Human immunodeficiency virus downregulates podocyte apoE expression

Abstract

Apolipoprotein E (apoE) has been demonstrated to play an important role in providing protection against mesangial cell injury. In the present study, we evaluated the role of apoE and its associated downstream effects in human immunodeficiency virus (HIV)-associated nephropathy (HIVAN). Control (n = 6) and age- and sex-matched HIV-1 transgenic mice (Tg26, n = 6) were evaluated for their renal cortical expression of apoE. Renal tissue from Tg26 mice not only showed decreased apoE expression but also displayed downregulation of perlecan mRNA expression. In in vitro studies, conditionally immortalized human podocytes (CIHPs) were transduced with either NL4-3HIV (an HIV-1 construct lacking gag and pol, used for the development of Tg26 mouse model; NL4-3/CIHP) or empty vector (EV/CIHP); NL4-3/CIHPs and EV/CIHPs were studied for apoE mRNA expression. NL4-3/CIHPs showed reduction in apoE expression compared with EV/CIHPs. To evaluate the role of HIV-1 genes in the modulation of apoE expression, conditionally immortalized mouse podocytes (CIMPs) were transduced with individual HIV-1 gene constructs. Only nef-transduced CIMPs showed a decrease in apoE expression. To confirm this effect of nef in CIHPs, microarray analysis was performed in stable colonies of nef/CIHPs and EV/CIHPs. nef/CIHPs showed a 60% decrease in apoE and a 90% reduction in heparan sulfate mRNA expression. Moreover, nef transgenic mice showed a decrease in renal tissue expression of both apoE and perlecan. Both Tg26 and nef transgenic mice also showed areas of mesangial cell proliferation. These findings suggest that HIV-1-induced reduction in podocyte apoE expression and associated downregulation of podocyte perlecan might be contributing to mesangial cell (MC) phenotype in HIVAN.

apolipoprotein e (apoE) is a 299-amino acid glycoprotein that plays a key role in the lipoprotein trafficking by both stabilizing and solubilizing lipoprotein particles (42). apoE, as a constituent of chylomicrons, VLDL, and HDL, performs as a ligand for the receptor-mediated clearance of these particles (20). Alteration in isoforms of apoE gene expression has been shown to predispose to development of accelerated atherosclerosis, diabetic kidney disease, and Alzheimer disease in humans (10, 40).

Lipid metabolism contributes to the progression of human immunodeficiency virus (HIV)-1-associated nephropathy (HIVAN) in multiple ways. Cellular cholesterol content contributes to HIV replication (5). HIV-infected patients have been reported to have increased risk of atherosclerosis in general and coronary artery disease (CAD) in particular (17, 27). Dyslipidemia, caused by protease inhibitors (PI), is a known risk factor for pathogenesis of CAD in HIV-1-infected patients (31). However, a number of clinical reports suggest a relationship between occurrence of heart disease and HIV viral load, even in the absence of PI treatment (46).

apoE-null mice have been reported to develop both atherosclerosis and glomerulosclerosis (32, 45). Glomerulosclerosis is also one of the predominant renal lesions in HIVAN (1, 36). Mesangial expansion in the form of mesangial cell (MC) proliferation and matrix accumulation has been considered to be the precursor of the development of glomerulosclerosis (12). We previously reported (6) that apoE provides protection against the development of mesangial expansion. On the other hand, mice lacking apoE are prone to develop MC proliferation and matrix accumulation and subsequent progressive glomerulosclerosis (6). Later on, other investigators also confirmed these findings (3, 43). Since patients with HIV-1 infection are prone to develop glomerulosclerosis, we hypothesize that apoE deficiency may be contributing to the pathogenesis of HIVAN by promoting MC proliferation and expansion by inhibiting perlecan production by podocytes.

In the present study, we evaluated the effect of HIV-1 on podocyte apoE expression both in vitro and in vivo. In addition, we studied podocyte expression of perlecan (a constituent of heparan sulfate proteoglycan), a product of apoE-induced downstream signaling. Since the apoE-deficient phenotype manifests in the form of MC proliferation, we studied the characteristics of the MC phenotype in mouse models of HIVAN as well.

MATERIALS AND METHODS

HIV transgenic mice.

This protocol was reviewed and approved by the Institutional Animal Care and Use Committee of the Feinstein Institute for Medical Research, and all experiments were conducted in accordance with institutional requirements. We used three models of HIVAN including Tg26, nef, and vpr transgenic mice. Respective breeding pairs were obtained for the development of colonies of Tg26 (kindly provided by Dr. Paul E. Klotman, Department of Medicine, Mount Sinai Medical Center, New York), vpr transgenic (kindly provided by Dr. Jeffery Kopp, National Institutes of Health, Bethesda, MD), and nef transgenic (from M. Husain) mice. The Tg26 transgenic animal has the proviral transgene pNL4-3:d1443, which encodes all the HIV-1 genes except gag and pol and therefore is noninfectious (19). nef-Expressing transgenic mice were generated by crossing nef-knockin animals with podocin/cre animals (transgenic animals that constitutively express the bacteriophage CRE protein from the podocin promoter; gift from Susan Quaggin, SLR Institute, Mount Sinai Hospital, Toronto, Canada) (26). vpr Transgenic animals were generated by crossing podocin/rtTA (constitutively expresses rtTA, which is a fusion protein comprised of the TetR repressor and the VP16 transactivation domain from the podocin promoter) mice with tetop/Vpr mice (TRE-regulated Vpr gene). These animals were fed with doxycycline in their drinking water to induce the expression of the Vpr gene (11). We are maintaining colonies of these animals in our animal facility. To genotype these animals, the tail was clipped, DNA was isolated, and PCR studies were carried out with the following primers: for Tg26, HIV-5F ACATGAGCAGTCAGTTCTGCCGCAGAC, HIV-3R CAAGGACTCTGATGCGCAGGTGTG; for Vpr, VPR sense (TPR 1) GGATGGAACAAGCCCCAG, VPR antisense (TPR 2) CTCTAGGATCTACTGGCT; for Nef [since Nef was tagged with green fluorescent protein (GFP) we evaluated for the presence of GFP], GFP sense CGGGATCCACCGGTCGCCACCATGGTGAGCAAGGGCGAG, GFP antisense GGGTACCTCATTACTAATCGATTTACTTGTACAGCTCGTC; for CRE, Cre F ATGTCCAATTTACTGACCG, Cre R CCGCATAACCAGTGAAAC.

Podocytes.

Conditionally immortalized human podocytes (CIHPs) were provided by M. A. Saleem. Human podocytes were conditionally immortalized by introducing temperature-sensitive SV40-T antigen by transfection (38). These cells proliferate at permissive temperature (PT, 33°C) and enter growth arrest after transfer to nonpermissive temperature (37°C). The growth medium contains RPMI 1640 supplemented with 10% fetal bovine serum (FBS), 1× Pen-Strep, 1 mM l-glutamine, and 1× insulin-transferrin-selenium (ITS) to promote expression of the T antigen. Conditionally immortalized mouse podocytes (CIMPs) were grown and maintained as described previously (16).

Production of pseudotyped retroviral supernatant.

Replication-defective viral supernatants were prepared as published previously (16). The parental construct (pNL4-3:ΔG/P-GFP) was used to produce VSV.G pseudotyped viruses to provide pleiotropism and high-titer virus stocks. Infectious viral supernatants were produced by transient transfection of 293T cells using Effectene (Qiagen) according to the manufacturer's instructions. The HIV-1 gag/pol and VSV.G envelope genes were provided in trans with pCMV R8.91 and pMD.G plasmids, respectively (gifts of Dr. Didier Trono, Salk Institute, La Jolla, CA). As a negative control, virus was also produced from pHR-CMV-IRES2-GFP-ΔB plasmid, which contains HIV-1 long terminal repeats (LTRs) and GFP. The viral stocks were titrated by infecting 293T cells with 10-fold serial dilution as reported previously (16). The reciprocal of the lowest dilution showing expression of GFP was defined as GFP-expressing units (GEU) per milliliter. Viral stocks ranging from 10⁵ to 10⁶ GEU/ml were obtained. Some low-titer viral stocks were further concentrated by ultracentrifugation.

Individual HIV-1 genes (env, vif, vpr, and tat) were cloned into pHR-CMV-IRES2-GFP-ΔB vector, and the expression was confirmed by Western blot as described previously (16). The nef gene was cloned in pBabe-puro retroviral vector, and expression was confirmed by Western blotting. We developed stable colonies of nef- and empty vector (EV)-expressing podocytes.

Podocyte transduction.

Cells were plated in 24-well plates at a density of 10,000 cells per well in 1.0 ml of growth medium at PT. The cells were first allowed to grow at PT on a type I collagen-coated surface to 90% confluence and then transferred to 37°C for 2 wk to inactivate temperature-sensitive T antigen. The cells were transduced with a multiplicity of infection of 0.5 GEU for 2 h.

Reverse transcription PCR analysis.

Four-week-old FVB/N (control, n = 3) and Tg26 (n = 3) mice were studied for renal cell expression of apoE and perlecan. Renal cortical tissue of Tg26/nef and FVB/N mice were harvested, and RNA was extracted with TRIzol (Invitrogen). For cDNA synthesis, 2 μg of the total RNA was preincubated with 2 nmol of random hexamer (Invitrogen) at 65°C for 5 min. Subsequently, 8 μl of the reverse transcription reaction mixture containing cloned avian myeloblastosis virus reverse transcriptase, 0.5 mmol each of the mixed nucleotides, 0.01 mol dithiothreitol, and 1,000 U/ml RNAsin (Invitrogen) was incubated at 42°C for 50 min. For a negative control, a reaction mixture without RNA or reverse transcription was used. Samples were subsequently incubated at 85°C for 5 min to inactivate the reverse transcriptase.

Quantitative PCR was carried out in an ABI Prism 7900HT sequence detection system with the following primers: apoE (forward), 5′-GGTACTGGGCACTGAGAACCGCTCCTTCCC; apoE (reverse), 5′-GCTCCTGAAGGAACTGGAGCACGTCCCAGC; perlecan (forward), 5′-ggagagtcctccatatgcca-3′; perlecan (reverse), 5′ggatggaagtgtcagggaga-3′.

SYBR Green was used as the detector and ROX as the passive reference gene. Results (means ± SD) represent three animals as described in Figs. 1, 4, and 5. Data were analyzed with the comparative C_T (ΔΔC_T, where C_T is threshold cycle) method. Differences in C_T are used to quantify the relative amount of PCR target contained within each well. Data were expressed as relative mRNA expression in reference to control, normalized to quantity of RNA input by performing measurements on an endogenous reference gene, GAPDH. A representative gel electrophoresis was also carried out with α-tubulin as the housekeeping gene. After agarose gel electrophoresis, a Polaroid Camera System was used to capture images. Image J (Research Services Branch, National Institutes of Health, Bethesda, MD) was used to carry out densitometric analysis of RT-PCR gels.

Microarray analysis.

CIHPs were transduced with either HIV-1 nef or EV, and stable colonies of nef/CIHPs and EV/CIHPs were developed. nef/CIHPs and EV/CIHPs were harvested and RNA was isolated, followed by microarray analysis. Microarray analysis was performed in duplicate samples with an Illumina Human V2-bead chip (Illumina, San Diego, CA) for 45,000 genes. In nef/CIHPs, a total of 5,000 genes showed either upregulation or downregulation compared with vector-expressing podocytes. After a thorough search of upregulated and downregulated genes in various national databases, we composed a list of the five widely downregulated candidate genes associated with lipid metabolism.

Immunohistochemical staining.

Renal cortical sections from control and HIV-1 transgenic (HIVAN) mice were deparaffinized, and antigen retrieval was done by microwave heating. Endogenous peroxidase was blocked with 0.3% hydrogen peroxide in methanol for 20 min at room temperature (RT). Sections were washed in PBS three times and incubated in blocking serum solution for 30 min at RT, followed by incubation with anti-perlecan antibody (1:100, Cell Signaling Technology) or anti-proliferating cell nuclear antigen (PCNA) antibody (1:200, Sigma) overnight at 4°C in a moist chamber. Each of the sections was washed three times with PBS and incubated in respective secondary antibodies at 1:250 dilutions at RT for 1 h. After washing with PBS three times, sections were incubated in ABC reagent (Vector Laboratories, Burlingame, CA) for 1 h. Sections were washed three times in PBS and then placed in diaminobenzidine (DAB)-hydrogen peroxide solution, counterstained with hematoxylin, dehydrated, and mounted with a xylene-based mounting medium (Permount, Fisher Scientific, Fair Lawn, NJ). Appropriate positive and negative controls were used.

Western blotting.

To confirm the expression of Nef protein by nef/CIHPs, proteins were isolated from confluent nef/CIHPs and EV/CIHPs, Western blots were prepared and probed for Nef [anti-Nef antibody, 1:100, National Institutes of Health (NIH) AIDS Research and Reference Reagent Program, Division of AIDS, National Institute of Allergy and Infectious Diseases (NIAID)] as described previously (22). Murine tissue lysates were prepared by homogenizing fresh tissue in lysis buffer [in mM: 20 Tris, pH 7.5, 100 NaCl, and 1 ethylenediaminetetraacetic acid, with 1% Triton X-100 and Complete protease inhibitors (Calbiochem, San Diego, CA)]. For immunoprecipitation, 300 μg of mouse kidney lysate was incubated with primary antibody for apoE (Santa Cruz Biotechnology, Santa Cruz, CA) for 4 h at 4°C before being left overnight with protein A/G plus agarose (Santa Cruz Biotechnology). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out in 15% gels. For Western blotting, proteins were transferred onto nitrocellulose membranes and blocked in 5% milk protein before incubation with the primary antibody (Santa Cruz Biotechnology) overnight at 4°C. An anti-goat IgG/horseradish peroxidase conjugate was used for detection. All immunoblots were visualized by chemiluminescence (Perkin Elmer).

Confocal microscopy to localize apoE in podocytes.

Confocal microscopy was performed on paraffin sections of FVB/N and Tg26 mice and on Nef⁺/CRE⁻ and Nef⁺/CRE⁺ mice to visualize apoE in podocytes in vivo. Sections were deparaffinized in xylene for 5 min and then dehydrated and rehydrated by sequential ethanol treatment followed by incubation in distilled water for 5 min. Antigen retrieval was carried out by heating slides for 10 min in 10 mM citrate buffer. Endogenous peroxidase activity was blocked by immersing the slides in 0.3% hydrogen peroxide solution. Sections were then incubated with avidin (Vector Laboratories) and biotin (Vector Laboratories) for 15 min each. Blocking was done with 1% BSA along with goat serum. Renal cortical sections were incubated with apoE antibody (1:100, Santa Cruz Biotechnology) for 90 min at 37°C, followed by repeated washing and labeling with fluorescein anti-goat (1:3,000, Vector Laboratories). Sections were washed three times again with PBS and counterstained with propidium iodide (1 μg/ml) to stain the nucleus. Sections were then mounted and viewed under the confocal microscope. The sections were viewed with a ×100 oil immersion objective, and the confocal aperture was adjusted to 3. Dual channel selection was done, and images were obtained with FLUOVIEW software utilizing FITC, rhodamine, and transmission light filters. The images were superimposed, keeping the photomultiplier tube and gain for all images constant to optimize signal intensity.

RESULTS

Attenuated renal tissue expression of apoE and perlecan in Tg26 mice.

To determine whether Tg26 mice show any alteration in apoE expression, RNA was isolated from six control and six age- and sex-matched Tg26 mice (4 wk old) and evaluated for expression of apoE by RT-PCR and real-time PCR. Renal tissue in Tg26 mice showed reduction in both mRNA (Fig. 1A, representative gel; Fig. 1B, real-time PCR) and protein (Fig. 1D, representative Western blot) expression of apoE.

Fig. 1.

Attenuated renal tissue expression of apolipoprotein E (apoE) and perlecan in Tg26 mice. Renal cortical tissue RNA was isolated from 6 control and 6 age- and sex-matched Tg26 mice. RNAs were evaluated for apoE, perlecan, and α-tubulin by RT-PCR and real-time PCR. A: representative gels represent data from 2 control and 2 Tg26 mice. Top: renal tissue expression of apoE by Tg26 (Tg1 and Tg2) and control (control 1 and control 2) mice. Middle: expression of perlecan by Tg26 and control mice. Bottom: α-tubulin content. B: real-time PCR data (means ± SD) showing relative apoE ratio from 3 control and 3 Tg26 mice. C: real-time PCR data (means ± SD) showing relative perlecan mRNA expression from 3 control and 3 Tg26 mice. D: representative gel showing immunoprecipitation studies carried out to probe renal tissue apoE expression in a control and a Tg26 mouse. Top: apoE expression by a control and a Tg26 mouse. Bottom: actin expression (same amount of protein loading) under same conditions.

To evaluate whether a decrease in renal tissue expression of apoE by Tg26 mice was also associated with downstream alteration in apoE-regulated transcription of perlecan, we probed RNAs of the above-mentioned mice for the expression of perlecan and α-tubulin by RT-PCR. As shown in Fig. 1A, (representative gel) and C (real-time PCR), Tg26 mice showed attenuated renal tissue expression of perlecan.

To evaluate glomerular localization of apoE, we carried out confocal studies after labeling renal cortical sections of FVB/N and Tg26 mice with anti-apoE antibody and propidium iodide. As shown in Fig. 2, A and B, podocytes in FVB/N mice showed uniform labeling for apoE. On the other hand, podocytes in Tg26 mice (Fig. 2, C and D) showed only scant labeling of apoE.

Fig. 2.

Glomerular apoE localization by confocal studies. Renal cortical sections of control, Tg26, and nef-positive (control) and nef transgenic mice were immunolabeled for apoE and examined under confocal microscope. A: representative microphotograph of a renal cortical section from a control mouse showing green bright fluorescence for apoE by podocytes (arrows). Nuclei are indicated by pink fluorescence (propidium iodide staining). B: same microphotograph as shown in A without nuclear localization, only podocyte apoE fluorescence. C: representative microphotograph of a renal cortical section from a Tg26 mouse showing minimal green fluorescence by podocytes. D: same microphotograph as shown in C without nuclear localization, only podocyte apoE fluorescence. E: representative microphotograph of a renal cortical section from a control mouse (Nef⁺/CRE⁻) showing green bright fluorescence for apoE by podocytes (arrows). Nuclei are indicated by pink fluorescence. F: same microphotograph as shown in E without nuclear localization, only podocyte apoE fluorescence. G: representative microphotograph of a renal cortical section from a nef (Nef⁺/CRE⁺) transgenic mouse showing only minimal green fluorescence by podocytes. H: same microphotograph as shown in G without nuclear localization, showing only podocyte apoE fluorescence.

To study the localization of the altered perlecan expression, we carried out immunohistochemical studies for labeling perlecan in renal cortical sections of control and Tg26 mice. As shown in Fig. 3, glomerular basement membrane (GBM) and mesangial matrix (MM) of Tg26 mice (Fig. 3B) showed decreased expression of perlecan compared with control mice (Fig. 3A).

Fig. 3.

Glomerular expression of perlecan and mesangial cell (MC) proliferation. Top: glomerular perlecan expression. Renal cortical sections of Tg26, nef (Nef⁺/CRE⁺) transgenic, and respective control mice were immunolabeled for perlecan. Representative microphotographs of renal cortical sections from Tg26 (B), nef (Nef⁺/CRE⁺) transgenic (D), and respective control (A and C) mice show perlecan localization predominantly in glomerular basement membrane (GBM) and scanty deposition in the mesangium. Middle: MC proliferation in Tg26 and nef (Nef⁺/CRE⁺) transgenic mice. Renal cortical sections from Tg26, nef transgenic, and respective control mice were stained with hematoxylin and eosin (H & E) and evaluated for MC proliferation and mesangial sclerosis under light microscopy. Representative photomicrographs show areas of increased mesangial cellularity (arrows) and mesangial sclerosis (arrowheads) both in Tg26 (F) and nef transgenic (H) mice. Bottom: glomerular proliferating cell nuclear antigen (PCNA)-positive cells. To confirm the proliferation phenotype, renal cortical sections of Tg26, nef transgenic, and respective control mice were immunolabeled for PCNA. Representative photomicrographs show PCNA-positive cells (arrows) in the mesangium of Tg26 (J) and nef transgenic (L) mice. Tg26 mice also show abundance PCNA-positive podocytes.

Enhanced MC proliferation in Tg26 mice.

To determine whether attenuated renal tissue expression of apoE and perlecan was associated with any alteration in MC phenotype, we carried out light microscopic studies on hematoxylin and eosin (H & E)-stained renal cortical sections of control and Tg26 mice to evaluate phenotype of MCs. As shown in Fig. 3, Tg26 mice (Fig. 3F) showed areas of increased MC proliferation compared with control mice (Fig. 3E). To confirm occurrence of MC proliferation, renal cortical sections of control and Tg26 mice were immunolabeled for PCNA. As shown in Fig. 3, renal cortical sections of Tg26 mice (Fig. 3J) showed a higher number of mesangial PCNA-positive cells (5.5 ± 1.2 per glomerulus) compared with control mice (1.3 ± 1.1 per glomerulus, Fig. 3I).

Attenuated expression of apoE in HIV-1-transduced podocytes.

Since podocytes are cells that maintain homeostasis in the maintenance of GBM heparan sulfate (HS) content in general and perlecan in particular, we studied the effect of HIV-1 infection on podocyte expression of apoE. RNA was extracted from NL4-3/CIHPs and EV/CIHPs. Equal amounts of RNAs were loaded and probed for the expression of apoE by RT-PCR and real-time PCR. As shown in Fig. 4A (real-time PCR), NL4-3/CIHPs showed attenuated expression of apoE compared with EV/CIHPs. To confirm whether the attenuated apoE expression in NL4-3/CIHPs was also associated with alteration in podocyte perlecan expression, RNA isolated from NL4-3/CIHPs and EV/CIHPs was also probed for perlecan by real-time PCR. As shown in Fig. 4B, NL4-3/CIHPs also showed attenuated expression of perlecan compared with EV/CIHPs.

Fig. 4.

Effect of human immunodeficiency virus (HIV)-1 and Nef expression on podocyte apoE and perlecan expression. Conditionally immortalized human podocytes (CIHPs) were transduced either with NL4-3 or empty vector and evaluated for mRNA expression for apoE and perlecan (n = 5) by real-time PCR. A: real-time PCR data showing attenuated expression of apoE by HIV-1-transduced podocytes (NL4-3/CIHPs) when compared with empty vector-transduced podocytes (EV/CIHPs). B: real-time PCR data showing diminished expression of perlecan by NL4-3/CIHPs and EV/CIHPs. C: proteins were isolated from stably transfected nef and vector podocytes and probed for expression of Nef. nef-Transduced podocytes showed protein expression of Nef. D: representative gel showing immunoprecipitation studies carried out to probe apoE in stably transfected nef and vector podocytes. Top: apoE expression by vector- and nef-transduced podocytes. Bottom: actin expression (equal amount of protein loading) in vector- and nef-transduced podocytes.

HIV-1-mediated podocyte apoE expression is mediated through nef.

To determine the role of HIV-1 gene in the modulation of podocyte apoE expression, RNA was isolated from nef/CIHPs, tat/CIHPs, gp120/CIHPs, and vpr/CIHPs and probed by RT-PCR for apoE expression. Only nef/CIHPs showed attenuated expression of apoE (data not shown).

To confirm the role of nef in modulation of apoE expression in humans, stable colonies of nef/CIHPs and EV/CIHPs were established. To confirm the expression of Nef protein by nef/CIHPs, proteins were isolated from confluent nef/CIHPs and EV/CIHPs and Western blots were prepared and probed for Nef. As shown in Fig. 4C, only nef/CIHPs showed expression of Nef.

To evaluate the effect of Nef expression on podocytes, RNA was extracted from nef/CIHPs and EV/CIHPs and microarray analysis was performed. Gene chip data showed the results of several thousand genes and the effect of nef versus vector. Lipid related genes were searched for in the National Center for Biotechnology Information (NCBI) database. Genes were cross matched, and the ratios of the five most downregulated lipid-related genes between nef/CIHPs and EV/CIHPs were calculated (Table 1). Microarray analysis revealed 60% reduction in apoE and 90% reduction in HS expression in nef/CIHPs compared with EV/CIHPs. To confirm the effect of Nef on podocyte expression of apoE, protein was extracted from nef/CIHPs and EV/CIHPs, and immunoprecipitation studies were carried out to determine the expression of apoE by podocytes. As shown in Fig. 4D, nef/CIHPs showed a decrease in apoE expression.

Table 1.

Effect of Nef on podocyte gene expression pertaining to lipid metabolism

Gene ID	Encodes	Action	% Decrease vs. Vector
NM00410	FABP3	Fatty acid transport	50
NM00023	Lipoprotein lipase	Lipid degradation	45
NM00552	Heparan sulfate	Inhibits mesangial cell proliferation	90
NM00254	OLR1	Targets oxidized LDL	70
NM00041	apoE	Lipid metabolism	60

nef-Transduced (nef/CIHPs) and empty vector-transduced (EV/CIHPs) conditionally immortalized human podocytes were harvested in 2 sets of experiments. RNA was isolated, followed by microarray analysis. Gene chip data showed the results of several thousand genes and the effect of nef vs. vector. Lipid-related genes were searched for in the National Center for Biotechnology Information (NCBI) database. Genes were cross matched, and ratios of the 5 lipid-related genes that were downregulated and considered as candidate genes for altered lipid metabolism [which might be contributing to human immunodeficiency virus-associated nephropathy (HIVAN) phenotype] between nef/CIHPs and EV/CIHPs were calculated.

Next we wanted to know whether nef also modulated renal tissue apoE expression in vivo. We isolated RNAs from renal cortical tissues of control (Nef⁺/CRE⁻) and nef transgenic (Nef⁺/CRE⁺) mice to determine effects on apoE and perlecan expression. A representative gel of two control and two nef transgenic mice is shown in Fig. 5A; renal tissues from nef transgenic mice showed attenuated expression not only of apoE but also of perlecan compared with control mice. This effect of nef transgene expression on renal tissue expression of apoE (Fig. 5B) and perlecan (Fig. 5C) was confirmed by real-time PCR studies. Similarly, renal tissue of nef transgenic mice showed attenuated expression of the apoE protein (Fig. 5D).

Fig. 5.

Attenuated renal tissue apoE and perlecan expression in control (Nef⁺/CRE⁻) and nef-expressing (Nef⁺/CRE⁺) transgenic mice. RNA was isolated from 4 nef transgenic (Nef⁺/CRE⁺) and control (Nef^+/CRE⁻) mice. RNAs were evaluated by RT-PCR for apoE and perlecan expression. Real-time PCR studies were carried out for quantitative analysis in 3 control and 3 nef transgenic mice. A: representative gels of 2 nef transgenic and 2 respective controls showing renal tissue expression of apoE and perlecan. Top: renal tissue expression of apoE by nef transgenic (Nef 1 and Nef 2) and control mice (C1 and C2). Middle: renal tissue perlecan expression by nef transgenic and control mice. Bottom: renal tissue content of α-tubulin of the respective RNAs. B: real-time PCR data (means ± SD) showing relative renal tissue mRNA expression of apoE from 3 control and 3 nef transgenic mice. C: real-time PCR data (means ± SD) showing relative renal tissue mRNA expression of perlecan from 3 control and 3 nef transgenic mice. D: representative gel showing immunoprecipitation studies carried out to probe apoE from renal tissue of a control and a nef transgenic mouse. Top: apoE expression of a control and a nef transgenic mouse. Bottom: actin expression (same amount of protein loading) by a control and a nef transgenic mouse.

Effect of nef on glomerular apoE/perlecan expression and MC phenotype.

To evaluate glomerular localization of apoE in nef transgenic mice, we carried out confocal studies after labeling renal cortical sections of control and nef transgenic mice with anti-apoE antibody and propidium iodide. Control mice showed uniform labeling of apoE (Fig. 2, E and F), whereas podocytes in nef transgenic mice (Fig. 2, G and H) showed only scant labeling of apoE.

We then asked whether attenuated renal tissue apoE expression was associated with an alteration in glomerular perlecan localization in nef transgenic mice. Renal cortical sections of nef transgenic and control mice were immunolabeled for perlecan by both immunohistochemical and immunofluorescence techniques. As shown in Fig. 3, perlecan was localized predominantly in GBM and partly in the mesangium in both control and nef transgenic mice. However, glomerular expression of perlecan was attenuated in nef transgenic mice (Fig. 3D) compared with control mice (Fig. 3C).

To determine whether altered expression of perlecan had any consequence on MC phenotype on nef transgenic mice, renal cortical sections (stained with H & E) of control and nef transgenic mice were evaluated for MC proliferation. As shown in Fig. 3, glomeruli of nef transgenic mice (Fig. 3H) showed mild MC proliferation. Moreover, nef transgenic mice (Fig. 3L) also showed an increased number of PCNA-positive cells compared with control mice.

Vpr does not modulate renal tissue apoE expression.

To find out whether apoE and/or perlecan attenuation is also playing a role in the HIVAN phenotype of Vpr transgenic mice, RNA was isolated from renal cortical tissues of two Vpr transgenic mice receiving either doxycycline or vehicle (control) in drinking water for 6 wk. RNA was evaluated for the expression of apoE, perlecan, and α-tubulin. Vpr transgenic mice did not show any alteration in renal tissue apoE or perlecan expression (data not shown).

DISCUSSION

The present study demonstrates that Tg26 mice have attenuated renal tissue expression of both apoE and perlecan. Both NL4-3/CIHPs and NL4-3/CIMPs showed attenuated expression of apoE. Microarray analysis carried out in nef/CIHPs showed a 60% decrease in apoE and a 90% decrease in HS expression. Renal cortical tissue of nef transgenic mice also showed decreased expression of apoE. Moreover, nef transgenic mice showed decreased GBM and MM expression of perlecan. These findings suggest that HIV-1-mediated podocyte expression of apoE may be mediated through nef.

ApoE has been reported to modulate HIV-AIDS pathogenesis in multiple ways. There is increasing evidence suggesting that the amphipathic helical domains of apolipoproteins may act as viral fusion inhibitors, because of their homology with the fusogenic domains of viral fusion proteins (12, 24, 29, 35, 39). Thus the amphipathic helix domains of apoE may inhibit HIV infection in a manner analogous to the clinical HIV fusion inhibitor Enfurvitide (2, 25). However, recently Burt et al. (4), in in vitro studies, showed that certain isoforms of apoE promote HIV infection by facilitating a greater frequency of fusion events. These observations provided a possible mechanism underlying the in vivo report suggesting accelerated HIV-1 disease progression in humans expressing certain apoE isoforms (4, 22, 23, 42). apoE also has potential to modulate HIV-1 attachment through heparan sulfate proteoglycans—a rate-limiting step in HIV infection. Interestingly, tandem repeat peptides synthesized based on the heparin-binding domain of apoE have been reported to contain notable antiviral activity when present in cell culture during HIV infection; this effect of apoE has been attributed to blockade at the level of attachment (12, 29). Heparan sulfate proteoglycan-bound apoE is abundant on many cell types and has been known to be carried with HIV particles that bud from these cells (8). Although apoE isoforms may have similar receptor-binding capacities, they have very different lipid-binding characteristics (21). It is possible that differing affinities for lipid may convey differential abilities to bind to the HIV envelope or to complex with cell-surface cholesterol-phospholipid rafts, through which HIV is believed to penetrate target cells (33, 34). Although the role of apoE in HIV-1 entry into podocytes has not been investigated, preliminary studies in our laboratory indicate that HIV-1 entry into podocyte is lipid raft dependent (unpublished observations). Thus it appears that apoE deficiency may also contribute to HIVAN pathogenesis by modulating HIV-1 entry into podocytes.

In various models of glomerulosclerosis, deficient production of HS has been demonstrated to contribute to the development of renal lesions (9, 37). HS is not only an integral glomerular basement protein but also an important component of MM (44). It inhibits MC proliferation and is perhaps responsible for maintaining MCs in a quiescent state in vivo (6). Moreover, HS also regulates assembly of latent transforming growth factor (TGF)-β binding protein (7). HS in general, and its specific component perlecan in particular, are synthesized by podocytes and MCs. Podocytes maintain homeostasis of HS content of GBM. Lack of HS in GBM has been demonstrated to contribute to proteinuria in various models of glomerulosclerosis (9, 37). Moreover, in diabetic proteinuric patients, therapeutic supplementation of HS has been reported to reduce the severity of proteinuria (13, 14). These findings not only indicate a causal relationship between proteinuric states and deficiency of HS but also suggest its therapeutic importance for future studies.

In the present study, HIVAN mice not only showed attenuated expression of apoE but also showed its downstream effects in the form of reduced glomerular expression of perlecan. We previously reported (6) a causal relationship between apoE and MC proliferation in both in vitro and in vivo studies. Interestingly, in the present study, the attenuated expression of perlecan in Tg26 mice was associated with MC proliferation and accumulation of MM. Since mesangial expansion has been shown to be a precursor of glomerulosclerosis, one may speculate that lack of apoE may be partly contributing to the development of glomerulosclerosis in HIVAN mice.

In the kidney, perlecan is usually produced by podocytes and MCs (6, 41). Both podocytes and MCs maintain homeostasis of perlecan content in GBM and MM under physiological conditions. However, deficient production of perlecan is likely to occur in conditions associated with podocyte injury. Since perlecan has an inhibitory effect on MC proliferation, we propose that attenuated production of perlecan by HIV-infected podocytes may be contributing to the MC proliferation phenotype in HIVAN.

In our in vitro studies, both NL4-3/CIMPs and NL4-3/CIHPs showed attenuated expression not only of apoE but also of HS. If we extend these findings to HIVAN mice, it is likely that HIV-infected podocytes are contributing to the attenuated expression of renal tissue expression of apoE and the associated downstream reduction in HS expression.

nef Transgenic mice have been shown to develop MC proliferation and mesangial expansion (15, 47). We asked whether apoE is playing a role in the MC phenotype in nef transgenic mice. As expected, nef transgenic mice not only showed attenuated expression of ApoE but also showed decreased expression of perlecan. To confirm the role of nef in the modulation of apoE expression, we carried out microarray analysis on EV/CIHP and nef/CIHP showed a 60% reduction in apoE expression. Moreover, we observed not only attenuated glomerular expression of perlecan but also areas of MC proliferation in nef transgenic mice.

Recent data indicate that HIV-1 infection is an independent risk factor for CAD (28). Mujawar et al. (30) reported that HIV-1 infection and transfection with nef caused redistribution of ABC1 to the plasma membrane and inhibited internalization of apoA-1 in macrophages. HIV-infected and nef-transfected macrophages accumulated a substantial amount of lipids and resembled foam cells. These investigators also showed HIV-positive foam cells in atherosclerotic plaques of HIV-infected patients. Foam cells are key players in the formation of atheromatous plaque (31). The majority of foam cells are macrophages loaded with cholesterol. Dyslipidemia and/or impairment of intracellular cholesterol metabolism promotes foam cell formation (31). The findings in the present study suggest that HIV proteins in general and nef in particular might also contribute to the altered lipid metabolism by decreasing the transcription of renal tissue apoE.

Earlier reports have documented the role of nef in podocyte dedifferentiation and proliferation both in vitro and in vivo (15, 16). Those studies also suggested that in addition to nef other factors are also required for the podocyte phenotype seen in the HIVAN mouse model (15). The present study presents an additional role of nef, manifested in the form of MC proliferation in mouse models of HIVAN.

GRANTS

This study in part was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-083931.