Sidekick-1 Is Upregulated in Glomeruli in HIV-Associated Nephropathy
Lewis Kaufman,
Kayo Hayashi,
Michael J. Ross,
Michael D. Ross and
Paul E. Klotman
Department of Medicine, Division of Nephrology, Mount Sinai School of Medicine, New York, New York
Correspondence to Dr. Lewis Kaufman, Mount Sinai School of Medicine, Annenberg Building, Room 23-40, Box 1243, One Gustave L. Levy Place, New York, NY 10029. Phone: 212-241-7088; Fax: 212-987-0389; E-mail: lewis.kaufman{at}mssm.edu
ABSTRACT. Infection of podocytes by HIV-1 induces unique changesin phenotype, which contribute to the pathogenesis of glomerulardisease in HIV-associated nephropathy (HIVAN). The host geneticpathways altered by HIV-1 infection that are responsible forthese phenotypic changes are largely unknown. For identifyingsuch pathways, representational difference analysis was performedcomparing cDNA from HIV-1 transgenic podocytes with nontransgeniccontrols. In this way, a gene named sidekick-1 (sdk-1) was cloned,a transmembrane protein of the Ig superfamily that is highlyupregulated in HIV-1 transgenic podocytes. Sdk-1 and its ortholog,sidekick-2 (sdk-2), were recently shown to guide axonal terminalsto specific synapses in developing neurons. Their presence androle in other organs, including the kidney, has not been described.The current study demonstrates developmental expression of bothsdk-1 and sdk-2 and a tight spatial and temporal regulationof these genes in kidney. During nephrogenesis, sidekick expressionwas observed first in ureteric bud and ureteric bud–derivedtissues in a pattern similar to other genes known to play importantroles in branching morphogenesis. In adult murine renal tissue,sidekick proteins were seen in glomeruli at low levels, andexpression of sdk-1 was greatly upregulated in diseased HIV-1transgenic mouse kidneys. In a human HIVAN kidney biopsy, sidekickexpression was increased in glomeruli in a pattern consistentwith the mouse model. It is proposed that the dysregulationof sdk-1 protein may play an important role in HIVAN pathogenesis.
HIV-associated nephropathy (HIVAN) is the third leading causeof ESRD in black individuals between the ages of 20 and 64 inthe United States (1,2). The clinical and pathologic findingsof HIVAN include focal and segmental glomerulosclerosis, oftenof the collapsing variant, combined with microcystic tubulointerstitialdisease (3,4). The disease untreated is rapidly progressive,often leading to renal failure within weeks to months.
HIVAN occurs almost exclusively in blacks and Hispanics (5);of all causes of ESRD, only sickle cell disease is more closelyassociated with black race than HIVAN (6). This striking racialpredilection suggests that genetic factors play an importantrole in disease susceptibility. Moreover, recent studies inthe HIV-1 transgenic mouse model suggest that host genetic responseslikely determine both susceptibility and progression of HIVAN(7,8). Thus, events downstream of infection are critical. Thegoal of the current studies was to identify unique pathwaysinduced by HIV-1 infection that could determine susceptibilityto disease. Furthermore, as blacks are at much higher risk ofdeveloping all forms of renal disease compared with the generalpopulation, pathways that are important in HIVAN pathogenesismay also explain the strong racial predilection for renal diseasein this population in general.
Recent evidence demonstrates that HIV-1 directly infects renalepithelium (9) and replicates within this unique compartment(10). In individuals with HIVAN, the impact of viral expressionis manifested by a profound change in the cellular phenotype.The predominant glomerular target of HIV-1 is the glomerularvisceral epithelial cell, or podocyte. Podocytes are highlyspecialized cells that are critical in forming the kidney’sultrafiltration barrier. Instead of remaining in a relativelyquiescent differentiated state, podocytes respond to HIV-1 infectionby dedifferentiating (11,12) and proliferating (13,14). Suchproliferation, which is unique to HIVAN and idiopathic collapsingglomerulopathy (CG), is manifested by increased podocyte number,expression of Ki-67 and cyclin A, and downregulation of cyclin-dependentkinase inhibitors such as p27 and p57 (12,14). Podocyte dedifferentiation,also uniquely characteristic of HIVAN and CG, is typified byloss of typical differentiation markers such as synaptopodinand WT-1 (11,12).
HIV-1 infection seems to induce a characteristic proliferativeresponse in podocytes, and we are just beginning to understandthe mechanisms responsible. We now know, for example, that expressionof HIV-1 as a transgene (13,15) or infection of podocytes byHIV-1 is required to induce these changes. In addition, recentstudies have defined the Nef gene as the likely gene productresponsible (16,17). More specifically, an interaction betweenthe proline-rich domain of Nef with Src kinases seems to initiatethe cascade of events that leads to cellular proliferation.What is less well understood, however, are the host cellularand genetic responses to expression of HIV-1 in podocytes. Recently,we identified a novel small leucine-rich protein that is dramaticallyincreased in infected podocytes (18). This protein, named podocan,seems to localize to the glomerular basement membrane and isgreatly increased as a component of sclerotic glomerular lesionsin experimental HIVAN.
In the current studies, we used representational differenceanalysis (RDA) of cDNA to isolate differentially expressed genesin HIV-1 transgenic podocytes. In this way, we identified arecently cloned gene that responds to HIV-1 infection namedsidekick-1 (sdk-1). Sidekick was initially identified in Drosophilamelanogaster as being important in specifying photoreceptorcell fate in the retina (19). More recently, sdk-1 and its ortholog,sidekick-2 (sdk-2), were shown to be important adhesion moleculesin directing synapse formation in developing chicken retinasby guiding axon terminals to specific lamalae (20). However,the potential role for sidekicks in the development of otherorgan systems, including the kidney, is completely unknown.In the current article, we describe the regulation and expressionof sidekicks in developing renal epithelium and demonstratea dramatic upregulation of sdk-1 in the setting of HIV-1 infection.We propose that sdk-1 upregulation may play an important rolein podocyte dysfunction in HIVAN.
Representational Difference Analysis of Podocyte cDNA
Temperature-sensitive conditionally immortalized HIV-1 transgenicand wild-type podocyte cell lines were previously establishedby breeding heterozygous HIV-1 transgenic mice (Tg26; FVB/N,pNL4-3d1443 [-gag, -pol]) with H-2Kb –tsA58 Immortomice;Charles River Laboratories, Wilmington, MA) (13). Murine podocyteswere isolated from the HIV-1 transgenic mice and nontransgeniclittermates at the onset of detectable disease in the HIV-1transgenic mice as described previously (13). Before the RDAwas performed, these wild-type and HIV-1 transgenic podocyteswere cultured for 2 wk under "nonpermissive" conditions to inducedegradation of T antigen and allow for maximal differentiation.RDA of cDNA was then performed using the original protocol describedby Hubank and Schatz (21) and as previously reported by Rosset al. (18). RDA was carried out through the generation of thesecond difference product. The nontransgenic driver was supplementedwith 4 µg of the plasmid DNA used to generate the HIV-1transgenic line (pNL4-3:d1443) digested with DpnI.
Cloning Murine Sdk-1 and Sdk-2
Twenty-nine genes were differentially expressed and confirmedby Northern blot analysis to be increased or decreased in responseto HIV-1. An 1.0-kb difference product identified by RDA waspredicted to be dramatically upregulated in the HIV-1 transgenicpodocytes. This difference product was homologous to the Drosophilamelanogaster sidekick gene. Using this 1.0-kb difference productas a probe, we screened a -ZapII (Stratagene) cDNA phage librarygenerated from size-selected polyA+ RNA from the HIV-1 transgenicpodocytes. In this way, we cloned the carboxyl terminal of sdk-1.The 5' portion of the 6.5-kb transcript was then cloned usingthe 5' rapid amplification of cDNA ends (5' RACE; Ambion) ontotal RNA extracted from embryonic day 17 (E17) mouse kidneysas per the manufacturer’s protocol. Reverse transcriptionwas performed using the sequence specific primer TTCGATTCCACACGGTGAGCATGG.Amplification steps were done using the nested primers CGTGTTGGCTGCATAGCAGGTGTAfollowed by TTGCCAGCATCCTGGATGAAGACGG using pfu turbo DNA polymerase(Stratagene). The 5' portion of the longer 8-kb transcript wasalso cloned by 5' RACE using HIV-1 transgenic podocyte RNA asthe template. The same primers were used as for cloning the6.5-kb transcript. Amplification was done using platinum pfxDNA polymerase (Invitrogen).
Sdk-2 was cloned via PCR using E17 kidney cDNA as a template.Primers were designed using available mouse genomic DNA andan EST sequence from Genbank (accession no. BB026945). The c-terminal3.5-kb portion of the gene was cloned using the primer TTTGGGAGGCGGCAAAGGAAas the 3' untranslated sequence specific primer for RT followedby PCR amplification using primers CCCACCCAGTCAACCTTCTA andTCAAACGAAGGATGAGAACC. The N-terminal 3.3-kb of the gene wascloned next using CAGCTGCTCAATAGTGTATT as the sequence-specificprimer for RT followed by PCR amplification using primers AGTCTCCCGGGACTCCAAAG(upstream noncoding) and CACGTTGGCTGGAGCTATGT. RT reactionswere performed using Superscript II (Invitrogen), and PCR reactionswere done using the Expand Long Template PCR System (Roche).PCR conditions were per the manufacturer’s protocol. Thetwo overlapping PCR products were both cut with XbaI and thenligated together into an expression vector to form full-lengthsdk-2 cDNA.
Analysis of Sidekick RNA Expression
Total RNA was extracted from cultured podocytes using TriReagent(Sigma). Messenger RNA was purified using the FastTrack mRNAIsolation Kit (Invitrogen). Two micrograms of mRNA was separatedby agarose gel electrophoresis, transferred to a 0.45-µmBiodyne A nylon membrane (Pall Gelman Laboratory), and probedwith the P32-labeled 1.0-kb difference product.
For performing multitissue analysis of sdk-1 and sdk-2, organswere harvested from both E17 embryos and 4-mo-old adult mice.For analyzing the temporal regulation of sdk-1 and -2, kidneyswere isolated from embryos and harvested at embryonic days 12to 18 and from postnatal mice ages 0 to 64 d at various intervals.Northern blot analysis was performed as above using 12 µg(multitissue northern) and 15 µg (temporal northern) oftotal RNA per sample. Full-length sdk-1 and the N-terminal 3.2kb of sdk-2 were used as the probes.
Production and Characterization of Anti-Sidekick Antibodies
The StuI-EcoNI fragment of sdk-1 cDNA was subjected to Klenowpolymerase fill in reaction and then cloned into the ApaI siteof the PGEX-4T2 bacterial expression vector (Amersham Pharmacia).The resulting glutathione S-transferase fusion protein was producedin BL21 cells and purified using B-Per GST Protein PurificationKit (Pierce) as per the manufacturer’s protocol. The purifiedfusion protein was then provided to Research Genetics (Invitrogen)for production of rabbit anti-sidekick polyclonal antiserum.
The BamHI-BamHI fragment of murine sdk-1was cloned into theBamHI site of pcDNA3.1 (Invitrogen), producing a full-lengthsdk-1 expression vector. The EcoRV-EcoRI fragment of murinesdk-2 was cloned into the KpnI (blunted with T4 DNA polymerase)and EcoRI sites of pcDNA4/HisMax (Invitrogen), resulting ina carboxyl terminal (900 bases) sdk-2 expression vector. Sdk-1,partial sdk-2, and corresponding control vectors were transfectedinto HEK-293T cells using Lipofectamine 2000 (Invitrogen). After48 h, the cells were washed three times in 1x PBS and lysedin Bug Buster (Novagen) supplemented with 1x complete proteaseinhibitor cocktail (Roche) and 1 µl/ml nuclease (Novagen).The samples were separated by SDS-PAGE and transferred to Immobilon-Ptransfer membranes (Millipore). Western blots were performedusing 1:5000 dilutions of anti-sidekick and preimmune sera anda 1:10,000 dilution of horseradish peroxidase (HRP)–labeledgoat anti-rabbit IgG (Kirkegaard and Perry).
Immunolocalization of Sidekicks in Kidney
Kidneys were isolated from E12.5 and E15.5 mice, fixed for 2h in 4% paraformaldehyde in 1x PBS, and then paraffin-embedded.Paraffin-embedded human fetal kidney (24 wk) sections were similarlyprepared. After dewaxing and rehydration through graded ethanols,endogenous peroxidase activity was blocked by incubation in7.5% H2O2 for 10 min. Anti-sidekick and preimmune sera wereused at 1:500 dilutions with a 1-h incubation at room temperature.A biotinylated goat anti-rabbit antibody (Kirkegaard and Perry)was used as the secondary antibody at a 1:200 dilution. Streptavidin/biotin-HRPcomplex was introduced using Vectastain Elite ABC Kit (VectorLabs). Color reaction was created using AEC Substrate Kit forHorseradish Peroxidase (Vector Labs).
Kidneys from adult HIV-1 transgenic and age-matched wild-typemice were harvested and immediately frozen in OCT Compound (Tissue-Tek).Thin sections (5 µm) were cut using the CryoJane Tape-TransferSystem (Instrumedics) and fixed for 7 min in ice-cold acetone.Endogenous peroxide was blocked by incubation in 7.5% H2O2 for10 min. Tissue staining was performed using the TSA Cy3 System(Perkin Elmer) as per the manufacturer’s protocol. Theanti-sidekick and preimmune sera were used at 1:500 dilutionfor 2 h at room temperature.
Frozen kidney sections (8 µm) were prepared from an adultpatient with biopsy-confirmed HIVAN and from a block of adultnormal kidney. Endogenous peroxide was blocked in 0.3% H2O2in methanol for 60 min. Endogenous biotin was also blocked withAvidin-Biotin Blocking Kit (Vector Labs). Tissue staining wasperformed using the TSA HRP System (Perkin Elmer) as per themanufacturer’s protocol at 1:500 dilutions of primaryantibody. Color reaction was generated using AEC Substrate Kit(Vector Labs).
For performing semiquantitative RT-PCR, total kidney RNA wasextracted from age-matched HIV-1 transgenic and wild-type adultmice using Tri-Reagent (Sigma). RT (Superscript; Invitrogen)was performed on 1 µg of both RNA samples using an oligo-dTprimer. With the use of consecutive dilutions of cDNA as templates,input RT was normalized using primers specific for the housekeepinggene glyceraldehyde-3 phosphate dehydrogenase. PCR amplificationof sdk-1 and sdk-2 was then performed using the normalized samples.Primers were designed to amplify either sdk-1 or sdk-2 specifically(sdk-1 primers, CCCACCCAGTCAACCTTCTA and CTCAGGGTTGCCGTTGTATT;sdk-2 primers, GCCCAAGATGATGTACCGCC and AGAGGGTGCTCTGCTGACTC).PCR without added RT product (RT–) was performed as anegative control.
Identification and Cloning of Mouse Sidekicks
Representational difference analysis was performed on cDNA fromconditionally immortalized HIV-1 transgenic and nontransgenicpodocyte cell lines (13) (see Materials and Methods). More than250 difference products were characterized corresponding to29 differentially expressed genes between the two cell populations.The differentially expressed genes included known extracellularmatrix components, adhesion molecules, transcription factors,and previously uncharacterized genes. Among the cDNA predictedto be most dramatically upregulated in HIV-1 transgenic podocytescompared with controls was a 1-kb difference product that waspredicted to encode a protein bearing homology to the Drosophilamelanogaster sidekick gene. To confirm the results of the RDA,we used this 1-kb difference product as a probe for Northernblot analysis of messenger RNA isolated from the transgenicand wild-type podocytes (Figure 1). Three distinct transcripts(6.5, 8, and 10 kb) were each highly expressed in the HIV transgenicpodocytes compared with wild-type controls, thus verifying thedifferential expression.
Figure 1. Expression of sidekick-1 (sdk-1) RNA in cultured podocytes. Northern blot analysis of sdk-1 mRNA demonstrates expression of sdk-1 in HIV-1 transgenic podocytes but not in nontransgenic control podocytes. The 1-kb difference product isolated from the representational difference analysis was used as the probe. Sdk-1 seems to be expressed as three distinct transcripts ranging in size from 6.5 to 10 kb. Hybridization with glyceraldehyde-3 phosphate dehydrogenase (GAPDH) was performed to demonstrate equivalent loading of RNA.
We then cloned the 6.5- and 8-kb transcripts by combining cDNAphage library screening and 5' RACE PCR. The novel gene, namedmouse sidekick-1 (sdk-1), is 35% homologous at the amino acidlevel to the Drosophila sidekick gene and 80% identical to themore recently described chicken sdk-1. The cloned 8-kb cDNAencodes a putative type I transmembrane protein that containsa large extracellular domain composed of 6 Ig motifs followedby 13 fibronectin type III repeats, a short transmembrane domain,and a cytoplasmic tail of 200 amino acids (Figure 2). This sequenceis consistent with the predicted domain organization of othersidekicks in other species. The coding region of the 6.5-kbtranscript is virtually identical to that of the 8-kb transcriptexcept that it contains a truncation of the terminal two-Igdomains and includes five unique amino acids at its amino terminus.The four- and six-Ig variants each contain entirely distinctupstream noncoding sequences derived from different exons. Thesix-Ig but not the four-Ig transcript contains a predicted signalsequence in this upstream region. Of note, the extreme carboxylsequence GFSSFV constitutes a potential PDZ binding domain andis 100% conserved across all known sidekicks from all species.
Figure 2. Alignment of mouse sdk-1 and sdk-2. Amino acid sequences of mouse sdk-1 (8-kb transcript) and sdk-2 as predicted from cDNA sequences. Identical residues are shown by dots. The predicted signal sequences are underlined, Ig C2 domains are shown in gray, FN-III repeats are shown in yellow, the transmembrane domain is green, and the conserved C-terminal PDZ binding domain is red. Sequence corresponding to the 6.5-kb transcript of sdk-1 starts at the red arrow, preceded by the unique peptide sequence MDRSG. The 6.5-kb transcript contains a unique upstream noncoding sequence (not shown).
Mouse sdk-2, 86% homologous to chicken sdk-2 at the amino acidlevel, was cloned via a PCR-based approach using embryonic kidneycDNA as a template and primers designed from both mouse genomicsequences and a predicted EST. Mouse sdk-1 and sdk-2 are 57%identical at the amino acid level and are similar in domainorganization (Figure 2). Sdk-2 is expressed as a single transcriptof 8 kb encoding a protein of similar topology to the 8-kb transcriptof sdk-1. By Northern blot analysis, it was difficult to detectthe presence of sdk-2 in both wild-type and HIV-1 transgenicpodocytes, suggesting that unlike for sdk-1, sdk-2 expressionis not upregulated by HIV-1 gene expression (data not shown).The nucleotide sequences of mouse sdk-1 and sdk-2 cDNA havebeen deposited in the GeneBank database under Genbank AccessionNumbers AY 353236 and AY 351699.
Sidekick Expression and Regulation during Kidney Development
Multitissue Northern blot analysis of total RNA from both adultand fetal (E17) mouse organs revealed nearly identical expressionpatterns for sdk-1 and sdk-2. Both genes are highly expressedin many fetal tissues, including kidney (Figure 3A), but showmarkedly lower expression levels in adult organs (data not shown).To look more closely at the developmental regulation of thesegenes in kidney, we evaluated expression of sdk-1 and sdk-2mRNA at multiple time points during nephrogenesis (Figure 3B).This revealed sdk-1 and sdk-2 gene expression to be high throughoutdevelopment with maximal expression occurring near birth. Althoughsdk-1 and sdk-2 RNA expression in adult tissue seems minimalin Figure 3, prolonged exposures clearly demonstrate their presence.
Figure 3. Expression of sdk-1 and sdk-2 mRNA. (A) Total RNA was isolated from embryonic day 17 (E17) mouse organs and examined by Northern blot analysis using full-length sdk-1 and the N-terminal 3.2 kb of sdk-2 as probes. Both sdk-1 and sdk-2 similarly demonstrate widespread expression during development. GAPDH hybridization was performed to normalize RNA loading. (B) Analysis of the temporal regulation of sdk-1 and sdk-2 was performed using kidneys harvested from embryos on E12 to E18, postnatal day 1 (D1), D10, and adult (4 mo). Developmental Northern blot analysis was performed on whole-kidney RNA. Both sdk-1 and sdk-2 have similar patterns of temporal regulation with high expression during nephrogenesis and low expression in adult kidney. RNA loading is again normalized to GAPDH.
Sidekicks Are Expressed in the Ureteric Bud during Nephrogenesis
To localize sidekick protein, we generated polyclonal antibodiesto a glutathione S-transferase fusion protein containing thecarboxyl terminal 83 amino acids of sdk-1. To characterize theantiserum, we performed Western blot analysis on HEK293T cellsthat were transfected with full-length sdk-1 or partial-lengthsdk-2 expression vectors (Figure 4). We found single predominantbands corresponding to appropriate predicted molecular weightsin transfected cells for both of these expression vectors. Noreactivity was seen in control-transfected lysates or when usingpreimmune serum in both transfected and control-transfectedcells. These results show that our antiserum is specific forsidekick proteins but cannot differentiate between sdk-1 andsdk-2.
Figure 4. Characterizing anti-sidekick serum. Reactivity of anti-sidekick serum to sdk-1 (A) and sdk-2 (B) was demonstrated by Western blot analysis performed on lysate (L) and membrane (M) fractions of HEK-293T cells that were transfected with a full-length sdk-1 expression vector or a partial-length sdk-2 expression vector. Control lysate and membrane fractions were made from HEK-293T cells transfected with a control vector. Sdk-1 migrated at 220 kD. Partial-length sdk-2 migrated at its predicted molecular weight of 35 kD. No reactivity was detected using preimmune serum.
We used the antiserum to perform immunohistochemistry on paraffin-embeddedsections of mouse kidneys at various stages of nephrogenesis(Figure 5, A, B, D, and E). We found sidekick protein to beexpressed abundantly in the ureteric bud and collecting ducts,which are the principle derivatives of the ureteric bud. AtE12.5, sidekicks were expressed specifically throughout theentire ureteric bud, including both its trunk and its distaltips (Figure 5A). The proteins were not detected in inducedmesenchymal tissue. As predicted from its amino acid sequence,the protein localizes to cell membranes (Figure 5B). By E15.5,however, expression in the tips of the ureteric bud is diminished,whereas prominent expression remains in the medullary uretericbud and collecting ducts. Of note, at E15.5, sidekicks werenot detected in developing glomeruli, although we cannot ruleout some expression in developing tubules. No staining was detectedusing the preimmune serum from the animal in which the anti-sidekickantibodies were raised (Figure 5, C and F).
Figure 5. Sidekicks are expressed in ureteric bud during nephrogenesis. (A and B) Paraffin sections of E12.5 mouse kidney were stained for sidekick. Immunoreactivity is specific for the ureteric bud, including both stalks and tips. (C) Staining with preimmune serum is negative. U, ureteric bud stalk; arrows, ureteric bud tips; NZ, nephrogenic zone. (D and E) Paraffin sections of E15.5 mouse kidney were stained for sidekick. Immunoreactivity remains in the ureteric bud but with lower expression in ureteric bud tips relative to medullary branches. (F) No immunoreactivity is present when using preimmune serum. Abbreviations are identical to above.
Homology between the human and mouse sdk-1 proteins is predictedto be 86% at the amino acid level. As a result, our antiserumis also capable of recognizing human antigens in immunohistochemistry.In human fetal tissue at 24 wk of gestation (Figure 6), expressionparallels what is seen at earlier stages in the mouse fetus;namely, sidekicks are expressed in the developing collectingducts with significantly higher levels in the renal medullarelative to cortex. Expression was greatest in ureteric bud–derivedtissue, whereas sidekicks were undetectable in glomeruli. Furthermore,sidekick expression seemed to be present equally on all ductalmembrane segments (apical, basolateral, and cell to cell junctions).Again, all preimmune controls were negative (data not shown).
Figure 6. Sidekicks are expressed in developing medullary collecting ducts. (A through C) Paraffin sections of human fetal kidney (24 wk) were stained for sidekick. In A, sidekick immunoreactivity seems much higher in the medulla relative to the cortex. Sidekick expression seems specific for medullary collecting ducts (B). All glomeruli and most mesenchymally derived structures fail to exhibit any sidekick reactivity (C).
Sidekicks Are Upregulated in Glomeruli in HIVAN
Initial attempts to localize sidekick proteins in paraffin-embeddedadult tissue sections were unsuccessful despite continued, albeitlower, expression of RNA in adult kidneys. Therefore, immunohistochemistrywas performed using frozen sections with a tyramide amplificationprotocol to allow for increased sensitivity. Under these conditions,sidekicks in wild-type adult kidney are detected in a patternconsistent with glomerular mesangial cells but few other cells(Figure 7A). Clearly, this represents a striking change fromthe expression pattern seen during kidney development. To determinewhether glomerular expression of sidekicks can also be detectedduring nephrogenesis using this more sensitive staining technique,frozen embryonic (E15.5) kidney sections were also stained usingtyramide amplification. Using this sensitive technique, we wereunable to detect staining in developing glomeruli (data notshown). Thus, expression in glomeruli seems to be a late eventin renal development and glomerular maturation. In diseasedHIV-1 transgenic mouse kidneys, sidekick proteins seem to beincreased overall, and they accumulate in sclerotic and collapsingglomeruli (Figure 7B). Thus, glomerular expression is increasedin response to HIV-1 but is normally not present or is presentonly at very low levels in development. Control immunostainingwith preimmune serum using an identical protocol was negativein glomeruli (Figure 7E).
Figure 7. Sidekicks are upregulated in adult glomeruli in HIV-associated nephropathy (HIVAN). (A and B) Frozen sections of adult wild-type and HIV-1–transgenic kidneys were harvested and immunostained for sidekick proteins. In adult wild-type kidney, sidekick is expressed in a pattern consistent with mesangial expression. In HIV-1–transgenic glomeruli, sidekick protein expression is increased throughout sclerotic or collapsed glomeruli. (C and D) Representative glomeruli (shown at x100) from wild-type– and HIV-1–transgenic mice were immunostained for sidekick proteins. Identical staining procedures were performed in parallel and visualized by confocal microscopy using identical parameters. Sidekick staining in HIV-1–transgenic glomeruli is clearly increased when compared with nontransgenic controls, and sidekick expression in HIV-1–transgenic glomeruli is more diffuse and no longer limited to the mesangium. (E) An identical immuostaining protocol using preimmune serum is negative. Gl, glomeruli. (F) Semiquantitative RT-PCR on kidney RNA extracted from age-matched adult wild-type– and HIV-1–transgenic mice using primers specific for sdk-1 and sdk-2. RT input was normalized using primers specific for GAPDH. Whereas both sdk-1 and sdk-2 are detectable by PCR in adult kidney, sdk-1 but not sdk-2 seems to be upregulated by the presence of HIV-1.
To compare staining of sidekicks in HIV-1 transgenic and nontransgenickidneys, we performed identical staining procedures in paralleland the sections were visualized by confocal microscopy performedwith identical parameters (Figure 7, C and D). In diseased glomeruli,sidekicks were significantly overexpressed. Expression was seenin a diffuse glomerular pattern, which was no longer restrictedto mesangial cells but seemed to include podocytes and parietalepithelial cells as well.
As previously stated, our antiserum does not differentiate betweensdk-1 and sdk-2. Therefore, to determine whether the accumulationof sidekick proteins in HIV-1 transgenic kidney represents upregulationof sdk-1, sdk-2, or both, we performed semiquantitative RT-PCRto compare expression levels in HIV-1 transgenic and wild-typekidneys (Figure 7F). RT input was closely normalized to glyceraldehyde-3phosphate dehydrogenase before proceeding to PCR with sdk-1–and sdk-2–specific primers. Whereas both sdk-1 and sdk-2are detectable by PCR in adult kidney, only sdk-1 was significantlyupregulated (at least fivefold) by HIV-1 expression. Sdk-2 expressionwas unchanged. These results are similar to what we observedin HIV-1 transgenic podocytes and suggest that the accumulationof sidekick protein induced by HIV-1 expression is largely thatof sdk-1 rather than sdk-2.
To confirm that sidekick upregulation is truly relevant in humandisease, we performed immunohistochemistry to compare sidekickexpression in HIVAN and normal human control kidneys. We useda kidney biopsy specimen from HIVAN patient MS 111, whose pathologyincluded CG, podocyte hypertrophy, and tubular microcystic disease.Immunohistochemistry was performed in parallel with identicalparameters using tyramide amplification. Sidekicks were increasedin expression in many glomeruli in the HIVAN biopsy but wereabsent from glomeruli of normal kidney tissue (Figure 8). Thepattern of glomerular sidekick expression was consistent withwhat was seen in the mouse model. Some staining in the renalinterstitium was present in HIVAN sections when staining witheither anti-sidekick or preimmune serums. This interstitialstaining was also observed in normal kidney but was much lessprominent. Preimmune serum on serial sections lacked any glomerularstaining.
Figure 8. Sidekicks are overexpressed in human HIVAN. Immunostaining for sidekick was performed on frozen kidney sections from a patient with HIVAN (MS 111) and a normal human control subject. Sidekicks are detected only in diseased glomeruli.
In most chronic renal diseases, the inability of podocytes toundergo regenerative cell division is a key factor that leadsto nephron loss. In HIVAN, however, podocytes become "dysregulated"such that they become dedifferentiated and are able to reenterthe cell cycle (11–14). In the current work, we analyzedalterations in gene expression in HIV-1 transgenic podocytesin an effort to explore how HIV-1 induces these changes in hostcells. We demonstrated that HIV-1 infection induces overexpressionof sdk-1 and proposed sdk-1 as a candidate gene for contributingto the phenotypic changes seen in HIVAN. In contrast, sdk-2is not significantly overexpressed either in HIV-1 transgenicpodocytes or in HIV-1 transgenic mouse kidneys, suggesting thatthe regulation of sdk-1 and sdk-2 is different.
Sidekicks have been shown to play a role in directing synapseformation during neuronal development (20). Sdk-1 and sdk-2localize specifically to the synaptic cleft in largely nonoverlappingsubsets of retinal neurons. Sdk-1–expressing presynapticcells form synapses with laminae that also express sdk-1; likewise,presynaptic cells that express sdk-2 target sdk-2–positivelaminae. Ectopic expression of sdk-1 in sdk-negative cells redirectedthe processes of these cells to sdk-1–positive sublaminae.Furthermore, in vitro, sidekicks mediate homophilic adhesionsuch that each sidekick is able to bind only to its own kind.These data suggest that sidekick adhesion across the synapseplays an important role in determining synapse specificity.
In the current studies, we demonstrate that sidekick expressionis by no means specific to the neuron but is in fact widespreadduring development. The function of sidekicks in nonneuronalcells is unknown, although, on the basis of the available functionaldata for these proteins, they likely mediate homophilic intercellularadhesion (20). In the fetal kidney at E12.5, we show that sidekicksare specifically expressed throughout the ureteric bud, includingthe stalk and tips, but by E15.5, expression is reduced in thetips. Furthermore, by postnatal day 1, expression is specificfor medullary collecting ducts. This expression pattern correlateswith several other genes known to have important roles in branchingmorphogenesis, including sonic hedgehog (22,23) and collagenXVIII (24). In general, the absence of high levels of expressionin the nephrogenic zone is most consistent with sidekicks’having a regulatory function in the complex cell division andmigration required for branching tubulogenesis. Such a rolein controlling cellular migration would be consistent with whatis known about sidekick function in developing neurons. We arecurrently generating the reagents necessary to distinguish betweensdk-1 and sdk-2 proteins so that their exact roles in kidneydevelopment can be better defined.
In normal adult kidney, sidekicks are expressed in a patternconsistent with glomerular mesangial cells, albeit at much lowerlevels than in the embryonic collecting system. The functionallink between ureteric bud and adult mesangial cell expressionis currently unclear. Consistent with in vitro data, wild-typepodocytes do not seem to express sidekick proteins or do soat low levels. In HIVAN, however, sdk-1 expression not onlyis significantly upregulated in mesangial cells but also seemsto be expressed in other cell types within the glomerulus, includingparietal epithelial cells and podocytes. It is unclear whetherthe upregulation and/or ectopic expression of sdk-1 proteinrepresents a pathologic host response to HIV-1 infection oran attempt by podocytes to maintain the adhesion and structuralintegrity necessary to maintain the filtration barrier. Othergroups have shown that podocyte injury is capable of inducingexpression of genes that are not typically expressed in thepodocyte, including desmin (11) and certain macrophage-specificmarkers (25). Ongoing experiments are focused on determiningthe exact role that sidekick dysregulation plays in the pathogenesisof HIVAN.
Acknowledgments
This work was funded by National Institutes of Health Grant1 PO1 DK56492. K.H. is supported by a fund from the Japan HealthScience Fund. Microscopy was performed at the MSSM-MicroscopyShared Resource Facility, supported, in part, with funding fromNational Institutes of Health–National Cancer Instituteshared resources grant 1 R24 CA095823-01.
We thank Dr. Leslie Bruggeman and Dr. Patricia Wilson for theirhelpful discussions and scientific input.
D’Agati V, Appel GB: HIV infection and the kidney. J Am Soc Nephrol 8: 138–152, 1997[Abstract]
Winston JA, Klotman PE: Are we missing an epidemic of HIV-associated nephropathy? J Am Soc Nephrol 7: 1–7, 1996[Abstract]
Cohen AH, Nast CC: HIV-associated nephropathy. A unique combined glomerular, tubular, and interstitial lesion. Mod Pathol 1: 87–97, 1988[Medline]
D’Agati V, Suh JI, Carbone L, Cheng JT, Appel G: Pathology of HIV-associated nephropathy: A detailed morphologic and comparative study. Kidney Int 35: 1358–1370, 1989[Medline]
Monahan M, Tanji N, Klotman PE: HIV-associated nephropathy: An urban epidemic. Semin Nephrol 21: 394–402, 2001[CrossRef][Medline]
Abbott KC, Hypolite I, Welch PG, Agodoa LY: Human immunodeficiency virus/acquired immunodeficiency syndrome-associated nephropathy at end-stage renal disease in the United States: Patient characteristics and survival in the pre highly active antiretroviral therapy era. J Nephrol 14: 377–383, 2001[Medline]
Gharavi AG, Ahmad T, Wong RD, Hooshyar R, Vaughn J, Oller S, Frankel RZ, Bruggeman LA, D’Agati VD, Klotman PE, Lifton RP: Mapping a locus for susceptibility to HIV-1-associated nephropathy to mouse chromosome 3. Proc Natl Acad Sci U S A 101: 2488–2493, 2004[Abstract/Free Full Text]
Gharavi A, Oller S, Bruggeman L, Lifton R, Klotman P: Genetic background modifies the development of renal disease in HIV-1 transgenic mice. J Am Soc Nephrol 10: 404A, 1999
Bruggeman LA, Ross MD, Tanji N, Cara A, Dikman S, Gordon RE, Burns GC, D’Agati VD, Winston JA, Klotman ME, Klotman PE: Renal epithelium is a previously unrecognized site of HIV-1 infection. J Am Soc Nephrol 11: 2079–2087, 2000[Abstract/Free Full Text]
Marras D, Bruggeman LA, Gao F, Tanji N, Mansukhani MM, Cara A, Ross MD, Gusella GL, Benson G, D’Agati VD, Hahn BH, Klotman ME, Klotman PE: Replication and compartmentalization of HIV-1 in kidney epithelium of patients with HIV-associated nephropathy. Nat Med 8: 522–526, 2002[CrossRef][Medline]
Barisoni L, Bruggeman LA, Mundel P, D’Agati VD, Klotman PE: HIV-1 induces renal epithelial dedifferentiation in a transgenic model of HIV-associated nephropathy. Kidney Int 58: 173–181, 2000[CrossRef][Medline]
Barisoni L, Kriz W, Mundel P, D’Agati V: The dysregulated podocyte phenotype: A novel concept in the pathogenesis of collapsing idiopathic focal segmental glomerulosclerosis and HIV-associated nephropathy. J Am Soc Nephrol 10: 51–61, 1999[Abstract/Free Full Text]
Schwartz EJ, Cara A, Snoeck H, Ross MD, Sunamoto M, Reiser J, Mundel P, Klotman PE: Human immunodeficiency virus-1 induces loss of contact inhibition in podocytes. J Am Soc Nephrol 12: 1677–1684, 2001[Abstract/Free Full Text]
Shankland SJ, Eitner F, Hudkins KL, Goodpaster T, D’Agati V, Alpers CE: Differential expression of cyclin-dependent kinase inhibitors in human glomerular disease: Role in podocyte proliferation and maturation. Kidney Int 58: 674–683, 2000[CrossRef][Medline]
Bruggeman LA, Dikman S, Meng C, Quaggin SE, Coffman TM, Klotman PE: Nephropathy in human immunodeficiency virus-1 transgenic mice is due to renal transgene expression. J Clin Invest 100: 84–92, 1997[Medline]
Husain M, Gusella GL, Klotman ME, Gelman IH, Ross MD, Schwartz EJ, Cara A, Klotman PE: HIV-1 Nef induces proliferation and anchorage-independent growth in podocytes. J Am Soc Nephrol 13: 1806–1815, 2002[Abstract/Free Full Text]
Simard MC, Chrobak P, Kay DG, Hanna Z, Jothy S, Jolicoeur P: Expression of simian immunodeficiency virus nef in immune cells of transgenic mice leads to a severe AIDS-like disease. J Virol 76: 3981–3995, 2002[Abstract/Free Full Text]
Ross MD, Bruggeman LA, Hanss B, Sunamoto M, Marras D, Klotman ME, Klotman PE: Podocan, a novel small leucine-rich repeat protein expressed in the sclerotic glomerular lesion of experimental HIV-associated nephropathy. J Biol Chem 278: 33248–33255, 2003[Abstract/Free Full Text]
Nguyen DN, Liu Y, Litsky ML, Reinke R: The sidekick gene, a member of the immunoglobulin superfamily, is required for pattern formation in the Drosophila eye. Development 124: 3303–3312, 1997[Abstract]
Yamagata M, Weiner JA, Sanes JR: Sidekicks: Synaptic adhesion molecules that promote lamina-specific connectivity in the retina. Cell 110: 649–660, 2002[CrossRef][Medline]
Hubank M, Schatz DG: Identifying differences in mRNA expression by representational difference analysis of cDNA. Nucleic Acids Res 22: 5640–5648, 1994[Abstract/Free Full Text]
Yu J, Carroll TJ, McMahon AP: Sonic hedgehog regulates proliferation and differentiation of mesenchymal cells in the mouse metanephric kidney. Development 129: 5301–5312, 2002
Bitgood MJ, McMahon AP: Hedgehog and Bmp genes are coexpressed at many diverse sites of cell-cell interaction in the mouse embryo. Dev Biol 172: 126–138, 1995[CrossRef][Medline]
Lin Y, Zhang S, Rehn M, Itaranta P, Tuukkanen J, Heljasvaara R, Peltoketo H, Pihlajaniemi T, Vainio S: Induced repatterning of type XVIII collagen expression in ureter bud from kidney to lung type: Association with sonic hedgehog and ectopic surfactant protein C. Development 128: 1573–1585, 2001[Abstract]
Bariety J, Nochy D, Mandet C, Jacquot C, Glotz D, Meyrier A: Podocytes undergo phenotypic changes and express macrophagic-associated markers in idiopathic collapsing glomerulopathy. Kidney Int 53: 918–925, 1998[CrossRef][Medline]
Received for publication September 3, 2003.
Accepted for publication March 24, 2004.
This article has been cited by other articles:
L. Kaufman, G. Yang, K. Hayashi, J. R. Ashby, L. Huang, M. J. Ross, M. E. Klotman, and P. E. Klotman The homophilic adhesion molecule sidekick-1 contributes to augmented podocyte aggregation in HIV-associated nephropathy
FASEB J,
May 1, 2007;
21(7):
1367 - 1375.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
1.0-kb difference product identified by RDA was predicted to be dramatically upregulated in the HIV-1 transgenic podocytes. This difference product was homologous to the Drosophila melanogaster sidekick gene. Using this 1.0-kb difference product as a probe, we screened a 
-ZapII (Stratagene) cDNA phage library generated from size-selected polyA+ RNA from the HIV-1 transgenic podocytes. In this way, we cloned the carboxyl terminal of sdk-1. The 5' portion of the 6.5-kb transcript was then cloned using the 5' rapid amplification of cDNA ends (5' RACE; Ambion) on total RNA extracted from embryonic day 17 (E17) mouse kidneys as per the manufacturer’s protocol. Reverse transcription was performed using the sequence specific primer TTCGATTCCACACGGTGAGCATGG. Amplification steps were done using the nested primers CGTGTTGGCTGCATAGCAGGTGTA followed by TTGCCAGCATCCTGGATGAAGACGG using pfu turbo DNA polymerase (Stratagene). The 5' portion of the longer 8-kb transcript was also cloned by 5' RACE using HIV-1 transgenic podocyte RNA as the template. The same primers were used as for cloning the 6.5-kb transcript. Amplification was done using platinum pfx DNA polymerase (Invitrogen). 
