Slide 1

Thank you very much and I’d like to thank the organisers for giving the privilege to be here in Barcelona.

Slide 2

What is allo-immunisation? This process is characterised by an immune response raised against antigens from a genetically distinct organism of the same species. In the human species there are three main triggers. The first is transplantation, the second is transfusion and the third is pregnancy.

Slide 3

Graft rejection is the paradigm of disease because of allo-immunization. However, because immune tolerance during pregnancy is only partial, allo-immunization can induce antenatal or neonatal disease, including hemolytic anemia mostly due to Rhesus incompatibility, neonatal thrombocytopenia, and even diseases affecting various organs, mostly the kidney.

Slide 4

In the patient with a kidney graft, 3 types of allo-immune antibodies can develop. First allo-immune antibodies can develop against a normal antigen of the donor kidney when this antigen is absent in the recipient because of a genetic defect. The best example is anti-GBM nephritis occurring in the patient with Alport syndrome. Second the recipient commonly develops antibodies against  allovariant antigens mostly but not exclusively HLA antigens. Last but not least allo-antibodies can be produced against hidden or modified antigens that are exposed in the donor kidney by episodes of acute rejection or after oxidative stress, following the renal ischemic period before transplantation. All three types of antibodies concur to chronic rejection. Many observations suggest that non-HLA allo-immunity plays a major role in the outcome of transplantation.

Slide 5

One of the best example of allo-immune disease occurring in the renal graft is anti-GBM nephritis observed after transplantation in patients with Alport’s syndrome. In the spontaneously occurring auto-immune disease, antibodies are produced against epitopes of the NC domain of the a3 chain of the type IV collagen that are unmasked by oxidant or toxic injury. In most Alport patients that have a mutation in the a5 chain of the type IV collagen, the a3 a4 a5 trimer is not expressed in the GBM, and therefore the recipient’s immune system is first exposed to a3, a4 and a5 epitopes in the GBM and TBM of the grafted kidney, which triggers an allo-immune response. Contrary to auto-antibodies, anti-GBM allo-antibodies are almost always directed against multiple accessible epitopes of the a3, a4 and a5 NC1 domains.

Slide 6

The question that next arises can an allo-immune response induce glomerular diseases in native kidneys? Not in grafted kidneys but in native kidneys?

Slide 7

The case who led to the first identification of an allo-immune renal disease on native kidneys is a male infant born at 38 weeks of gestation who presented with respiratory distress and oligoanuria followed by  nephrotic range proteinuria and increased blood pressure. A kidney biopsy was performed at 4 weeks of age. Tests for syphilis, toxoplasmosis, and viral infections were negative. The Coombs’test was also negative, and the infant had normal levels of complement.

Slide 8

The infant’s biopsy specimen showed a severe form of MN, with abundant electron-dense deposits on the outer aspect of the GBM and heavy subepithelial deposits of IgG of C5b-9. Because of the early development of MN in the infant, we suspected pregnancy induced immunization of the mother and tranplacental passage of antibodies. This hypothesis was tested by indirect IF using normal kidneys. The mother’s serum sampled before this pregnancy was negative whereas a serum sampled after delivery reacted with the glomerular capillary walls and the brush border. The target antigen could be immunoprecipitated from the renal brush-border with the mother’s serum and identified as NEP by WB and enzymatic assay. Anti-NEP IgG produced by the mother were delivered transplacentally to the fetus who then developed severe antenatal renal disease.

Slide 9

We then asked why the infant’s mother became immunized against NEP although NEP has a broad tissue distribution. Because the mother had no renal abnormalities despite high serum titers of anti-NEP antibodies persisting more than 2 years after pregnancy, we hypothesized that she might be deficient in NEP. NEP is normally expressed on granulocytes. Ne reactivity was observed by Western blotting when the mother’s granulocytes were incubated with monoclonal and polyclonal anti-NEP antibodies, whereas NEP was detected in the father’s granulocyte extracts. Note that the mother’s serum reacted with the father’s granulocytes. This important finding confirmed allo-immunization that probably occurred at the time of a mother’s miscarriage 2 months before this pregnancy. The mother’s lymphocytes were then massively exposed to NEP antigen which is heavily expressed on fetus’s syncytiotrophoblastic cells.

Slide 10

In addition to the Portuguese family that we first reported, we have identified a Dutch and a Moroccan family with FMAIG, as well as the truncating mutations in the MME gene coding for NEP that are responsible for the disease. This gene contains 24 exons and encodes a protein with a short cytoplasmic domain, a membrane spanning region and a large extracellular domain with a zinc binding motif which is essential for enzymatic activity. We identified 2 mutations in the 3 families. Those two mutations predict highly truncated proteins that we failed to detect by Western blot analysis which indicates that the mutated MME gene is functionally KO. The Portuguese mother is a compound heterozygote (and has inherited the exon 15 mutation from her mother and the exon 7 mutation from her father). Mothers from Dutch and Moroccan pedigrees are homozygous for the same deletion mutation in exon 7. Despite the absence of NEP protein in the five mothers and in a male individual, the subjects aged 20 to 46, are surprisingly healthy. Contrary to KO mice, they have normal blood pressure and lymphocyte phenotypes and functions. All heterozygous subjects also showed no phenotype with the exception of the (heterozygous) children born with MN due to transplacental passage of nephritogenic anti-NEP antibodies, which confirms that renal manifestations are immune mediated.

Slide 11

We were particularly interested in the outcome of second pregnancies in NEP-deficient mothers because we hypothesized that this pregnancy might present a higher risk to the fetus. The first mother showed a dramatic increase in anti-NEP antibodies, but she refused any treatment until the 7th month when anti-NEP titer reached 160, 000. IvIg then induced a significant decrease of the anti-NEP titer. The neonate presented at birth with severe nephrotic syndrome, together with marked proximal tubule impairment which may be related to binding of anti-NEP antibodies to the BB, and with osteopenia. In the second mother, treatment could be given earlier when the titer was 16, 000. The effect of IvIg was completed by adding corticosteroids which induced a further reduction of antibody titer. Unfortunately death occurred in utero due to mother’s thrombophilia but only scarce subepithelial deposits of IgG were seen. This slide is  important for the care of membranous nephropathy in general because it shows that when you have a good biomarker, then you can treat the patients with a greater efficacy and  monitor them very precisely.

Slide 12

Because NEP deficiency is asymptomatic in humans, NEP-deficient graft recipients are not identified prior to transplant. These individuals are most likely to raise an anti-NEP allo-immune response when their immune system is exposed to NEP in the donor kidney. To investigate this hypothesis, human podocyte extracts were analysed by Western blot with sera of 3 patients with de novo MN. Three different bands of reactivity were observed, but with apparent MW at about 110 kD and 40 kD different from the one of NEP which is close to 90 kD. Those “new” antigens are being identified by mass spectrometry.

Slide 13

We have recently obtained some evidence that allo-immune nephropathies could occur in children’s native kidneys after the neonatal period. This child developed MN at the age of 3 years. The child’s serum was unreactive with normal kidney whereas the mother’s serum strongly reacted with endothelial cells in the glomerulus and interstitial capillaries in some but not all normal kidney sections of the panel which may denote allo-immunization. Previous experiments by Andres et al clearly showed that antibodies to endothelial antigens including ACE could induce MN. Studies are in process to identify the target antigen in this patient.

Slide 14

What about the adult? The view that allo-immune nephropathies occur in adult native kidneys is supported by observations made in patients receiving a bone marrow transplant or allogenic blood stem cells, often with GVHD. GVHD is a typical example of microchimerism that fulfills the classical definition, that is presence of small population of cells (or DNA) in one individual that is derived from another genetically distinct individual.

Slide 15

In addition, to iatrogenic chimerism, cell trafficking between mother and fetus during pregnancy can result in long-term persistence of fetal cells in the mother, and of maternal cells in her progeny.

Maternal or fetal microchimerism have been implicated in so-called auto-immune diseases including systemic sclerosis which resembles GVHD and dermatomyositis. Their implication in adult MN and other types of nephropathy remains to be investigated.

Slide 16

In clinical settings compatible with allo-immune reactions, the following investigations should be performed :

- indirect IF using a panel of human kidney sections from different individuals

- WB analysis of podocyte extracts

- followed by identification of antigen by mass spectrometry

Slide 17

In conclusion, we have identified a new disease entity, FMAIG (Feto Maternal AlloImmune Glomerulopathy), that  extends the concept of alloimmune disease to encompass nephropathies of the native kidney. This concept may also apply to glomerulopathies in alloimmune settings at any life stage. A subset of autoimmune processes that target the kidney might be triggered by an initial alloimmune response, followed by exposure of cryptic epitopes and autoimmunity in response to organ injury.

Slide 18

Finally, I would like to thank my collaborators who performed this at Tenon hospital and especially Hanna Debiec, as well as the paediatricians and families who were kind enough to participate in this study.

Slide 19

Question: If this syndrome is caused by the antibodies of the mother, the disease should disappear after 6 months or so?

Prof Ronco: Yes, the disease  seems to disappear rapidly. The infants were born with severe nephrotic syndrome and/or acute renal failure, then the renal manifestation disappeared within 4 or 6 weeks. But when you look carefully at the patient outcome, some of them will keep proteinuria and the eldest patient now aged 20 from the Dutch family has developed postponed renal failure with nephrotic syndrome. Unfortunately, he refused a repeat biopsy and we do not know whether this postponed renal disease is due to nephron loss during pregnancy or to the persistence of an autoimmune process that would relay the initial alloimmune aggression.

Chairman: Another question.

Question: Pierre I think that’s one of the most beautiful applications of bench to bedside and bedside to bench in nephrology, really fascinating story.

Prof Ronco: Thank you.

Question: There’s one way actually to detect microchimerisms in the woman  carrying a male child by detecting Y chromosome positive cells. Would that be one way to get closer to monitoring microchimerism and – further understanding these diseases in skin biopsies for example?

Prof Ronco: Sure, this is a very nice suggestion and we are currentlt investigating this possibility.

Chairman: Thank you again Doctor Ronco.