Protein Kinase X Activates Ureteric Bud Branching Morphogenesis in Developing Mouse Metanephric Kidney
Xiaohong Li,
Deborah P. Hyink,
Katalin Polgar,
G. Luca Gusella,
Patricia D. Wilson and
Christopher R. Burrow
Department of Medicine, Division of Nephrology, Mount Sinai School of Medicine, New York, New York
Address correspondence to: Dr. Patricia D. Wilson, Department of Medicine, Division of Nephrology, Mount Sinai School of Medicine, 1425 Madison Avenue, New York, NY 10029. Phone: 212-659-9376; Fax: 212-849-2434; pat.wilson{at}mssm.edu
Received for publication March 2, 2005.
Accepted for publication September 4, 2005.
The human protein kinase X (PRKX) gene was identified previouslyas a cAMP-dependent serine/threonine kinase that is aberrantlyexpressed in autosomal dominant polycystic disease kidneys andnormally expressed in fetal kidneys. The PRKX kinase belongsto a serine/threonine kinase family that is phylogeneticallyand functionally distinct from classical protein kinase A kinases.Expression of PRKX activates cAMP-dependent renal epithelialcell migration and tubular morphogenesis in cell culture, suggestingthat it might regulate branching growth of the collecting ductsystem in the fetal kidney. With the use of a mouse embryonickidney organ culture system that recapitulates early kidneydevelopment in vitro, it is demonstrated that lentiviral vector-drivenexpression of a constitutively active, cAMP-independent PRKXin the ureteric bud epithelium stimulates branching morphogenesisand results in a 2.5-fold increase in glomerular number. Theseresults suggest that PRKX stimulates epithelial branching morphogenesisby activating cell migration and support a role for this kinasein the regulation of nephrogenesis and of collecting systemdevelopment in the fetal kidney.
Mammalian kidney development requires a complex series of inductiveand morphogenetic events to produce normal branching growthof the ureteric bud to give rise to the collecting duct systemand induction of mesenchymal to epithelial cell conversion andtubulogenesis to give rise to nephrons. Developmental disruptionof these events cause congenital renal malformations and canlead to adult kidney disease. Targeted gene disruption and tissueculture experiments have provided information about the molecularsignaling mechanisms underlying the reciprocal inductive tissueinteractions that are essential for kidney development and suggestthat epithelial tubular morphogenesis depends largely on thedirectional outgrowth of cells that are controlled by glial-derivedneurotrophic factor (1); bone morphogenetic proteins (BMP),including BMP2 and BMP7 (2); and other signaling molecules thatactivate receptor tyrosine kinase receptors, including fibroblastgrowth factor, hepatocyte growth factor (3), and epidermal growthfactor (4). With the use of both cell culture model systemsand metanephric organ culture, BMP regulation of epithelialbranching morphogenesis has been shown to depend, at least inpart, on functionally defined cAMP-dependent protein kinases(2,5).
As well as soluble growth factors, insoluble matrix factor interactionsplay important roles in metanephric morphogenesis. Recent studieshave demonstrated that the PKD1-encoded protein polycystin-1plays an essential role in the regulation of tubulogenesis inkidney development (6), even though symptomatic disease mostcommonly presents during adult life. PKD1 gene mutations areresponsible for 85% of cases of autosomal dominant polycystickidney disease (ADPKD) and result in progressive cystic dilationof all nephron segments, which often leads to ESRD. Polycystin-1forms multiprotein complexes with proteins of the focal adhesionplaque at the basal cell–matrix interface in the fetalkidney, with the cell–cell adherens complex at cell–celljunctions in adult kidney, and with protein complexes that arefound in the primary cilium at the apical–lumen interface(7–10). Recent studies suggest that polycystin complexesmay act as mechanosensors receiving signals from the extracellularmatrix (via focal adhesions), adjacent cells (via cell junctions),and the tubule lumen (via cilia) followed by transduction intointracellular responses. Polycystin-1 signaling is thought toexert regulatory control of gene transcription via its translocatedC-terminal tail and likely modulates cellular adhesion and migrationin a way that is essential to the control epithelial cell shape,tubular diameter, and volume during kidney development (11–14).
In studies that were designed to elucidate protein kinase genesthat are aberrantly activated in ADPKD and associated regulationof tubular morphogenesis in the developing kidney, the PRKXgene (15,16) was identified. The PRKX kinase belongs to a subfamilyof serine/threonine kinases that are implicated in the regulationof cellular migration and morphogenesis (17) and other developmentalprocesses, including control of granulocyte/macrophage differentiation(18,19). The PRKX gene encodes a cAMP-dependent kinase, is expressedin fetal and ADPKD kidneys, but is transcriptionally silentin the adult kidney (15,17). In situ hybridization analysesof normal fetal kidneys showed that PRKX mRNA expression wasrestricted to the fetal ureteric bud epithelium (17). Expressionof the PRKX kinase markedly activates cAMP-dependent migrationof cultured renal epithelial cells and induces branching morphogenesisof MDCK cells in collagen gels even in the absence of hepatocytegrowth factor or other stimulatory factors, an effect not producedby expression of the protein kinase A (PKA) kinase (17). Takentogether with the ureteric bud localization of PRKX in fetalkidneys, these observations suggested a role for PRKX in theregulation of epithelial morphogenesis during mammalian kidneydevelopment. These studies also emphasized the importance ofdistinguishing PRKX from PKA kinases when seeking to identifycAMP-dependent kinase genes that modulate BMP or Hedgehog-dependentresponses on branching morphogenesis of the ureteric bud.
In studies reported here, we show that PRKX mRNA is broadlyexpressed in mesoderm-derived tissues in the fetus and thatexpression of PRKX stimulates the migration of human fetal collectingtubule (HFCT) epithelia in culture. Using an embryonic kidneyorgan culture system that recapitulates early kidney developmentin vitro and a newly developed technique for viral vector genetransduction by microinjection into ureteric bud epithelialcells, we demonstrate that PRKX kinase expression stimulatestwo distinct aspects of renal development: Ureteric bud branchingand induction of glomeruli. These results suggest that PRKXmight play an important role of these processes during normalkidney development and that persistent activation of this kinasemay have important effects on the abnormal cystic tubular phenotypefound in ADPKD.
Construction of PRKX Expression Vectors
The generation of the recombinant peGFP/PRKX, pFLAG/PRKX, andthe kinase dead pFLAG/PRKX/K78R fusion protein constructs hasbeen described previously (17). The kinase dead GFP fusion proteinencoding construct peGFP/PRKX/K78R was derived by point mutagenesisof peGFP/PRKX. The constitutively active PRKX kinase constructpeGFP/PRKX.ca was engineered by point mutagenesis (His 93 changedto Gln, and Trp 202 changed to Arg) as described for PKA kinase(20). PRKX viral–expressing constructs were generatedusing the VVC self-inactivating lentiviral vector. The VVC lentivectorwas derived from the substitution of the U3 region of the 5'LTR of the pHR/CMV vector (21) with the human early cytomegaloviruspromoter. PRKX transgenes were expressed by cloning into thepolylinker downstream of an internal constitutively active CMVpromoter. The functionalities and titers of the PRKX viral constructswere tested on cultured HeLa-Tat cells. Infectious viral supernatantswith a titer of >1 x 107 transducing units (TU)/ml were producedby transient transfection of 293T cells using Lipofectamine2000 (Invitrogen, San Diego, CA) with 8 µg of VVC-peGFP/PRKX,VVC-pFLAG/PRKX/K78R, or VVC-peGFP/PRKX.ca co-transfected with5.0 µg of the pCMVR8.2 plasmid producing the viral proteins,and 2.5 µg of the pMD.G plasmid expressing the pseudotypingVSV glycoprotein (VSVG) (22). After 24 h, the medium was replacedwith complete medium supplemented with 4 mM sodium butyrate,and supernatants were collected 48 h later. The titer (TU/ml)was determined 48 h after infection of 3 x 104 HeLa-Tat cells/wellin 24-well plates by scoring the number of eGFP expressing cellsusing ultraviolet microscopy.
Northern Blot Analysis
Total RNA was fractionated on agarose/formaldehyde gels, andthe integrity of RNA was monitored with ethidium bromide. RNAthen was transferred to GeneScreen membranes using 25 mM sodiumphosphate (pH 6.5) and mRNA sized by comparison with 28S and18S rRNA. The filters were prehybridized overnight at 42°Cin 50% formamide, 0.04% polyvinlypyrrolidone (PVP), 0.04% BSA,0.04% Ficoll, 1% SDS, 0.75 M NaCl, 0.075 M sodium citrate, anddenatured salmon sperm DNA (100 mg/ml). Hybridization with PRKXand 14S RNA 32P-labeled cDNA probes was carried out in 0.02%PVP, 0.02% BSA, and 0.02% Ficoll at 42°C for 24 h followedby 4x stringent washes of SSC + 0.1% SDS at 65°C. The RNAbands were visualized by autoradiography.
In Situ Hybridization
Deparaffinized, dehydrated human kidney sections were treatedwith proteinase K, prehybridized with triethanolamine/aceticanhydride, and hybridized overnight with digoxigenin-substitutedPRKX antisense or sense 220-bp PRKX PCR fragments as probes.After washing in 0.1x SSC (0.15 M sodium chloride/0.015 M sodiumcitrate [pH 7.0]), sections were incubated in antidigoxigeninantibody/alkaline phosphatase and color-developed with NitroBluetetrazolium reagent (Roche Molecular Biochemicals, Indianapolis,IN).
Cell Culture
Human renal epithelial cells that were derived from normal humanfetal collecting tubule epithelia were grown according to standardprotocols devised in our laboratory (23,24). Briefly, microdissectedhuman fetal (19 wk) collecting tubules (HFCT) were conditionallyimmortalized using a temperature-sensitive T antigen expressingretroviral vector (25) and clonal cells lines derived by limitingdilution. For studies described here, the HFCT clone 7F wasgrown to 70% confluence at 33°C then transferred to thenonpermissive temperature of 37°C for 7 to 10 d before usein adhesion and migration assays.
Migration Assay
HFCT cells were transfected with peGFP/PRKX.ca or with peGFP/PRKX/K78Rwith Lipofectamine 2000 (Invitrogen), washed, and cultured for24 h before FACS sorting to isolate GFP-positive cells. GFP-positivecells from each transfection or control nontransfected cellswere washed with serum-free medium and then labeled with thefluorescence probe calcein AM (Molecular Probes, Eugene, OR).Two thousand labeled cells from each group then were suspendedin 0.3 ml of DMEM plus 1% FBS and plated onto transwells with8-µM pores. The lower well contained 0.8 ml of DMEM plus1% FBS. After 4 h, the transwells were transferred to new wellsthat contained 15% FBS to establish a serum gradient. The numberof labeled cells that migrated through the pore to the lowerchamber was measured by the appearance of fluorescence at 1,2, 4, 6, 8, 16, 20, and 24 h after treatment using an HTS 7000microfluorimeter (Perkin Elmer Cetus, Wellesley, MA).
Mouse Embryonic Kidney Organ Culture and Viral Vector Microinjection
Embryonic day (E) 11 to 12 kidneys were retrieved from timedpregnant CD-1 mice (Charles Rivers Laboratories, Wilmington,MA), and paired kidneys were harvested by microdissection usinga dissecting stereomicroscope (Olympus, Japan). Kidneys wereplaced on uncoated 24-mm Transwell Clear membrane inserts, 0.4mm pore size, in a six-well cluster (Corning Inc., Corning,NY), and 1 ml of supplemented serum-free organ culture mediumthat contained 50% DMEM, 50% Ham F12, 15 mM HEPES and l-glutamine(Cellgro; Mediatech, Herndon, VA), 4.5 g/L glucose, 45 mM sodiumbicarbonate, 1x Insulin/Transferrin/Selenium (Sigma, St. Louis,MO), 2 x 10–9 M T3, and 7 x 10–8 M prostaglandinE1 was added to the lower chamber of each well. Plates wereplaced into a tissue culture incubator (5% CO2) for 3 to 5 dwith daily media changes. For the microinjection of the VVCvector into the ureteric bud lumen, sterile transfer tips witha 15-µm internal diameter (ES; Eppendorf, Hamburg, Germany)were used. On day 0, an Eppendorf Micromanipulator 5171 andtransjector 5246 were used to inject 20 to 200 nl of the VSVGpseudotyped packaged VVC PRKX lentiviral stocks (titers rangedfrom 1 to 4 x 107 TU/ml), VVC vector alone, or PBS control (sham)into the ureteric bud of day 11 to 12 embryonic mouse kidneys,and comparisons were made with untreated kidneys. Kidneys werecultured for 3 to 5 d after microinjection and GFP fluorescenceobserved daily by ultraviolet microscopy to confirm uretericbud expression. Cultured kidneys were also fixed in 95% coldmethanol for 15 min at 4°C followed by a 2-h incubationat 37°C using a mouse monoclonal anti-calbindin antibody(1:500 dilution, Sigma clone D28K), a rabbit anti–WT-1antibody (1:100 dilution, C19 antibody; Santa Cruz Biotechnology,Santa Cruz, CA), or a rabbit anti-GFP antibody (1:250; MolecularProbes) in PBS plus 5% FBS. After two rinses in PBS-Tween andone in PBS for 10 min, the secondary antibodies donkey anti-mouse–Alexa488or donkey anti-rabbit–Alexa 568 (Molecular Probes) wereused at 1:250 dilution and incubation was carried out for 2h at 37°C. After three washes in PBS, some kidneys weresubjected to Tyramide amplification (anti-rabbit Tyramide amplificationkit, 1:100; Molecular Probes) according to the manufacturer’sinstructions for 15 min at room temperature, washed three timesin PBS, and then stored in the dark until examination by confocalmicroscopy.
Microscopy and Morphometric Data Analysis
Morphologic changes were documented daily using an Olympus dissectingmicroscope equipped with Scion computer imaging system. Immunofluorescenceanalysis was carried out using a multi-photon laser scanningmicroscope with inverted configuration using immersion lenses(BioRad Radiance 2000, MSSM core facilities). Single imageswere taken of whole-kidney preparations, or stacks of multipleimages were compressed for Image J analysis (National Institutesof Health, Bethesda, MD), an open domain Java image processingsystem that was used to calculate area and pixel value statistics,to measure distances and volumes and to create skeletonizedimages.
Developmental Regulation of PRKX mRNA Expression in Kidney and Other Organs
Northern blot analysis of PRKX gene expression demonstratedthe presence of the 6-kb PRKX mRNA in human fetal muscle, heart,brain, lung, liver, and kidney (Figure 1A), but no PRKX mRNAwas detected in the same organs of human adult (data not shown).The expression of PRKX mRNA in human fetal kidneys was detectedfrom 12 to 24 wk of gestation with an apparent peak in PRKXmRNA abundance at 16.5 wk (Figure 1B). As previously reported,no PRKX gene expression was found in the adult kidney (17).Immunohistochemical analysis showed low levels of expressionin the collecting duct epithelia of 32-wk fetal kidneys andno expression in 4- or 16-yr-old human kidney sections (datanot shown).
Figure 1. Northern blot analysis of protein kinase X (PRKX). (A) Northern blot assays were carried out on mRNA extracted from fetal organs (16 wk). Lane 1, skeletal muscle; lane 2, heart; lane 3, brain; lane 4, lung; lane 5, liver; lane 6, kidney. Arrow, 6-kb PRKX mRNA; arrowhead, 14S rRNA control. (B) Northern blot assay carried out on mRNA that was extracted from fetal metanephric kidneys of different gestational ages and adult kidneys. Lane 1, 12W; lane 2, 14W; lane 3, 16.5W; lane 4, 23W; lane 5, 27-yr-old adult; lane 6, 42-yr-old adult. Arrow, 6-kb PRKX mRNA; arrowhead, 14S rRNA control. (C) Immunohistochemical localization of PRKX in 34-wk human fetal kidney. Inner medullary collecting tubules at the tip of the papilla are stained. (D) Immunohistochemical localization of PRKX in 15 mo-old human kidney. Light staining is seen in some inner medullary collecting ducts. (E) Immunohistochemical localization of PRKX in 4-yr-old human kidney. No staining is seen.
PRKX Is Expressed throughout the Fetal Kidney Ureteric Bud to the Branching Tips In situ hybridization using the digoxigenin-substituted 220-bpPRKX PCR fragment as probe (Figure 2A through D) showed thatPRKX mRNA transcripts were specifically localized throughoutthe ureteric bud epithelial cells of 14- to 17-wk human fetalkidneys from the medullary truncal regions through to the corticalbranching tips, which also expressed the characteristic uretericbud epithelial markers cytokeratin (Figure 2E) and polycystin-1(Figure 2F) (24,25). PRKX mRNA and protein are also highly expressedin primary and differentiated conditionally immortalized humanrenal HFCT epithelial cell lines derived from microdissectedfetal kidneys (Figure 3, A and B).
Figure 2. PRKX mRNA is localized throughout the ureteric bud of human fetal metanephric kidneys. In situ hybridization analysis with digoxigenin-substituted antisense PRKX PCR fragment riboprobes was carried out on human fetal kidney sections and visualized by anti–digoxigenin-alkaline phosphatase immunohistochemical staining. (A) Low-power micrograph of outer cortex of 14-wk human fetal kidney. Arrows denote positively stained ureteric bud structures. (B) Low-power micrograph of inner medulla of 14-wk human fetal kidney. Arrows denote positively stained ureteric bud structures. (C) High-power micrograph of nephrogenic zone of the outer cortex of 17-wk human fetal kidney. Arrows denote positively stained ureteric bud structures. (D) High-power micrograph of nephrogenic zone of the inner medulla of 17-wk human fetal kidney. Arrows denote positively stained ureteric bud structures. (E) Low-power micrograph of immunohistochemical localization of cytokeratin in the cortex of 17-wk human fetal kidney. (F) High-power micrograph of immunohistochemical localization of polycystin-1 in the medulla of 17-wk human fetal kidney.
Figure 3. PRKX stimulates human fetal collecting tubule (HFCT) cell migration. (A) Digoxigenin–anti-digoxigenin in situ hybridization staining of PRKX mRNA in HFCT primary culture. (B) Western immunoblot analysis of PRKX in differentiated conditionally immortalized HFCT cells (1:300). (C) HFCT cells without transfection (control, blue line) or transfected with peGFP/PRKX.ca encoding a constitutively active PRKX kinase (pink line) or transfected with peGFP/PRKX/K78R encoding a kinase-dead PRKX kinase (KD-PRKX; yellow line) and GFP-positive cells were isolated by FACS, labeled with calcein acetoxymethyl ester, and plated (2000 cells/well density) in Fluoroblock cell migration chamber wells. After 4 h of attachment in 1% FBS-containing media, cell migration through the occluding filter into the lower chamber that contained 5% FBS was measured. The fluorescence shown on the ordinate is proportional to the number of migrated cells and was determined after 1, 2, 4, 6, 8, 16, 20, and 24 h of cell plating. The values are mean ± SEM from two independent experiments in triplicate. **P < 0.05; *P < 0.01.
PRKX Stimulates Migration of HFCT Epithelia
In previously published studies (17), we showed that PRKX butnot PKA activated epithelial cell migration only when cellswere treated with 8-bromo-cAMP. To determine whether this stimulationof cell migration depended only on activation of the PRKX kinase,we generated a constitutively active form of PRKX by mutagenesisof amino acid residues required for binding the regulator subunitI regulator subunit (see Materials and Methods). In contrastto the cytoplasmic localization of wild-type PRKX in the absenceof 8-bromo-cAMP cell treatment (17), we found that the constitutivelyactive PRKX mutant showed predominantly nuclear localizationin transfected LLC-PK1 epithelial cells (data not shown), consistentwith a loss of regulator subunit I subunit binding. Furthermore,we found that constitutively active PRKX had significantly increasedkinase activity in cell lysates that were prepared from transfectedcells (312 ± 25.9 pmol of 32P incorporated per min permg of protein) compared with wild-type PRKX (58.6 ± 4.5pmol of 32P incorporated per min per mg of protein) in the absenceof 8-bromo-cAMP. To analyze the cell migration effects of themutant, constitutively active PRKX kinase, we conducted modifiedBoyden chamber assays using HFCT cells, and migration was measuredat different time points for 24 h. Under the basal conditionsof this assay, 2.5% of HFCT cells are migratory. A significantincrease in cell migration was seen in GFP-positive cells thatwere transfected with the constitutively active peGFP/PRKX.caconstruct after 6 h and was maintained for 24 h (Figure 3C).By contrast, no significant increase in migration was seen inuntransfected cells or in GFP-positive cells that were transfectedwith the kinase-dead PRKX mutant construct peGFP/PRKX/K78R.
Expression of PRKX Kinase Activates Branching Morphogenesis and Nephron Formation in Cultured Metanephric Kidneys
Because our previous studies had shown that PRKX mRNA was expressedin the ureteric bud of human fetal kidneys, proviral supernatantswere introduced into the ureteric bud lumen of E11 mouse kidneysby microinjection of VVC-peGFP/PRKX.ca provirus encoding a constitutivelyactive PRKX kinase and cultured for 3 d (Figure 4A). After 3d in culture, GFP fluorescence analysis showed that the peGFP/PRKXfusion protein was expressed throughout the ureteric bud–derivedcollecting system as far as the extreme tips (Figure 4B). ThispeGFP/PRKX expression was detectable from day 2 after injection(data not shown). It is interesting that expression was seennot only in the cytoplasm of all cells but also in the nucleiof some cells at the extreme tips of the transduced uretericbud (Figure 4C). No fluorescence was seen in kidneys that wereinjected with empty VVC vector alone (Figure 4D). Paired kidneysfrom E11 mice were injected with constitutively active PRKXexpressing lentiviral vector VVC-peGFP/PRKX.ca or the emptyviral vector alone (control) and cultured for 3 d. When comparedwith the control kidneys that were injected with the empty viralvector, kidneys that expressed the constitutively active formof PRKX had increased numbers of induced glomeruli as determinedby WT-1 staining (Figure 5, A versus B). Quantitative analysisusing IMAGE J software further showed that the constitutivelyactive PRKX stimulated the induction of glomeruli by >2.5-foldcompared with the control group (18.2 ± 4.32 [PRKX] versus6.8 ± 1.9 [control]; P = 0.042; Figure 5C). By contrast,when kinase dead PRKX was injected into the ureteric bud, nosignificant change in glomerular induction was seen, althougha trend to decrease was noted (Figure 5, D versus E, and F).
Figure 4. Expression of pEGFP/PRKX in the cultured embryonic day 11 (E11) kidney. (A) VVC-peGFP/PRKX proviral stocks were titered on HeLa-TAT cells and delivered by microinjection into the lumen of the ureteric bud of E11 to E12 mouse metanephric kidneys. At these early stages of organogenesis, the ureteric bud lumen can be filled with nanoliter quantities of proviral supernatants, ensuring a high multiplicity of infection favoring high-efficiency transduction of the epithelial cell progenitors of the developing collecting duct system over the ensuing 3 to 5 d of organ culture. (B) A single epifluorescence image of an E11.5 mouse embryonic kidney after microinjection with the VVC-peGFP/PRKX.ca viral vector encoding a constitutively active PRKX kinase and 3 d of organ culture. Expression (GFP fluorescence) can be seen throughout the ureteric bud branching structure (compare with ureteric bud marker, calbindin, stained kidney in Figure 6E). (C) A high-power image of the cells in the region of the VVC-peGFP/PRKX.ca microinjected ureteric bud denoted by the arrow in B shows green fluorescence in the cytoplasm of all cells and also in the nuclei of some cells at the outer surface of the bud tip (arrowheads). (D) E11 mouse embryonic kidney, microinjected with VVC (empty vector) alone and cultured for 3 d in organ culture, showed no detectable fluorescence by confocal microscopy.
Figure 5. PRKX stimulates glomerular induction in E11 cultured fetal kidneys. (A and B) Paired E12 fetal kidneys were injected with the VVC-pEGFP/PRKX.ca viral vector encoding a constitutively active PRKX kinase (A) or the empty viral vector alone (control; B). Kidneys were cultured for 3 d and then labeled by WT-1 for visualization of glomeruli by confocal microscopy. (C) Quantitative analysis of glomerular numbers in WT-1–labeled kidneys showed significant increases after VVC-peGFP/PRKX.ca injection compared with empty vector controls. Confocal image stacks were converted to Z projections. The values are means ± SEM from six paired kidneys. (D and E) Paired E12 fetal kidneys that were injected with kinase dead VVC-pEGFP/PRKX-KD viral vector (D) or empty viral vector alone (E). Kidneys were culture for 3 d and then labeled with WT-1. (F) Quantitative analysis of glomeruli as in C shows no significant differences.
The effect of expression of PRKX on ureteric bud branching morphogenesisthat was produced by viral transduction with VVC-peGFP/PRKX.cawas also analyzed. Compared with the control kidneys that wereinjected with the empty lentiviral vector, constitutively activePRKX overexpression resulted in increased ureteric bud branchingas determined by anti-calbindin staining with confocal imageanalysis of the branched ureteric bud (Figure 6, A versus B).Using IMAGE J software, we quantitatively analyzed numbers ofureteric bud branch tips and branch points, the first and secondureteric bud branch lengths, and ureteric bud branch diameters.Our results showed that PRKX expression significantly increasedthe number of ureteric branch tips and branch points comparedwith the control group (tips 34.2 ± 1.43 [PRKX] versus19.8 ± 2.89 [control], P = 0.002; branch points 31.2± 2.48 [PRKX] versus 17.6 ± 2.38 [control], P= 0.004; Figure 6C). PRKX expression also significantly increasedthe second ureteric branch lengths compared with the controlgroup (130.2 ± 10.7 [PRKX] versus 99.1 ± 10.3µm [control]; P = 0.045). However, PRKX expression didnot produce significant effects on either the first uretericbranch lengths or ureteric bud diameters (first branch length124.8 ± 14.8 [PRKX] versus 146.7 ± 19.4 µm[control], P = 0.38; branch diameter 64.6 ± 2.73 [PRKX]versus 56.1 ± 3.76 µm [control], P = 0.07; Figure6D). By contrast, the microinjection of kinase dead PRKX didnot stimulate and tended to decrease parameters of uretericbud development (Figure 6, E versus F, and G). Similar trendsto reduction in ureteric bud branching were seen in the presenceof PRKX siRNA that reduced mRNA and protein levels by >70%.(data not shown).
Figure 6. PRKX stimulates ureteric bud branching in E11 cultured fetal kidneys. (A and B) Paired E12 fetal kidneys were injected with the VVC-pEGFP/PRKX.ca viral vector (A) or the empty viral vector (B) as a control. Kidneys were cultured for 3 d and then labeled with anti-calbindin for visualization of ureteric bud branches and examined by confocal microscopy. Multiple (10 µm) images were taken through the kidneys and compressed into single images. (A and B, left) Confocal image stacks of labeled kidneys were converted to Z projections. The Z projection from the green channel (anti-calbindin) was skeletonized using IMAGE J software to create a ureteric tree (A and B, right), which was used to determine the diameter and length of ureteric branches as well as to count the numbers of ureteric branch points and tips (C). The values are mean ± SEM from six paired kidneys. *Significant difference between PRKX and control values (P < 0.005 for branch tips, P < 0.005 for branch points, and P < 0.05 for second ureteric branch lengths; D). (E) Paired E12 fetal kidneys that were microinjected with kinase dead VVC-pEGFP/PRKX-KD viral vector (E) or empty viral vector (F) stained with anti-calbindin. (G) Quantitative analysis of ureteric bud tips and points as in C shows no significant differences.
The PRKX kinase is a mammalian cAMP-dependent protein kinasethat seems to have multiple functions in cellular differentiationand epithelial morphogenesis (26,27). Phylogenetic analysisbased on sequence alignments of the core catalytic domains haveshown that the human PRKX gene belongs to a unique and ancientfamily that includes Caenorhabditis elegans, Drosophila melanogaster,Dictyostelium discoideum, Ascaris suum, and Mus musculusPRKXhomologous genes rather than to the classic PKA gene family(17). The biologic function of the PRKX gene family member Dictyosteliumkinase PKA-C has been studied intensively, and this kinase hasbeen shown to play an essential role in the cell-shape changesand cell migration necessary for cell sorting during morphogenesisas well as in the transcriptional regulation of later cell fatedifferentiation during development (28–31). We thereforehypothesized that PRKX may play a role in the regulation ofepithelial cell morphogenesis in the developing mammalian kidney.Studies by other laboratories had shown that PRKX was expressedin macrophages and granulocytes, and inhibition of PRKX expressionblocked terminal differentiation in both the leukemic HL-60cells and normal peripheral blood monocytes (19). Our previousstudies demonstrated that PRKX does activate renal epithelialcell migration and tubulogenesis in vitro (17). Taken together,these results suggested that PRKX might have multiple rolesin mammalian development, including tissue-specific cell differentiationin hematopoietic cells and epithelial cell morphogenesis inthe developing kidney.
PRKX mRNA is expressed in the ureteric bud of fetal kidneysand the cyst epithelia of ADPKD kidneys but not in the normaladult kidney (17). To explore further the distribution of PRKXmRNA in other fetal organs as well as its expression patternin developing kidneys, we carried out high-stringency Northernblot analysis. Our results show that PRKX mRNA was also expressedin fetal heart, brain, lung, and skeletal muscle in additionto kidney, although very little expression was found in adultorgans (and none in adult kidney) (15). Northern blot analysisof PRKX mRNA in developing human fetal kidneys demonstratedexpression from 12 to 24 wk gestational age, a gestational periodwhen ureteric bud branching morphogenesis is highly active.Here, we further demonstrate that PRKX is expressed all of theway through the ureteric bud branching tubular structure fromthe inner stalk to the outer branching tips, suggesting a possiblerole in the direction and maintenance branching morphogenesisof the ureteric bud in the early phases of metanephric kidneydevelopment.
To determine whether PRKX can activate epithelial morphogenesisin the fetal kidney, we introduced a constitutively active formof PRKX into the ureteric bud lumen of E11 metanephric kidneysand analyzed its influence on development by organ culture.This was made possible by using high-titer VSVG pseudotypedlentiviral vectors. Use of this system allowed us to monitorthe early growth and morphogenesis of kidneys continuously (3to 7 d) after microinjection with PRKX lentiviral constructs.Whole-mount confocal microscopy demonstrated that the lentiviralvector–encoded PRKX was expressed throughout the entireureteric bud–derived branched collecting system with cytoplasmiclocalization in all cells and nuclear localization in some cellsat the extreme tips of the bud. Because this constitutivelyactive mutant PRKX kinase is found predominantly in the nucleusof transfected LLC-PK1 renal epithelial cells, the largely cytoplasmiclocalization of this mutant PRKX kinase in most regions of theureteric bud suggests that cAMP-independent mechanisms may regulatePRKX in vivo.
Intriguingly, this study shows that overexpression of constitutivelyactive PRKX in the fetal kidney dramatically stimulates uretericbud branching as well as promotes induction of increased numbersof glomeruli. The numbers of ureteric branch points and tipsand the length of second branch segments all significantly increasedin the kidneys that overexpressed constitutively active PRKX,compared with the paired control group kidneys that were microinjectedwith empty VVC vector alone. These studies identify a potentialrole for PRKX in renal development as a regulatory kinase controllingureteric bud elongation and branching growth as well as nephronnumber. In this regard, our previous studies as well as studiesof HFCT cells in vitro suggest that PRKX may have a moderatelystimulatory effect on cell proliferation. The localization ofPRKX mRNA in the inner deep portions of the ureteric bud branchingstalk as well as throughout the structure to the outer branchingtips shown by in situ hybridization (17) (Figure 2) suggestthat PRKX may play an important role in directing branchingmorphogenesis at an early time in mammalian metanephric kidneydevelopment, when there is rapid growth of the collecting ductsystem.
To investigate whether the PRKX kinase affects the migrationof tubular kidney epithelial cells, processes that are essentialfor the control of renal tubule branching morphogenesis, weconducted assays on HFCT epithelial cells that were transfectedwith PRKX. Our results showed that PRKX significantly increasesHFCT cell migration in a modified Boyden chamber assay. Theseobservations suggest that the activation of ureteric bud branchingmorphogenesis in the metanephric kidney by PRKX may be due atleast in part to the activation of epithelial cell migrationin the ureteric bud and its collecting duct derivatives.
The downstream phosphorylation targets of the PRKX kinase thatmediate its effects on epithelial cell migration and branchingmorphogenesis have yet to be identified, although PRKX activityto phosphorylate the C-terminus of polycystin-1 (X.L., C.R.B.,P.D.W., unpublished observations, 2004) suggests the C-terminusof polycystin-1 as one potential target. Focal adhesion complexproteins, including FAK, paxillin, c-src, p130cas, vinculin,talin, and -actinin, play an important role in cell adhesionand migration, and phosphorylation of components of this complexis a key regulatory mechanism for these processes (32–34).It is interesting that polycystin-1 has also been shown to formmultiprotein complexes with proteins of the focal adhesion plaque,as well as with the cell–cell adherens protein complexes,and can also be found in apical collecting duct cilia (7–10).Recent studies have also suggested that polycystin-1 is likelyinvolved in the regulation of epithelial tubulogenesis becauseexpression of polycystin-1 can trigger branching morphogenesisof tubular kidney epithelial cells in vitro (35), and overexpressionof a 202–amino acid residue C-terminal polycystin-1 polypeptidedisrupts branching morphogenesis in cultured mouse metanephrickidneys in organ culture (14). Identification of the downstreamtargets of the PRKX kinase that mediate its effects on epithelialmorphogenesis therefore may be important for elucidation ofsignaling pathways that normally regulate kidney organogenesisand are aberrantly activated in polycystic kidney diseases.
Acknowledgments
We gratefully acknowledge Barbara Bloswick for expert technicalassistance and Jaime Pei for help with preparation of the figures.These studies were supported by National Institutes of Health(NIH) National Research Service Award F32DK10130 to X.L. andNIH PO1 DK62345 to P.D.W. The BioRad Radiance 2000 MSSM-MicroscopyShared Research Facility is supported in part by the HowardHughes Medical Institute-Biomedical Research Support Programaward to Mount Sinai School of Medicine and NIH–NationalCancer Institute shared resources grant R24CA095823.
Pichel JG, Shen L, Sheng HZ, Granholm A-C, Drago J, Gingberg A, Lee EJ, Huang SP, Saarma M, Hoffer BJ, Sariola H, Westphal H: Defects in enteric innervation and kidney development in mice lacking GDNF.
Nature382
:73
–76,1996[CrossRef][Medline]
Gupta IR, Piscione TD, Grisaru S, Phan T, Macias-Silva M, Zhou X, Whiteside C, Wrana JL, Rosenblum ND: Protein kinase A is a negative regulator of renal branching morphogenesis and modulates inhibitory and stimulatory bone morphogenetic proteins.
J Biol Chem274
:26305
–26314,1999[Abstract/Free Full Text]
Montesano R, Matsumoto K, Nakamura T, Orci L: Identification of a fibroblast-derived epithelial morphogen as hepatocyte growth factor.
Cell67
:901
–908,1991[CrossRef][Medline]
Sakurai H, Tsukamoto T, Kjelsberg CA, Cantley LG, Nigam SK: EGF receptor ligands are a large fraction of in vitro branching morphogens secreted by embryonic kidney.
Am J Physiol273
:F463
–F472,1997
Piscione TD, Yager TD, Gupta IR, Grinfeld B, Pei, Attisano L, Wrana JL, Rosenblum ND: BMP-2 and OP-1 exert direct and opposite effects on renal branching morphogenesis.
Am J Physiol273
:F961
–F975,1997
Lu WPB, Babakhanlow H, Pavlova A, Geng L, Fan X, Larson C, Brent G, Zhou J: Perinatal lethality with kidney and pancreae defects in mice with a targeted Pkd1 mutation.
Nat Genet17
:179
–181,1997[CrossRef][Medline]
Wilson PD, Geng L, Li X, Burrow CR: The PKD1 gene product, "polycystin-1," is a tyrosine-phosphorylated protein that colocalizes with alpha2beta1-integrin in focal clusters in adherent renal epithelia.
Lab Invest79
:1311
–1323,1999[Medline]
Geng L, Burrow CR, Li H, Wilson PD: Modification of the composition of polycystin-1 multiprotein complexes by calcium and tyrosine phosphorylation.
Biochem Biophys Acta1535
:21
–35,2000[Medline]
Yoder BK, Hou X, Guay-Woodford LM: The polycystic kidney disease proteins, polycystin-1, polycystin-2, polaris, and cystin, are co-localized in renal cilia.
J Am Soc Nephrol13
:2508
–2516,2002[Abstract/Free Full Text]
Huan Y, van Adelsberg J: Polycystin-1, the PKD1 gene product, is in a complex containing E-cadherin and the catenins.
J Clin Invest104
:1459
–1468,1999[Medline]
Kim E, Arnould T, Sellin LK, Benzing T, Fan KJ, Gruning W, Sokol SY, Drummond I, Walz G: The polycystic kidney disease 1 gene product modulates Wnt signaling.
J Biol Chem274
:4947
–4953,1999[Abstract/Free Full Text]
Wilson PD: Polycystin: New aspects of structure, function and regulation.
J Am Soc Nephrol12
:834
–845,2001[Abstract/Free Full Text]
Bhunia AK, Boletta APK: PKD1 induces p21 (wf1) and regulation of the cell cycle via direct activation of the JAK-STAT pathway in a process requiring PKD2.
Cell:109
:57
–168,2002[CrossRef]
Polgar K, Li X, Hyink D, Burrow CR, Wilson PD: Functional analysis of polycystin-1 in renal development: Microinjection of VVC lentiviral vector engineered PKD1 gene into metanephroi of mouse embryos [Abstract].
J Am Soc Nephrol13
:112a
,2002[CrossRef]
Klink A, Schiebel K, Winkelmann M, Rao E, Horsthemke B, Ludecke HJ, Claussen U, Scherer G, Rappold G: The human protein kinase gene PKX1 on Xp22.3 displays Xp/Yp homology and is a site of chromosomal instability.
Hum Mol Genet4
:869
–878,1995[Abstract/Free Full Text]
Schiebel K, Mertz A, Winkelmann M, Glaser B, Schempp W, Rappold G: FISH localization of the human Y-homolog of protein kinase PRKX (PRKY) to Yp11.2 and two pseudogenes to 15q26 and Xq12->q13.
Cytogenet Cell Genet76
:49
–52,1997[Medline]
Li X, Li HP, Amsler K, Hyink D, Wilson PD, Burrow CR: PRKX, a phylogenetically and functionally distinct cAMP-dependent protein kinase, activates renal epithelial cell migration and morphogenesis.
Proc Natl Acad Sci U S A99
:9260
–9265,2002[Abstract/Free Full Text]
Zimmermann B, Chiorini JA, Ma Y, Kotin RM and Herberg FW: PRKX is a novel catalytic subunit of the cAMP-dependent protein kinase regulated by the regulatory subunit type I.
J Biol Chem274
:5370
–5378,1999[Abstract/Free Full Text]
Semizarov D, Glesne D, Laouar A, Schiebel K, Huberman E: A lineage-specific protein kinase crucial for myeloid maturation.
Proc Natl Acad Sci U S A95
:15412
–15417,1998[Abstract/Free Full Text]
Orellana SA, Amieux PS. Zhao X, McKnight GS: Mutations in the catalytic subunit of the cAMP-dependent protein kinase interfere with holoenzyme formation without disrupting inhibition by protein kinase inhibitor.
J Biol Chem268
:6843
–6846,1993[Abstract/Free Full Text]
Naldini L, Blomer U, Gallay P, Ory D, Mulligan R, Gage FH, Verma IM, Trono D: In vivo gene delivery and stable transduction of nondividing cells by a lentiviral.
Science272
:263
–267,1996[Abstract]
Wilson PD, Dillingham MA, Breckon R, Anderson RJ: Defined human renal tubular epithelia in culture: Growth, characterization, and hormonal response.
Am J Physiol248
:F436
–F443,1985
Wilson P: In vitro methods in renal research. In:
Pediatric Nephrology, 5th Ed., edited by Niaudet P, Lippincott Philadelphia, Williams & Wilkins,2004
, pp317
–333
Racusen LC, Wilson PD, Hartz PA, Fivush BA, Burrow CR: Renal proximal tubular epithelium from patients with nephropathic cystinosis: immortalized cell lines as in vitro model systems.
Kidney Int48
:536
–543,1995[Medline]
Wilson PD, Hreniuk D, Gabow PA: Abnormal extracellular matrix and excessive growth of human adult polycystic kidney disease epithelia.
J Cell Physiol150
:360
–369,1992[CrossRef][Medline]
Schiebel K, Winkelmann M, Mertz A, Xu X, Page DC, Weil D, Petit C, Rappold GA: FISH localization of the human Y-homolog of protein kinase PRKX (PRKY) to Yp11.2 and two pseudogenes to 15q26 and Xq12->q13.
Hum Mol Genet6
:1985
–1989,1997[Abstract/Free Full Text]
Primpke G, Iassonidou V, Nellen W, Wetterauer B: Role of cAMP-dependent protein kinase during growth and early development of Dictyostelium discoideum.
Dev Biol221
:101
–111,2000[CrossRef][Medline]
Aubry L, Firtel R: Integration of signaling networks that regulate Dictyostelium differentiation.
Annu Rev Cell Dev Biol15
:469
–517,1999[CrossRef][Medline]
Jung S, Hoffmann R, Rodriguez PH, Mutzel R, Hofer HW: The catalytic subunit of cAMP-dependent protein kinase from Ascaris suum. The cloning and structure of a novel subtype of protein kinase A.
Eur J Biochem232
:111
–117,1995[Medline]
Blaschke RJ, Monaghan AP, Bock D, Rappold GA: A novel murine PKA-related protein kinase involved in neuronal differentiation.
Genomics64
:187
–194,2000[CrossRef][Medline]
Cary LA, Guan JL: Focal adhesion kinase in integrin mediated signaling.
Front Biosci4
:D102
–D113,1999[Medline]
Brown MC, Perrotta JA, Turner CE: Serine and threonine phosphorylation of the paxillin LIM domains regulates paxillin focal adhesion localization and cell adhesion to fibronectin.
Mol Biol Cell9
:1803
–1816,1998[Abstract/Free Full Text]
Sakai RNT, Ozawa K, Aizawa S, Hirai H: Characterization of the kinase activity essential for tyrosine phosphorylation of p130CAS in fibroblasts.
Oncogene14
:1419
–1426,1997[CrossRef][Medline]
Nickel C, Benzing T, Sellin L, Gerke P, Karihaloo A, Liu ZX, Cantley LG, Walz G: The polycystin-1 C-terminal fragment triggers branching morphogenesis and migration of tubular kidney epithelial cells.
J Clin Invest109
:481
–489,2002[CrossRef][Medline]
Related Article
This Month’s Highlights
J. Am. Soc. Nephrol. 2005 16: 3447-3448.
[Full Text][PDF]
Top
Abstract
Introduction
(PKA
R8.2 plasmid producing the viral proteins, and 2.5 µg of the pMD.G plasmid expressing the pseudotyping VSV glycoprotein (VSVG) (22). After 24 h, the medium was replaced with complete medium supplemented with 4 mM sodium butyrate, and supernatants were collected 48 h later. The titer (TU/ml) was determined 48 h after infection of 3 x 104 HeLa-Tat cells/well in 24-well plates by scoring the number of eGFP expressing cells using ultraviolet microscopy. 
