Podocyte and food antigens in membranous nephropathy Pierre Ronco INSERM Unit 702 and Division of Nephrology, Tenon hospital, Paris, France ECP, Helsinki, 2011 August 30th

Membranous Nephropathy Major cause of nephrotic syndrome and chronic renal failure

Aetiologies of membranous nephropathy 30%associated with: - infections - cancers -autoimmune diseases -drugs 70%« idiopathic forms » Proteinuriaiscomplement-dependent,andmay involve productionofoxygenfreeradicalsand metalloproteases

Membranous Nephropathy: IgG Subclass Distribution IgG 1 IgG 2 IgG 3 IgG 4 Idiopathic++±++++ Lupus+++ ++± Neoplasia Ohtani et al, Nephrol Dial Transplant, 2004, 19:574

Possible mechanisms of the formation of subepithelial deposits Glassock, New Engl J Med 2009, 361:81 Serum SicknessHeymann nephritis« Planted » antigen

Heymann nephritis Renal BB IgG deposits Megalin, the target antigen of HN In situ formation of immune deposits Y Y Y Y Y Y Y Podocytes Endothelium Y Y Rat: megalin Human? Y GBM Proteinuria Activation of complement Kerjaschki and Farquhar

From rats to men : Allo-immune neonatal MN

The case : Hugo D. (male gender) 34 Wks of gestation : oligohydramnios and enlarged kidneys 38 Wks : birth 1st Days of life : respiratory distress and oligoanuria, followed by nephrotic range proteinuria and increased blood pressure 4 Wks : CT-guided kidney biopsy Negative tests for syphilis, toxoplasmosis, cytomegalovirus and hepatitis -B virus infections Negative Coombs’ test. Normal levels of complement components at day 35

IgG Infant born with MN and nephrotic syndrome PLACENTA C5b-9 MOTHER FETUS Debiec et al. N Engl J Med 2002, 346:2053 NEP anti-NEP IgG before after Blot: anti-NEP mAb mother control

Potential mechanisms of albuminuria green - IgG red - NEP Green IgG Red NEP Blue C5b-9

Human IgGNEP NEWBORN RABBIT KIDNEY Induction of neonatal renal disease in pregnant rabbit by injection of human maternal IgG

Why did the mother become immunized without developing the renal disease ? IgG (Debiec et al. Lancet. 2004) NEP deficient anti-NEP IgG Genetic defect M C PLACENTA MOTHER NEP C5b-9 IgG Endothelium GBM Y Y Y Y Y Y Y Y Podocytes FETUS Infant born with MN and nephrotic syndrome  Feto-Maternal Allo-Immunization with antenatal Glomerulopathies (FMAIG)

Western blotting of glomerular proteins with serum from patients with iMN Beck et al, New Engl J Med, 2009, 361:11

Expression of PLA 2 R in normal kidney and glomeruli Beck et al, New Engl J Med, 2009, 361:11

GWAS analysis of patients with idiopathic MN Three cohorts Controls : ethnically matched Illumina platform : 300,000 SNPs Bioinformatics analysis PatientsControls French75157 Dutch1461,832 British Total5562,338 Stanescu et al, New Engl J Med, 2011, 364: 616

Single Nucleotide Polymorphisms (SNPs) Single base mutation in DNA Most simple form of genetic polymorphism 90% of all human DNA polymorphisms Occur per every base pairs Not uniformly distributed > 1, 000, 000 SNPs identified SNP in a coding region can be – Synonymous (silent mutation) – Non-synonymous (missense or nonsense mutation)

Principle of pangenomic (GWAS) studies

A risk HLA-DQA1 allele is associated with iMN and may interact with PLA2R alleles French (n=75 ; c=157) Dutch (n=146 ; c=1832) British (n=335 ; c=349)All patients (n=556; c=2338) Stanescu et al, New Engl J Med, 2011, 364: 616

Conclusions An HLA-DQA1 allele on chromosome 6p21 is most closely associated with idiopathic membranous nephropathy in persons of white ancestry This allele may facilitate an autoimmune response against targets such as variants of PLA2R1 Our findings suggest a basis for understanding this disease and illuminate how adaptive immunity is regulated by HLA The risk of iMN is higher with the HLA-DQA1 allele than with the PLA2R1 allele, suggesting that the HLA-DQ1 allele might favor autoantibody targeting also other antigens

PLA2R autoantibodies and PLA2R glomerular deposits in membranous nephropathy Debiec and Ronco, New Engl J Med, 2011, 364 :689

Are there other antigens on the horizon in patients with idiopathic MN ?

Screening of MN sera reactivity with podocyte « proteome » Human podocyte lysates Western blot analysis Different autoantibody profiles Patients’ sera kDa

Identification of antigens Human podocyte lysate Differential extraction Ion exchange chromatography Silver staining Western immunoblotting with patient’s serum 2D-electrophoresis Identification MALDI-TOF MS LC-MS/MS

Candidate autoantigens recognized by autoantibodies in sera from MN patients « Metabolic » proteins Cytoskeletal proteins Molecular chaperones > 10 Cytoplasm/cytoskeletal proteins

J Am Soc Nephrol 21 : , 2010

Mechanisms of immune complex formation and podocyte injury in MN Y Y Y Y Y Y Y Y Activation of complement GBM Podocytes C5b-9 ROS Proteases Proteinuria Cytoskeletal changes degradation Detachment DNA damage Apoptosis Lack of proliferation Podocytopenia Endothelium Y Y Initiation Progression Stress proteins Altered intracellular transport PLA2R Renal failure Enolase (Wakui et al, 1999) Cytoskeletal proteins NEP Progression SOD2 Aldose reductase

A role for nonnative antigens? Glassock, New Engl J Med 2009, 361:81 Serum SicknessHeymann nephritis« Planted » antigen

Anti-BSA antibodies in patients with MN Debiec et al, NEJM 2011, 364 : 2101

IgG binding of BSA synthetic peptides CDEFKADEKKFWGKYLY CTAFHDNEETFLKKYLY BSA HSA AB MNMMNM BSAHSA Peptide MN patientsControls Debiec et al, NEJM 2011, 364 : 2101

ELISA BSA B 2D electrophoresis and immunoblot Circulating BSA in patients with MN

Colocalization of BSA and IgG in immune deposits BSAIgG Green BSA Red IgG

IgG1 IgG2 IgG3IgG4 1 4 BSA HSA Biopsy specimen stained for IgG subclasses Reactivity of eluted IgG Patient with BSA in biopsy Patients without BSA in biopsy Debiec et al, NEJM 2011, 364 : 2101

Y APC B cell YYY Y Y Y Y Y Anti-BSA Antibodies Y Y Y Y Processing Digestion BSA T cell Absorption of antigen from the intestinal tract Role of BSA in the pathophysiology of MN

Proteinuria g/l IgG subclass initial disease relapse remission Circulating BSA ng/ml 2008 Association with disease activity 2005

What do these results teach us? The repertoire of antigens involved in « idiopathic » MN may include nonglomerular antigens Food cationized BSA and/or abnormal processing of BSA in the GI tract in young children may lead to passage in the blood of cationic BSA, immunization, and in situ formation of immune complexes (Border et al, J Clin Invest, 1982, 69:451) These cases point to a role for environmental factors in the pathogenesis of MN

The spectrum of human membranous nephropathies Neonatal, alloimmune : NEP Early childhood MN : BSA « Idiopathic » MN % : PLA 2 R (+ other specificities:AR, SOD2, enolase..?) % : other Ags including food/environmental Ag (BSA) « Secondary » MN Ags to be identified Graft MN - Recurrent : PLA 2 R (and other antigens) - de novo : allo-immune

Acknowledgments TENON CENTER OF NEPHROLOGY Bergamo (I) G. Remuzzi M. Abbate Nijmegen (NL) J. Wetzels J. Hofstra UK R. Kleta (London) P. Mathieson (Bristol) A.Rees (Vienna) CNG (Evry) D. Bacq-Daian Paediatricians V. Guigonis (Limoges, F) A. Bensman (Paris, F) G. Deschênes (Paris, F) T. Ulinski (Paris, F) J. Nauta (Rotterdam, NL) F. Janssen (Brussels, B) J. Nortier (Brussels, B) D. Bockenhauer (London, UK) M. Kemper (Hamburg, D) B. Stengel (Paris, F)

iMNLN-MNCancerHBVGvH DCHC N= % Europe + Asia Positive for anti-PLA2R (%) N=45 4.4% N=14 28% N=18 17% N=13 15% N=90 0% N=153 0% Secondary MN Anti-PLA2R antibody in membranous nephropathy Debiec,Tesar and Ronco; Hoxha et al, NDT 2011; Qin et al, JASN 2011

iMN Italy N=16 43% iMN Japan N=142 65% iMN Dutch iMN France N=42 57% N=60 80% iMN China iMN Germany N=100 52% N=81 72% iMN Czech IFA N=25 68% Positive for anti-PLA2R (%) Anti-PLA2R antibody in idiopathic membranous nephropathy Debiec,Tesar and Ronco; Hoxha et al, NDT 2011; Qin et al, JASN 2011