hyaluronan and proximal tubular cell migration

Kidney International, Vol. 65 (2004), pp. 823–833

CELL BIOLOGY – IMMUNOLOGY – PATHOLOGY

Hyaluronan and proximal tubular cell migration

TAKAFUMI ITO, JOHN D. WILLIAMS, SAPHWAN AL-ASSAF, GLYN O. PHILLIPS, and ALED O. PHILLIPS

Institute of Nephrology, University of Wales College of Medicine, Cardiff, Wales; and The North East Wales Institute,Centre for Water Soluble Polymers, Wrexham, Wales

Hyaluronan and proximal tubular cell migration.Background. The ubiquitous polysaccharide hyaluronan has

been associated with both acute renal injury and progressiverenal disease. The aim of this study was to examine the effectof hyaluronan on proximal tubular cell migration.

Methods. The proximal tubular cell line, HK-2 cells, weregrown in monolayer culture, and cell migration following addi-tion of hyaluronan characterized in an in vitro model of injurythat we have previously developed and characterized.

Results. Addition of well-defined preparations of exogenoushyaluronan increased cell migration; however, optimum en-hancement of migration was seen with hyaluronan of highmolecular weight. Activation of the mitogen-activated proteinkinase (MAPK) signaling cascade, as assessed by increased ex-pression of the dually phosphorylated active form of MAPK,could be demonstrated following addition of hyaluronan. Thiswas blocked by the addition of a specific antibody to thehyaluronan receptor, CD44. Hyaluronan-dependent enhancedmigration was also abrogated by addition the CD44 blockingantibody, and by inhibition of MAPK kinase (MEK) activity.Generation of a denuded area also led to increased synthesis ofendogenous hyaluronan and activation of MAPK, and blockageof either CD44 or MAPK activation inhibited proximal tubulecell (PTC) migration and re-epithelialization under nonstimu-lated conditions.

Conclusion. We have demonstrated that hyaluronan activa-tion of the MAPK pathway through binding to its receptorCD44, enhances proximal tubule cell (PTC) migration. In ad-dition, the results suggest that mechanical injury of PTC stim-ulated hyaluronan generation. These observations may haveimplications for both recovery from acute tubular injury andprogressive renal fibrosis.

The importance of pathologic changes in the renalcortical interstitium is now well recognized, suggestingthat interstitial fibrosis represents a crucial developmentleading to progressive renal dysfunction [1–6]. With in-creasing awareness of the importance of these pathologicinterstitial changes, interest has focused on the role ofcells, such as the epithelial cells of the proximal tubule

Key words: hyaluronan, migration, MAPK, CD44.

Received for publication June 11, 2003and in revised form July 23, 2003Accepted for publication September 26, 2003

C© 2004 by the International Society of Nephrology

(PTC) or the interstitial fibroblast, in the initiation of afibrotic response. We have demonstrated that PTC maycontribute to the pathogenesis of renal fibrosis directlyby alterations in the production of components of theextracellular matrix and indirectly by the production ofprofibrotic cytokines [7–11]. Hyaluronan is an ubiquitousconnective tissue polysaccharide, which in vivo is presentas a high molecular mass component of the extracellularmatrix. In addition to its role in providing cellular sup-port, it is now known that under normal circumstances,hyaluronan regulates cell-cell adhesion, migration, prolif-eration, differentiation, and the movement of interstitialfluid and macromolecules (reviewed in [12]). As a result,it is likely to be an important contributor to, and a regula-tor of, wound healing and tissue remodeling. In the nor-mal kidney, hyaluronan is expressed in the interstitiumof the renal papilla only, and alteration in papillary inter-stitial hyaluronan has been implicated in regulating renalwater handling by affecting physiochemical characteris-tics of the papillary interstitial matrix and influencing theinterstitial hydrostatic pressure [13]. Increased glomeru-lar production of hyaluronan has been implicated in thepathogenesis of diabetic nephropathy [14, 15]. Althoughhyaluronan is not a major constituent of the normal renalcorticointerstitium, it is known to be expressed aroundPTC following renal injury caused by diverse diseases[16–19]. Increased deposition of interstitial hyaluronanhas also been correlated with renal function in progres-sive renal disease associated with IgA nephropathy [20].These observations suggest that hyaluronan may be in-volved in repair following tubular injury, and also in themechanisms, which underlie progressive fibrosis.

PTC migration is common to both tubular repair andprogressive renal fibrosis. More specifically followingtubular injury, recovery of renal function is accompaniedby and may be dependent on the restoration of PTC in-tegrity and function. This process is dependent on cellproliferation as well as numerous other events such asthe organized migration and rearrangement of cells alonga modified tubular basement. In terms of progressive fi-brosis, the presence of myofibroblast in the renal intersti-tium has been identified as one of the best predictors ofprogression [21, 22] and furthermore there is increasing

823

824 Ito et al: Hyaluronan enhances PTC migration

evidence that these cells derive at least in part throughtransdifferentiation of PTC [23–26]. Although it is clearthat cytokines such as transforming growth factor-b1(TGF-b1) are critical in regulation of PTC phenotype,less is known regarding regulation of their migration, aprocess that is essential for the “transdifferentiated” cellsto enter the interstitial compartment [27]. The aim ofthe current study was therefore to examine the effectsof hyaluronan on PTC migration, and the mechanism bywhich it mediated these effects.

METHODS

Cell culture and quantification of cell migration

HK-2 cells [human renal proximal tubular epithelialcells immortalized by transduction with human papil-loma virus (HPV) 16 E6/E7 genes [28]] were cultured inDulbeccco’s modified Eagle’s medium (DMEM)/Ham’sF12 medium (Invitrogen Ltd., Paisley, UK) supple-mented with 10% fetal calf serum (FCS) (Biological In-dustries Ltd., Cumbernauld, UK), 2 lmol/L glutamine(Life Technologies Ltd., Paisley, UK), 20 mmol/L Hepesbuffer (Gibco BRL, Paisley, UK), 5 lg/mL insulin,5 lg/mL transferrin, and 5 ng/mL sodium selenite (Sigma,Poole, UK). Cells were grown at 37◦C in 5% CO2 and95% air. Fresh growth medium was added to cells ev-ery 3 to 4 days until confluent, and serum deprived for48 hours prior to experimental manipulation. All exper-iments were performed under serum-free conditions. Inall aspects of cell biology that we have studied previously,HK-2 cells respond in an identical fashion to primarycultures of human proximal tubular cells [29–32]. Fur-thermore, in contrast to malignant transformed cells [33],HK-2 cells exhibit an antiproliferative and antimigratoryresponse to TGF-b1 [34].

They are, therefore, a good model from which generalconclusions can be drawn in terms of PTC biology. How-ever, to confirm that PTC immortalization per se did notaffect migration responses, additional experiments wereperformed using primary cultures of human renal prox-imal tubular cells (HPTC) isolated and characterized aspreviously described [35]. Cell migration was examined aspreviously described [36, 37]. Briefly, a denuded area wasgenerated on quiescent cell monolayers of HK-2 cells orHPTC by scratching with a sterile pipette tip. The mono-layer was washed twice with phosphate-buffered saline(PBS) (Gibco BRL) and then incubated with serum-freemedium, or serum-free medium to which hyaluronan,prepared as described below, was added.

To quantify re-epithelialization, an intersecting area ofdenuded cells was generated as described previously [36,37], and closure of the denuded area monitored usingan Axiovert 100M inverted microscope fitted with a dig-ital camera (ORCA-1394, Hamamatsu Photonics, K.K.Hamamatsu, Japan.), and images of the denuded area

10.0

8.0

6.0

4.0

2.0

0.01.0 × 103 1.0 × 104

5.44 × 104

9.65 × 104

4.33 × 105

2.0 × 106

(Control)

1.0 × 105 1.0 × 106 1.0 × 107

Diff

eren

tial w

eigh

t fra

ctio

n

Molar mass, g/mol

Fig. 1. Molecular weight distribution of the hyaluronan samples ofeach weight average molecular weight shown as determined by gel per-meation chromatography coupled to an online multi angle laser lightscattering system. Each line represents the molecular weight distribu-tion of a separate hyaluronan preparation from which the weight av-erage molecular weight (molecular weight per gram of molecule) isderived.

captured as a digitalized sequence. The rate of motilityof cells was calculated as the number of cells entering thecentral denuded area. Cell number was expressed as cellsper mm2 of original denuded area. Cell proliferation incell culture monolayers was evaluated by incorporationof bromodeoxyuridine (BrdU) (Sigma) into the DNA ofcells as previously described [37].

Preparation and characterization of hyaluronanof defined molecular weight

Hyaluronan samples (Lot No. F1750762), in the form offreeze dried white powder was kindly provided by DenkiKagaku Kogyo, Kabushiki Kaisha, Japan. A solution of0.15 mol/L NaCl was prepared using pathogen-free wa-ter. The solution was autoclaved at 128◦C for 20 minutesand subsequently used to prepare 0.1% (wt/vol) stocksolution of hyaluronan. Two portions (∼20 mL) of thestock solution were introduced into a clean vial and son-icated for 1 and 5 minutes, then autoclaved for 20 min-utes at 128◦C. The third portion was only autoclaved for20 minutes at 128◦C. Two milliliters of the respective sam-ple was withdrawn using a sterile syringe. The final con-centration was accurately measured using the carbazolemethod [38].

The weight average molecular weight and molecularweight distribution was determined using gel permeationchromatography coupled to an online multiangle laserlight scattering system (GPC-MALLS) (Fig. 1). Full de-tails of the GPC-MALLS system are given elsewhere [39].For a polydisperse rod-like polymer such as hyaluronan,the weight average molecular weight is defined by:

Mw = �ciMi

where Mw is molecular weight, ci is the molar concentra-tion, and Mi the molecular weight of each of the individualmolecules.

Ito et al: Hyaluronan enhances PTC migration 825

Immunocytochemistry

Immunocytochemistry for examination of the focal ad-hesion components and cytoskeleton organization wasperformed on cells grown in 8-well glass chamber slides(Nunc, Gibco/BRL Life Technologies, Ltd., Paisley, UK).Cells were grown to confluence and stimulated underserum-free conditions with low- (5.4 × 104) or high-(2.0 × 106) molecular-weight hyaluronan for up to 5 days.Culture medium was subsequently removed and the cellmonolayer washed with sterile PBS. Cells were fixed with3.5% paraformaldehyde for 10 minutes at room temper-ature, and permeabilized with 0.1% Tween in PBS for5 minutes at room temperature. Following fixation, cellswere blocked with 1% bovine serum albumin (BSA) for1 hour prior to a further washing step with PBS. Sub-sequently, slides were incubated with the primary anti-body, diluted in 1%BSA/PBS overnight at 4◦C. Finally,cells were incubated with the appropriate fluoresceinisothiocyanate (FITC)-conjugated secondary antibody.Specimens were then mounted and analyzed by confo-cal laser scanning microscopy (TCS-40) (Leica Microsys-tems, Cambridge, UK).

Antibodies (and working dilutions) used for immuno-cytochemistry were purchased as follows: rabbit anti-human b-catenin (Santa Cruz, Wiltshire, UK) (dilution1:100); rabbit antihuman ZO-1 (Zymed Laboratories,Ltd., CA, USA) (dilution 1:100); mouse antihuman pax-illin (Zymed Laboratories Ltd.) (dilution 1:100); mouseTRITC-conjugated phalloidin antibody (Sigma) (dilu-tion 1:50); FITC-conjugated goat antirabbit (Sigma) (di-lution 1:100); FITC-conjugated rabbit antimouse (DakoCytomation Ltd., Cambridgeshire, UK) (dilution 1:100);and mouse monoclonal blocking antibody to CD44 (TheBinding Site, Birmingham, UK).

Western blot analysis

Association of focal adhesion components with theactin cytoskeleton was indirectly assessed by examiningtheir triton solubility as previously described [40]. Briefly,confluent monolayers were washed once with cold PBS,scraped and rinsed into 5 mL cold PBS. After centrifu-gation at 2500 rpm for 10 minutes, cells pellets were ex-tracted in buffer (150 mmol/L NaCl, 50 mmol/L Tris-HCl,0.01% NaN3, 2 mmol/L ethylenediaminetetraacetic acid(EDTA), 1 mmol/L sodium orthovandadate, 10 lg/mLleupeptin, and 25 lg/mL aprotinin) containing 1% Tri-ton X-100 (TX buffer) for 30 minutes on ice. Sampleswere centrifuged at 12,500 rpm for 30 minutes and thenthe supernatant (Triton-soluble components, includingmembrane and cytosolic fraction) was transferred to aseparate tube and kept at –70◦C until use. Western blotanalysis was subsequently performed by standardmethodologies. Proteins were visualized using enhancedchemiluminescence (Amersham Pharmacia Biotech UK,

Ltd., Buckinghamshire, UK) according to the manufac-turer’s instructions.

Antibodies (and working dilutions) used for Westernblot analysis were purchased as follows: goat antihumanvinculin (Santa Cruz) (dilution 1:1000); mouse antihu-man paxillin (Zymed Laboratories, Ltd.) (dilution 1:500);horseradish peroxidase (HRP)-conjugated secondary an-tibodies (Sigma) (dilution 1:10,000).

Detection of MAPK kinase (MEK) activation status

Confluent cell layers or cell layers in which a denudedarea had been generated were washed once with coldPBS, scraped, and rinsed into 250 lL reducing buffer.After boiling at 95◦C for 5 minutes, equal amounts ofsamples were loaded onto 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) andsubjected to Western blot analysis as described aboveusing an anti-active MAPK antibody (Promega, Madi-son, WI, USA), which recognizes the dually phosphory-lated active form of MAPK (p44/ERK1 and p42/ERK2)(final dilution 1:5000). Total MAPK expression was de-termined using an antitotal MAPK antibody (Sigma)(dilution 1:5000).

Fluorescence-activated cell sorter analysis

Cell surface CD44 expression was examined in cellsstimulated by hyaluronan preparations of differentmolecular weight. Following cell harvesting, each wasincubated with an anti-CD44 common region antibody(Calibiochem, San Diego, CA, USA) for 30 minutes at4◦C (antibody dilution 1:1000). FITC-labeled secondaryantibody (Sigma) (dilution 1:100) was then added for afurther 30 minutes at 4◦C. Cells were subsequently re-suspended in FACS buffer (Dulbecco’s Ca2+/Mg2+ freePBS, 5% BSA, 10 mmol/L EDTA, and 15 mmol/L sodiumazide). Data were collected using a Becton Dickinson(Oxford, UK) FACSCalibur 4CaTM, and analyzed usingCellQuest ProTM software.

Quantification hyaluronan concentrationby migrating cells

Generation of a denuded area was carried out asdescribed previously [34, 37]. Forty-eight hours later,hyaluronan concentration in the cell culture supernatantwas determined by an enzyme-linked binding proteinassay (hyaluronic acid; “Chugai” quantitative test kit;Chugai Diagnostics, Tokyo, Japan). To ensure that equalnumbers of cells were studied in all experiments, totalcell protein was determined using a micro bicinchonicicacid kit (Pierce-Wariner, Chester, UK). There were nodifferences in cell protein in any experiments. All experi-ments were performed using the same volume of medium;


therefore, the hyaluronan results are expressed as an ab-solute concentration of hyaluronan.

Statistical analysis

Statistical analysis was performed using the unpairedStudent t test, with a value of P < 0.05 considered to rep-resent a significant difference. The data are presented asmeans ± SD of number of experiments. For each indi-vidual experiment the mean of duplicate determinationswas calculated.

RESULTS

Hyaluronan and re-epithelialization

Characterization of the monolayer scratch system hasbeen described previously [34, 37]. Following the gener-ation of a denuded area, the progress of a typical experi-ment, under serum-free conditions, is shown in Figure 2A[i]. Under nonstimulated serum-free conditions, the de-nuded area was repopulated in all control experiments5 days following injury. Previous studies have demon-strated that FCS accelerated the kinetics of closure [34,37], addition of 0.1% FCS was therefore used as a pos-itive control (Fig. 2A [iv]). Addition of hyaluronan (25lg/mL) led to increased cell migration as compared tothe migration seen in control experiments in HPTC (Fig.2B). Although both low- (molecular weight 5.4 × 104) andhigh- (molecular weight 2.0 × 106) molecular-weightpreparations led to increased migration of cells, thepromigratory effect was significantly greater with thehigher-molecular-weight preparation at all time pointsbeyond 24 hours. A significant increase in migratorycell number following addition of hyaluronan was seen24 hours after addition of high-molecular-weight hyaluro-nan. In contrast, a significant increase in cell migra-tion in response to the low-molecular-weight preparationwas only seen 120 hours after its addition to scratchedmonolayers.

For HK-2 cells both dose- and time-dependent effectof hyaluronan on cell migration were studied (Fig. 3).As with primary cultures at all time points, the promi-gratory effects of hyaluronan were greatest with thehigh-molecular-weight hyaluronan preparation (molec-ular weight 2.0 × 106). For the high-molecular-weightpreparation a significant promigratory effect was seenafter 72 hours. For the low-molecular-weight prepara-tion (molecular weight 5.4 × 104), significant stimula-tion of cell migration was seen after 120 hours. Forboth hyaluronan preparations, stimulation of cell mi-gration was directly dependent on the concentration ofhyaluronan used. At all time points studied the greateststimulation of cell migration was seen at the highest con-centration of hyaluronan used.

Using this system we have previously demonstratedthat re-epithelialization of the denuded area is not de-

A

B1000

750

500

250

00 12 24 48 72 96 120

Time, hours

Cel

l num

ber/

mm

2

**

**

*

*

*

*

Fig. 2. Hyaluronan enhances re-epithelialization. (A) Phase contrastmicroscopy of HK-2 cells migrating into the denuded area of the scratchwound. Quiescent cell monolayers were injured by scratching with asterile pipette tip to generate an area denuded of cells. All images weretaken 48 hours following injury to demonstrate relative efficiency of re-epithelialization under control serum-free conditions [i], following ad-dition of 25 lg/mL of low-molecular-weight hyaluronan, 5.4 × 104 [ii] orfollowing addition of 25 lg/mL of high-molecular-weight hyaluronan,2.0 × 106 [iii]. For comparison, 0.1% fetal calf serum (FCS) was usedas a positive control to stimulate increased migration [iv]. (B) Quantifi-cation of wound closure following addition of 25 lg/mL of high- (2.0 ×106) (�) or low-molecular-weight (5.4 × 104) (�) hyaluronan to primarycultures/human renal proximal tubular cells (HPTC). In control experi-ments (�) cells were incubated with serum-free medium alone. Conflu-ent monolayers of HPTC were scratched as described in the Methodssection to produce an intersecting area denuded of cell. Subsequently,the rate of cell migration under each of the experimental conditionsshown was assessed by directly counting the number of cells migratinginto the intersecting denuded area at each of the time points indicated.The data are expressed as the number of cells per mm2 of denuded area.Data represent the mean ± SD of four individual experiments. ∗P <

0.05; ∗∗P < 0.005.

pendent on cell proliferation but rather on cell migration[34]. Furthermore, addition of hyaluronan to growth ar-rested HK-2 cells did not influence cell proliferation inthe wounded monolayer system using BrdU labeling as amarker of proliferating cells. Figure 4A to C demonstrateslittle uptake of BrdU at the edge of the denuded areafor up to 96 hours. Moreover, there were no differencesin BrdU uptake under serum-free conditions (Fig. 4A),or following addition of either low- (Fig. 4B) or high-(Fig. 4C) molecular-weight hyaluronan under serum-free


600

500

400

300

200

100

00 12 24 48 72 96 12

0

Time, hours

A

600

500

400

300

200

100

00 12 24 48 72 96 12

0

Time, hours

B

Cel

l num

ber,

mm

2

Fig. 3. Quantification of dose and time dependent wound closure fol-lowing addition high-molecular-weight hyaluronan (2.0 × 106) (A) orlow-molecular-weight hyaluronan (5.4 × 104) (B). Confluent monolay-ers of HK-2 cells were scratched as described in the Methods sectionto produce an intersecting area denuded of cell. Subsequently, the rateof cell migration was assessed by directly counting the number of cellsmigrating into the intersecting denuded area at each of the time pointsindicated following addition of hyaluronan at concentrations of 1 lg/mL(�), 10 lg/mL (�), or 25 lg/mL (�), or under control conditions (�).The data are expressed as the number of cells per mm2 of denuded areaand represent the mean ± SD of four individual experiments ∗P < 0.05,stimulated vs. control.

Fig. 4. Immunostaining of HK-2 cells with mouse anti-bromodeoxyuridine (BrdU) to demonstrate the proliferativeresponse to wounding under serum-free conditions, or followingaddition of 25 lg/mL of either low-molecular-weight hyaluronan (5.4 ×104) or high-molecular-weight hyaluronan (2 × 106). (A, B, and C)Cells at the edge of the scratched area 96 hours after mechanical injury.(D, E, and F) BrdU labeling of cells from the unaffected layer awayfrom the “scratched” edge.

conditions. This is in contrast to assessment of cell prolif-eration away from the denuded area in which prominentcell proliferation was seen (Fig. 4D to F). Uptake of BrdUwas, however, no different from control (Fig. 4D) follow-ing addition of either low- (Fig. 4E) or high- (Fig. 4F)molecular-weight hyaluronan.

Fig. 5. Hyaluronan does not affect cell-cell contact, focal adhesion for-mation or F-actin cytoskeleton organization. Adherens junction andtight junction organization was examined by immunocytochemical anal-ysis of the expression of b-catenin (A, B, and C) and ZO-1 (D, E,and F), respectively. Cells were stimulated with 25 lg/mL of eitherlow-molecular-weight hyaluronan (B and E) or high-molecular-weighthyaluronan (C and F). After 48 hours, cells were fixed with 3.5%paraformaldehyde, prior to visualization of b-catenin, ZO-1 as de-scribed in the Methods section. In control experiments, cells were ex-posed to serumfree medium only for 48 hours prior to fixation (A andD). Focal adhesion formation and F-actin cytoskeleton organizationwas examined in confluent monolayers of HK-2 cells were stimulatedwith 25 lg/mL of high-molecular-weight hyaluronan (H and K) (2.0 ×106) under serum-free conditions. Serum-free medium alone was addedin control experiments (G and J) and addition of recombinant trans-forming growth factor-b1 (TGF-b1) (10 ng/mL) was used as the positivecontrol to demonstrate up-regulation of focal adhesion formation andF-actin reorganization as previously described (I and L). After 48 hours,cells were fixed with 3.5% paraformaldehyde, prior to visualization ofpaxillin (G, H, and I) or filamentous actin (J, K, and L) as described inthe Methods section. One representative experiment of four individualreplicate experiments is shown.

Mechanism of accelerated epithelial cell migrationin response to hyaluronan

Our previous studies demonstrated modification ofHK-2 cell migration following addition of TGF-b1 orsodium orthovanadate was associated with alterationin cell-cell contact, focal adhesion plaque formationand their association with the F-actin cytoskeleton, andchanges in b3 integrin expression [34]. Addition ofhyaluronan did not influence cell-cell contact as assessedby visualization of the adherens junction component b-catenin (Fig. 5A to C) nor the tight junction componentZO-1 (Fig. 5D to F). Furthermore, hyaluronan prepara-tions, regardless of their molecular weight, did not al-ter the expression the focal adhesion component paxillin(Fig. 5G to I), nor organization of the F-actin cytoskele-ton (Fig. 5J to L), as assessed by immunocytochemistry.We have previously demonstrated that monolayer scratchwounding per se did not affect expression or distribution


of focal adhesion components nor the organization of theF-actin cytoskeleton [34].

Modulation of cell function by hyaluronan may be reg-ulated by alterations of its cell surface binding and in-ternalization, which is mediated by its principal receptorCD44. Cell surface CD44 expression examined by flowcytometry using an antibody to the common region ofCD44 was no different to control following stimulationby hyaluronan (data not shown). However, cell migrationand re-epithelialization in response to addition of exoge-nous hyaluronan was inhibited by the addition of a mon-oclonal blocking antibody to human CD44 (Clone BU52)(The Binding Site Ltd.) (Fig. 6). Addition of the blockingantibody was effective at abrogation of the promigratoryeffect of both the high (Fig. 6B) and the low (Fig. 6C)hyaluronan preparations.

Previous studies have demonstrated that CD44 medi-ated alteration in cell function may be associated with theMAPK signaling cascade [41]. In our epithelial mono-layer wounding system, cell migration following additionof exogenous hyaluronan, either the high- (2.0 × 106) orlow- (5.4 × 104) molecular-weight hyaluronan prepara-tion, was inhibited by the addition of the MEK inhibitorPD98059 (Calbiochem) (Fig. 7B and C). Furthermore,addition of hyaluronan led to activation of MAPK asassessed by MAPK phosphorylation status (Fig. 8), andhyaluronan-mediated activation of MEK was abrogatedby the addition of CD44 antibody (Fig. 8).

The role of endogenous hyaluronanin re-epithelialization

Previous work has demonstrated that mechanical in-jury of peritoneal mesothelial cells in vitro is associ-ated with induction of hyaluronan synthesis [42]. In thecurrent study, utilizing HK-2 cells, a significant increasein hyaluronan concentration in the cell culture super-natant was seen 48 hours following monolayer wounding(Fig. 9). This represented a twofold increase over the con-trol value (control, 91.1 ± 43.5 ng/mL; scratch, 180.3 ±72.4 ng/mL; N = 12; P < 0.05). Addition of an anti-CD44antibody caused an inhibition of cell migration (Fig. 10A)below that seen in the unstimulated controls followingthe generation of a scratch. This could also mimicked bythe addition of the MEK inhibitor PD98059 (Fig. 10B).In addition, activation of MEK could be demonstratedfollowing “scratching” of the epithelial monolayer, andthis was also inhibited by the addition of the anti-CD44antibody (Fig. 11).

DISCUSSION

Hyaluronan has long been associated with alteration ofcell migration of cultured cells [43]. Furthermore, numer-ous studies suggest that hyaluronan may be involved in

Control HMW-HA HMW-HA+ CD44Ab

A

B1500

1000

500

024 48 72 960

HMW-HA

Cel

l num

ber/

mm

2

Time, hours

**

**

C1000

500

750

250

24 48 72 96

LMW-HA

Time, hours

00

**

Fig. 6. Inhibition of hyaluronan stimulated cell migration by blockadeof CD44. (A) Phase contrast microscopy of HK-2 cells migrating into thedenuded area of the scratch wound. Quiescent cell monolayers were in-jured by scratching with a sterile pipette tip to generate an area denudedof cells. All images have been taken 48 hours following injury to demon-strate relative efficiency of re-epithelialization under control serum-freeconditions, following addition of 25 lg/mL of high-molecular-weighthyaluronan (HMW-HA) (2.0 × 106) or following addition of high-molecular-weight hyaluronan in the presence of a blocking antibodyto CD44 (HMW-HA + CD44Ab) (5 lg/mL). One representative ex-periment of four replicates is shown. Quantification of wound closurefollowing addition of either high-molecular-weight hyaluronan (HMW-HA) (2.0 × 106) (B) or low-molecular-weight hyaluronan (LMW-HA)(5.4 × 104) (C) in the presence (•) and absence (�) of blocking antibodyto CD44 (5 lg/mL). In control experiments (�) cell migration was as-sessed in the absence of hyaluronan or anti-CD44 antibody. Confluentmonolayers of HK-2 cells were wounded as described in the Methodssection to produce an intersecting area denuded of cell. Subsequently,the rate of cell migration under each of the experimental conditions de-scribed was assessed by directly counting the number of cells migratinginto the intersecting denuded area at each of the time points indicated.The data are expressed as the number of cells per mm2 of denuded areaand represent the mean ± SD of four individual experiments. ∗P < 0.05;∗∗P < 0.005, for hyaluronan stimulated ± anti CD44 antibody.

epithelial cancer cell migration and metastatic potential[44, 45]. The increased expression of hyaluronan asso-ciated with either tubular injury and progressive renalfibrosis led us to postulate that it may also modu-late PTC migration. We have previously described asimple and reproducible model with which to investi-gate PTC migration. Re-epithelialization in this modeloccurred in the absence of proliferation at the lead-ing edge, thus the model is ideally suited to examine


Control HMW-HA HMW-HA+ PD98059

A

300

250

200

150

100

50

00 24 48 72 96

Time, hours

Cel

l num

ber/

mm

2

B

*

*HMW-HA

00 24 48 72 96

Time, hours

C

*

*400

300

200

100

LMW-HA

Fig. 7. Hyaluronan-stimulated cell migration is abrogated by mitogen-activated protein kinase (MAPK) inhibition. (A) Phase-contrast mi-croscopy of HK-2 cells migrating into the denuded area of the scratchwound. Quiescent cell monolayers were injured by scratching with asterile pipette tip to generate an area denuded of cells. All images havebeen taken 48 hours following injury to demonstrate relative efficiencyof re-epithelialization under control serum-free conditions, followingaddition of 25 lg/mL of high-molecular-weight hyaluronan (HMW-HA)(2.0 × 106) or following addition of high-molecular-weight hyaluro-nan in the presence of PD98059 (HMW-HA + PD98059) (10 lmol/L).One representative experiment of four individual replicate experimentsis shown. Quantification of wound closure following addition of ei-ther high-molecular-weight hyaluronan (HMW-HA) (2.0 × 106) (B)or low-molecular-weight hyaluronan (LMW-HA) (5.4 × 104) (C) in thepresence (•) and absence (�) of PD98059 (10 lmol/L). In control ex-periments (�) cell migration was assessed in the absence of hyaluronanor PD98059. Confluent monolayers of HK-2 cells were scratched as de-scribed in the Methods section to produce an intersecting area denudedof cell. Subsequently, the rate of cell migration under each of the ex-perimental conditions described was assessed by directly counting thenumber of cells migrating into the intersecting denuded area at eachof the time points indicated. The data are expressed as the numberof cells per mm2 of denuded area, and represent the mean ± SD offour individual experiments. ∗P < 0.005, for hyaluronan stimulated ±PD98059.

factors regulating PTC migration [34]. In this study, wehave clearly demonstrated that, in vitro, hyaluronan pro-motes PTC migration. Recent studies suggest that high-molecular-weight hyaluronan and low-molecular-weighthyaluronan oligosaccharides present different signals tocells. In general, high-molecular-weight hyaluronan rep-resents the normal homeostatic state whereas the gen-eration of low-molecular-weight hyaluronan fragmentssignals a disruption of the normal homeostatic environ-

Active MAP kinase

Total MAP kinase

LMW-HAHMW-HACD44-AB

p44/ERK1p42/ERK2

− − + + − −− − − − + +− + − + − +

Fig. 8. Hyaluronan induced activation of mitogen-actived protein(MAP) kinase is inhibited by CD44 blockade. Confluent monolayersof HK-2 cells were stimulated with 25 lg/mL of either low-molecular-weight hyaluronan (LMW-HA) (5.4 × 104) or high-molecular-weighthyaluronan (HMW-HA) (2.0 × 106) for 48 hours in the presence orabsence of anti CD44 antibody (CD44-AB) (5 lg/mL). In control ex-periments, cells were exposed to serum-free medium alone. Whole cellextracts subjected to sodium dodecyl sulfate-polyacrylamide gel elec-trophoresis (SDS-PAGE) and Western blotting were probed with anti-active MAP kinase antibody, which recognizes the dually phosphory-lated active form of MAP kinase. The blot was reprobed with an an-titotal MAP kinase antibody to ensure no change in the expression ofwhole cell MAP kinase expression. One representative of three indi-vidual experiments is shown.

300

250

200

150

100

50

0Control Scratch

HA

con

cent

ratio

n ng

/mL

Fig. 9. Mechanical injury of the HK-2 cell monolayer increaseshyaluronan generation. Quiescent cell monolayers were injured byscratching with a sterile pipette tip to generate an area denuded of cells.After 48 hours cell supernatant samples were collected for quantitationof hyaluronan concentration as described in the Methods section. Thedata represent the mean ± SD of three separate experiments. ∗P < 0.05.

ment. Several workers have reported that hyaluronan-oligosaccharides may stimulate gene expression andprotein synthesis of chemokines [46] and interstitial colla-gens [47]. In contrast, high-molecular-weight hyaluronaninhibits the “bioactivity” of TGF-b and stimulate the se-cretion of tissue inhibitors of metalloproteinases [48, 49].The data in the current study support the notion that PTCresponse to hyaluronan is also related to its molecularweight in that, although all preparations of hyaluronanare promigratory, this effect was most pronounced withthe highest-molecular-weight form.

To date, little is known regarding the functional signifi-cance of alterations in hyaluronan in the renal cortex andhow might these results relate to renal disease. From our


800

600

400

200

00 24 48 72 96

Time, hours

Cel

l num

ber/

mm

2

A250

0

50

100

150

200

0 24 48 72 96Time, hours

B

Fig. 10. Re-epithelialization under serum-free conditions is inhibitedby CD44 blockade (A) or inhibition of mitogen-activated protein(MAP) kinase (B). Confluent monolayers of HK-2 cells were scratchedas described in the Methods section to produce an intersecting area de-nuded of cell. Subsequently, the rate of cell migration under serum-freecondition (�) was compared to the rate of cell migration either in thepresence of 5 lg/mL blocking antibody to CD44 (�) or in the presenceof 10 lmol/L PD98059 (�). The data are expressed as the number ofcells per mm2 of denuded area, and represent the mean ± SD of four in-dividual experiments. ∗P < 0.05, control vs. either anti-CD44 antibodyor PD98059.

Active MAP kinase

Total MAP kinase

WoundingCD-44 Ab

p44/ERK1p42/ERK2

− − + +− + − +

Fig. 11. Mitogen-activated protein (MAP) kinase activation followingmechanical wounding is inhibited by blockade of CD44. Quiescent cellmonolayers were injured by scratching with a sterile pipette tip to gener-ate an area denuded of cells either in the presence or absence of a block-ing antibody to CD44 (5 lg/mL). After 48 hours, total cell lysates wereanalyzed for both activated and total MAP kinase activity by Westernblot analysis as described in the Methods section. One representativeexperiment of four individual replicate experiments is shown.

data, we speculate that the net effect on altered hyaluro-nan generation in the kidney is dependent on the diseasecontext. For example, promigratory effects of hyaluronanrelated to tubular injury, may suggest that it may be in-volved in the post-mitotic remodeling phase of recovery[16–19].

Following tubular injury such as that seen following re-nal ischemia, restoration of functional integrity is depen-dent on PTC proliferation followed by migration alonga modified tubular basement membrane. Although theproliferative phase has been extensively studied [50–53],very little is known regarding the phase of remodeling,which precedes restoration of normal renal function. Al-though the kidney exhibits a remarkable ability to recoverfrom acute ischemic injury, it is also clear that, following

severe ischemic injury, kidney function may not com-pletely return to normal.

Furthermore, studies in rats have shown that incom-plete recovery from acute injury may be followed by pro-gressive renal injury [54–56]. Reports that impairment ofgraft function immediately after transplantation is asso-ciated with increased risk of late graft loss suggest thata similar predisposition to progressive renal disease mayalso occur in humans following severe ischemic injury [57,58]. Increased expression of hyaluronan around injuredcortical tubules has recently been documented followingischemic injury. This coupled to the promigratory effectsdocumented in this manuscript suggest that increased ex-pression of hyaluronan in acute injury may be a mecha-nisms to limit injury and facilitate restoration of function.

In contrast to the potential benefit of hyaluronan inacute tubular injury, the promigratory effect would sug-gest that in the context of chronic renal disease, thathyaluronan may have detrimental, profibrotic effects. It isnow clear that one of the key events during interstitial fi-brogenesis is the de novo activation of a-smooth muscleactin (a-SMA)-positive myofibroblasts. These cells areprimarily responsible for interstitial matrix accumulationand deposition in chronically diseased states. Resident a-SMA–positive myofibroblasts are sparse in the corticoin-terstitium of the normal kidney, although their numbercorrelates with the degree of fibrosis in progressive renaldiseases [21, 22, 59].

Myofibroblasts were traditionally presumed to origi-nate from resident fibroblasts. Recent evidence suggeststhat a small number of these cells migrate to the inter-stitial space from bone marrow [60]. In addition, theymay also arise in large numbers from tubular epithelialcells via epithelial to mesenchymal transdifferentiation(EMT) [24, 61, 62]. This process involves acquisition afibroblastic phenotype, as assessed by the expression offibroblast specific markers in vitro and in vivo [24]. Tubu-lar epithelial to myofibroblast transition is proposed tobe an orchestrated, highly regulated involving four keysteps: (1) loss of epithelial cell adhesion, (2) de novo ac-quisition of fibroblastic markers and reorganization of theactin cytoskeleton, (3) disruption of the tubular basementmembrane, and (4) enhanced cell migration [27]. The keyprofibrotic cytokine TGF-b1 is a critical regulator of PTCphenotype and function [25, 26, 63]. However, althoughTGF-b1 is an important factor in regulating these pro-cesses, it is likely that transdifferentiation is not solelydependent on the generation of TGF-b1. For example,for PTC to reside in the corticointerstitium, they needto adopt a promigratory phenotype. Although TGF-b1may be associated with a migratory phenotype in the con-text of malignant/metastatic phenotype of epithelial cells[64], addition of TGF-b1 to PTC by increasing the avid-ity of cell-matrix interactions inhibited migration [34].There is now increasing evidence that although TGF-b1 is


central to the process, that multiple factors may coop-eratively regulate epithelial-mesenchymal transition inthe kidney. It is interesting to speculate therefore thatincreased TGF-b1 and hyaluronan seen in progressiverenal fibrosis [20] act cooperatively driving the fibroticprocess, at least in part, through facilitating PTC transd-ifferentiation and the formation of an expanding popula-tion of renal interstitial myofibroblasts.

Numerous different mechanisms could explain the ef-fect of hyaluronan on PTC migration. First, hyaluronanthrough its ability to bind water forms highly hydratedchannels within the extracellular matrix, which facilitatemigration [65]. A second possibility would involve mod-ification of the cell’s interaction with the extracellularmatrix. Previously we have demonstrated that regula-tion of PTC-matrix interactions mediate alterations incell migration following addition of the profibrotic cy-tokine TGF-b1 [34]. Addition of TGF-b1 led to a markedinhibition of cell migration, increased expression of pax-illin and vinculin and their incorporation into dense focaladhesion plaques. This was associated with increased as-sociation of focal adhesion components with the F-actincytoskeleton.

There was also increased b3 integrin expression andincreased synthesis of the matrix component fibronectin.The data in the current study, however, clearly demon-strate that the mechanism by which hyaluronan modifiedcell motility was unrelated to focal adhesion assembly. Athird possible mechanism by which cell migration may bemodified is by alteration of cell-cell adhesion. Ectopic ex-pression of each of the three isoforms of hyaluronan syn-thase (HAS1, HAS2, or HAS3) led to reduced cadherin-mediated cell adhesion and reorganization of filamentousactin [66]. This is, however, unlikely to be the mode ofaction of exogenous hyaluronan in our experimental sys-tem, as we were unable to demonstrate any alterationin the expression of components of either adherens ortight junctions. That loss of cell-cell contract per se doesnot equate with enhanced PTC motility is consistent withour previous work demonstrating association of adherensand tight junction complex disassembly with the antimi-gratory effect of TGF-b1 [34]. The fourth possible mecha-nism, by which exogenous hyaluronan may enhance PTCmigration, is through its interaction with its transmem-brane receptors such as CD44, which is the principal re-ceptor for hyaluronan. In our experimental system, en-hanced migration following addition of hyaluronan wasabrogated by the addition of antibodies to this receptor. Avariety of signaling pathways, including the MAPK/ERKpathway, have been reported to associate with CD44, andpostulated to account for hyaluronan-CD44–mediatedalterations in cell migration [41, 67–71]. In the currentstudy, enhanced cell migration was also reduced by in-hibition of MEK, suggesting that exogenous hyaluronanthrough its interaction with CD44 resulted in the subse-

quent activation of the MAPK cascade and increased cellmigration. It is of note that CD44 seems to have a rela-tively low affinity for hyaluronan, requiring clustering oroligomerization of CD44 for down stream signaling [72].The maximal promigratory effect of hyaluronan seen withthe highest-molecular-weight preparation in the currentstudy would therefore be consistent with this observation.

Our observation that addition of exogenous hyaluro-nan enhanced re-epithelialization led us to examining theregulation of hyaluronan in the same system and de-termining whether endogenous hyaluronan may in anautocrine manner also facilitate re-epithelialization ofthe denuded area. It is clear that the effect of endoge-nous hyaluronan on cell migration may be dependentof the cellular background. Increased hyaluronan syn-thesis mediated enhanced migration in melanoma cells[73], mesothelioma cells [74], and keratinocytes [75].In contrast increased hyaluronan generation followingHAS transfection, inhibited migration of CHO cells [76].Wounding of the PTC cell monolayer in our experimen-tal system increased hyaluronan generation, and also ac-tivation of MEK kinase. Furthermore, as with additionof exogenous hyaluronan, cell migration was inhibited tocontrol levels by either addition of antibodies to CD44,or inhibition of the MAPK cascade. This suggests thatre-epithelialization is indeed related to autocrine activa-tion of CD44 by hyaluronan, and is consistent with pre-vious observations from our laboratory demonstratingincreased hyaluronan synthesis restricted to the woundedge and cells entering the wound following mesothelialcell “wounding” [42].

We therefore postulate that exogenous hyaluronan ac-celerated re-epithelialization in our system by enhancingsignaling through this autocrine pathway. Overall, the re-sults presented in this study demonstrate that hyaluro-nan stimulates PTC migration through CD44-mediatedactivation of MAPK. We suggest that the functional sig-nificance of these findings is dependent on the diseasebackground such that altered hyaluronan in acute injurymay provide a protective mechanisms to limit renal in-jury, while in the chronic setting, in combination withother factors, hyaluronan may enhance transdifferenti-ation and the fibrotic response by facilitating migrationof PTC into the renal interstitium.

ACKNOWLEDGMENT

A.O.P. is in receipt of a Glaxo-Smith-Kline Senior Fellowship.

Reprint requests to Dr. A.O. Phillips, Institute of Nephrology, Univer-sity of Wales College of Medicine, Heath Park, Cardiff CF14 4XN.E-mail: [email protected]

REFERENCES

1. MACKENSEN-HAEN S, BOHLE A, CHRISTENSEN J, et al: The conse-quences for renal function of widening of the interstitium and


changes in the tubular epithelium of the renal cortex and outermedulla in various renal diseases. Clin Nephrol 37:70–77, 1992

2. WEHRMANN M, BOHLE A, HELD H, et al: Long-term prognosis offocal sclerosing glomerulonephritis. An analysis of 250 cases withparticular regard to tubointerstitial changes. Clin Nephrol 33:115–122, 1990

3. WEHRMANN M, BOHLE A, BOGENSCHUTZ O, et al: Long-term prog-nosis of chronic idiopathic membranous glomerulonephritis. ClinNephrol 31:67–76, 1989

4. MACKENSEN-HAEN S, BADER R, GRUND KE, et al: Correlationsbetween renal cortical interstitial fibrosis, atrophy of the proxi-mal tubules and impairment of the glomerular filtration rate. ClinNephrol 15:167–1721, 1981

5. BOGENSCHUTZ O, BOHLE A, BATZ C, et al: IgA Nephritis: On theimportance of morphological and clinical parameters in the long-term prognosis of 239 patients. Am J Nephrol 10:137–147, 1990

6. BADER R, BADER H, GRUND KE, et al: Structure and function of thekidney in diabetic glomerulosclerosis: Correlations between mor-phological and functional parameters. Path Res Pract 167:204–216,1980

7. PHILLIPS AO, TOPLEY N, STEADMAN R, et al: Induction of TGF -b1synthesis in D-glucose primed human proximal tubular cells: Dif-ferential stimulation by the macrophage derived pro-inflammatorycytokines IL-1b and TNF-a. Kidney Int 50:1546–1554, 1996

8. PHILLIPS AO, MORRISEY R, STEADMAN R, et al: Polarity of stimulationand secretion of TGF-b1 by proximal tubular cells. Am J Pathol150:1101–1111, 1997

9. PHILLIPS AO, MORRISEY K, STEADMAN R, et al: Decreased degra-dation of collagen and fibronectin following exposure of proximaltubular cells to glucose. Exp Nephrol 7:449–462, 1999

10. MORRISEY K, STEADMAN R, WILLIAMS JD, et al: Renal proximal tubu-lar cell fibronectin accumulation in response to glucose is polyolpathway dependent. Kidney Int 55:160–167, 1999

11. JONES SG, MORRISEY T, WILLIAMS JD, et al: TGF-b1 stimulates re-lease of pre-formed bFGF from renal proximal tubular cells. KidneyInt 56:83–91, 1999

12. LAURENT TC, FRASER RE: Hyaluronan. FASEB 6:2397–2404, 199213. HANSELL P, GORANSSON V, ODLIND C, et al: Hyaluronan content in

the kidney in different states of body hydration. Kidney Int 58:2061–2068, 2000

14. MAHADEVAN P, LARKINS RG, FRASER JRE, et al: Increased hyaluro-nan production in the glomeruli from diabetic rats: Link betweenglucose induced prostaglandin production and reduced sulphatedproteoglycans. Diabetologia 38:298–305, 1995

15. MAHADEVAN P, LARKINS RG, FRASER RE, et al: Effect ofprostaglandin E2 and hyaluronan on mesangial cell proliferation.Diabetes 45:44–50, 1996

16. LEWINGTON AJP, PADANILAM BJ, MARTIN DR, et al: Expression ofCD44 in kidney after acute ischemic injury in rats. Am J Physiol278:R247–R254, 2000

17. SIBALIC V, FAN X, LOFFING J, et al: Up- regulated renal tubular CD44,hyaluronan, and osteopontin in kdkd mice with interstitial nephritis.Nephrol Dial Transplant 12:1344–1353, 1997

18. WELLS A, LARSSON E, HANAS E, et al: Increased hyaluronan inacutely rejecting human kidney grafts. Transplantation 55:1346–1349, 1993

19. WELLS AF, LARSSON E, TENGBLAD A, et al: The localisation ofhyaluronan in normal and rejected human kidneys. Transplantation50:240–243, 1990

20. SANO N, KITAZAWA K, SUGISAKI T: Localization and roles of CD44,hyaluronic acid and osteopontin in IgA nephropathy. Nephron89:416–21, 2001

21. ESSAWY M, SOYLEMEZOGLU O, MUCHANETA-KUBARA EC, et al: My-ofibroblasts and the progression of diabetic nephropathy. NephrolDial Transplant 12:43–50, 1997

22. GOUMENOS DS, BROWN CB, SHORTLAND J, et al: Myofibroblasts, pre-dictors of progression of mesangial IgA nephropathy? Nephrol DialTransplant 9:1418–1425, 1994

23. NG YY, HUANG TP, YANG WC, et al: Tubular epithelial myofibrob-last transdifferentiation in progressive tubulointerstitial fibrosis in5/6 nephrectomized rats. Kidney Int 54:864–876, 1998

24. STRUTZ F: Novel aspects of renal fibrogenesis. Nephrol Dial Trans-plant 10:1526–1532, 1995

25. FAN JM, NG YY, HILL PA, et al: Transforming growth factor b reg-ulates tubular epithelial myofibroblast transdifferentiation in vitro.Kidney Int 56:1455–1467, 1999

26. TIAN YC, FRASER DJ, ATTISANO L, et al: TGF-b1 mediated alter-ations renal proximal tubular epithelial cell phenotype. Am J Phys-iol 285:F130–F142, 2003

27. YANG J, LIU Y: Dissection of key events in tubular epithelial to my-ofibroblast transition and its implications in renal interstitial fibrosis.Am J Pathol 159:1465–75, 2001

28. RYAN MJ, JOHNSON G, KIRK J, et al: HK-2: An immortalized proximaltubule epithelial cell line from normal adult human kidney. KidneyInt 45:48–57, 1994

29. FRASER DJ, WAKEFIELD L, PHILLIPS AO: Independent regulation oftransforming growth factor-b1 transcription and translation by glu-cose and platelet-derived growth factor. Am J Pathol 161:1039–1049,2002

30. JANSSEN U, THOMAS G, GLANT T, et al: Regulation of inter-alpha-trypsin inhibitor (IaI) and tumour necrosis factor-stimulated gene6 (TSG-6) expression in human renal proximal tubular epithelialcells. Kidney Int 60:126–136, 2001

31. JONES SG, JONES S, PHILLIPS AO: Regulation of renal proximal tubu-lar epithelial cell hylauronan generation: Implications for diabeticnephropathy. Kidney Int 59:1739–1749, 2001

32. MORRISEY K, EVANS RA, WAKEFIELD L, et al: Translational regula-tion of renal proximal tubular epithelial cell transforming growthfactorb1 generation by insulin. Am J Pathol 159:1905–1915, 2001

33. BHOWMICK NA, GHIASSI M, BAKIN A, et al: Transforming growthfactor-beta1 mediates epithelial to mesenchymal transdifferentia-tion through a RhoA-dependent mechanism. Mol Biol Cell 12:27–36, 2001

34. TIAN YC, PHILLIPS AO: TGF-b1 mediated inhibition of HK-2 cellmigration. J Am Soc Nephrol 14:631–640, 2003

35. PHILLIPS AO, STEADMAN R, TOPLEY N, et al: Elevated D-glucoseconcentrations modulate TGF-b1 synthesis by human cultured re-nal proximal tubular cells: The permissive role of platelet derivedgrowth factor. Am J Pathol 147:362–374, 1995

36. YUNG S, DAVIES M: Mechanical injury of human peritoneal mesothe-lial cells is accompanied by an increase in hyaluronan synthesis. JAm Soc Nephrol 9:531, 1998

37. YUNG S, DAVIES M: Response of the human peritoneal mesothelialcell to injury: An in vitro model of peritoneal wound healing. KidneyInt 554:2160–2169, 1998

38. BITTER T, MUIR HM: A modified uronic acid carbazole reaction.Anal Biochem 4:330–331, 1962

39. AL-ASSAF S, WILLIAMS PA, PHILLIPS GO: The Molecular Charac-terisation of Hyaluronan and Hylan, Cambridge, UK, WoodheadPublishing, 2002, pp 1, 55–65

40. BEN-ZE’EV A, DUERR A, SOLOMON F, et al: The outer boundary of thecytoskeleton: A lamina derived from plasma membrane proteins.Cell 17:859–865, 1979

41. SLEVIN M, KRUPINSKI J, KUMAR S, et al: Angiogenic oligosaccharidesof hyaluronan induce protein tyrosine kinase activity in endothelialcells and activate a cytoplasmic signal transduction pathway result-ing in proliferation. Lab Invest 78:987–1003, 1998

42. YUNG S, THOMAS GJ, DAVIES M: Induction of hyaluronanmetabolism after mechanical injury of human peritoneal mesothe-lial cells in vitro. Kidney Int 58:1953–1962, 2000

43. SCHOR SL, SCHOR AM, GREY AM, et al: Mechanism of action of themigration stimulating factor produced by fetal and cancer patientfibroblasts: Effect on hyaluronic and synthesis. In vitro Cell DevBiol 25:737–746, 1989

44. ANTTILA MA, TAMMI RH, TAMMI MI, et al: High levels of stromalhyaluronan predict poor disease outcome in epithelial ovarian can-cer. Cancer Res 60:150–155, 2000

45. AUVINEN PK, PARKKINEN JJ, JOHANSSON RT, et al: Expression ofhyaluronan in benign and malignant breast lesions. Int J Cancer74:477–481, 1997

46. MCKEE CM, PENNO MB, COWMAN M, et al: Hyaluronan (HA) frag-ments induce chemokine gene expression in alveolar macrophages.J Clin Invest 98:2403–2413, 1996

47. ROONEY P, WANG M, KUMAR P, et al: Angiogenic oligosaccharidesof hyaluronan enhance the production of collagens by endothelialcells. J Cell Sci 105:213–218, 1993


48. LOCCI P, MARINUCCI L, LILLI C, et al: Transforming growth factorb1-hyaluronic acid interaction. Cell Tissue Res 281:317–324, 1995

49. YASUI T, AKATSUKA M, TOBETTO K, et al: Effects of hyaluronan on theproduction of stromolysin and tissue inhibitor of metalloproteinase-1 (TIMP-1) in bovine articular chondrocytes. Biomed Res 13:343–348, 1992

50. ICHIMURA T, MAIER JAM, MACIAG T, et al: FGF-1 in normal andregenerating kidney: expression in mononuclear, interstitial and re-generating epithelial cells. Am J Physiol 269:F653–F662, 1995

51. IGAWA T, MATSUMOTO K, KANDA S, et al: Hepatocyte growth factormay function as a renotropic factor for regeneration in rats withacute renal injury. Am J Physiol 265:F61–F69, 1993

52. MATEJKA GL, JENNISCHE E: IGF-1 binding and IGF-1 mRNA ex-pression in the postischemic regenerating rat kidney. Kidney Int42:1113–1123, 1992

53. HOMMA T, SAKAI M, CHENG HF, et al: Induction of heparin-bindingepidermal growth factor like growth factor mRNA in rat kidneyafter acute injury. J Clin Invest 96:1018–1025, 1995

54. PAGTALUNAN ME, OLSON JL, TILNEY NL, et al: Late consequencesof acute ischemic injury to a solitary kidney. J Am Soc Nephrol10:366–73, 1999

55. PAGTALUNAN ME, OLSON JL, MEYER TW: Contribution of an-giotensin II to late renal injury after acute ischemia. J Am SocNephrol 11:1278–86, 2000

56. BASILE DP, DONOHOE D, ROETHE K, et al: Renal ischemic injury re-sults in permanent damage to peri-tubular capillaries and influenceslong-term function. Am J Physiol 281:F887–F899, 2001

57. TILNEY NL, GUTTMANN RD: Effects of initial ischemia/reperfusioninjury on the transplanted kidney. Transplantation 64:945–947, 1997

58. SAMANIEGO M, BALDWIN WM, SANFILIPPO F: Delayed graft function:Immediate and late impact. Curr Opin Nephrol Hypertens 6:533–537, 1997

59. MUCHANETA-KUBARA EC, EL NAHAS AM: Myofibroblast pheno-types expression in experimental renal scarring. Nephrol Dial Trans-plant 12:904–915, 1997

60. IWANO M, PLIETH D, DANOFF TM, et al: Evidence that fibroblastsderive from epithelium during tissue fibrosis. J Clin Invest 110:341–50, 2002

61. RASTALDI MP, FERRARIO F, GIARDINO L, et al: Epithelial-mesenchymal transition of tubular epithelial cells in human renalbiopsies. Kidney Int 62:137–146, 2002

62. STRUTZ F, OKADA H, LO CW, et al: Identification and characterisa-tion of a fibroblast marker: FSP1. J Cell Biol 130:393–405, 1995

63. TIAN YC, PHILLIPS AO: Interaction between the TGF-b type II re-ceptor/Smad pathway and b-catenin during TGF-b1 mediated ad-herens junction disassembly. Am J Pathol 160:1619–1628, 2002

64. DUMONT N, BAKIN A, ARTEAGA CL: Autocrine transforming growthfactor-b mediates Smad-independent motility in human cancer cells.J Biol Chem 278:3275–3285, 2003

65. TAMMI MI, DAY AJ, TURLEY EA: Hyaluronan and homeostasis: Abalancing act. J Biol Chem 277:4581–4584, 2002

66. ITANO N, ATSUMI F, SAWAI T, et al: Abnormal accumulation ofhyaluronan matrix diminishes contact inhibition of cell growth andpromotes cell migration. Proc Natl Acad Sci USA 99:3609–3614,2002

67. ILANGUMARAN S, BRIOL A, HOESSLI DC: CD44 selectively associateswith active Src family protein tyrosine kinases Lck and Fyn inglycosphingolipid-rich plasma membrane domains of human pe-ripheral blood lymphocytes. Blood 91:3901–3908, 1988

68. TAHER TEI, SMIT L, GRIFFIOEN AW, et al: Signalling through CD44is mediated by tyrosine kinases. J Biol Chem 271:2863–2867, 1996

69. FITZGERALD KA, BOWIE AG, SKEFFINGTON BS, et al: Ras, proteinkinase C and IkB kinase 1 and 2 are downstream effectors of CD44during the activation of NF-kB by hyaluronic acid fragments in T-24carcinoma cells. J Immunol 164:2053–2063, 2000

70. LEGG JW, LEWIS CA, PARSONS M, et al: A novel PKC-regulatedmechanism controls CD44 ezrin association and directional cellmotility. Nature Cell Biol 4:399–407, 2002

71. BOURGUIGNON LY, ZHU H, SHAO L, et al: CD44 interaction with c-Src kinase promotes cortactin-mediated cytoskeleton function andhyaluronic acid-dependent ovarian tumor cell migration. J BiolChem 276:7327–7336, 2001

72. LESLEY J, HASCALL VC, TAMMI M, et al: Hyaluronan binding by cellsurface CD44. J Biol Chem 275:26967–26975, 2000

73. ICHIKAWA T, ITANO N, SAWAI T, et al: Increased synthesis ofhyaluronate enhances motility of human melanoma cells. J InvestDermatol 113:935–939, 1999

74. LI Y, HELDIN P: Hyaluronan production increases the malignantproperties of mesothelioma cells. Br J Cancer 85:600–607, 2001

75. RILLA K, LAMMI J, SIRONEN R, et al: Changed lamellipodial exten-sion, adhesion plaques and migration in epidermal keratinocytescontaining constitutively expressed sense and antisense hyaluronansynthase 2 genes. J Cell Sci 115:3633–3643, 2002

76. BRINCK J, HELDIN P: Expression of recombinant hyaluronan syn-thase isoforms in CHO cells reduces cell migration and cell surfaceCD44. Exp Cell Res 252:342–351, 1999

hyaluronan and proximal tubular cell migration

Documents