HIV infection changes glomerular podocyte cytoskeletal composition and results in distinct cellular mechanical properties.

In addition to forming the selective filtration barrier for the renal glomerulus, podocytes maintain glomerular capillary architecture by opposing distending hemodynamic forces. To understand the relationship of cytoskeletal properties and the mechanical characteristics of podocytes, we studied filamin expression and distribution and measured cell membrane deformability in conditionally immortalized wild-type (WT) mouse podocytes, and in podocytes derived from a mouse model of HIV-associated nephropathy (HIVAN). In the WT cells, filamin and F-actin were localized at the periphery and in prominent stress fibers. In the HIVAN cells, filamin expression was reduced, and stress fibers were sparse. In a microaspiration assay, HIVAN cells ruptured under minimal negative pressure. Atomic force microscopy demonstrated that the WT cells had a stiffness of 17 kPa, whereas the value for the HIVAN cells was 4 kPa. These results demonstrate that the mechanical properties of WT and HIVAN podocytes are markedly different in a manner that is consistent with differences in the composition and arrangement of their cytoskeletons. The mechanical properties of the WT podocytes suggest that these cells can better maintain capillary integrity than the HIVAN podocytes and implicate pathological assembly of the cytoskeleton as a mechanism of HIVAN.

Detection and localization of HIV-1 DNA in renal tissues by in situ polymerase chain reaction.

The localization of HIV-1 DNA in renal tissues is critically important for understanding pathogenesis of HIV-associated nephropathy (HIVAN), but the clarification has been technically challenging. We applied in situ polymerase chain reaction (IS-PCR) to human renal tissues to demonstrate viral entry into the renal epithelial cells in vivo. To test the specificity of this method and to determine the cell types infected, we used IS-PCR followed by in situ hybridization (ISH) and IS-PCR followed by immunohistochemistry and histochemical counterstains. Brief 2 hour fixation in 4% paraformaldehyde had 92.9% sensitivity and 100% specificity for detection of viral DNA in renal biopsies of HIVAN patients, compared to 70.8% sensitivity and 66.7% specificity in renal biopsies fixed overnight in 10% formalin. Under optimized conditions, the only signals detectable in HIV-1 seronegative cases were false positives attributable to renal tubular apoptosis. In HIVAN cases, positive signal was observed in podocytes, parietal cells, renal tubular cells, and interstitial leukocytes. Immunohistochemical co-labeling for pan-T cell and macrophage markers revealed that the interstitial leukocytes with positivity for HIV-1 DNA included both T cells and macrophages. Application of ISH after IS-PCR showed the same distribution of signal as observed using IS-PCR alone, confirming the specificity of the technique. IS-PCR is a powerful technique to detect viral DNA in human tissue sections, but requires proper use of negative controls to set optimal fixation, protein digestion, and amplification conditions.

Nuclear factor-kappa B binding to the HIV-1 LTR in kidney: implications for HIV-associated nephropathy.

BACKGROUND: We have recently shown that renal epithelium is infected by HIV-1 and supports HIV-1 transcription in seropositive patients with renal disease. To investigate the regulation of HIV-1 gene expression in kidney, an HIV-1 transgenic mouse model was used to analyze the host transcriptional proteins that bind the 5' long-terminal repeat (LTR). METHODS: Viral gene expression was assessed in transgenic mouse tissue using Northern blotting and mRNA in situ hybridization. The transcription factors involved in LTR binding were determined using electrophoretic mobility shift assays. Cytoplasmic and nuclear extracts were prepared from tissues with varied levels of transgene expression. The binding of transcription factors to specific LTR fragments was determined using DNA competition experiments and supershifts with transcription factor-specific antibodies. RESULTS: Tissue-specific expression of the transgene was variable, with viral gene expression in the kidney at an intermediate level as compared with other tissues. Overall, the level of transgene expression directly correlated with abundance of nuclear factor-kappa B (NF-kappa B) in the nuclear extracts. High expressing tissue, however, had a constitutively active form of NF-kappa B. In contrast, the kidney contained an inducible NF-kappa B, which bound the LTR in combination with Sp1, suggesting a requirement for an activating event in renal HIV-1 expression of the LTR. CONCLUSIONS: These studies indicate that the regulation of the HIV-1 LTR in the kidney is similar to lymphoid tissues, and may explain, in part, why the HIV-1 life cycle is supported in kidney.

HIV-1 induces renal epithelial dedifferentiation in a transgenic model of HIV-associated nephropathy.

BACKGROUND: Human immunodeficiency virus-associated nephropathy (HIVAN) is the most common cause of renal failure in HIV-1-seropositive patients. Recent studies using an HIV-1 transgenic mouse model have demonstrated that expression of HIV-1 in the kidney is required for the development of HIVAN. What has remained unclear, however, is the renal cell type responsible for pathogenesis and the essential pathological process. METHODS: To address these issues, we used a transgenic murine model of HIVAN. We identified the cell types in kidney in which HIV transgene expression occurs using in situ hybridization. We evaluated evidence of proliferation by immunocytochemical analysis using an antibody to Ki-67 and cell type-specific markers, including WT-1, synaptopodin, Na+,K+-ATPase, adducin, and desmin. TUNEL assay was used to evaluate apoptosis. RESULTS: We found that glomerular and tubular epithelial cells express the HIV-1 transgene early in the disease process when renal architecture is well preserved. Transgene expression is lost, however, in tubular epithelial cells when they lose their differentiated cuboidal phenotype. In glomerular epithelial cells, dedifferentiation occurs with reduced expression of WT-1 and synaptopodin, in association with activation of desmin expression. Tubular microcysts also form with mislocalization of Na+,K+-ATPase expression to the lateral and apical cellular membranes. CONCLUSIONS: These studies support the hypothesis that the glomerular and renal epithelial cells are the primary targets of HIV-1 pathogenesis in the kidney. The essential pathologic process is dysregulation of the epithelial cell cycle with increased proliferation, apoptosis, cellular dedifferentiation, and altered cellular polarity.

Adeno-associated virus gene transfer into renal cells: potential for in vivo gene delivery.

The human parvovirus adeno-associated virus (AAV), type 2, has a number of features that make it an attractive choice as a vector for gene delivery to the kidney. AAV vectors permit long-term gene expression in vivo by integration into the host genome, have potential for site-specific integration on chromosome 19, do not express viral genes or generate a cellular immune response, and demonstrate enhancement of gene expression by chemotherapeutic agents that are approved for use in vivo. These properties confer advantages to AAV over other viral and nonviral methods for gene transfer. Preliminary experiments in our laboratory suggest that AAV is able to transfer genes to both renal cells in culture and the kidney in vivo. Thus, AAV has the potential to be an important gene transfer vector for the kidney in vivo.

Identification and characterization of a cell membrane nucleic acid channel.

We have identified a 45-kDa protein purified from rat renal brush border membrane that binds short single-stranded nucleic acid sequences. This activity was purified, reconstituted in proteoliposomes, and then fused with model planar lipid bilayers. In voltage-clamp experiments, the reconstituted 45-kDa protein functioned as a gated channel that allows the passage of nucleic acids. Channel activity was observed immediately after addition of oligonucleotide. Channel activity was not observed in the absence of purified protein or of oligonucleotide or when protein was heat-inactivated prior to forming proteoliposomes. In the presence of symmetrical buffered solution and oligonucleotide, current passed linearly over the range of holding potentials tested. Conductance was 10.4 +/- 0.4 picosiemens (pS) and reversal potential was 0.2 +/- 1.7 mV. There was no difference in channel conductance or reversal potential between phosphodiester and phosphorothioate oligonucleotides. Ion-substitution experiments documented a shift in reversal potential only when a concentration gradient for oligonucleotide was established, indicating that movement of oligonucleotide alone was responsible for current. Movement of oligonucleotide across the bilayer was confirmed by using 32P-labeled oligonucleotides. Channel open probability decreased significantly in the presence of heparan sulfate. These studies provide evidence for a cell surface channel that conducts nucleic acids.

Nephropathy in human immunodeficiency virus-1 transgenic mice is due to renal transgene expression.

HIV-associated nephropathy (HIVAN) is a progressive glomerular and tubular disease that is increasingly common in AIDS patients and one of the leading causes of end stage renal disease in African Americans. A major unresolved issue in the pathogenesis of HIVAN is whether the kidney disease is due to renal cell infection or a "bystander" phenomenon mediated by systemically dysregulated cytokines. To address this issue, we have used two different experimental approaches and an HIV-1 transgenic mouse line that develops a progressive renal disease histologically similar to HIVAN in humans. In the murine model, kidney tissue expresses the transgene and in heterozygous adults, renal disease develops shortly thereafter. We demonstrate by terminal deoxynucleotide transferase-mediated dUTP-biotin nick-end labeling assay that similar to the disease in humans, apoptosis of renal tubular epithelial cells is a component of the molecular pathogenesis. To determine whether apoptosis is due to transgene expression or environmental factors, we treated fetal kidney explants (normal and transgenic) with UV light to induce transgene expression. Apoptosis occurred in transgenic but not normal littermates after stimulation of transgene expression. To confirm a direct effect of HIV expression on the production of HIVAN, we transplanted kidneys between normal and transgenic mice. HIVAN developed in transgenic kidneys transplanted into nontransgenic littermates. Normal kidneys remained disease free when transplanted into transgenic littermates. Thus, the renal disease in the murine model is intrinsic to the kidney. Using two different experimental approaches, we demonstrate a direct effect of transgene expression on the development of HIVAN in the mouse. These studies suggest that in humans, a direct effect of HIV-1 expression is likely the essential cause of HIVAN, rather than an indirect effect of cytokine dysregulation.

Molecular therapy for renal diseases.

The introduction of molecular therapy through the delivery of nucleic acids either as oligonucleotides or genetic constructs holds enormous promise for the treatment of renal disease. Significant barriers remain, however, before successful organ-specific molecular therapy can be applied to the kidney. These include the development of methods to target the kidney selectively, the definition of vectors that transduce renal tissue, the identification of appropriate molecular targets, the development of constructs that are regulated and expressed for long periods of time, the demonstration of efficacy in vivo, and the demonstration of safety in humans. As the genetic and pathophysiologic basis of renal disease is clarified, obvious targets for therapy will be defined, for example, polycystin in polycystic kidney disease, human immunodeficiency virus (HIV) type 1 in HIV-associated nephropathy, alpha-galactosidase A in Fabry's disease, insulin in diabetic nephropathy, and the "minor" collagen IV chains in Alport's syndrome. In addition, several potential mediators of progressive renal disease may be amenable to molecular therapeutic strategies, such as interleukin-6, basic fibroblast growth factor (bFGF), platelet-derived growth factor (PDGF), and transforming growth factor-beta(TGF-beta). To test the in vivo efficacy of molecular therapy, appropriate animal models for these disease states must be developed, an area that has received too little attention. For the successful delivery of genetic constructs to the kidney, both viral and nonviral vector systems will be required. The kidney has a major advantage over other solid organs since it is accessible by many routes, including intrarenal artery infusion, retrograde delivery through the uroexcretory pathways, and ex vivo during transplantation. To further restrict expression to the kidney, tropic vectors and tissue-specific promoters also must be developed. For the purpose of inhibition of endogenous or exogenous genes, current therapeutic modalities include the delivery of antisense oligodeoxynucleotides or ribozymes. For these approaches to succeed, we must gain a much better understanding of the nature of their transport into the kidney, requirements for specificity, and in vivo mechanisms of action. The danger of a rush to clinical application is that superficial approaches to these issues will likely fail and enthusiasm will be lost for an area that should be one of the most exciting developments in therapeutics in the next decade.

Transport of phosphorothioate oligonucleotides in kidney: implications for molecular therapy.

The systemic administration of phosphorothioated antisense oligonucleotides has been demonstrated to be an effective strategy for the control of gene expression. Because previous studies have suggested both hepatic and renal accumulation of systemically administered oligonucleotides, we explored whether the kidney might be a site of free DNA transport. [32P]-phosphorothioate oligonucleotides (20 mers) were excreted in urine but cleared at only 30% of glomerular filtration rate. Plasma clearance of the label was very rapid (t1/2 approximately 5 min) but the half life of labeled S-deoxynucleotide excreted in urine was much slower (28 min). Infused oligonucleotide appeared in urine with little degradation. By autoradiography of renal tissue, labeled antisense oligonucleotides appeared within Bowman's capsule and the proximal tubule lumen. DNA was detected in association with brush border membrane and within tubular epithelial cells. Brush border membrane preparations from rat kidney contained oligonucleotide binding proteins as determined by gel mobility shift and UV cross linking assays. Because renal epithelial cells efficiently take up phosphorothioate oligonucleotides without apparent degradation, the kidney appears to be an excellent target for site-directed antisense therapy, but may be a site of antisense toxicity as well.

Transgenic models of HIV-1.

Transgenic technology has been very successful at providing insights into possible processes involved in HIV-induced pathogenesis. The availability of these small animal models for the study of HIV-related syndromes including KS, epidermal proliferative lesions, HIV-associated nephropathy, AIDS-related growth failure and cachexia may well facilitate the development of novel therapies for these complications. Other phenotypes created in mice, such as cataracts and hepatic cancer [59], may not have human analogies but may still provide insight into pathogenesis. Thus, transgenic models have already provided resources to study many manifestations of AIDS and others are likely to be developed. The optimal strategy for designing future transgenic animals, however, is less clear. No transgenic mouse model has been generated to date that will provide an avenue for vaccine development. This advance awaits the further discovery of the host factors that facilitate the virus replicative cycle in humans and a better understanding of these pathways in the mouse. For the development of molecular-based therapy, however, the currently available models may well be adequate to test molecular inhibitors of transcription [7,60,61] and post-transcriptional processing of viral mRNA [62]. Whether single or multigenic constructs under the control of the LTR are better or worse for this purpose is a debatable issue. Transgenic technology may yet make an additional contribution to the development of molecular therapy for AIDS. The best method of demonstrating that a gene therapeutic strategy is safe to administer to patients has not been determined. By introducing potentially therapeutic constructs into mice as transgenes, their safety can be assessed in many different cell types in vivo, analogous to toxicological testing in rodents for systemically administered drugs. Thus, transgenic technology has already provided insights into the pathogenesis of HIV-1. While it has not yet proven its utility for vaccine development, transgenic technology holds the promise of being an active participant in the development of both safe and effective gene therapy approaches for the treatment of AIDS.

bFGF and its low affinity receptors in the pathogenesis of HIV-associated nephropathy in transgenic mice.

HIV-associated nephropathy is characterized by extensive tubulointerstitial disease with epithelial cell injury, microcystic proliferation, and tubular regeneration with glomerulosclerosis. To explore the role of bFGF as a mediator of HIV-induced interstitial disease, we utilized an HIV transgenic mouse model that manifests clinical and histological features observed in patients. In transgenic mice, simultaneous renal epithelial cell proliferation and injury were detected in vivo. In areas of microcystic proliferation, immunoreactive bFGF colocalized with extracellular matrix. Kidneys from transgenic mice had increased bFGF low affinity binding sites, particularly in the renal interstitium. In vitro, transgenic renal tubular epithelial cells proliferated more rapidly and generated tubular structures spontaneously, in marked contrast to nontransgenic renal cells where these pathologic features could be mimicked by exogenous bFGF. These studies suggest that renal bFGF and its receptors play an important role in the pathogenesis of HIV-associated nephropathy.

Patterns of HIV-1 mRNA expression in transgenic mice are tissue-dependent.

To explore tissue-specific factors that may be important in HIV-1 transcriptional and post-transcriptional regulation, we examined a transgenic mouse model containing a mutant provirus deleted in the gag and pol region. The level of transgene expression was tissue-dependent. Skin, muscle, and tail consistently expressed the transgene abundantly; intestine, kidney, and thymus exhibited variable but generally low levels of expression; while liver expression was undetectable by Northern analysis. Individual mRNAs within the family of singly and multiply spliced messages were determined by reverse transcription (rt) of RNA samples from mouse tissues, polymerase chain reaction (PCR) amplification, and Southern hybridization with exon-specific probes. The exact percentage of Tat-coding mRNA that was multiply spliced was also determined by competitive rtPCR. When 2-, 4-, or 7-kb (full-length) mRNA species were calculated as a percentage of the total mRNA, two phenotypes of distribution were detected. Lymphoid tissue (thymus and spleen) and kidney had significantly greater amounts of unspliced message (P < 0.001) regardless of the level of expression. All other tissues expressed the multiply spliced messages encoding Tat, Rev, and Nef predominantly. Furthermore, utilization of the three major second exon splice acceptor sites for tat, rev, and nef was the same in transgenic mice as has been demonstrated in human cells but the splice acceptor site for the vpu/env was different in murine tissue. The marked tissue-dependent patterns of HIV mRNA expression suggest a potential mechanism for the organ-specific manifestations of AIDS.

Growth failure and AIDS-like cachexia syndrome in HIV-1 transgenic mice.

The mechanisms which predispose to growth failure in infants and children infected with immunodeficiency virus type-1 (HIV-1) are not fully understood. The contributions of viral replication and CD4+ T cell depletion to growth failure in an HIV-1 transgenic mouse model were investigated. Mice homozygous for the transgene, a gag-pol deletion mutant of the HIV-1 provirus pNL4-3, exhibited marked cachexia, growth retardation, lymphoproliferation with a reduction in the percentage of CD4+ T cells but an increase in the absolute number of splenic CD4+ and CD8+ T cells, thymic hypoplasia, and early death. Despite the absence of T cells, athymic nude mice, homozygous for the HIV transgene, displayed comparable growth failure. The results indicate that AIDS-like cachexia may be produced by expression of viral envelope or accessory genes, need not be accompanied by absolute depletion of CD4+ T cells, and may occur independent of T cell function.

Angiotensin II stimulates human fetal mesangial cell proliferation and fibronectin biosynthesis by binding to AT1 receptors.

The renin-angiotensin system is activated during vascular development and injury. Furthermore, angiotensin II (Ang II) is a comitogen for fetal mesangial cells in vitro and it may be important in vascular smooth cell growth in disease states. Since fibronectin is an important extracellular matrix protein for vascular development and it too is overexpressed in the mesangium of diseased glomeruli, we explored the interrelationship of fibronectin and Ang II in fetal mesangial cell growth. In human fetal kidney, Ang II type 2 receptors (AT2) were detected in abundance by ex vivo autoradiography. When mesangial cells were isolated from fetal kidney and grown in culture, Ang II type 1 receptors (AT1) were also detected. To explore the mitogenic properties Ang II and fibronectin as well as the effects of Ang II on fibronectin metabolism, studies were performed in vitro, isolated from the potentially confounding variables of hemodynamic influence and circulating growth factors and cytokines. In vitro, mesangial cells expressed a single class of AT1 receptors that were not altered by growth on various substrates. Ang II (10(-7) M) significantly increased thymidine incorporation by confluent human fetal mesangial cells (twofold). When subconfluent, Ang II-stimulated proliferation was greater (fourfold). Ang II significantly increased cell-associated and secreted fibronectin as determined by immunoprecipitation at concentrations that also stimulate mitogenesis. Both of these Ang II-mediated responses were inhibited by the AT1 receptor antagonist DuP-753 (10(-5) M) but not by AT2 receptor antagonist.(ABSTRACT TRUNCATED AT 250 WORDS)

Extracellular matrix gene expression in experimental glomerulonephritis.

Recent studies using animal models of glomerulonephritis indicate that steady-state levels of mRNAs encoding basement membrane proteins are frequently, but not universally, increased in parallel and that mRNAs encoding interstitial matrix proteins represent a separate set of genes that may also display coordinate regulation. A variety of maneuvers that ameliorate progressive glomerulosclerosis may act, at least in part, by suppressing glomerular cell matrix protein gene expression. The observed coordinate regulation of matrix genes may be a consequence of shared genetic regulatory sequences. Future therapies designed to retard glomerulosclerosis may take advantage of similarities among the extracellular mediators and regulatory sequences that influence matrix gene expression.

Thromboxane and prostacyclin differentially regulate murine extracellular matrix gene expression.

Alterations in the arachidonic acid metabolites thromboxane and prostacyclin are known to contribute to hemodynamic changes observed in certain models of acute and chronic renal failure. We have previously shown that thromboxane may have an important role in mediating glomerulosclerosis by stimulating the expression of certain extracellular matrix proteins. In the present study, we compared the effects of thromboxane and prostacyclin on the expression of genes encoding basement membrane proteins using a murine teratocarcinoma cell line, that when differentiated to an endodermal phenotype synthesizes abundant extracellular matrix. Incubation of these cells with stable analogs of thromboxane and prostacyclin for four hours resulted in changes in basement membrane gene expression. Thromboxane increased steady-state mRNA levels for all three laminin chains, type IV collagen, and fibronectin, but decreased the level of mRNA for heparan sulfate proteoglycan. In contrast, incubation with carbo-prostacyclin, a stable analog of prostacyclin, decreased the steady-state mRNA level for the laminin A and B1 chains, type IV collagen and fibronectin, and increased the mRNA level for heparan sulfate proteoglycan and laminin B2. Carbo-prostacyclin did not affect cellular proliferation or thymidine incorporation. These results indicate that eicosanoids directly modulate matrix gene expression independently of hemodynamic influence, and independently of effects mediated by platelets, or mitogenesis. Furthermore, these findings suggest that the alterations in renal eicosanoid metabolism may directly participate in the pathogenesis of glomerulosclerosis and thus provide a rationale for therapy directed toward the specific inhibition of thromboxane in the treatment of progressive glomerular sclerosis.

Identification of an activating transcription factor (ATF) binding site in the human transforming growth factor-beta 2 promoter.

Transforming growth factor TGF-beta 2 is encoded by multiple mRNA transcripts of 5.8, 5.1, 4.0, 3.8, and 2.8 kilobase pairs (kb) that are expressed in various human and monkey cells. Northern blot analysis using genomic fragments of DNA was used to demonstrate that some of this size heterogeneity is due to differences in the length of the 5'-untranslated region. Probes that were colinear with the first 600 nucleotides of the 5'-untranslated region detected only the 5.8-, 4.0-, and 3.8-kb transcripts. In order to identify DNA elements that regulate the transcription of these mRNA transcripts, deletion constructs of 5'-flanking DNA were ligated to the coding region for chloramphenicol acetyltransferase (CAT) and analyzed for promoter activity in several cell lines. Sequences responsible for putative enhancer and silencer regions were identified between -778 and -40 relative to the transcription initiation site. Addition of a cyclic AMP-responsive element/activating transcription factor-like element at -74 resulted in a 5-10-fold increase in CAT activity over that expressed with a construct that contained only the TATA box. This increase in CAT activity was suppressed by the addition of DNA sequences between -257 and -187, whereas sequences between -778 and -257 stimulated CAT activity. Point mutations within the ATF binding site at -74 resulted in a marked decrease in CAT expression. Cotransfection with ATF-1 or ATF-2 expression plasmids resulted in both dose-dependent stimulatory and inhibitory activities that were cell type-dependent. These studies identify multiple transcription initiation sites for TGF-beta 2 and demonstrate that transcription from one of these promoters is dependent upon an ATF binding site located 5' of the TATA box.