Tubular Shear Stress and Phenotype of Renal Proximal Tubular Cells
Marie Essig and
Gérard Friedlander
INSERM U 426 and Department of Physiology, Faculté de Médecine Xavier Bichat, Université Denis Diderot–Paris 7, Paris, France.
Correspondence to Marie Essig, Inserm U 426, Faculté de Médecine Xavier Bichat, 16, rue Henri Huchard, F-75018, Paris, France; Phone: 33-1-44-85-62-70; Fax: 33-1-42-28-15-64;
ABSTRACT. Phenotypic alterations resulting from flow-inducedmechanical strains is a growing field of research in many celltypes such as vascular endothelial and smooth muscle cells,chondrocytes, and osteocytes. Because renal mass reduction isfollowed by a dramatic increase in GFR in the remaining nephron,modulation of tubular cell phenotype by flow-induced mechanicalstrains could be one of the events initiating the deleteriouspathways that lead to the destruction of renal parenchyma afterrenal mass reduction. This study demonstrates that increasedflow induced, in vitro and in vivo, a reinforcement of the apicaldomain of actin cytoskeleton and an inhibition of plasminogenactivator expression. These effects of flow on plasminogen activatorexpression were prevented by blocking the reorganization ofactin cytoskeleton and were associated with an increase in ashear-stress responsive element binding activity. These resultsconfirm that tubular flow affects the phenotype of renal epithelialcells and suggest that flow-induced mechanical strains couldbe one determinant of tubulointerstitial lesions during theprogression of renal diseases. E-mail: essig@bichat.inserm.fr
One of the hallmarks of renal mass reduction is the progressivedestruction of remaining functional nephrons occurring evenafter the apparent resolution of the initial injury. This deteriorationof renal structures is observed in a large number of renal diseasesand involves glomerular and tubulointerstitial damage that resultin part from an imbalance between extracellular matrix (ECM)production and proteolysis. Numerous molecules such as growthfactors, proteases, and cytokines have been implicated in thisphenomenon. However mechanisms that initiate these deleteriouspathways are still unknown.
It has been known for several decades that renal mass reductionis rapidly followed by a dramatic increase in tubular flow.Indeed, micropuncture studies evidenced a threefold increasein GFR in the remaining nephrons soon after subtotal nephrectomybefore any remodeling of proximal tubule occurs (1). Althoughit has been acknowledged that tubular flow is a main determinantof tubular behavior in terms of vectorial transport of waterand solutes, the effect of flow on other characteristics oftubular cell phenotype was ignored until recently. Some yearsago, Kaysen et al. (2) demonstrated in the rotating vessel wallmodel that shear stress modifies the expression of numerousgenes in proximal cells such as cubulin, megalin, villin, intercellularadhesion molecule, vascular cell adhesion molecule, and manganesesuperoxide dismutase. Furthermore, several studies on variouscell types such as vascular endothelial and smooth muscle cells,chondrocytes, and osteocytes (3) have focused on the potentrole of flow in regulating cell functions by modulation of themechanical stress applied to the cells. Mechanical stress isnow involved in bone response to gravity or in the progressionof various diseases such as atheroma and arthrosis (4–7).
On the basis of these observations, we hypothesized that tubularflow could modify the phenotype of proximal tubular cells andcould be involved in the modification of ECM remodeling observedafter renal mass reduction. To address this question, we usedtwo different approaches. In vitro, we applied a laminar flowon proximal tubular cells cultured on glass slides in a laminarflow chamber. In vivo, we used subtotal nephrectomy (Nx) toinduce a large increase in proximal flow and unilateral ureteralobstruction (UUO) to stop urinary flow. These experiments letus confirm that tubular flow has pleiotropic effects on proximaltubular cells and could be one of the events underlying tubulardamage that occurs after renal mass reduction.
Cellular actin network in proximal tubules is organized in vivoin long thick fibers oriented along the axis of the brush border,in thin fibers in the terminal web, and in basolateral stressfibers anchored in focal adhesion contacts. This pattern iscommon to all epithelial cells that display a brush border andcontrasts with the aspect of cultured epithelial cells. Indeed,epithelial cells from various origins when cultured under stillconditions demonstrated numerous pronounced cytosolic actinstress fibers that have been related to the dedifferentiationof the cells (8). However, proximal tubular cells submittedto a laminar flow show a disappearance of the cytosolic actinstress fibers and a reinforcement of the lateral actin networkand the brush border, an aspect obviously closer to the organizationof the cytoskeleton observed in vivo in epithelial cells. Invivo modifications of tubular flow also affected the organizationof the apical domain of the cytoskeleton in proximal cells.Indeed, high flow conditions after Nx induced a densificationof actin fluorescence in the brush border and in the terminalweb that became punctuated and striated, suggesting a furtherreinforcement of the apical network of actin cytoskeleton. Onthe contrary, decrease in tubular flow induced by UUO was associatedwith the disappearance of the brush border and the formationof cytosolic thin actin fibers.
It is interesting that the reorganization of the cytoskeletonobserved in epithelial cells is not identical to that observedin endothelial cells, where the major event is the alignmentof actin stress fibers, which had been evidenced in vitro andin vivo in different vessels (9). Such an alignment is not observedin proximal cells. In contrast, tubular epithelial cells submittedto flow seem to reinforce the apical and lateral domains ofactin filaments, which are specific elements of the cytoskeletonin epithelial cells, suggesting that the pattern of the reorganizationof the cytoskeleton induced by flow depends on the functionof the cells rather than on a general effect of flow.
Tubular cells are involved in the remodeling of ECM and aresuspected to play a major role in the fibrosis observed afterrenal mass reduction. One of the proteases involved in ECM remodelingis plasmin that is formed from plasminogen after cleavage bythe two plasminogen activators, tissue-type plasminogen activator(tPA) and urokinase (uPA), along the fibrinolytic pathway. Inhibitionof the fibrinolytic pathway has been shown to be associatedwith glomerulosclerosis or interstitial fibrosis in variousrenal diseases (10) and to correlate with the severity of impairmentof renal function during chronic renal failure (11). Furthermore,vascular flow is known to modulate tPA and PAI-1 expressionin endothelial and vascular smooth muscle cells (5,7). Tubularflow was also shown to affect the fibrinolytic activity of proximalcells and induced a major inhibition of tPA and uPA at the levelof mRNA and protein abundances and activity. The changes occurredearly after the beginning of flow. Experiments using variouslevels of flow (300 µl/min, 600 µl/min, and 1 ml/min)and static conditions revealed that only the highest value offlow (1 ml/min) corresponding to tubular flow observed aftersubtotal Nx induced a significant decrease in tPA or uPA mRNA.Finally, this phenotypic modification was demonstrated to bereversible because inhibition of tPA activities induced by 12h of shear stress was reversed when cells returned to staticconditions. Does the same phenomenon occur in vivo? Zymographiesperformed on kidney slices demonstrated that as compared withsham-operated animals (sham), Nx was associated with an inhibitionof the renal fibrinolytic activity, which occurred as earlyas 1 d after surgery and became significant 8 d after surgery.Furthermore, this inhibition was associated with a decreasein uPA mRNA content in proximal tubules from Nx animals as comparedwith sham animals. In contrast, UUO was associated with a dramaticincrease in renal fibrinolytic activity, which reached 180%of sham animals.
Are Tubular Mechanical Strains Exerted by Tubular Flow Similar to Endothelial Mechanical Strains?
Mechanical strains resulting from flow include stretch stress,which results from hydrostatic pressure, and shear stress, whichdepends on the viscosity of the fluid, the value of the flow,and the internal ray of the structure. In vascular beds, shearstress ranges from 5 to 100 dynes/cm2 (4) but diminishes rapidlyall along vascular thickness. Smooth muscle cells thus are submittedto lower mechanical strains ranging under 5 dynes/cm2.
In proximal tubules, as in blood vessels, stretch stress andshear stress may happen. However, it is very likely that, inmost circumstances except severe ureteral obstruction, stretchstress is of very low magnitude. Moreover, the accurate valueof these mechanical strains could not be calculated as easilyas in vascular beds. In fact, only the flow in the initial portionof the proximal tubule, corresponding to the single nephronGFR, could be known because urinary flow decreases along theproximal tubule in consequence of the tubular reabsorption.Whatever the accurate level of shear stress in proximal tubule,it has been demonstrated by Bonvalet et al. (12) that the internalray of proximal tubule also decreases along this nephron segmentand that the linear velocity of the tubular fluid is constantin the proximal tubules. On the basis of these observations,it could be postulated that the threefold increase in GFR observedafter renal mass reduction results initially in a similar threefoldincrease in the linear velocity of tubular flow. These previousresults allowed us to evaluate the tubular fluid velocity resultingfrom various single-nephron GFR and to submit cultured cellsto a laminar flow of similar fluid velocity. The highest valueof flow leads in a laminar flow chamber to a mechanical stressof 0.17 dynes/cm2 that is far lower than values used in endothelialor vascular smooth muscle cells studies.
Intracellular Pathway Involved in Tubular Flow Effects
Mechanisms underlying the phenotypic modifications induced bytubular flow are not elucidated, but it is likely that the reorganizationof the cytoskeleton, which was observed under flow condition,is instrumental. Indeed, flow-induced inhibition of the fibrinolyticsystem was associated in vivo and in vitro with a reorganizationof the cytoskeleton evidenced by immunostaining of the actinnetwork. Furthermore, blocking this reorganization by the useof cytochalasin D prevented the decrease in uPA and tPA mRNA.These results illustrate the close relationship between cytoskeletonreinforcement and the inhibition of the fibrinolytic system,but the underlying intracellular pathways involved in thesephenomena remain to be identified. On the basis of recent experiments,one could speculate that cytoskeletal reorganization affectsthe stability of mRNAs. Indeed, several 3'-regulatory sequencesaffecting uPA mRNA stability have been demonstrated in LLC-PK1cells (13,14), and a protein destabilizing uPA mRNA and actingthrough the binding to the 3' end has recently been identifiedin pulmonary epithelial cells (15).
Tubular flow could also modify the activity of specific transcriptionfactors. In endothelial cells, flow-induced increase in PDGFsynthesis was related to an increased binding of transcriptionalfactors to a specific DNA sequence called shear-stress responsiveelement (SSRE) (16). Activation of this element has been shownto be instrumental in the activation of PDGF and intercellularadhesion molecule-1 genes by vascular flow and to result fromthe binding of a p50-p65 heterodimer of NF-kB (17). The presenceof a SSRE binding activity composed in part by p50 and p65 proteinscould be evidenced in vitro in flow-stimulated proximal cellsand in vivo in nuclear extracts of proximal tubules. However,numerous genes have been demonstrated to contain an SSRE sequencein their promoter region, and the targets of the SSRE bindingproteins in nuclear extracts of stimulated proximal cells couldbe different from those identified in endothelial cells andstill need to be elucidated.
Tubular flow should now be viewed as a potent modulator of proximalcell phenotype. By affecting the organization of the cytoskeletonand the brush border, it could affect the polarity of the celland modify various cellular functions such as solute reabsorptionand EXM remodeling. By extrapolating the effect of shear stresson endothelial cells, one could hypothesize that tubular flowcould also affect proximal cell proliferation and response togrowth factors. However, many questions remain to be elucidated,particularly the precise role of the brush border in mechanotransduction.Indeed, the brush border is a highly specialized domain of theproximal cell containing specific epithelial actin-binding proteinssuch as villin that could lead to unique response of epithelialcells to mechanical strains.
Hostetter TH, Olson JL, Rennke HG, Venkatachalam MA, Brenner BM: Hyperfiltration in remnant nephrons: A potentially adverse response to renal ablation. Am J Physiol 241: F85–F93, 1981
Kaysen JH, Campbell WC, Majewski RR, Goda FO, Navar GL, Lewis FC, Goodwin TJ, Hammond TG: Select de novo gene and protein expression during renal epithelial cell culture in rotating wall vessels is shear stress dependent. J Membr Biol 168: 77–89, 1999[CrossRef][Medline]
Nomura S, Takano-Yamamoto T: Molecular events caused by mechanical stress in bone. Matrix Biol 19: 91–96, 2000[CrossRef][Medline]
Diamond SL, Eskin SG, McIntire LV: Fluid flow stimulates tissue plasminogen activator secretion by cultured human endothelial cells. Science 243: 1483–1485, 1989[Abstract/Free Full Text]
Kawai Y, Matsumoto Y, Watanabe K, Yamamoto H, Satoh K, Murata M, Handa M, Ikeda Y: Hemodynamic forces modulates the effects of cytokines on fibrinolytic activity of endothelial cells. Blood 87: 2314–2321, 1996[Abstract/Free Full Text]
Papadaki M, Ruef J, Nguyen K, Li F, Patterson C, Eskin S, McIntire L, Runge M: Differential regulation of protease activated receptor-1 and tissue plasminogen activator expression by shear stress in vascular smooth muscle cells. Circ Res 83: 1027–1034, 1998[Abstract/Free Full Text]
Brown D, Stow J: Protein trafficking and polarity in kidney epithelium: From cell biology to physiology. Physiol Rev 76: 245–297, 1996[Abstract/Free Full Text]
Eddy AA: Interstitial inflammation and fibrosis in rats with diet-induced hypercholesterolemia. Kidney Int 50: 1139–1149, 1996[Medline]
Colucci M, Semerano N, Montemurro P, Chiumarulo P, Triggiani R, Morrone LF, Schena FP: Urinary procoagulant and fibrinolytic activity in human glomerulonephritis. Relationship with renal function. Kidney Int 39: 1213–1217, 1991[Medline]
Bonvalet J-P, de Rouffignac C: Distribution of ferrocyanide along the proximal tubular lumen of the rat kidney: Its implication upon hydrodynamics. J Physiol 318: 85–98, 1981[Abstract/Free Full Text]
Montero L, Nagamine Y: Regulation by p38 mitogen-activated protein kinase of adenylate- and uridylate-rich element-mediated urokinase-type plasminogen activator (uPA) messenger RNA stability and uPA-dependent in vitro cell invasion. Cancer Res 59: 5286–5293, 1999[Abstract/Free Full Text]
Koziczak M, Montero L, Maurer F, Nagamine Y: Emerging regulatory mechanisms for fibrinolytic gene expression. Fibrinol Proteol 14: 146–154, 2000[CrossRef]
Shetty S, Idell S: Post-transcriptional regulation of urokinase mRNA. J Biol Chem 275: 13771–13779, 2000[Abstract/Free Full Text]
Resnick N, Collins T, Atkinson W, Bonthron D, Dewey C, Gimbrone M: Platelet-derived growth factor B chain promoter contains a cis-acting fluid shear-stress-responsive element. Proc Natl Acad Sci U S A 90: 4591–4595, 1993[Abstract/Free Full Text]
Khachigian LM, Resnick N, Gimbrone MAJ, Collins T: Nuclear factor-B interacts functionally with the platelet-derived growth factor B-chain shear stress response element in vascular endothelial cells exposed to fluid shear stress. J Clin Invest 96: 1169–1175, 1995
Top
Abstract
Introduction
B interacts functionally with the platelet-derived growth factor B-chain shear stress response element in vascular endothelial cells exposed to fluid shear stress. J Clin Invest 96: 1169–1175, 1995
