideas in pathology. Focal and segmental glomerulosclerosis.
Mod Pathol 4
: 383
–391, 1991[Medline]
Fries JW, Sandstrom DJ, Meyer TW, Rennke HG: Glomerular hypertrophy and epithelial cell injury modulate progressive glomerulosclerosis in the rat.
Lab Invest 60
: 205
–218, 1989[Medline]
Nagata M, Kriz W: Glomerular damage after uninephrectomy in young rats. II. Mechanical stress on podocytes as a pathway to sclerosis.
Kidney Int 42
: 148
–160, 1992[Medline]
Unemori EN, Ferrara N, Bauer EA, Amento EP: Vascular endothelial growth factor induces interstitial collagenase expression in human endothelial cells.
J Cell Physiol 153
: 557
–562, 1992[CrossRef][Medline]
Mundel P, Kriz W: Structure and function of podocytes: An update.
Anat Embryol (Berl) 192
: 385
–397, 1995[Medline]
de Vriese AS, Tilton RG, Elger M, Stephan CC, Kriz W, Lameire NH: Antibodies against vascular endothelial growth factor improve early renal dysfunction in experimental diabetes.
J Am Soc Nephrol 12
: 993
–1000, 2001[Abstract/Free Full Text]
Flyvbjerg A, Dagnaes-Hansen F, De Vriese AS, Schrijvers BF, Tilton RG, Rasch R: Amelioration of long-term renal changes in obese type 2 diabetic mice by a neutralizing vascular endothelial growth factor antibody.
Diabetes 51
: 3090
–3094, 2002[Abstract/Free Full Text]
Chen S, Lee JS, Iglesias-de la Cruz MC, Wang A, Izquierdo-Lahuerta A, Gandhi NK, Danesh FR, Wolf G, Ziyadeh FN: Angiotensin II stimulates alpha3(IV) collagen production in mouse podocytes via TGF-beta and VEGF signalling: Implications for diabetic glomerulopathy.
Nephrol Dial Transplant 20
: 1320
–1328, 2005[Abstract/Free Full Text]
Zachary I: Vascular endothelial growth factor: How it transmits its signal.
Exp Nephrol 6
: 480
–487, 1998[CrossRef][Medline]
Kim NH, Jung HH, Cha DR, Choi DS: Expression of vascular endothelial growth factor in response to high glucose in rat mesangial cells.
J Endocrinol 165
: 617
–624, 2000[Abstract]
Hoshi S, Nomoto K, Kuromitsu J, Tomari S, Nagata M: High glucose induced VEGF expression via PKC and ERK in glomerular podocytes.
Biochem Biophys Res Commun 290
: 177
–184, 2002[CrossRef][Medline]
Hoshi S, Shu Y, Yoshida F, Inagaki T, Sonoda J, Watanabe T, Nomoto K, Nagata M: Podocyte injury promotes progressive nephropathy in Zucker diabetic fatty rats.
Lab Invest 82
: 25
–35, 2002[Medline]
Garcia I, Miyazaki Y, Araki K, Araki M, Lucas R, Grau GE, Milon G, Belkaid Y, Montixi C, Lesslauer W, et al.: Transgenic mice expressing high levels of soluble TNF-R1 fusion protein are protected from lethal septic shock and cerebral malaria, and are highly sensitive to Listeria monocytogenes and Leishmania major infections.
Eur J Immunol 25
: 2401
–2407, 1995[Medline]
Kitajima S, Liu E, Morimoto M, Koike T, Yu Y, Watanabe T, Imagawa S, Fan J: Transgenic rabbits with increased VEGF expression develop hemangiomas in the liver: A new model for Kasabach-Merritt syndrome.
Lab Invest 85
: 1517
–1527, 2005[Medline]
Sun H, Koike T, Ichikawa T, Hatakeyama K, Shiomi M, Zhang B, Kitajima S, Morimoto M, Watanabe T, Asada Y, Chen YE, Fan J: C-reactive protein in atherosclerotic lesions: Its origin and pathophysiological significance.
Am J Pathol 167
: 1139
–1148, 2005[Abstract/Free Full Text]
Helwig JJ, Zachary D, Bollack C: Isolation of glomeruli and tubular fragments from rabbit kidney.
Urol Res 2
: 55
–59, 1974[Medline]
Nolin TD, Colaizzi IV, Palevsky PM, Matzke GR, Frye RF: Rapid microtiter plate assay for determination of inulin in human plasma and dialysate.
J Pharm Biomed Anal 28
: 209
–215, 2002[CrossRef][Medline]
Brown SA: Evaluation of a single-injection method for estimating glomerular filtration rate in dogs with reduced renal function.
Am J Vet Res 55
: 1470
–1473, 1994[Medline]
Liu E, Kitajima S, Higaki Y, Morimoto M, Sun H, Watanabe T, Yamada N, Fan J: High lipoprotein lipase activity increases insulin sensitivity in transgenic rabbits.
Metabolism 54
: 132
–138, 2005[CrossRef][Medline]
Kanemoto K, Takahashi S, Shu Y, Usui J, Tomari S, Yan K, Hamazaki Y, Nagata M: Variable expression of podocyte-related markers in the glomeruloid bodies in Wilms tumor.
Pathol Int 53
: 596
–601, 2003[CrossRef][Medline]
Inada A, Nagai K, Arai H, Miyazaki J, Nomura K, Kanamori H, Toyokuni S, Yamada Y, Bonner-Weir S, Weir GC, Fukatsu A, Seino Y: Establishment of a diabetic mouse model with progressive diabetic nephropathy.
Am J Pathol 167
: 327
–336, 2005[Abstract/Free Full Text]
Numaguchi Y, Okumura K, Harada M, Naruse K, Yamada M, Osanai H, Matsui H, Ito M, Murohara T: Catheter-based prostacyclin synthase gene transfer prevents in-stent restenosis in rabbit atheromatous arteries.
Cardiovasc Res 61
: 177
–185, 2004[Abstract/Free Full Text]
Liang J, Liu E, Koike T, Yu Y, Kitajima S, Yingji J, Morimoto M, Watanabe T, Sasaguri Y, Watanabe S, Fan J: Macrophage elastase (MMP-12) accelerates the progression of atherosclerosis in transgenic rabbits.
Circulation 113
: 1993
–2001, 2006[Abstract/Free Full Text]
This article has been cited by other articles:
P. Vieitez, O. Gomez, E. R Uceda, M. E. Vera, and E. Molina-Holgado Systemic and local effects of angiotensin II blockade in experimental diabetic nephropathy
Journal of Renin-Angiotensin-Aldosterone System,
June 1, 2008;
9(2):
96 - 102.
[Abstract][PDF]
Top
Abstract
Introduction
-1-antitrypsin promoter were generated and characterized. With the use of heterozygous Tg rabbits and their littermates aged 8 to 55 wk, renal functions and structures were investigated. Compared with control rabbits, Tg rabbits exhibited progressive proteinuria with increased GFR at the early stage and decreased GFR at the later stage. Histologic examinations revealed that Tg rabbit kidneys were characterized by considerable glomerular hypertrophy as a result of increased proliferation of both glomerular capillaries and mesangial cells accompanied by prominent podocyte hypertrophy. With increasing age starting from 20 wk, Tg rabbit kidneys showed prominent formation of microaneurysms and capillary proliferation at the vascular pole area. At a later stage (55 wk), many glomeruli showed sclerosis and tuft collapse with the formation of glomerular cysts on a background of tubular atrophy and interstitial fibrosis. This study provides the first evidence that increased expression of VEGF in glomeruli directly causes the glomerular hypertrophy that is associated with proteinuria, suggesting that VEGF exerts multiple effects on the glomerular pathophysiologic processes. 
