Discoidin Domain Receptor 1 Null Mice Are Protected against Hypertension-Induced Renal Disease
Martin Flamant*,
Sandrine Placier*,
Anita Rodenas,
Cyrile Anne Curat,
Wolfgang F. Vogel,
Christos Chatziantoniou* and
Jean-Claude Dussaule*,||
* INSERM U702, Tenon Hospital, Pierre et Marie Curie University, Department of Pathology, Tenon Hospital, and || AP-HP, Department of Physiology, School of Medicine St. Antoine, Pierre et Marie Curie University, Paris, France; Institute of Cardiovascular Physiology, J.W. Goethe University, Frankfurt, Germany; and Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada
Address correspondence to: Dr. Christos Chatziantoniou, INSERM U702, Hopital Tenon, 4 rue de la Chine, Paris 75020, France. Phone: +331-5601-6653; Fax: +331-4364-5448; E-mail: christos.chatziantoniou{at}tnn.ap-hop-paris.fr
Received for publication June 29, 2006.
Accepted for publication September 12, 2006.
A frequent complication of hypertension is the development ofchronic renal failure. This pathology usually is initiated byinflammatory events and is characterized by the abnormal accumulationof collagens within the renal tissue. The purpose of this studywas to investigate the role of discoidin domain receptor 1 (DDR1),a nonintegrin collagen receptor that displays tyrosine-kinaseactivity, in the development of renal fibrosis. To this end,hypertension was induced with angiotensin in mice that weregenetically deficient of DDR1 and in wild-type controls. After4 or 6 wk of angiotensin II administration, wild-type mice developedhypertension that was associated with perivascular inflammation,glomerular sclerosis, and proteinuria. Systolic pressure increasewas similar in the DDR1-deficient mice, but the histologic lesionsof glomerular fibrosis and inflammation were significantly bluntedand proteinuria was markedly prevented. Immunostaining for lymphocytes,macrophages, and collagens I and IV was prominent in the renalcortex of wild-type mice but substantially reduced in DDR1 nullmice. In separate experiments, renal cortical slices of DDR1null mice showed a blunted response of chemokines to LPS thatwas accompanied by a considerable protection against the LPS-inducedmortality. These results indicate the importance of DDR1 inmediating inflammation and fibrosis. Use of DDR1 inhibitorscould provide a completely novel therapeutic approach againstdiseases that have these combined pathologies.
Hypertension frequently is complicated by the development ofchronic renal failure, a complex pathology that is initiatedby inflammatory events that evolve to increased synthesis andaccumulation of extracellular matrix (ECM; mainly collagens)within the renal tissue and lead over time to loss of functionand ESRD. To date, no efficient treatment that can stop or,even more desirable, reverse the decline of renal function exists.Therefore, the understanding of the systems and/or mechanismsthat are involved in the development of renal vascular inflammationand fibrosis will provide valuable information to design specificpharmacologic targets to treat this incurable disease.
Important advancements have been made regarding the mechanismsthat are involved in the development of chronic renal failure.These studies focused mainly in the systems or agents that promoteECM synthesis and progression of renal disease. We and otherinvestigators, for instance, clearly identified and characterizedthe signaling pathways that vasoconstrictor peptides are usingto activate collagen synthesis (1–4). Less is known aboutthe mechanisms regarding the postsynthesis regulation of ECM,such as matrix anchoring and interactions with the cell membrane.
Among the systems that interact with the ECM are the discoidindomain receptors (DDR). They are the first identified receptortyrosine kinases that bind directly to the ECM (5). DDR1 bindsall types of collagens and is widely expressed in a varietyof tissues, including vascular smooth muscle, mesangial, andrenal epithelial cells and macrophages (6–8). Aortic smoothmuscle cells that were cultivated from DDR1 null mice showeddecreased proliferation, collagen attachment, migration and,matrix metalloproteinase-2/9 activity compared with cell culturesfrom wild-type controls (6,9). In addition, neointimal developmentwas severalfold reduced in DDR1 null mice compared with wild-typecontrols after vascular injury in carotids. It is interestingthat an important part of this reduction was due to a dramaticdecrease of collagen deposition in the neointima (6).
On the basis of these results, we hypothesized that DDR1 couldbe involved in the mechanisms of the hypertension-associatedrenal fibrosis. To test this hypothesis, we examined the developmentand the severity of renal vascular and glomerular lesions inDDR1 null mice and compared them with wild-type controls, usingan experimental model of hypertension-induced renal disease(angiotensin II [AngII] infusion). We found that DDR1 null miceare protected against the development of renal failure becauseof negligible perivascular and glomerular infiltration accompaniedby reduced levels of the abnormal accumulation of collagensI and IV.
Treatment
Male transgenic mice that weighed 30 to 35 g (4 to 6 mo of age)at the time of the experiments were fed high-NaCl (5%) mousefood with water available ad libitum. The higher-than-normalsalt diet accelerates the development of hypertension and aggravatesthe degree of the renal and vascular lesions. The generationand genotyping of mice was described previously (6,10). Theoriginal background of the DDR1-null mice was a mix of 129/Svwith CD1. These mice have been backcrossed five times to 129/Sv.No difference of the genetic background was found between DDR1–/–and wild-type controls after microsatellite analysis of DNAsamples from 26 mice (13 DDR1–/– and 13 wild type).The breeding couples that were used in our protocol were heterozygotes,and experiments were performed using DDR1-null mice and wild-typelittermates. All animal procedures were in accordance with theEuropean Union Guidelines for the Care and use of LaboratoryAnimals.
AngII (Sigma Chemical, St. Louis, MO) was infused subcutaneously(1 µg/kg per min) using osmotic minipumps (Model 1004;Alzet, Cupertino, CA) for 4 or 6 wk. No mortality was observedin DDR1-null and wild-type littermates for these periods oftime. In preliminary experiments, we established that this infusionrate of AngII was gradually increasing BP (from day 3) and wasproducing glomerular and vascular lesions (from day 14). A totalof 74 DDR1 null and 76 wild-type control mice were used.
Systolic BP was measured twice per week by the tail-cuff methodadapted to the mouse as described previously using the Chartmodule of the MacLab software (1,2). To avoid variations inBP as a result of day cycle, all measurements were carried outbetween 9 and 11 a.m. Eight measurements from each mouse weretaken at 2-min intervals, and a mean value was determined.
Isolation of Renal Cortical Slices
The technique to isolate renal cortical slices from mouse kidneywas similar to that previously described (1,2). The corticaltissue was used for morphology, immunocytochemistry, or cytokineevaluation according to the different protocols described next.
Renal Histology
Kidneys from at least 10 mice from each group were immersedin Dubosq solution. After fixation, cortical slices of eachkidney were embedded in paraffin after conventional processing(alcohol dehydration), and 3-µm-thick sections were stainedwith Masson trichromic solution for staining of ECM proteins.
Morphologic Evaluation
Sections of kidneys were examined on a blinded basis for thelevel of glomerular ischemia, glomerular sclerosis, and periglomerularand perivascular infiltration using a 0 to 4+ injury scale asdescribed previously (1–3). At least 200 glomeruli werescored to estimate the sclerotic index of an animal.
Immunohistochemistry for DDR1, Collagens I and IV, CD3, and F4–80
Four-microgram-thick cryostat sections of renal cortex werefixed with acetone for 7 min. After blockade of endogenous peroxidase,they were immunostained with an anti–collagen I or anti–collagenIV (both at 10 µg/ml; Chemicon, Temecula, CA) or anti-DDR1(4 µg/ml; Santa Cruz Biotechnology, Santa Cruz, CA) andthe Envision kit (Dako, Carpinteria, CA) was applied for 30min at room temperature. Staining was revealed by applying DABkit (Dako), hematoxylin QS (Vector, Burlingame, CA), and PermanentMounting Media Aqueous based (Innovex, Richmond, VA).
For staining of inflammatory cells, 4-µm-thick sectionsof paraffin-embedded kidneys were dewaxed, heated in citricacid solution, and incubated with a polyclonal rabbit anti-humanCD3 (Dako) or a biotinylated polyclonal rat anti-mouse F4–80(Serotec, Oxford, UK) at a concentration of 3 and 10 µg/ml,respectively. For CD3 immunostaining, sections first were treatedwith biotinylated anti-rabbit IgG, followed by AB solution treatment.The development was performed using 3,3-diaminobenzidine–glucoseoxidase and light counterstaining with hematoxylin.
The double-stain experiments with monocytes/macrophages wereperformed using frozen sections fixed to acetone for 7 min andthen washed with PBS using anti-DDR1 (C-20; Santa Cruz Biotechnology),anti-IgG rabbit TRITC (Jackson Immunoresearch, West Grove, PA),and rat anti-mouse F4–80–FITC (Serotec). Immunofluorescencemicrographs were obtained using an Olympus BX 51 camera DP70(Olympus, Rungis, France).
Blood Cell Count
White blood cells were counted and identified using the ADVIA120 Hematology System (Bayer Diagnostics, Puteaux, France),a technology that uses peroxidase staining and is based on cytochemicallight scatter and light absorption measurements.
Measurement of Urinary Albumin Excretion
The day before the mice were killed, they were transferred intometabolic cages and urine samples were collected for a 24-hperiod. Measurements of microalbuminuria were performed usingthe Olympus System Reagent (ref OSR6167) and an Olympus AU 400apparatus. Urinary albumin concentration was normalized to urinarycreatinine concentration, and values were expressed as mg albumin/µmolcreatinine.
LPS Administration
Endotoxemic shock was produced in DDR1–/– and wild-typecontrols (n = 10 per strain) by intraperitoneal injections ofLPS (10 mg/kg). Survival curves were established for a 36-hperiod. In additional in vitro experiments, renal cortical slicesthat were freshly isolated from both strains of mice were stimulatedby LPS at the concentrations of 100 and 1000 ng/ml. Incubationlasted 4 h in RPMI at 37°C, and monocyte chemoattractantprotein-1 (MCP-1) concentration was measured in the supernatantsusing a commercial ELISA kit (R&D System, Minneapolis, MN).
Statistical Analyses
Statistical analyses were performed using ANOVA followed byProtected Least Significance Difference Fisher test of the Statviewsoftware package. Results with P < 0.05 were considered statisticallysignificant. All values are means ± SE.
DDR1 Protein Expression Is Increased in Renal Cortex and Is Accompanied by Severe Nephroangio- and Glomerulosclerosis during AngII-Induced Hypertension
Continuous perfusion of AngII gradually increased BP in wild-typecontrols (Figure 1). At the same period, DDR1 expression wasincreased in renal vessels and within glomeruli as evidencedby immunocytochemistry using an antibody that was specific toDDR1 (Figure 2A). In untreated mice, however, DDR1 immunostainingwas present at a lesser degree in renal vessels and was almostnegligible in glomeruli (Figure 2B). In agreement with the literature,AngII produced severe vascular and glomerular lesions and profoundlyaltered the structure of the renal vasculature. These alterationswere characterized mainly by the appearance of sclerotic glomerulias evidenced by the abnormal deposition of ECM, the presenceof periglomerular and perivascular infiltrates, the formationof fibrin within the vascular wall, and the deposition of proteinaggregates in tubular lumen (Figures 3, A, D, and E, and 4).At least 50% of vessels showed fibrin-like lesions (Figure 3E).Almost all inflammatory cells were positive to anti-CD3 antibody(specific marker of lymphocytes) and were massively localizedaround damaged vessels and glomeruli (Figure 5A). A minor fractionof these infiltrating cells were positive for anti-F4–80antibody staining, a specific marker of monocytes and macrophages(Figure 5C). Very little staining was observed in or aroundthe tubular interstitium. In contrast, control tissue showeda very small number of positive cells (Figure 5, E and F).
Figure 1. Systolic BP increase after angiotensin II (AngII) infusion for 42 d in wild-type and discoidin domain receptor 1 (DDR1) null mice. Values are means ± SEM; n = 10; **P < 0.01 versus control.
Figure 2. Representative examples of DDR1 expression revealed by immunocytochemistry in the renal cortex of mice that received a continuous infusion of AngII (A) or placebo (B) for 28 d. G, glomerulus; V, renal vessel. Note the increased expression of DDR1 within glomeruli and renal vessels in the AngII-treated group. Bar = 20 µm.
Figure 3. Representative examples of renal cortical morphology (bar = 20 µm) revealed by Masson’s trichrome stain in wild-type (A, D, and E) or DDR1 null mice (B and F) that were treated for 28 d with AngII. (C) Wild-type mouse infused with placebo for 28 d. In A through C the magnification is lower to allow an overall view of the renal cortex. Note the lower levels of extracellular matrix deposition (green), formation of intratubular protein precipitation (red), and the quasi-absence of infiltrating cells in DDR1 null mice (B and F). The cellular infiltrates are mainly perivascular and periglomerular in the wild-type mice (A and D, arrows); the arrows in E show fibrin-like formation within renal vessel walls.
Figure 4. Quantitative analysis showing the percentage of sclerotic glomeruli (A) and the degree of infiltration (B) in wild-type and DDR1 null mice after 28-d infusion with AngII. Values are means ± SEM; n = 10; *P < 0.05 versus control; **P < 0.01 versus control; #P < 0.05 versus wild type.
Figure 5. Immunostaining with antibodies specific to lymphocytes (anti-CD3; A and B) and macrophages (anti F4–80; C and D) in the renal cortex of wild-type (A and C) and DDR1 null (B and D) mice that received an infusion of AngII for 28 d (bar = 20 µm). Note the intense staining of lymphocytes in wild-type (A) that contrasts with the sparse staining seen in DDR1 null (B) mice. Macrophages also were present in wild-type mice (C); in contrast, they were completely absent in DDR1 null mice (D). (E and F) CD3 and F4–80 staining in wild-type controls. (G) Representative experiment of double staining of DDR1 (anti-DDR1 in red) and CD3 lymphocytes (anti-CD3 in brown). Note that the DDR1 staining is restricted almost exclusively to renal vessels, whereas CD3-positive cells do not stain for DDR1. (H) DDR1 staining (red) in the absence of anti-CD3 antibody. (I) Representative experiment of double staining of DDR1 (anti-DDR1 in red) and F4–80 macrophages (anti-F4–80 in green). Note that the DDR1 staining is mainly in renal vessels and mesangial cells, whereas F4–80–positive cells do not stain for DDR1.
To test whether infiltrating cells were DDR1 positive, we performedexperiments to examine whether there was co-localization ofDDR1 expression with leukocytes in the renal cortical tissueof wild-type mice. As shown in Figure 5, G through I, inflammatorycells did not stain with anti-DDR1 antibody, whereas DDR1 wasexpressed mainly on smooth muscle cells of renal vessels. Thesemorphologic alterations were accompanied by the appearance ofalbuminuria (Figure 6).
Figure 6. Parameters of renal function (A: microalbuminuria in mg/mmol creatinine; B: plasma creatinine in µmol/L) measured in DDR1 null mice and wild-type controls after 4 or 6 wk of AngII infusion. Values are means ± SEM; n = 6; **P < 0.01 versus control; #P < 0.05 versus wild type.
DDR1 Null Mice Showed Reduced Structural and Functional Alterations during AngII-Induced Hypertension
Baseline systolic pressure of DDR1 null mice was not differentfrom that of wild-type controls and increased during AngII treatmentto a similar level compared with hypertensive wild-type mice(Figure 1). In addition, levels of leukocytes were not differentbetween DDR1–/– and wild-type mice under controlconditions (Table 1). Despite the similarity of the pressureresponse, the AngII-induced renal lesions were markedly decreasedin DDR1 null mice (Figures 3, B and F, and 4). In addition,immunostaining for lymphocytes or macrophages was negligiblein the renal cortex of the hypertensive DDR1 null mice (Figure 5,B and D). This preservation in the renal cortical structurewas accompanied by a relative protection of renal function (Figure 6A).
Table 1. White blood cell count in wild-type and DDR1 null mice under control conditions
To investigate whether the protection that was observed in DDR1–/–was transient, potentially as a result of a delay in the inflammatoryresponse, we performed an additional series of experiments inwhich the AngII infusion was prolonged for up to 42 d (Figure 1).Renal function continued to deteriorate in wild-type mice (Figure 6A).In contrast, urine concentration of albumin did not increaseand remained at low levels, similar to those observed after28 d of AngII in DDR1–/– mice (Figure 6A). The declineof renal function in wild-type mice was accompanied by severealterations of the renal cortical structure as evidenced bythe presence of sclerotic glomeruli and periglomerular and perivascularinfiltrates (Figure 7A). A major part of infiltrating cellswas positive to anti-CD3 antibody (Figure 7, C and E). In sharpcontrast, DDR1 null mice showed little ECM accumulation (Figure 7B)or leukocyte infiltration (Figure 7, D and F).
Figure 7. (A and B) Representative examples of renal cortical morphology (bar = 20 µm) revealed by Masson’s trichrome stain in wild-type (A) or DDR1 null mice (B) that were treated for 42 d with AngII. (C through F) Immunostaining with antibodies specific to lymphocytes (anti-CD3; C and D) and macrophages (anti–F4–80; E and F) in the renal cortex of wild-type (C and E) and DDR1 null (D and F) mice that receive AngII infusion for 42 d. Note the nearly complete absence of inflammatory cells in DDR1 null mice.
DDR1 Null Mice Exhibited Decreased Protein Expression of Collagens I and IV during AngII-Induced Hypertension
Because an important part of the exaggerated ECM formation duringrenal failure is due to the abnormal accumulation of collagens(mainly I and IV) and because DDR1 bind collagens, we investigatedwhether the protection of the renal structure in DDR1 null micecould be attributed to decreased levels of these collagens.As expected, collagen I was almost absent within the renal vasculatureunder normal conditions in both the wild-type (Figure 8A) andthe DDR1 null mice (data not shown). After AngII treatment for28 d, increased collagen I staining was observed in renal vesselsand glomeruli of wild-type mice (Figure 8B). In sharp contrast,very little specific collagen I staining was observed withinthe renal cortex of DDR1 null mice (Figure 8C). A strong stainingfor collagen IV was observed in wild-type (Figure 8D) and DDR1null mice (data not shown) under control conditions. This resultwas expected because collagen IV is the main component of renalbasal membrane. After AngII treatment, the accumulation of collagenIV was increased (thickening of basal membrane) in the renalcortex of wild-type mice (Figure 8E); again, the increase ofcollagen IV protein expression was limited in DDR1 null mice(Figure 8F) and resembled that of the untreated controls.
Figure 8. Representative examples of immunostaining with collagen I (A through C) or collagen IV (D through F) antibodies in wild-type controls (A and D) and mice that were treated with AngII for 28 d, either of wild-type (B and E) or deficient for DDR1 (C and F; bar = 20 µm). Note the exaggerated collagen I and IV staining after AngII treatment in the wild-type mice (B and E) compared with DDR1 null mice (C and F).
DDR1 Null Mice Are Protected against LPS-Induced Endotoxemic Shock
Next, we tested whether the strain difference in the inflammatoryresponse was specific to the hypertension-induced renal fibrosis.To this end, the LPS-induced endotoxemic shock (an acute modelof inflammation not related to hypertension and/or chronic renaldisease) was applied to a subgroup of DDR1 null mice and theirwild-type controls. Twenty-four hours after LPS intraperitonealinjection, the mortality rate accounted for 10% in the DDR1null mice and 60% in the wild-type littermates (Figure 9A).In addition, LPS was administered ex vivo in freshly isolatedrenal cortical slices in a dose-dependent manner. As shown inFigure 9B, renal cortical slices from wild-type mice respondedas expected by increasing MCP-1 secretion into the supernatantsin a dose-dependent manner. In contrast, MCP-1 secretion fromthe renal cortex of DDR1 null mice remained at low baselinevalues independent of the dosage of LPS (Figure 9B).
Figure 9. (A) Survival rate after intraperitoneal injection of LPS in DDR1 null mice and wild-type littermates (n = 21 for each strain). (B) Monocyte chemoattractant protein-1 concentration in the supernatant of cortical slices that were freshly isolated from DDR1-deficient mice and wild-type controls and incubated in the presence of increasing dosages of LPS during 4 h (n = 6 mice per strain per condition).
The objective of our study was to investigate the mechanismsof progression of renal fibrosis by examining the role of DDR1,a collagen receptor that displays tyrosine kinase activity (5–8,11)and is expressed in the kidney (12,13) in the development ofrenal fibrosis. Our initial working hypothesis was that if DDR1is involved in the fibrogenic process (acting as a collagenreceptor and displaying mitogenic properties), then mice thatlack functional DDR1 should be protected against renal fibrosiscompared with wild-type controls. Indeed, a major novel findingof our study is that the renal vasculature of DDR1 null micedisplayed a markedly blunted degree of collagen I and IV expressionsduring AngII infusion and thus was preserved from the structuraland functional alterations that are observed in this model.An additional, noninitially anticipated, important finding wasthe almost complete absence of infiltrating cells in the renalvessels and glomeruli of DDR1-deficient mice. It is interestingthat the lack of response is not limited to hypertension-inducedvascular inflammation but seems to apply to a much broader spectrumof inflammatory stimuli. These results show in an in vivo settingthat DDR1 is an important mediator of inflammation and fibrogenesisand suggest the possibility of using DDR1 blockers to treatinflammatory and/or fibrotic pathologies.
First, we observed that DDR1 expression was low in the kidneyunder normal conditions, whereas it was upregulated during AngII-inducedhypertension. The increased expression was specific to renalvessels and glomeruli, the two main renal compartments thatfibrogenesis (and the subsequent abnormal collagen formation)is originating in this experimental model. Similar to our findings,negligible or little expression of DDR1 in glomeruli or renalarterioles, respectively, under normal conditions was reportedrecently (13). As is the case with our experiments, DDR1 expressionwas induced in glomeruli during the development of renal failurein the remnant kidney model of nephropathy. In addition, otherinvestigators reported that DDR1 immunostaining was increasedand co-localized with the scar in the arterial wall of rat carotidsafter balloon injury (6).
That DDR1 expression was increased specifically in the damagedtissue does not necessarily mean that DDR1 is involved in themechanisms of renal vascular fibrosis. To corroborate this hypothesis,we used mice that do not express functional DDR1. Contrary tothe wild type, the degree of exaggerated collagen I and IV depositionwas inhibited and the progression toward renal failure (as evidencedby morphology and proteinuria) was significantly blunted inDDR1 null mice that were challenged with AngII for 4 or 6 wk.An analogous role for arterial DDR1 has been observed in thevascular remodeling after balloon injury. As with the data presentedhere, DDR1 expression was severalfold increased and co-localizedwith collagen formation and scar in carotids after vascularinjury in control animals; again, the structure of the arterialwall of carotids of DDR1 null mice was protected and showeddecreased collagen formation (6,9). The strain difference inthe degree of renoprotection cannot be attributed to the model(AngII-induced hypertension), because a similar strain differencewas observed when animals were submitted to another hypertensiveprotocol (L-NAME model, data not shown) in which endothelin-1plays a major role. Difference in the degree of hypertensionalso can be excluded because there was no strain differencein BP increase.
The contrast between the strain similarity in the BP increaseand the strain difference in the degree of renal vascular andglomerular damage suggests that systemic pressure increase anddevelopment of renal fibrosis are not always associated. Thisresult adds to our previous observations in which the developmentor prevention of renal structure was independent of the BP increaseduring hypertension (1,2,14,15). There is, however, a new elementthat distinguishes the present from the previous studies: Theprevention and/or protection seen previously was observed duringpharmacologic antagonism of endothelin or AngII and was attributedto the difference of the local-renal versus systemic activationof the vasoconstrictor’s receptor. This is not the casein these studies. A possible explanation is that collagen andDDR1 interact in a positive feedback manner to amplify the fibrogeniceffect of AngII. In agreement with this hypothesis, DDR1 isactivated in vascular smooth muscle cells (VSMC) that grow incollagen substrate, and VSMC that display functional DDR1 proliferatein a collagen substrate faster than cells that are deficientof DDR (6,9,16). In addition, we found that AngII and endothelin-1activate collagen I and IV genes and induce fibrosis in therenal vasculature (1,2,17). Therefore, we propose that AngIIpromotes collagen, which in turn upregulates DDR1 expressionin the renal vasculature; once activated, DDR1 promotes remodeling,cellular proliferation, and excessive matrix deposition. WhenDDR1 is absent, the feedback is broken at the step of vasoconstrictor-inducedcollagen formation. This hypothesis also provides an explanationfor why the exaggerated formation of collagens was blunted butnot completely normalized in DDR1 null mice. Alternatively,AngII could transactivate DDR1 independent of collagen formation,as it does with some other families of tyrosine kinase receptorssuch as PDGF and EGF receptors (18,19).
The other major phenotypic difference in our study concernsthe inflammatory response. The prolonged AngII action on vesselsfrequently is accompanied by the recruitment of infiltratingcells and the induction of vascular inflammation (20,21). Therefore,it was not surprising to see infiltrating cells around renalvessels and glomeruli in the wild-type mice. This contrastedwith the almost complete absence of inflammatory cells in therenal vasculature of mice that lacked DDR1. This differencecannot be secondary to impaired basal levels of leukocytes becauseno strain difference was observed in the white cell count undernormal conditions. Furthermore, that DDR1 null mice showed nocell infiltration even at prolonged time of AngII administration(6 wk) in which renal function worsened further in wild-typecontrols indicates that DDR1 deficiency prevented rather thandelayed leukocyte infiltration. The double-stain experimentssuggest that the interaction between DDR1 and cell infiltrationis mediated by the DDR1 expressed on the VSMC of renal vesselsin the model of hypertensive nephropathy.
Several in vitro studies have suggested that DDR1 could be involvedin the inflammatory response. DDR1 is expressed during differentiationof monocytes to macrophages, and this differentiation is facilitatedby collagens (22). A monocyte cell line that was transfectedwith DDR1 underwent differentiation and responded to inflammatorystimuli, whereas it remained undifferentiated and nonresponsiveto inflammation in the absence of DDR1 expression (8,23). Anamplifier role for DDR1 was proposed: Proinflammatory agentsinduce expression of DDR1 in the scarred tissue; interactionof DDR1 with collagen of the ECM in turn promotes differentiationof monocytes to macrophages and upregulation of cytokine secretionthrough a signaling cascade involving p38 mitogen-activatedprotein kinase and NF-B (8,22–24). These in vitro studiessuggested a possible role of DDR1 in mediating inflammatoryresponses. Our study is among the first reports to show thatDDR1 indeed is a major mediator of the inflammation, and itsabsence is accompanied by a deficient inflammatory responsein an in vivo pathology. In agreement with our results, a recentstudy observed that short-term administration of small interferenceRNA against DDR1 significantly inhibited expression of DDR1in bronchoepithelial cells and protected animals from the developmentof bleomycin-induced lung damage (25). It therefore seems thatthe role of DDR1 as mediator of inflammatory response can beextended to a more generalized inflammatory event, because itapplies in pathologies that range from chronic models of hypertensionto acute models of inflammation (bleomycin, LPS). We proposethat DDR1 participates in fibrosis as an amplifier of the AngII-inducedcollagen synthesis and in inflammation as an attractant or facilitatorof cellular infiltration and cytokine secretion. When thesetwo physiopathologic mechanisms are met (as is the case withthe renal chronic failure), DDR1 becomes a crucial factor ofthe progression of the disease.
This is among the first in vivo studies to investigate the roleof DDR1, a collagen receptor, in the physiopathologic mechanism(s)of renal fibrotic disease. Mice that lacked DDR1 showed decreasedcollagen formation and an absence of inflammatory cell infiltrationand were protected against the development of hypertension-associatedchronic renal failure. Development of inhibitors or blockersof systems that, like DDR1, mediate both fibrosis and inflammationcould provide a completely novel therapeutic approach againstdiseases with these combined pathologies.
Acknowledgments
This work was financially supported by the "Institut Nationalde la Santé et de la Recherche Médicale," the"Faculté de Médecine Pierre et Marie Curie," andan ACI grant from the "Ministère de la Recherche."
M.F. was research fellow of INSERM.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Chatziantoniou C, Boffa JJ, Ardaillou R, Dussaule JC: Nitric oxide inhibition induces early activation of type I collagen gene in renal resistance vessels and glomeruli in transgenic mice: Role of endothelin.
J Clin Invest 101
: 2780
–2789, 1998[Medline]
Boffa JJ, Tharaux PL, Placier S, Ardaillou R, Dussaule JC, Chatziantoniou C: Angiotensin II activates collagen type I gene in the renal vasculature of transgenic mice during inhibition of nitric oxide synthesis: Evidence for an endothelin-mediated mechanism.
Circulation 100
: 1901
–1908, 1999
Francois H, Placier S, Flamant M, Tharaux PL, Chansel D, Dussaule JC, Chatziantoniou C: Prevention of renal vascular and glomerular fibrosis by epidermal growth factor receptor inhibition.
FASEB J 18
: 926
–928, 2004[Abstract/Free Full Text]
Chatziantoniou C, Dussaule JC: Insights in the mechanisms of renal fibrosis: Is it possible to achieve regression?
Am J Physiol Renal Physiol 289
: F227
–F234, 2005[Abstract/Free Full Text]
Vogel W, Gish GD, Alves F, Pawson T: The discoidin domain receptor tyrosine kinases are activated by collagen.
Mol Cell 1
: 13
–23, 1997[CrossRef][Medline]
Curat AC, Vogel WF: Discoidin domain receptor 1 controls growth and adhesion of mesangial cells.
J Am Soc Nephrol 13
: 2648
–2656, 2002[Abstract/Free Full Text]
Matsuyama W, Wang L, Farrar WL, Faure M, Yoshimura T: Activation of discoidin domain receptor 1 isoform b with collagen up-regulates chemokine production in human macrophages: Role of p38 mitogen-activated protein kinase and NF-kappa B.
J Immunol 172
: 2332
–2340, 2004[Abstract/Free Full Text]
Hou G, Vogel WF, Bendeck MP: Tyrosine kinase activity of discoidin domain receptor 1 is necessary for smooth muscle cell migration and matrix metalloproteinase expression.
Circ Res 90
: 1147
–1149, 2002[Abstract/Free Full Text]
Vogel WF, Aszodi A, Alves F, Pawson T: Discoidin domain receptor 1 tyrosine kinase has an essential role in mammary gland development.
Mol Cell Biol 21
: 2906
–2917, 2001[Abstract/Free Full Text]
Shrivastava A, Radziejewski C, Campbell E, Kovac L, McGlynn M, Ryan TE, Davis S, Goldfarb MP, Glass DJ, Lemke G, Yancopoulos GD: An orphan receptor tyrosine kinase family whose members serve as nonintegrin collagen receptors.
Mol Cell 1
: 25
–34, 1997[CrossRef][Medline]
Gross O, Beirowski B, Harvey SJ, McFadden C, Chen D, Tam S, Thorner PS, Smyth N, Addicks K, Bloch W, Ninomiya Y, Sado Y, Weber M, Vogel WF: DDR1-deficient mice show localized subepithelial GBM thickening with focal loss of slit diaphragms and proteinuria.
Kidney Int 66
: 102
–111, 2004[CrossRef][Medline]
Lee R, Eidman KE, Kren SM, Hostetter TH, Segal Y: Localization of discoidin domain receptors in rat kidney.
Nephron Exp Nephrol 97
: e62
–e70, 2004[CrossRef][Medline]
Ying L, Flamant M, Vandermeersch S, Boffa JJ, Chatziantoniou C, Dussaule JC, Chansel D: Renal effects of omapatrilat and captopril in salt-loaded nitric oxide deficient rats.
Hypertension 42
: 937
–944, 2003[Abstract/Free Full Text]
Boffa JJ, Tharaux PL, Dussaule JC, Chatziantoniou C: Regression of renal vascular fibrosis by endothelin receptor antagonism.
Hypertension 37
: 490
–496, 2001[Abstract/Free Full Text]
Boffa JJ, Ying L, Placier S, Stefanski A, Dussaule JC, Chatziantoniou C: Regression of renal vascular and glomerular fibrosis: Role of angiotensin II receptor antagonism and metalloproteinases.
J Am Soc Nephrol 14
: 1132
–1144, 2003[Abstract/Free Full Text]
Inagami T, Eguchi S, Numaguchi K, Motley ED, Tang H, Matsumoto T, Yamakawa T: Cross-talk between angiotensin II receptors and the tyrosine kinases and phosphatases.
J Am Soc Nephrol 10[Suppl 11]
: S57
–S61, 1999
Saito Y, Berk BC: Transactivation: A novel signaling pathway from angiotensin II to tyrosine kinase receptors.
J Mol Cell Cardiol 33
: 3
–7, 2001[CrossRef][Medline]
Fliser D, Buchholz K, Haller H; EUropean Trial on Olmesartan and Pravastatin in Inflammation and Atherosclerosis (EUTOPIA) Investigators: Anti-inflammatory effects of angiotensin II subtype 1 receptor blockade in hypertensive patients with microinflammation.
Circulation 110
: 1103
–1107, 2004
Ruiz-Ortega M, Lorenzo O, Suzuki Y, Ruperez M, Egido J: Proinflammatory actions of angiotensins.
Curr Opin Nephrol Hypertens 10
: 321
–329, 2001[CrossRef][Medline]
Matsuyama W, Faure M, Yoshimura T: Activation of discoidin domain receptor 1 facilitates the maturation of human monocyte-derived dendritic cells through the TNF receptor associated factor 6/TGF-beta-activated protein kinase 1 binding protein 1 beta/p38 alpha mitogen-activated protein kinase signaling cascade.
J Immunol 171
: 3520
–3532, 2003[Abstract/Free Full Text]
Matsuyama W, Kamohara H, Galligan C, Faure M, Yoshimura T: Interaction of discoidin domain receptor 1 isoform beta (DDR1beta) with collagen activates p38 mitogen-activated protein kinase and promotes differentiation of macrophages.
FASEB J 17
: 1286
–1288, 2003[Abstract/Free Full Text]
Matsuyama W, Watanabe M, Shirahama Y, Hirano R, Mitsuyama H, Higashimoto I, Osame M, Arimura K: Suppression of discoidin domain receptor 1 by RNA interference attenuates lung inflammation.
J Immunol 176
: 1928
–1936, 2006[Abstract/Free Full Text]
This article has been cited by other articles:
C. Franco, G. Hou, P. J. Ahmad, E. Y.K. Fu, L. Koh, W. F. Vogel, and M. P. Bendeck Discoidin Domain Receptor 1 (Ddr1) Deletion Decreases Atherosclerosis by Accelerating Matrix Accumulation and Reducing Inflammation in Low-Density Lipoprotein Receptor-Deficient Mice
Circ. Res.,
May 23, 2008;
102(10):
1202 - 1211.
[Abstract][Full Text][PDF]
A. D. Konitsiotis, N. Raynal, D. Bihan, E. Hohenester, R. W. Farndale, and B. Leitinger Characterization of High Affinity Binding Motifs for the Discoidin Domain Receptor DDR2 in Collagen
J. Biol. Chem.,
March 14, 2008;
283(11):
6861 - 6868.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
B (8,22–24). These in vitro studies suggested a possible role of DDR1 in mediating inflammatory responses. Our study is among the first reports to show that DDR1 indeed is a major mediator of the inflammation, and its absence is accompanied by a deficient inflammatory response in an in vivo pathology. In agreement with our results, a recent study observed that short-term administration of small interference RNA against DDR1 significantly inhibited expression of DDR1 in bronchoepithelial cells and protected animals from the development of bleomycin-induced lung damage (25). It therefore seems that the role of DDR1 as mediator of inflammatory response can be extended to a more generalized inflammatory event, because it applies in pathologies that range from chronic models of hypertension to acute models of inflammation (bleomycin, LPS). We propose that DDR1 participates in fibrosis as an amplifier of the AngII-induced collagen synthesis and in inflammation as an attractant or facilitator of cellular infiltration and cytokine secretion. When these two physiopathologic mechanisms are met (as is the case with the renal chronic failure), DDR1 becomes a crucial factor of the progression of the disease. 
