Neutralization of Circulating Vascular Endothelial Growth Factor (VEGF) by Anti-VEGF Antibodies and Soluble VEGF Receptor 1 (sFlt-1) Induces Proteinuria

Abstract

There are about 2.5 million glomeruli in the kidneys each consisting of a barrel of glomerular basement membrane surrounded by glomerular endothelial cells on the inside and glomerular epithelial cells with established foot processes (podocytes) on the outside. Defects in this filtration apparatus lead to glomerular vascular leak or proteinuria. The role of vascular endothelial growth factor (VEGF) in the regulation of glomerular vascular permeability is still unclear. Recent studies indicate that patients receiving anti-VEGF antibody therapy may have an increased incidence of proteinuria. In a different setting, pregnancies complicated by preeclampsia are associated with elevated soluble VEGF receptor 1 protein (sFlt-1), endothelial cell dysfunction and proteinuria. These studies suggest that neutralization of physiologic levels of VEGF, a key endothelial survival factor, may lead to proteinuria. In the present study, we evaluated the potential of anti-VEGF neutralizing antibodies and sFlt-1 in the induction of proteinuria. Our studies demonstrate that anti-VEGF antibodies and sFlt-1 cause rapid glomerular endothelial cell detachment and hypertrophy, in association with down-regulation of nephrin, a key epithelial protein in the glomerular filtration apparatus. These studies suggest that down-regulation or neutralization of circulating VEGF may play an important role in the induction of proteinuria in various kidney diseases, some forms of cancer therapy and also in women with preeclampsia. There are about 2.5 million glomeruli in the kidneys each consisting of a barrel of glomerular basement membrane surrounded by glomerular endothelial cells on the inside and glomerular epithelial cells with established foot processes (podocytes) on the outside. Defects in this filtration apparatus lead to glomerular vascular leak or proteinuria. The role of vascular endothelial growth factor (VEGF) in the regulation of glomerular vascular permeability is still unclear. Recent studies indicate that patients receiving anti-VEGF antibody therapy may have an increased incidence of proteinuria. In a different setting, pregnancies complicated by preeclampsia are associated with elevated soluble VEGF receptor 1 protein (sFlt-1), endothelial cell dysfunction and proteinuria. These studies suggest that neutralization of physiologic levels of VEGF, a key endothelial survival factor, may lead to proteinuria. In the present study, we evaluated the potential of anti-VEGF neutralizing antibodies and sFlt-1 in the induction of proteinuria. Our studies demonstrate that anti-VEGF antibodies and sFlt-1 cause rapid glomerular endothelial cell detachment and hypertrophy, in association with down-regulation of nephrin, a key epithelial protein in the glomerular filtration apparatus. These studies suggest that down-regulation or neutralization of circulating VEGF may play an important role in the induction of proteinuria in various kidney diseases, some forms of cancer therapy and also in women with preeclampsia. vascular endothelial growth factor antibody enzyme-linked immunosorbent assay glomerular basement membrane phosphate-buffered saline soluble vascular endothelial growth factor receptor 1 Vascular endothelial growth factor is a 43–46-kDa glycoprotein that serves as a key survival factor for vascular endothelium (1Ferrara N. Nat. Rev. Cancer. 2002; 2: 795-803Crossref PubMed Scopus (1277) Google Scholar, 2Watanabe Y. Dvorak H.F. Exp. Cell Res. 1997; 233: 340-349Crossref PubMed Scopus (83) Google Scholar, 3Alon T. Hemo I. Itin A. Pe'er J. Stone J. Keshet E. Nat. Med. 1995; 1: 1024-1028Crossref PubMed Scopus (1417) Google Scholar, 4Ferrara N. Houck K. Jakeman L. Leung D.W. Endocr. Rev. 1992; 13: 18-32Crossref PubMed Scopus (1554) Google Scholar). Several splice variants of human VEGF1 have been reported to exist, and VEGF165 along with VEGF121 are key circulating forms (5Tischer E. Mitchell R. Hartman T. Silva M. Gospodarowicz D. Fiddes J.C. Abraham J.A. J. Biol. Chem. 1991; 266: 11947-11954Abstract Full Text PDF PubMed Google Scholar, 6Ferrara N. Houck K.A. Jakeman L.B. Winer J. Leung D.W. J. Cell. Biochem. 1991; 47: 211-218Crossref PubMed Scopus (539) Google Scholar, 7Brown L.F. Detmar M. Claffey K. Nagy J.A. Feng D. Dvorak A.M. Dvorak H.F. Exs. 1997; 79: 233-269PubMed Google Scholar). Physiological levels of VEGF in the serum are considered pivotal for maintaining vascular endothelial cell homeostasis (8Gerber H.P. Hillan K.J. Ryan A.M. Kowalski J. Keller G.A. Rangell L. Wright B.D. Radtke F. Aguet M. Ferrara N. Development (Camb.). 1999; 126: 1149-1159Crossref PubMed Google Scholar, 9Ferrara N. Kidney Int. 1999; 56: 794-814Abstract Full Text Full Text PDF PubMed Scopus (660) Google Scholar, 10Wong A.K. Alfert M. Castrillon D.H. Shen Q. Holash J. Yancopoulos G.D. Chin L. Proc. Natl. Acad. Sci. U. S. A. 2001; 98: 7481-7486Crossref PubMed Scopus (64) Google Scholar). The importance of VEGF is further illustrated by the fact the deletion of even a single allele of VEGF in mice leads to embryonic lethality (11Carmeliet P. Ferreira V. Breier G. Pollefeyt S. Kieckens L. Gertsenstein M. Fahrig M. Vandenhoeck A. Harpal K. Eberhardt C. Declercq C. Pawling J. Moons L. Collen D. Risau W. Nagy A. Nature. 1996; 380: 435-439Crossref PubMed Scopus (3459) Google Scholar). While the importance of VEGF as a survival factor for vascular endothelium is well appreciated, its role in the regulation of kidney glomerular vascular endothelium behavior is poorly understood. Some studies have suggested that VEGF expression is reduced during renal injury (12Fan L. Wakayama T. Yokoyama S. Amano O. Iseki S. Nephron. 2002; 90: 95-102Crossref PubMed Scopus (29) Google Scholar, 13Honkanen E.O. Teppo A.M. Gronhagen-Riska C. Kidney Int. 2000; 57: 2343-2349Abstract Full Text Full Text PDF PubMed Scopus (47) Google Scholar, 14Kang D.H. Joly A.H. Oh S.W. Hugo C. Kerjaschki D. Gordon K.L. Mazzali M. Jefferson J.A. Hughes J. Madsen K.M. Schreiner G.F. Johnson R.J. J. Am. Soc. Nephrol. 2001; 12: 1434-1447PubMed Google Scholar) and supplementing the injured kidney with exogenous VEGF121 was shown to provide some improvement in renal function and histology (15Kang D.H. Hughes J. Mazzali M. Schreiner G.F. Johnson R.J. J. Am. Soc. Nephrol. 2001; 12: 1448-1457PubMed Google Scholar). Preeclampsia is also associated with proteinuria, hypertension, and endothelial cell dysfunction (16Pridjian G. Puschett J.B. Obstet. Gynecol. Surv. 2002; 57: 598-618Crossref PubMed Scopus (115) Google Scholar, 17Roberts J.M. Cooper D.W. Lancet. 2001; 357: 53-56Abstract Full Text Full Text PDF PubMed Scopus (1066) Google Scholar, 18Roberts J.M. Taylor R.N. Musci T.J. Rodgers G.M. Hubel C.A. McLaughlin M.K. Am. J. Obstet. Gynecol. 1989; 161: 1200-1204Abstract Full Text PDF PubMed Scopus (1668) Google Scholar). In this regard, Vuorela et al. (19Vuorela-Vepsalainen P. Alfthan H. Orpana A. Alitalo K. Stenman U.H. Halmesmaki E. Hum. Reprod. 1999; 14: 1346-1351Crossref PubMed Scopus (49) Google Scholar) demonstrate that VEGF is bound by a circulating serum protein in the amniotic fluid and maternal serum. Later, the same group demonstrates that soluble VEGF receptor 1 protein (sFlt-1) is significantly elevated in the amniotic fluid of peeclamptic women, again suggesting a pathogenic significance for the decrease of VEGF levels in the serum and amniotic fluid in this condition (20Vuorela P. Helske S. Hornig C. Alitalo K. Weich H. Halmesmaki E. Obstet. Gynecol. 2000; 95: 353-357Crossref PubMed Scopus (118) Google Scholar). In a related study, Zhou et al. (21Zhou Y. McMaster M. Woo K. Janatpour M. Perry J. Karpanen T. Alitalo K. Damsky C. Fisher S.J. Am. J. Pathol. 2002; 160: 1405-1423Abstract Full Text Full Text PDF PubMed Scopus (536) Google Scholar) demonstrate that cytotrophoblasts from placentas of women whose pregnancies were complicated by preeclampsia produce higher levels of sFlt-1 in vitro as compared with control cells. Such separate studies collectively suggest a causal connection between proteinuria/hypertension and endothelial cell dysfunction, mediated by sFlt-1. The significance of physiological circulating levels of VEGF in the regulation of glomerular vascular leak or proteinuria is important to evaluate as anti-VEGF antibody therapy, and VEGF receptor antagonists are being used in several clinical cancer trials (www.nci.nih.gov). In this regard, recent reports from Phase I and Phase II human clinical cancer trials using anti-VEGF antibodies (Bevacizumab) suggest that proteinuria was associated with this treatment protocol (22Kabbinavar F. Hurwitz H.I. Fehrenbacher L. Meropol N.J. Novotny W.F. Lieberman G. Griffing S. Bergsland E. J. Clin. Oncol. 2003; 21: 60-65Crossref PubMed Scopus (1582) Google Scholar). These results suggest that anti-VEGF antibodies may induce proteinuria. In the present study we evaluated the capacity of anti-mouse VEGF neutralizing antibodies and sFlt-1 to induce proteinuria. Our results indicate that a single intravenous infusion of anti-VEGF antibodies into normal healthy mice results in excessive albumin excretion in the urine (proteinuria) via massive glomerular endothelial cell detachment/damage and suppression of the glomerular epithelial slit diaphragm apparatus-associated protein, nephrin. Thus we propose that circulating physiological levels of VEGF are important for the proper function and survival of glomerular endothelial cells and appropriate filtration of blood in the kidney glomeruli. Additionally, these results suggest that long term proteinuria may be a significant side effect of chronic anti-VEGF therapy. Anti-VEGF antibody and sFlt-1 bolus injection studies were performed in wild-type CD1 mice. Each mice were injected with a single intravenous injection of anti-VEGF antibody or sFlt-1/Fc at a concentration of 3.25 and 32.5 pm (picomole per liter). This concentration corresponds to equivalent molar concentration to that of 65 pg/ml (3.25 pm) of normal plasma VEGF (1:1 ratio) and 10 times molar excess to that of 65 pg/ml (3.25 pm) of normal plasma VEGF concentration (1:10 ratio). Mouse anti-VEGF antibody (IgG1) and sFlt-1/Fc chimera were purchased from NeoMarkers (Fremont, CA) and R&D Systems (Minneapolis, MN), respectively. For the in vivo inhibition experiments, human recombinant VEGF165 was injected at a concentration of 32.5 pm to counteract the anti-VEGF antibodies injected at the same concentration. Since timed urine collection was essential to assess protienuria, we used 10 μg of furosemide in 200 μl of PBS for injection after the infusion of anti-VEGF antibodies or sFlt-1. Mice injected with control IgG1 served a control. One-hundred microliters of urine was collected 0, 1, 3, 5, and 24 h after the initial injection. Injections of furosemide were repeated every 1 h before the collection of the urine. Albumin and creatinine concentrations in the urine were estimated using a colorimetric assay according to the manufacturer's recommendations (Sigma). Urine albumin excretion was estimated as the quotient of urine albumin and urine creatinine (23Kristal B. Shasha S.M. Labin L. Cohen A. Am. J. Nephrol. 1988; 8: 198-203Crossref PubMed Scopus (38) Google Scholar). For these experiments, five mice per each group were used. Some mice were sacrificed 5 h after to collect kidneys for immunohistochemistry. The entire experiment was repeated three times. Immunofluorescence staining was performed as described previously (24Hamano Y. Grunkemeyer J.A. Sudhakar A. Zeisberg M. Cosgrove D. Morello R. Lee B. Sugimoto H. Kalluri R. J. Biol. Chem. 2002; 277: 31154-31162Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar). Briefly, 4-μm cryosections were fixed in acetone (−20 °C) for 3 min and dried at room temperature. After incubation with primary antibodies to nephrin, CD2AP, podocin, and α-actinin-4 for 2 h at room temperature, the sections were washed three times with PBS and incubated with fluorescein isothiocyanate-labeled secondary antibodies. After washing with PBS the sections were covered with glass slips using Vectashield mounting media (Vector Laboratories, Burlingame, CA). The staining was analyzed using a fluorescence microscope Eclipse TE300 (Nikon, Tokyo, Japan). Antibodies to nephrin, CD2AP, podocin, and α-actinin-4 were reported by our laboratory in a previous publication (24Hamano Y. Grunkemeyer J.A. Sudhakar A. Zeisberg M. Cosgrove D. Morello R. Lee B. Sugimoto H. Kalluri R. J. Biol. Chem. 2002; 277: 31154-31162Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar). Equal weights of cortical portions of kidneys from each groups were homogenized in liquid nitrogen and they were solubilized in the lysis buffer (0.05 m Hepes, pH 7.5, 0.01 m CaCl2, 4 mmN-ethylmaleimide, 5 mm benzamidine HCl, 1 mm phenylmethanesulfonyl fluoride, and 25 mmε-aminohexanoic acid). The lysates were dialyzed with PBS and electrophoresed on a 10% SDS-polyacrylamide gel. Immunoblots were blocked with 5% skim milk-containing TBST buffer (0.1% Tween 20, 20 mm Tris, pH 7.6, 140 mm NaCl). Western blotting was performed by incubating with indicated antibodies in TBST buffer followed by secondary antibodies conjugated with horseradish peroxidase and then developed by ECL kit as an enhanced chemiluminescence system (Amersham Biosciences). Transmission electron microscopy was performed as described previously (25Cosgrove D. Rodgers K. Meehan D. Miller C. Bovard K. Gilroy A. Gardner H. Kotelianski V. Gotwals P. Amatucci A. Kalluri R. Am. J. Pathol. 2000; 157: 1649-1659Abstract Full Text Full Text PDF PubMed Scopus (167) Google Scholar). Analysis of variance was used to determine statistical differences. As needed, further analysis was carried out using t test with Bonferroni correction to identify significant differences. A p value <0.05 was considered statistically significant. Circulating physiological levels of free VEGF in the normal mouse plasma (VEGF164 and VEGF120) is about 65 pg/ml (3.25 pm) as determined by ELISA (data not shown). We used neutralizing mouse anti-VEGF antibodies at two different concentrations, equivalent molar amount to the free serum VEGF in normal mice according to our measurement (3.25 pm) and also 10-fold molar excess (32.5 pm), to evaluate their capacity to induce proteinuria (as measured by albumin content in the urine). The amount of protein in the urine was estimated as the ratio of urine protein to urine creatinine (Fig. 1). Our results indicate that starting 3 h after the intravenous injection of anti-VEGF antibody, the mice develop significant proteinuria and maintain the same level of proteinuria for next 7 h and gradually after 24 h the albumin content in the urine returns to normal levels (Fig. 1A; data not shown). The amount of proteinuria in mice with 1:1, VEGF:anti-VEGF antibody ratio (1:1 ratio), is less compared with 1:10 ratio (Fig. 1A). The likely explanation for this is the potential lack of sensitivity on part of the ELISA assay to detect all of the circulating VEGF in the plasma. Nevertheless, these experiments suggest that very low amounts of anti-VEGF antibody (3.25 pm) can induce proteinuria. To further establish the specificity of this antibody, we performed the antibody infusion (1:10 ratio) experiments in conjunction with infusion of equivalent molar amounts of exogenous human VEGF165(32.5 pm). Such experiments show that human VEGF165 can inhibit proteinuria inducing capacity of anti-VEGF antibody, providing further proof that anti-VEGF antibody specifically induces proteinuria (Fig. 1A). Recent studies propose that sFlt-1 can also neutralize circulating VEGF and suggest the use of this protein as an anti-cancer agent (26Bainbridge J.W. Mistry A. De Alwis M. Paleolog E. Baker A. Thrasher A.J. Ali R.R. Gene Ther. 2002; 9: 320-326Crossref PubMed Scopus (156) Google Scholar, 27Lai C.M. Brankov M. Zaknich T. Lai Y.K. Shen W.Y. Constable I.J. Kovesdi I. Rakoczy P.E. Hum. Gene Ther. 2001; 12: 1299-1310Crossref PubMed Scopus (99) Google Scholar, 28Goldman C.K. Kendall R.L. Cabrera G. Soroceanu L. Heike Y. Gillespie G.Y. Siegal G.P. Mao X. Bett A.J. Huckle W.R. Thomas K.A. Curiel D.T. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: PubMed Scopus Google Scholar, Y. H. M. T. K. A. M. H. T. M. Y. K. Res. 2002; Google Scholar, W. S. A. K. T. N. T. H. M. Res. 2001; Google Scholar). in the present study we used mouse sFlt-1/Fc protein to experiments as using anti-VEGF antibodies. with anti-VEGF antibody experiments, we show that sFlt-1 can induce proteinuria in mice 3 h of intravenous injection and this effect by 24 h (Fig. data not shown). Additionally, as shown exogenous VEGF at equivalent molar concentration of sFlt-1 (32.5 pm) the effect of sFlt-1 (Fig. These results collectively show that intravenous injection of a single of anti-VEGF antibodies and sFlt-1 to neutralize circulating VEGF leads to proteinuria. We next performed electron microscopy analysis of the kidney glomerular sections from mice developed proteinuria treatment with anti-VEGF antibodies and sFlt-1. The control kidney 5 h after injected with 32.5 pm of normal histology with well endothelial to the glomerular basement membrane proper of foot processes and slit (Fig. 1, and 3 h and even at 24 anti-VEGF antibodies treatment glomerular endothelial (Fig. (Fig. endothelial cell detachment from (Fig. and of slit (Fig. 1, results are kidney from sFlt-1 injected mice are analyzed (data not shown). These results suggest that anti-VEGF antibody and sFlt-1 infusion leads to glomerular endothelial cell and also significant glomerular epithelial cell Such in proteinuria as shown by recent studies (22Kabbinavar F. Hurwitz H.I. Fehrenbacher L. Meropol N.J. Novotny W.F. Lieberman G. Griffing S. Bergsland E. J. Clin. Oncol. 2003; 21: 60-65Crossref PubMed Scopus (1582) Google Scholar, Y. Grunkemeyer J.A. Sudhakar A. Zeisberg M. Cosgrove D. Morello R. Lee B. Sugimoto H. Kalluri R. J. Biol. Chem. 2002; 277: 31154-31162Abstract Full Text Full Text PDF PubMed Scopus (106) Google T. U. F. A. H. Kerjaschki D. N. J. J. Clin. 1999; PubMed Scopus Google Scholar). associated with glomerular endothelial cells are a of the kidney Google Scholar, G. J. Clin. PubMed Scopus Google Scholar). Such are considered to for the filtration of the J. Am. Soc. Nephrol. 1999; Google Scholar, J. Exp. Med. PubMed Scopus Google Scholar, Kidney Int. Full Text PDF PubMed Scopus Google Scholar, C. E. J. Exp. Med. PubMed Scopus Google Scholar). is well established that endothelium not albumin and in can the J. Am. Soc. Nephrol. 1999; Google Scholar, C. E. J. Exp. Med. PubMed Scopus Google Scholar, Proc. Google Scholar). of albumin in the urine to be to a in the glomerular basement membrane or slit diaphragm associated with glomerular epithelial cells J. Am. Soc. Nephrol. 1999; Google Scholar, J. Exp. Med. PubMed Scopus Google Scholar, J. Exp. Med. PubMed Scopus Google Scholar, J. Cell Biol. PubMed Scopus Google Scholar, R. J. Cell Biol. PubMed Scopus Google Scholar, Kidney Int. 8: Full Text PDF PubMed Scopus Google Scholar, B.D. Am. J. 2001; PubMed Google Scholar). In the anti-VEGF antibody and sFlt-1 experiments, the glomerular basement membrane is in the glomerular can be (Fig. 1, we glomerular considered to be important for glomerular in nephrin, podocin, and α-actinin-4 in kidney associated with proteinuria J.M. H. J.C. A.H. Nat. 2000; PubMed Scopus Google Scholar, N. O. S. F. Lee H. A. K. P. C. Nat. 2000; PubMed Scopus Google Scholar, M. U. M. J. P. H. V. T. M. R. L. C. A. K. Cell. 1998; 1: Full Text Full Text PDF PubMed Scopus Google Scholar). been as for glomerular function J. V. A. O. 1999; PubMed Scopus Google Scholar). Mice in 2 after associated with massive proteinuria and kidney (24Hamano Y. Grunkemeyer J.A. Sudhakar A. Zeisberg M. Cosgrove D. Morello R. Lee B. Sugimoto H. Kalluri R. J. Biol. Chem. 2002; 277: 31154-31162Abstract Full Text Full Text PDF PubMed Scopus (106) Google Scholar, H. R. P. J. K. Hum. 2001; PubMed Scopus Google Scholar, M. T. A. H. S. E. T. F. W. P. Kerjaschki D. H. J. Am. Soc. Nephrol. 2002; 13: PubMed Scopus Google Scholar). studies have shown that is key of glomerular filtration apparatus H. H. S. T. P. A. Kerjaschki D. Am. J. Pathol. 1999; Full Text Full Text PDF PubMed Scopus Google Scholar, V. P. J. U. M. H. C. K. Proc. Natl. Acad. Sci. U. S. A. 1999; PubMed Scopus Google Scholar). in this study we the effect of anti-VEGF antibodies and sFlt-1 on the expression of glomerular slit associated The expression of nephrin, CD2AP, podocin, and α-actinin-4 were in the kidneys from mice with anti-VEGF antibodies (1:10 ratio). and Western using kidney sections or kidney we demonstrate that anti-VEGF antibodies significantly the expression of glomerular slit diaphragm associated protein, (Fig. The expression of CD2AP, podocin, and α-actinin-4 was in these kidneys (Fig. results are also kidneys from sFlt-1 injected mice were analyzed (data not shown). These results suggest that significant decrease in the expression of nephrin, a of glomerular slit diaphragm an important to serum as is a key of anti-VEGF antibody and sFlt-1 proteinuria in mice. These results are with reports suggest that expression be for induction of proteinuria A.J. H. P. N. 2002; PubMed Scopus Google Scholar, A. S. E. C. Grunkemeyer J.A. Kalluri R. G. J. Am. Soc. Nephrol. 2001; 12: PubMed Google Scholar). these experiments provide for the importance for circulating physiological levels VEGF in the homeostasis of kidney The glomerular endothelial cells are to VEGF 1 and glomerular epithelial cells not these M. W. Hornig C. H. Weich E. A. J. Am. Soc. Nephrol. 1998; 9: Google Scholar, D. Nagy J.A. A. K. R. P.E. Dvorak H.F. Dvorak A.M. J. 2000; PubMed Scopus Google Scholar). Thus the effect on glomerular epithelial cells is and from the of glomerular endothelial cells to the lack of survival from circulating VEGF (Fig. This study also the and proof that anti-VEGF antibodies and sFlt-1 can cause proteinuria, a explanation for proteinuria in some cancer patients on anti-VEGF antibody therapy and also pregnancies complicated by preeclampsia (16Pridjian G. Puschett J.B. Obstet. Gynecol. Surv. 2002; 57: 598-618Crossref PubMed Scopus (115) Google Scholar, 17Roberts J.M. Cooper D.W. Lancet. 2001; 357: 53-56Abstract Full Text Full Text PDF PubMed Scopus (1066) Google Scholar, P. Alfthan H. Orpana A. Alitalo K. Stenman U.H. Halmesmaki E. Hum. Reprod. 1999; 14: 1346-1351Crossref PubMed Scopus (49) Google Scholar, P. Helske S. Hornig C. Alitalo K. Weich H. Halmesmaki E. Obstet. Gynecol. 2000; 95: 353-357Crossref PubMed Scopus (118) Google Scholar, Y. McMaster M. Woo K. Janatpour M. Perry J. Karpanen T. Alitalo K. Damsky C. Fisher S.J. Am. J. Pathol. 2002; 160: 1405-1423Abstract Full Text Full Text PDF PubMed Scopus (536) Google Scholar). In this regard, a to neutralize the of anti-VEGF antibodies and sFlt-1 to be We for in the of We also in the of this