Embed Size (px)
Citation preview
ImtHsp(iciaBslafm
DA
0©
S
Pathogenesis ofHIV-Associated Nephropathy
Jeremy S. Leventhal, MD, and Michael J. Ross, MD
Summary: Human immunodeficiency virus–associated nephropathy (HIVAN) is a leadingcause of end-stage renal disease in the HIV-1–seropositive population. HIVAN, which ischaracterized by heavy proteinuria and a rapid decline in renal function, is caused by infectionand subsequent expression of viral genes in renal epithelial cells, although the exact mech-anism of viral entry into these cells is unknown. The infected renal epithelium is a distinctcompartment that supports the evolution of viral strains that may diverge from those foundin the patient’s blood. Research using animal models and in vitro studies has shown that vprand nef are the HIV-1 genes most responsible for inducing the characteristic clinical andhistopathologic syndrome of HIVAN. Dysregulation of several host factors, including media-tors of inflammation, apoptosis, proliferation, transcription, and cell–cell interactions, arealso critical factors in determining whether infection of the renal epithelium will lead toHIVAN. Additional research is required to delineate the mechanisms of HIVAN pathogenesisfurther so that more effective interventions can be implemented to prevent and treat thisdisease.Semin Nephrol 28:523-534 © 2008 Elsevier Inc. All rights reserved.Keywords: HIV, HIVAN, glomerulosclerosis, pathogenesis
plmSoicato
H
HI
SwHmlcp(i
t has been nearly 25 years since the firstpublished case series describing the clinicaland histopathologic syndrome of human im-
unodeficiency syndrome–associated nephropa-hy (HIVAN).1 The classic clinical presentation ofIVAN is one of rapidly progressive azotemia as-
ociated with severe proteinuria, often in ne-hrotic range, and little or no peripheral edemareviewed by Wyatt et al in this issue of Seminarsn Nephrology, p. 513).2 HIVAN may not beharacterized simply as a glomerular or tubulo-nterstitial renal disease because of widespreadbnormalities present in the renal parenchyma.iopsy findings include focal glomerulosclero-is, often of the collapsing variant, with col-apse of the glomerular tuft and proliferationnd hypertrophy of podocytes, sometimesorming a cellular “pseudocrescent” in Bow-an’s space.3 Tubulointerstitial disease is also a
ivision of Nephrology, The Mount Sinai School of Medicine, New York, NY.ddress reprint requests to Jeremy S. Leventhal, MD, Fellow in Nephrology,Division of Nephrology, The Mount Sinai School of Medicine, 1 Gustave L.Levy Place, Box 1243, New York, NY 10029. E-mail: [email protected]
270-9295/00/$ - see front matter
s2008 Elsevier Inc. All rights reserved. doi:10.1016/j.semnephrol.2008.08.003eminars in Nephrology, Vol 28, No 6, November 2008, pp 52
rominent finding, and includes microcystic di-atation of renal tubules and pleiocytic inflam-
atory interstitial infiltrates and fibrosis.4,5
ince its original description, the pathogenesisf HIVAN has been studied intensely by many
nvestigators. In this review, we describe theurrent state of our understanding of the mech-nisms by which host and viral factors interacto create the clinical and pathologic syndromef HIVAN.
IV INFECTION OF THE KIDNEY
IV-Infected Renal Epitheliums a Distinct Viral Compartment
everal early studies focused on the question ofhether HIV infects renal parenchymal cells inIVAN with conflicting results.6-8 The issue re-ained unresolved until Bruggeman et al9 pub-
ished a study in which biopsy specimens wereollected prospectively from 21 HIV-positiveatients with renal disease including HIVANN � 16) and other renal diseases (N � 5). Thenvestigators used several techniques, RNA in
itu hybridization, and DNA in situ polymerase3-534 523
crfctHpdHonH1eamrsc
eppsiwtfptraeacg
totb
atDtcrhbtfr(tsPtpppslbsimo
M
T
FssP 02.13
524 J.S. Leventhal and M.J. Ross
hain reaction to detect HIV nucleic acid inenal biopsy specimens. They detected HIV in-ection of renal epithelial cells, including podo-ytes, glomerular parietal epithelial cells, andubular cells, in the majority of biopsy samples.IV was detected in kidney specimens fromatients with HIVAN and other forms of renalisease and further studies have shown thatIV infection and tubular microcystic dilatationccur in a focal distribution that can affect allephron segments.4 Renal tubular infection byIV-1 in a patient with HIVAN is shown in Fig.. These findings were confirmed recently andxtended by Tanji et al,10 who reported that inddition to renal epithelial cells, HIV-infectedacrophages and T cells are present in the
enal interstitium. No studies have definitelyhown infection of nonepithelial renal paren-hymal cells in vivo.
Interestingly, in the series from Bruggemant al,9 HIV-1 RNA was detected in each of the 4atients who had no detectable viral RNA in thelasma, suggesting that HIV can remain tran-criptionally active in renal epithelial cells evenn the presence of maximal viral suppression
ith antiretroviral therapy (ART). This persis-ence of renal HIV expression was highlightedurther by a report by Winston et al11 of aatient who presented with HIVAN in the set-ing of acute HIV-1 seroconversion. Despite aapid reduction of plasma HIV RNA to undetect-ble levels and resolution of clinical renal dis-ase, examination of renal biopsy tissue takenfter the response to treatment revealed nohange in the expression of HIV-1 RNA. To-
igure 1. Detection of viral nucleic acids in renal biopsypecimens by in situ hybridization. (a) Antisense riboproense riboprobe was used as a negative hybridization coublishers Ltd: Nat Med. 2002;8:522-526. Copyright 20
ether, these studies showed that HIV-1 infec- e
ion of the renal epithelium occurs in the majorityf HIV-infected patients with renal disease andhat the virus is not eradicated from the kidneyy currently available ART regimens.
The HIV-1 reverse transcriptase enzyme lacksproofreading function and is therefore prone
o making errors when synthesizing proviralNA from its RNA template, allowing the virus
o evolve rapidly and for distinct viral quasispe-ies to exist in each infected individual.12 Mar-as et al13 studied whether renal epithelial cellsarbor HIV quasispecies that are distinct fromlood-derived virus isolated from the same pa-ients. They compared viral sequences isolatedrom HIV-infected renal tubules with those de-ived from peripheral blood mononuclear cellsPBMCs) from the same patients. They foundhat in each patient, the kidney-derived HIVequences were similar to but distinct from theBMC-derived HIV sequences. There are 2 impor-ant conclusions to be drawn from this study: theresence of distinct renal HIV quasispeciesroved that renal tubular cells are capable of sup-orting HIV replication and that the kidney is aeparate viral compartment that may support evo-ution of divergent viral strains. Because testing oflood-derived HIV for mutations that predict re-istance to ART is an important strategy for guid-ng HIV treatment, it will be important to deter-
ine if the renal epithelium supports persistencef viral quasispecies that are resistant to ART.
echanism of HIV-1 Renal Infection
he mechanism by which HIV-1 enters renal
ens. Detection of viral messenger RNA in HIVAN biopsybridization shows positive tubular epithelial cells. (b) An serial section. Adapted by permission from Macmillan
specimbe hy
ntrol i
pithelial cells remains unknown. The virus in-
fatoCriHttpemett
niHtbRoomqcfkiCitlaBadeacirmflwlc
trftc
HI
Sbppuatumvtbhcpbglth
dratipHttgn
ratecctH
Pathogenesis of HIVAN 525
ects lymphocytes and macrophages via inter-ction of the viral envelope protein gp120 withhe cellular CD4 receptor and either the CXCR4r CCR5 co-receptor. Conaldi et al14 detectedXCR4 and CD4 in a subpopulation of cultured
enal epithelial cells, suggesting that HIV-1 maynfect renal epithelial cells via these receptors.owever, these studies were performed in cul-
ured cells, which are vulnerable to contamina-ion with nonepithelial cells and may expressroteins not present in vivo. In contrast, Eitnert al8,15 was unable to detect CXCR4 or CCR5essenger RNA by in situ hybridization in renal
pithelial cells in biopsy specimens from pa-ients with HIVAN, normal kidneys, and pa-ients with renal allograft rejection.
Ray et al16 reported infection of primary re-al tubular epithelial cells (RTECs) using HIV
solated from the PBMCs of children withIVAN, but not when using common labora-
ory HIV isolates. The addition of a CD4 anti-ody did not inhibit infection, suggesting thatTEC infection by renal tropic HIV virus mayccur via a CD4-independent mechanism. An-ther group characterized the tropism of chi-eric viruses containing HIV envelope se-
uence cloned from infected tubular epithelialells in HIVAN biopsy specimens and PBMCsrom the same patients. They found that theidney-derived viruses were capable of infect-ng cells expressing CD4 and either CXCR4 orCR5, whereas the blood-derived viruses could
nfect only CCR5-expressing cells.17 The inves-igators also showed that kidney-derived iso-ates were able to infect cell lines using thelternate HIV co-receptors BONZO/STRL33 andOB/GPR15. This study suggests that the mech-nism of viral entry into renal epithelial cells isependent on both host and viral factors. How-ver, because the in vivo expression of theselternate HIV co-receptors in renal epithelialells has not been well characterized, their rolen viral entry remains unknown. A recent studyeported that the C-type lectin DEC-205 canediate internalization and nonproductive in-
ection of the HK-2–immortalized tubular celline.18 It is not clear how to reconcile this study
ith the many others showing that HIV estab-ishes productive infection of tubular epithelial
ells. Taken together, these studies suggest that dhere may be renal tropic strains of HIV and theeceptor use of these HIV variants may differrom common HIV isolates. However, the ques-ion of how HIV gains entry into renal epithelialells remains unresolved.
IVAN Is Caused by HIVnfection of Renal Epithelial Cells
tudies using transgenic animal models haveeen invaluable for determining the viral andatient-related factors that contribute to theathogenesis of HIVAN. In the most extensivelysed transgenic model, known as Tg26, micere transgenic for an HIV provirus with dele-ions of the gag and pol genes that is expressednder control of the endogenous viral long ter-inal repeat (LTR) promoter. Tg26 mice de-
elop a clinical and histopathologic syndromehat is identical to HIVAN. Similar to humaneings, in whom susceptibility to HIVAN isighly influenced by genetic factors (see the arti-le by Wyatt et al in this issue, p. 513), the Tg26henotype is highly dependent on the geneticackground of the mice.19 Gharavi et al20 usedenome-wide linkage analysis to identify geneticoci that are associated with a risk for developinghe HIVAN phenotype, although the culprit genesave not yet been identified.
Investigators have used the Tg26 model toetermine whether the HIVAN phenotype is aesult of renal expression of HIV-1 genes or,lternatively, the effect of systemic factors onhe kidney. In reciprocal transplantation stud-es, kidneys from Tg26 mice that were trans-lanted into wild-type mice developed theIVAN phenotype whereas wild-type kidneys
ransplanted into Tg26 mice remained normal,hereby showing that renal expression of HIVenes is necessary to produce the HIVAN phe-otype.21
Expression of HIV-1 genes in vitro is able toecapitulate many of the cellular abnormalitiesssociated with HIVAN in vivo. One of the his-opathologic hallmarks of HIVAN is the pres-nce of podocyte proliferation. Normal podo-ytes are terminally differentiated quiescentells and in most forms of chronic renal diseasehe number of podocytes decreases.22 InIVAN, however, podocytes proliferate and un-
ergo dedifferentiation with loss of expression
osSfemmcvathhgha
V
Tcss
tn
tpepsacsotHwmmipppe
526 J.S. Leventhal and M.J. Ross
f podocyte-specific markers including CALLA,ynaptopodin, WT-1, and podocalyxin.23,24
chwartz et al25 showed that podocytes isolatedrom Tg26 mice have increased levels of prolif-ration and express lower levels of podocytearkers than podocytes isolated from wild-typeice. Furthermore, infection of wild-type podo-
ytes with HIV-1 induces these same changes initro.26 The availability of transgenic modelsnd in vitro cell-based assays that reliably modelhe in vivo histopathologic features of HIVANave allowed investigators to identify viral andost factors that are critical for HIVAN patho-enesis. The various transgenic models thatave been used to study HIVAN pathogenesisre summarized in Table 1.
IRAL FACTORS
he polycistronic 9-kb HIV-1 RNA genome en-odes 9 genes.27 Animal models and in vitrotudies have delineated the viral genes respon-ible for inducing derangements in the pheno-
Table 1. Summary of Transgenic Models of H
Species/ReferencePromoter Controlling
HIV Expression HIV Genes Exp
Mouse (Tg26)19 HIV LTR promoter All HIV genes excand pol
Mouse28 Nephrin promoter All HIV genes excpol, and env
Mouse31,32 CD4 promoter Whole HIV genowell as experimmutations of iHIV genes
Mouse34 Podocin promoter-drivenCre recombinaseactivates nefexpression
nef
Mouse73 HIV LTR promoter All HIV genes excpol, and nef
Mouse35 HIV LTR promoter Deletions of vprnef
Mouse35 c-fms promoter vpr
Mouse29 Nephrin promoter Transgenic linesexpressing indHIV genes (vif,vpu, nef, rev, oand vpr/nef dotransgenic
Rat74 HIV LTR promoter All HIV genes excand pol
Abbreviation: FSGS, focal segmental glomerulosclerosis.
ype of renal epithelial cells and the mecha-isms by which they produce renal pathology.
The Tg26 mouse model, which recapitulateshe full HIVAN phenotype, lacks the gag andol genes, making it unlikely that expression ofither of these genes is necessary for HIVANathogenesis.19 Moreover, Zhong et al28 createdeveral murine transgenic lines that expressedll HIV genes except gag, pol, and env underontrol of the nephrin promoter, thereby en-uring podocyte-specific gene expression. Mostf these mice developed proteinuria and theypical glomerular and tubular changes ofIVAN. The renal phenotype in these mice alsoas dependent on the genetic strain of theice, with a severe phenotype noted in FVB/Nice (same strain as Tg26) and minimal disease
n C57BL/6 mice. This study suggests that ex-ression of vif, vpu, vpr, tat, rev, and/or nef inodocytes is sufficient to induce the HIVANhenotype. Several additional transgenic mod-ls expressing combinations of HIV genes un-
Location ofTransgene Expression Renal Phenotype
Various tissues FSGS with tubulointerstitial diseaseand proteinuria
, Podocytes FSGS, tubulointerstitial disease,and proteinuria
l
CD4-positive cells Tubulointerstitial disease(dependent on intact nef SH3domain)
Podocytes Podocyte dedifferentiation andproliferation; no proteinuria orFSGS
, Various tissues FSGS, microcystic tubulardilatation, interstitial infiltrate,proteinuria
Various tissues FSGS, and proteinuria only in micewith intact vpr
Macrophages FSGS with evidence of epithelialcell proliferation but no tubulardisease
Podocytes Podocyte injury and FSGS in vprand nef transgenic mice,worsened glomerular disease innef/vpr mice
Various tissues FSGS, mesangial hypercellularity,microcystic dilatation, andinterstitial disease
IVAN
ressed
ept gag
ept gag
me asental
ndividua
ept gag
and/or
ividualvpr,r tat),uble
ept gag
dpgasuOdrrAtp
N
NptkHpptdaelirlnrirbptFhsnm
iitttpa
pfct
NtHacss3(MHNtvenn
t
FMcodhN(s
Pathogenesis of HIVAN 527
er control of the HIV and/or cell type–specificromoters have been created to determine theene(s) responsible for inducing HIVAN. Zuo etl29 created transgenic mouse lines expressing aingle HIV genes (vif, vpr, vpu, nef, rev, or tat)nder the control of the nephrin promoter.nly transgenic mice expressing vpr and/or nefeveloped proteinuria and focal glomeruloscle-osis, with the most severe phenotype occur-ing in mice transgenic for both nef and vpr.ccordingly, much recent research has cen-
ered on the role of these 2 genes in HIVANathogenesis.
EF
ef is a 206-amino acid protein with many re-orted functions, including reduction of CD4rafficking to the cell surface, effects on cyto-ine expression, and prevention of apoptosis.30
anna et al31 created transgenic mice that ex-ress nef under the control of the human CD4romoter. These mice developed several fea-ures of acquired immune deficiency syn-rome–like illness (loss of CD4� cells, wasting,nd so forth) and interstitial nephritis. The rel-vance of this model to HIVAN is limited by theack of a glomerular phenotype and because its unlikely that the transgene was expressed inenal epithelial cells. When the same groupater generated mice that were transgenic foref with a mutated SH3 binding domain, theesulting mice developed neither an acquiredmmune deficiency syndrome phenotype norenal disease.32 This SH3 binding domain im-ues Nef with the ability to interact with manyroteins, including those of the Src-family ofyrosine kinases, which include Hck, Lck, andyn. Breeding CD4 nef-transgenic mice withck-mutant mice did not prevent renal disease,uggesting that Nef-Hck interaction was notecessary for development of disease in thisodel.To determine the HIV-1 gene(s) necessary for
nducing podocyte proliferation, Husain et al26
nfected murine podocytes with a series of len-iviral vectors, each of which harbored a muta-ion in a single HIV gene. They found that muta-ion of nef completely eliminated podocyteroliferation and that infection of podocytes with
vector expressing nef alone induced podocyte troliferation and dedifferentiation. Nef is there-ore necessary and sufficient to induce podo-yte abnormalities in vitro that closely modelhose found in HIVAN.
He et al33 performed studies to elucidate theef-induced signaling pathways that result in
he characteristic podocyte abnormalities inIVAN. They showed that nef induced prolifer-tion, and dedifferentiation of murine podo-ytes is mediated via Src kinase activation, withubsequent phosphorylation and activation ofignal transducer and activator of transcription(Stat3) and mitogen-activated protein kinase
MAPK)1,2. Higher levels of phosphorylatedAPK1,2 and Stat3 also were found in humanIVAN samples and Tg26 mice. Mutation of theef SH3 domain, which is required for its ability
o interact with and activate Src kinase, pre-ented nef-induced proliferation and dediffer-ntiation. Conversely, expression of a domi-ant-negative Src also prevented the effects ofef expression (Fig. 2).33
Recently, 2 groups reported renal pheno-ypes in separate transgenic models in which
igure 2. Role of Src activation in Nef-inducedAPK1,2 and Stat3 phosphorylation and phenotypic
hanges in podocytes. Control vector-infected (Vector)r Nef-infected (Nef) podocytes were transfected withominant-negative Src (Src-DN) and then selected withygromycin B. Inhibition of Src by Src-DN inhibited (A)ef-induced activation of cyclin E, Stat3, and MAPK1,2,
B) as well as Nef-induced decrease in expression ofynaptopodin and (C) Nef-induced podocyte prolifera-
ion.33 Adapted with permission from He et al.33
mcerpemcKwntf(scu
easdnb
V
ViitfggvvsgcdlvvgdwcvrtV
ptt
iccasptcTcmiipn
T
TVpscCmdtopraiwIrv
tgttagapc
528 J.S. Leventhal and M.J. Ross
ice express nef under the control of podo-yte-specific promoters. In the study by Husaint al,34 the investigators did not detect clinicalenal disease (proteinuria or renal failure), butodocytes that expressed nef showed loss ofxpression of the podocyte differentiationarkers WT1 and synaptopodin, and had in-
reased expression of the proliferation markeri-67. Zuo et al29 also created mice in which nefas expressed specifically in podocytes using aephrin promoter. They found that the pheno-
ype of the mice was variable, with 4 of 11ounder lines developing glomerular diseasepredominantly focal and segmental glomerulo-clerosis). Further characterization revealed de-reased expression of synaptopodin in glomer-li of nef-transgenic mice.
Together, these studies provide compellingvidence that expression of nef in podocytes isn important component of HIVAN pathogene-is, with particular roles in inducing podocyteifferentiation and proliferation. The effect ofef expression in tubular epithelial cells has noteen well studied.
PR
pr is a 96–amino acid protein whose actionsn the HIV life cycle include facilitating nuclearmport of the HIV preintegration complex andransactivation of the viral LTR promoter. A roleor vpr in HIVAN pathogenesis was first sug-ested by Dickie et al,35 who created HIV trans-enic mice that had mutations in either nef orpr. They found that only mice with an intactpr gene developed proteinuria and glomerulo-clerosis. The investigators also created trans-enic mice expressing tat and vpr under theontrol of the LTR promoter. These animalseveloped significant proteinuria and glomeru-
osclerosis. The renal phenotype was more se-ere when these mice were crossed with thepr-mutant mice, suggesting that other HIVenes worsen the course of vpr-induced renalisease. Interestingly, another transgenic line inhich vpr was expressed under control of the
-fms (macrophage-specific) promoter also de-eloped modest proteinuria and glomeruloscle-osis. The mechanism of renal pathogenesis inhese mice is unclear, however, because free
pr protein can directly transduce cells, it is mossible that macrophage-derived Vpr wasaken up by renal epithelial cells, resulting inhe renal phenotype.
Few studies have addressed the role of vpr innducing tubular disease. Studies in nonrenalells have shown that vpr can induce severalellular effects, including G2/M cell-cycle arrestnd apoptosis.36 Recently, Rosenstiel et al37
howed that vpr expression in the HK-2 humanroximal tubular epithelial cell line impaired cy-okinesis and induced accumulation of multinu-leated cells. Moreover, they found that RTEC ing26 mice and human HIVAN biopsies had in-reased levels of epithelial cell hypertrophy andultinucleation. These findings suggest that the
n vitro abnormalities observed in vpr-express-ng HK-2 cells also are present in HIVAN, thusroviding important insights into the mecha-ism of tubulointerstitial disease in HIVAN.
AT and ENV
at is a 101–amino acid protein that, similar topr, can be detected in the serum of infectedatients. Tat is a critical activator of HIV tran-cription and can induce cellular changes in-luding cytokine production and apoptosis.38
onaldi et al39 reported that incubation of pri-ary podocytes with Tat protein induces dose-
ependent proliferation and loss of differentia-ion markers in vitro. Deciphering the relevancef these studies to disease pathogenesis is ham-ered by the fact that the podocytes were de-ived from Caucasian patients, who are not usu-lly susceptible to HIVAN, and because severalnvestigators have failed to detect renal disease
hen Tat is expressed alone in murine models.t is possible, however, that Tat may potentiateenal injury when present in addition to nef andpr.
The HIV-1 env gene encodes the gp160 pro-ein, which after proteolytic cleavage yields thep120 and gp41 proteins. Gp120 is present onhe surface of HIV virions and facilitates infec-ion of target cells by interacting with CD4 and
co-receptor. One group has reported thatp120 can induce aberrant proliferation andpoptosis of human mesangial cells40 and apo-tosis of both tubular and glomerular epithelialells.41,42 Because HIV is not known to infect
esangial cells in vivo, and mice expressing the
etr
H
Aigisdtsrhetdcti
FA
Iamawwf
ia
iscbirtHci
tfimmmiRHapvHstt
pgfFmvetnto
I
Thica
Fwl
Pathogenesis of HIVAN 529
nv gene alone do not develop renal disease,he role of gp120 in the pathobiology HIVANemains unclear.
OST FACTORS
lthough HIV infection of the renal epitheliums a necessary step in HIVAN pathogenesis, onlyenetically susceptible persons respond to thisnfection by developing HIVAN. The host re-ponse to HIV infection is therefore a criticaleterminant in the pathobiology of HIVAN. Al-hough genetic factors clearly contribute to theusceptibility of blacks to HIVAN, these factorsemain unknown. However, several studiesave identified features of the response of renalpithelial cells to HIV gene expression that con-ribute to the development of progressive renalisease. The major renal epithelial cellular pro-esses affected by HIV infection that lead tohe development of HIVAN are summarizedn Fig. 3.
actors Promotingpoptosis and Fibrogenesis
ncreased apoptosis of tubular epithelial cellsnd interstitial fibrosis are common findings inany nephropathies, including HIVAN. Bodi et
l43 compared biopsy specimens from patientsith HIVAN with HIV-seronegative patientsith focal segmental glomerulosclerosis and
ound a substantially higher rate of tubular ep-
igure 3. Diagrammatic depiction of biological path-ays altered by HIV infection of renal epithelial cells,
peading to progressive renal failure.
thelial apoptosis. Interestingly, no increase inpoptosis was observed in HIVAN glomeruli.
Conaldi et al14 showed the ability of HIV-1 tonfect human tubular cells and induce apopto-is in vitro. HIV infection induced extensiveaspase-dependent apoptosis of proximal tu-ular cells and, in further experiments, the
nvestigators found that the apoptosis-inducingeceptor, Fas, was up-regulated by HIV infec-ion. The importance of Fas up-regulation inIV-induced tubular cell apoptosis was notlear because addition of Fas-blocking antibod-es to the cells did not prevent apoptosis.
Transforming growth factor (TGF)-� is a cy-okine with important roles in promoting renalbrogenesis and apoptosis in several animalodels of renal disease.44-46 Quantitative poly-erase chain reaction analysis of biopsy speci-ens from patients with HIVAN showed signif-
cantly increased levels of TGF-� messengerNA when compared with biopsy specimens ofIV-positive patients without signs of HIVAN47
nd TGF-� protein levels are increased in HIV-ositive patients with glomerular disease.48 Initro cell culture models have suggested thatIV gene products may increase TGF-� expres-
ion in renal parenchymal cells,49,50 however,he effect of TGF-� inhibition has not beenested in in vivo models of HIVAN.
Ross et al51 reported that the ubiquitin-likerotein FAT10 is one of the most up-regulatedenes after HIV infection in a RTEC line derivedrom a patient with HIVAN. Expression ofAT10 also was increased in kidneys from Tg26ice and in HIVAN biopsy specimens, and pre-
ention of FAT10 expression using RNA interfer-nce prevented HIV-induced apoptosis. Althoughhese results suggest that FAT10 expression isecessary for HIV-induced RTEC apoptosis,52
he mechanism by which FAT10 facilitates ap-ptosis is not known.
nflammatory Mediators
ubulointerstitial inflammation is a prominentistopathologic finding in HIVAN.3 Accord-
ngly, several case series and retrospectivease-control studies have reported an associ-tion between corticosteroid treatment and im-
roved renal outcomes in patients of HIVAN.53-56
Tto
dmHcflmseTggnt
cHHkRtdhccnigrct
P
AiSacpiT(Hmmia
pwt
T
BRhsntommtatpiset
tpCaeglitit
MS
K(pcmSggdof
530 J.S. Leventhal and M.J. Ross
hese observations suggest that renal inflamma-ion is an important factor in the clinical coursef HIVAN.
Ross et al51 profiled the genes that wereifferentially expressed after infection of a hu-an RTEC line derived from a patient withIVAN. The most prominent response of theells to HIV infection was up-regulation of proin-ammatory mediators, including cytokines, che-okines, and adhesion molecules. Many of these
ame genes were found to be up-regulated mark-dly in kidneys from the Tg26 HIVAN model.hese data suggest that expression of HIVenes in RTEC up-regulates proinflammatoryenes that then recruit leukocytes to the kid-ey, resulting in the tubulointerstitial inflamma-ion that is a characteristic finding in HIVAN.
In another study, investigators measuredytokine concentrations in the renal tissue ofIV-seropositive patients with and withoutIVAN.57 Several cytokines, including interleu-in-8, monocyte chemoattractant protein-1, andANTES, were increased in HIV-infected pa-
ients irrespective of whether they had renalisease. Patients with HIVAN were found toave higher renal tissue levels of major histo-ompatibility complex class II, interferon-� re-eptor, and interferon-alfa, although levels didot correlate with the amount of interstitial
nflammatory cells. Together, these studies sug-est that HIV-infected patients have increasedenal levels of inflammatory mediators, whichan recruit inflammatory cells into the intersti-ium.
roliferation
s discussed previously, the HIV-1 nef genenduces podocyte proliferation via activation ofrc kinases and the transcription factors Stat3nd MAPK1,2.33 Dysregulated expression of cy-lins is also an important factor in promotingodocyte proliferation in HIVAN. HIV infection
ncreases cyclin E expression in podocytes.58
he expression of the cyclin-dependent kinaseCDK) inhibitors p27 and p57 are decreased inIVAN biopsy samples as compared with nor-al kidneys or those with membranous andinimal change disease, whereas p21 levels are
ncreased. Because CDK inhibitors suppress the
ctivity of pro-proliferative cyclin/CDK com- alexes, the increased levels of cyclin E, coupledith decreased p27 and p57, likely contribute
o podocyte proliferation in HIVAN.
ranscription Factors
ecause the HIV genome does not encode anNA polymerase, the virus must co-opt theost’s transcriptional machinery to replicate it-elf. In lymphocytes, the transcription factorsuclear factor �B (NF-�B) and Sp1 are impor-ant activators of HIV transcription, whereas inther cell types other transcription factors areore important. In studies using the Tg26odel of HIVAN, Bruggeman et al59 determined
hat NF-�B and Sp1 regulate HIV expression inmanner similar to that seen in lymphocytes. In
his model, there is constitutively increasedhosphorylation and degradation of the NF-kB
nhibitor, IkB�, freeing NF-�B to activate tran-cription of HIV and promote epithelial prolif-ration, apoptosis, and production of inflamma-ory mediators.60,61
CDK9 is a member of the protein complexhat activates transcription of the HIV-1 LTRromoter in the presence of Tat protein.62
DK9 inhibitors inhibited expression of HIVnd restored expression of differentiation mark-rs in murine podocytes after infection with aag/pol-deleted HIV virus.63 The same groupater showed that administration of CDK9 inhib-tors to Tg26 mice prevented development ofhe HIVAN phenotype, suggesting that CDK9nhibition may be a therapeutic strategy for thereatment and/or prevention of HIVAN.64
ediators of Cell Adhesion, Cellignaling, and Extracellular Matrix
aufman et al65 reported that the sidekick-1sdk-1) gene is up-regulated in HIV-infectedodocytes and levels of Sdk-1 protein are in-reased in vivo in podocytes in the Tg26 mouseodel of HIVAN and in HIVAN biopsy samples.
idekick proteins are members of the immuno-lobulin superfamily and have roles in neuronaluidance during retinal development.66 HIV-in-uced expression of SDK-1 leads to aggregationf podocytes in vitro and may contribute to theormation of podocyte “pseudocrescents” that
re observed commonly in HIVAN.67
ruPurgtpu
ggcdtttiHpsvSbpvf
C
InHlmTHtllppt
Haiid
in
suwfeem
R
1
1
1
Pathogenesis of HIVAN 531
Podocan, a recently described small leucineich protein, is up-regulated in sclerotic glomer-li of Tg26 HIV transgenic mice.68 The role ofodocan in HIVAN pathogenesis remains spec-lative. Expression of another small leucineich protein, Decorin, influences the course oflomerular disease in diabetic mice.69 It isherefore plausible that levels of Podocan ex-ression may affect the development of glomer-losclerosis in HIVAN.
Altered expression of vascular endothelialrowth factor (VEGF) contributes to the patho-enesis of many renal diseases.70 In mice, in-reased expression of the VEGF164 isoform in-uces a collapsing glomerulopathy similar tohat found in HIVAN.71 Korgaonkar et al72 de-ected increased levels of VEGF protein and itsranscriptional regulator (HIF-2�) in glomerulin the murine HIVAN model and in humanIVAN biopsy specimens. Infection of murineodocytes with HIV in vitro induced expres-ion of VEGF and HIF-2�, which was mediatedia nef-induced activation of Src kinase andtat3. Moreover, addition of neutralizing anti-odies against VEGFR2 reduced HIV-inducedodocyte proliferation and dedifferentiation initro. These studies suggest an important roleor VEGF in HIVAN pathogenesis.
ONCLUSIONS
n susceptible individuals, HIV infection of re-al epithelial cells leads to the development ofIVAN, which is characterized by histopatho-
ogic abnormalities that include collapsing glo-erulosclerosis with tubulointerstitial disease.his renal syndrome is caused by the effects ofIV gene expression in the kidney, although
he mechanism of HIV entry into renal epithe-ial cells remains unknown. The renal epithe-ium is a distinct viral compartment that sup-orts active viral transcription even when theatient is treated with aggressive antiretroviralherapy.
Studies using transgenic murine models ofIVAN and in vitro studies have implicated vprnd nef as the HIV genes most responsible fornducing HIVAN. Host cellular responses to HIVnfection are important determinants of renal
isease and several cellular genes have beenmplicated as contributors to the HIVAN phe-otype.
The prevalence of persons living with end-tage renal disease as a result of HIVAN contin-es to increase even in countries where ART isidely available. Additional research is there-
ore urgently needed to delineate the pathogen-sis of HIVAN and enable the design of moreffective strategies for the prevention and treat-ent of this disease.
EFERENCES1. Rao TK, Filippone EJ, Nicastri AD, Landesman SH,
Frank E, Chen CK, et al. Associated focal and segmen-tal glomerulosclerosis in the acquired immunodefi-ciency syndrome. N Engl J Med. 1984;310:669-73.
2. Ross MJ, Klotman PE. Recent progress in HIV-associ-ated nephropathy. J Am Soc Nephrol. 2002;13:2997-3004.
3. D’Agati V, Suh JI, Carbone L, Cheng JT, Appel G.Pathology of HIV-associated nephropathy: a detailedmorphologic and comparative study. Kidney Int.1989;35:1358-70.
4. Ross MJ, Bruggeman LA, Wilson PD, Klotman PE.Microcyst formation and HIV-1 gene expression oc-cur in multiple nephron segments in HIV-associatednephropathy. J Am Soc Nephrol. 2001;12:2645-51.
5. D’Agati V, Appel GB. HIV infection and the kidney.J Am Soc Nephrol. 1997;8:138-52.
6. Kimmel PL, Ferreira-Centeno A, Farkas-Szallasi T,Abraham AA, Garrett CT. Viral DNA in microdissectedrenal biopsy tissue from HIV infected patients withnephrotic syndrome. Kidney Int. 1993;43:1347-52.
7. Cohen AH, Sun NC, Shapshak P, Imagawa DT. Dem-onstration of human immunodeficiency virus in renalepithelium in HIV-associated nephropathy. ModPathol. 1989;2:125-8.
8. Eitner F, Cui Y, Hudkins KL, Stokes MB, Segerer S,Mack M, et al. Chemokine receptor CCR5 and CXCR4expression in HIV-associated kidney disease. J Am SocNephrol. 2000;11:856-67.
9. Bruggeman LA, Ross MD, Tanji N, Cara A, Dikman S,Gordon RE, et al. Renal epithelium is a previouslyunrecognized site of HIV-1 infection. J Am Soc Neph-rol. 2000;11:2079-87.
0. Tanji N, Ross MD, Tanji K, Bruggeman LA, Markow-itz GS, Klotman PE, et al. Detection and localizationof HIV-1 DNA in renal tissues by in situ polymerasechain reaction. Histol Histopathol. 2006;21:393-401.
1. Winston JA, Bruggeman LA, Ross MD, Jacobson J,Ross L, D’Agati VD, et al. Nephropathy and establish-ment of a renal reservoir of HIV type 1 during primaryinfection. N Engl J Med. 2001;344:1979-84.
2. Basavapathruni A, Anderson KS. Reverse transcrip-tion of the HIV-1 pandemic. FASEB J. 2007;21:3795-
808.
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
532 J.S. Leventhal and M.J. Ross
3. Marras D, Bruggeman LA, Gao F, Tanji N, MansukhaniMM, Cara A, et al. Replication and compartmentaliza-tion of HIV-1 in kidney epithelium of patients withHIV-associated nephropathy. Nat Med. 2002;8:522-56.
4. Conaldi PG, Biancone L, Bottelli A, Wade-Evans A,Racusen LC, Boccellino M, et al. HIV-1 kills renaltubular epithelial cells in vitro by triggering an apo-ptotic pathway involving caspase activation and Fasupregulation. J Clin Invest. 1998;102:2041-9.
5. Eitner F, Cui Y, Hudkins KL, Anderson DM, SchmidtA, Morton WR, et al. Chemokine receptor (CCR5)expression in human kidneys and in the HIV infectedmacaque. Kidney Int. 1998;54:1945-54.
6. Ray PE, Liu XH, Henry D, Dye L, 3rd, Xu L, Oren-stein JM, et al. Infection of human primary renalepithelial cells with HIV-1 from children with HIV-associated nephropathy. Kidney Int. 1998;53:1217-29.
7. Zerhouni-Layachi B, Husain M, Ross MJ, Marras D,Sunamoto M, Liu X, et al. Dual tropism of HIV-1envelopes derived from renal tubular epithelial cellsof patients with HIV-associated nephropathy. AIDS.2006;20:621-4.
8. Hatsukari I, Singh P, Hitosugi N, Messmer D, Valder-rama E, Teichberg S, et al. DEC-205-mediated inter-nalization of HIV-1 results in the establishment ofsilent infection in renal tubular cells. J Am Soc Neph-rol. 2007;18:780-7.
9. Dickie P, Felser J, Eckhaus M, Bryant J, Silver J, Mari-nos N, et al. HIV-associated nephropathy in trans-genic mice expressing HIV-1 genes. Virology. 1991;185:109-19.
0. Gharavi AG, Ahmad T, Wong RD, Hooshyar R,Vaughn J, Oller S, et al. Mapping a locus for suscep-tibility to HIV-1-associated nephropathy to mousechromosome 3. Proc Natl Acad Sci U S A. 2004;101:2488-93.
1. Bruggeman LA, Dikman S, Meng C, Quaggin SE, Coff-man TM, Klotman PE. Nephropathy in human immu-nodeficiency virus-1 transgenic mice is due to renaltransgene expression. J Clin Invest. 1997;100:84-92.
2. Shankland SJ. The podocyte’s response to injury: rolein proteinuria and glomerulosclerosis. Kidney Int.2006;69:2131-47.
3. Barisoni L, Kriz W, Mundel P, D’Agati V. The dysregu-lated podocyte phenotype: a novel concept in thepathogenesis of collapsing idiopathic focal segmentalglomerulosclerosis and HIV-associated nephropathy.J Am Soc Nephrol. 1999;10:51-61.
4. Barisoni L, Bruggeman LA, Mundel P, D’Agati VD,Klotman PE. HIV-1 induces renal epithelial dediffer-entiation in a transgenic model of HIV-associated ne-phropathy. Kidney Int. 2000;58:173-81.
5. Schwartz EJ, Cara A, Snoeck H, Ross MD, SunamotoM, Reiser J, et al. Human immunodeficiency virus-1induces loss of contact inhibition in podocytes. J AmSoc Nephrol. 2001;12:1677-84.
6. Husain M, Gusella GL, Klotman ME, Gelman IH, Ross
MD, Schwartz EJ, et al. HIV-1 nef induces prolifera-tion and anchorage-independent growth in podocytes.J Am Soc Nephrol. 2002;13:1806-115.
7. Frankel AD, Young JA. HIV-1: fifteen proteins and anRNA. Annu Rev Biochem. 1998;67:1-25.
8. Zhong J, Zuo Y, Ma J, Fogo AB, Jolicoeur P, Ichikawa I,et al. Expression of HIV-1 genes in podocytes alone canlead to the full spectrum of HIV-1-associated nephropa-thy. Kidney Int. 2005;68:1048-60.
9. Zuo Y, Matsusaka T, Zhong J, Ma J, Ma LJ, Hanna Z, etal. HIV-1 genes vpr and nef synergistically damagepodocytes, leading to glomerulosclerosis. J Am SocNephrol. 2006;17:2832-43.
0. Fackler OT, Baur AS. Live and let die: nef functionsbeyond HIV replication. Immunity. 2002;16:493-7.
1. Hanna Z, Kay DG, Rebai N, Guimond A, Jothy S,Jolicoeur P. Nef harbors a major determinant ofpathogenicity for an AIDS-like disease induced byHIV-1 in transgenic mice. Cell. 1998;95:163-75.
2. Hanna Z, Weng X, Kay DG, Poudrier J, Lowell C,Jolicoeur P. The pathogenicity of human immunode-ficiency virus (HIV) type 1 Nef in CD4C/HIV trans-genic mice is abolished by mutation of its SH3-bind-ing domain, and disease development is delayed inthe absence of Hck. J Virol. 2001;75:9378-92.
3. He JC, Husain M, Sunamoto M, D’Agati VD, KlotmanME, Iyengar R, et al. Nef stimulates proliferation ofglomerular podocytes through activation of Src-de-pendent Stat3 and MAPK1,2 pathways. J Clin Invest.2004;114:643-51.
4. Husain M, D’Agati VD, He JC, Klotman ME, KlotmanPE. HIV-1 Nef induces dedifferentiation of podocytesin vivo: a characteristic feature of HIVAN. AIDS. 2005;19:1975-80.
5. Dickie P, Roberts A, Uwiera R, Witmer J, Sharma K,Kopp JB. Focal glomerulosclerosis in proviral and c-fmstransgenic mice links Vpr expression to HIV-associatednephropathy. Virology. 2004;322:69-81.
6. Andersen JL, Planelles V. The role of Vpr in HIV-1pathogenesis. Curr HIV Res. 2005;3:43-51.
7. Rosenstiel P, Gruosso T, Letourneau A, Chan J, Mo-hammed Husain VP, D’Agati V, et al. HIV-1 vpr im-pairs cytokinesis in human proximal tubule cells lead-ing to multinucleation and hypertrophy: implicationsfor HIVAN pathogenesis. Kidney Int. 2008. In press.
8. Pugliese A, Vidotto V, Beltramo T, Petrini S, Torre D.A review of HIV-1 Tat protein biological effects. CellBiochem Funct. 2005;23:223-7.
9. Conaldi PG, Bottelli A, Baj A, Serra C, Fiore L, Fed-erico G, et al. Human immunodeficiency virus-1 tatinduces hyperproliferation and dysregulation of renalglomerular epithelial cells. Am J Pathol. 2002;161:53-61.
0. Singhal PC, Sharma P, Reddy K, Sanwal V, Rawal J,Gibbons N, et al. HIV-1 gp160 envelope protein mod-ulates proliferation and apoptosis in mesangial cells.Nephron. 1997;76:284-95.
1. Kapasi AA, Patel G, Franki N, Singhal PC. HIV-1
gp120-induced tubular epithelial cell apoptosis is me-
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
Pathogenesis of HIVAN 533
diated through p38-MAPK phosphorylation. Mol Med.2002;8:676-85.
2. Singhal PC, Reddy K, Franki N, Ding G. HIV-1 gp120envelope protein modulates proliferation of humanglomerular epithelial cells. J Cell Biochem. 1999;76:61-70.
3. Bodi I, Abraham AA, Kimmel PL. Apoptosis in humanimmunodeficiency virus-associated nephropathy. Am JKidney Dis. 1995;26:286-91.
4. Bottinger EP. TGF-beta in renal injury and disease.Semin Nephrol. 2007;27:309-20.
5. Ziyadeh FN, Hoffman BB, Han DC, Iglesias-De La CruzMC, Hong SW, Isono M, et al. Long-term preventionof renal insufficiency, excess matrix gene expression,and glomerular mesangial matrix expansion by treat-ment with monoclonal antitransforming growth fac-tor-beta antibody in db/db diabetic mice. Proc NatlAcad Sci U S A. 2000;97:8015-20.
6. Grygielko ET, Martin WM, Tweed C, Thornton P,Harling J, Brooks DP, et al. Inhibition of gene markersof fibrosis with a novel inhibitor of transforminggrowth factor-beta type I receptor kinase in puromy-cin-induced nephritis. J Pharmacol Exp Ther. 2005;313:943-51.
7. Yamamoto T, Noble NA, Miller DE, Gold LI, HishidaA, Nagase M, et al. Increased levels of transforminggrowth factor-beta in HIV-associated nephropathy.Kidney Int. 1999;55:579-92.
8. Bodi I, Kimmel PL, Abraham AA, Svetkey LP, KlotmanPE, Kopp JB. Renal TGF-beta in HIV-associated kidneydiseases. Kidney Int. 1997;51:1568-77.
9. Kapasi A, Bhat P, Singhal PC. Tubular cell and HIV-1gp120 interaction products promote migration ofmonocytes. Inflammation. 1998;22:137-44.
0. Singhal PC, Sharma P, Loona R, Gibbons N, Franki N,Klotman PE. Enhanced proliferation, apoptosis, andmatrix accumulation by mesangial cells derived fromHIV-1 transgenic mice. J Investig Med. 1998;46:297-302.
1. Ross MJ, Fan C, Ross MD, Chu TH, Shi Y, Kaufman L,et al. HIV-1 infection initiates an inflammatory cas-cade in human renal tubular epithelial cells. J AcquirImmune Defic Syndr. 2006;42:1-11.
2. Ross MJ, Wosnitzer MS, Ross MD, Granelli B, GusellaGL, Husain M, et al. Role of ubiquitin-like proteinFAT10 in epithelial apoptosis in renal disease. J AmSoc Nephrol. 2006;17:996-1004.
3. Szczech LA, Edwards LJ, Sanders LL, van der Horst C,Bartlett JA, Heald AE, et al. Protease inhibitors areassociated with a slowed progression of HIV-relatedrenal diseases. Clin Nephrol. 2002;57:336-41.
4. Eustace JA, Nuermberger E, Choi M, Scheel PJ, Jr.,Moore R, Briggs WA. Cohort study of the treatment ofsevere HIV-associated nephropathy with corticoste-roids. Kidney Int. 2000;58:1253-60.
5. Laradi A, Mallet A, Beaufils H, Allouache M, MartinezF. HIV-associated nephropathy: outcome and progno-sis factors. Groupe d’ Etudes Nephrologiques d’Ile de
France. J Am Soc Nephrol, 1998;9:2327-35.6. Smith MC, Austen JL, Carey JT, Emancipator SN, Her-bener T, Gripshover B, et al. Prednisone improvesrenal function and proteinuria in human immunode-ficiency virus-associated nephropathy. Am J Med.1996;101:41-8.
7. Kimmel PL, Cohen DJ, Abraham AA, Bodi I, SchwartzAM, Phillips TM. Upregulation of MHC class II, inter-feron-alpha and interferon-gamma receptor proteinexpression in HIV-associated nephropathy. NephrolDial Transplant. 2003;18:285-92.
8. Sunamoto M, Husain M, He JC, Schwartz EJ, Klot-man PE. Critical role for Nef in HIV-1-induced podo-cyte dedifferentiation. Kidney Int. 2003;64:1695-701.
9. Bruggeman LA, Adler SH, Klotman PE. Nuclear factor-kappa B binding to the HIV-1 LTR in kidney: implica-tions for HIV-associated nephropathy. Kidney Int.2001;59:2174-81.
0. Ross MJ, Martinka S, D’Agati VD, Bruggeman LA.NF-{kappa}B regulates Fas-mediated apoptosis in HIV-associated nephropathy. J Am Soc Nephrol. 2005;16:2403-11.
1. Martinka S, Bruggeman LA. Persistent NF-kappaB ac-tivation in renal epithelial cells in a mouse model ofHIV-associated nephropathy. Am J Physiol RenalPhysiol. 2006;290:F657-65.
2. Wei P, Garber ME, Fang SM, Fischer WH, Jones KA. Anovel CDK9-associated C-type cyclin interacts di-rectly with HIV-1 Tat and mediates its high-affinity,loop-specific binding to TAR RNA. Cell. 1998;92:451-62.
3. Nelson PJ, Gelman IH, Klotman PE. Suppression ofHIV-1 expression by inhibitors of cyclin-dependentkinases promotes differentiation of infected podo-cytes. J Am Soc Nephrol. 2001;12:2827-31.
4. Gherardi D, D’Agati V, Chu TH, Barnett A, Gianella-Borradori A, Gelman IH, et al. Reversal of collapsingglomerulopathy in mice with the cyclin-dependentkinase inhibitor CYC202. J Am Soc Nephrol. 2004;15:1212-22.
5. Kaufman L, Hayashi K, Ross MJ, Ross MD, KlotmanPE. Sidekick-1 is upregulated in glomeruli in HIV-associated nephropathy. J Am Soc Nephrol. 2004;15:1721-30.
6. Yamagata M, Weiner JA, Sanes JR. Sidekicks: synapticadhesion molecules that promote lamina-specific con-nectivity in the retina. Cell. 2002;110:649-60.
7. Kaufman L, Yang G, Hayashi K, Ashby JR, Huang L,Ross MJ, et al. The homophilic adhesion moleculesidekick-1 contributes to augmented podocyte aggre-gation in HIV-associated nephropathy. FASEB J. 2007;21:1367-75.
8. Ross MD, Bruggeman LA, Hanss B, Sunamoto M, Mar-ras D, Klotman ME, et al. Podocan, a novel smallleucine-rich repeat protein expressed in the scleroticglomerular lesion of experimental HIV-associated ne-phropathy. J Biol Chem. 2003;278:33248-55.
9. Williams KJ, Qiu G, Usui HK, Dunn SR, McCue P,
Bottinger E, et al. Decorin deficiency enhances pro-
7
7
7
7
7
534 J.S. Leventhal and M.J. Ross
gressive nephropathy in diabetic mice. Am J Pathol.2007;171:1441-50.
0. Schrijvers BF, Flyvbjerg A, De Vriese AS. The role ofvascular endothelial growth factor (VEGF) in renalpathophysiology. Kidney Int. 2004;65:2003-17.
1. Eremina V, Sood M, Haigh J, Nagy A, Lajoie G, FerraraN, et al. Glomerular-specific alterations of VEGF-Aexpression lead to distinct congenital and acquired re-nal diseases. J Clin Invest. 2003;111:707-16.
2. Korgaonkar SN, Feng X, Ross MD, Lu TC, D’Agati V,
Iyengar R, et al. HIV-1 upregulates VEGF inpodocytes. J Am Soc Nephrol. 2008;19:877-83.
3. Kajiyama W, Kopp JB, Marinos NJ, Klotman PE,Dickie P. Glomerulosclerosis and viral gene expres-sion in HIV-transgenic mice: role of nef. Kidney Int.2000;58:1148-59.
4. Reid W, Sadowska M, Denaro F, Rao S, Foulke J, Jr.,Hayes N, et al. An HIV-1 transgenic rat that developsHIV-related pathology and immunologic dysfunction.
Proc Natl Acad Sci U S A. 2001;98:9271-6.
