My lifetime in <scp>IgA</scp> nephropathy: An <i>unexpected</i> journey

My unexpected journey began when my wife, Rachel and I in 1975 pursued our post-doctoral medical training at the University of Kentucky, hers in psychiatry and mine in internal medicine. I planned to be a general practitioner in a small town. In 1978, Rachel unexpectedly decided to extend her training to 4 years, 1 year more than my three-year residency. Instead of a one-year position as an internist in the emergency department of a local hospital, I started a two-year fellowship in nephrology during which time I encountered many patients with IgA nephropathy. I worked with Robert J. Wyatt, a paediatric nephrologist at the University of Kentucky, who investigated complement's role in kidney disease. In the middle of my fellowship training, four nephrology faculty unexpectedly moved to the University of Alabama at Birmingham (UAB). I remained in Lexington rather than relocating to Birmingham to complete my fellowship because our first son arrived in May 1979. In 1980, after completion of my fellowship, Rachel and I explored multiple options for employment in several states. A new choice emerged unexpectedly when John H. Galla, who had been a mentor throughout my fellowship and had decided to join a new nephrology group in Lexington, changed his plans. After extensive discussions, he opted to reunite with former faculty colleagues in the Division of Nephrology at UAB. I took his place in the private-practice group. Over the next 2 years, I encountered many patients with IgA nephropathy. Through contact with Galla, I started to participate in clinical research with Jiri Mestecky at UAB, an expert on the immunobiology of IgA. In 1982, I biopsied a patient with IgA nephropathy who then informed me about her first cousin with the disease whom I had biopsied about 6 months earlier. Several other relatives had undefined kidney disease. Over the next year, I worked with Wyatt, Patricia A. Quiggins—a graduate student in anthropology, and Susan Y. Woodford—a research nurse in the Division of Nephrology at the University of Kentucky to document the familial relationships of the two first cousins with their relatives with kidney disease. We found that their pedigree included several more members with IgA nephropathy and others with clinical glomerulonephritis who had not undergone a kidney biopsy. In 1983, I attended the 1st International Symposium on IgA Nephropathy in Milan, Italy. I had purchased a ticket for ‘Lunch with the Experts’ and took the opportunity to show a hand-drawn pedigree to Jean Berger and Anthony R. Clarkson. Their enthusiasm for evidence of a genetic influence on IgA nephropathy was the driving force for me to devote more time to research. During the next few months, Wyatt, Quiggins, Woodford and I continued to define the relationships in the multiplex family. We identified 6 members with IgA nephropathy, 14 with clinical glomerulonephritis and 5 who had ‘chronic nephritis’ listed on their death certificates. In late 1983, Rachel, a board-certified psychiatrist, unexpectedly told me I was ‘crazy’ to continue research in a busy small private-practice group. Within a week, I called UAB Nephrology to ask about joining the faculty. In the summer of 1984, I sent a manuscript produced on a manual typewriter with the multiplex Kentucky pedigree to the New England Journal of Medicine for consideration for publication. One month later, Rachel and I joined the faculty at UAB. The paper with the pedigree was published in January 1985.1 During the next 2 years, we discovered that the expanded pedigree included 29 patients with IgA nephropathy.2 Remarkably, 48 of 80 patients born in eastern or central Kentucky had at least one relative with IgA nephropathy.2 During the ensuing 38 years, I collaborated with many gifted researchers and physicians at UAB, University of Tennessee-Memphis, and Columbia University, and in Canada, Czech Republic, France, United Kingdom, Italy, The Netherlands, China, Germany and Japan. The evidence of a genetically based influence in the development of IgA nephropathy in eastern Kentucky led to a multi-centre collaboration with Ali G. Gharavi that in 2000 showed linkage of the disease to chromosome 6q22-233; further investigations by Gharavi and Krzysztof Kiryluk with larger international cohorts over the next 23 years found 30 genetic loci associated with the risk for developing IgA nephropathy.4-6 These loci can have risk-associated or protective alleles. Twenty-four of the 30 loci reside outside of the major histocompatibility complex on chromosome 6 and are associated with various autoimmune diseases, the complement cascade, biosynthesis of galactose-deficient IgA1, cytokine–cytokine receptor signalling and innate and acquired immunity.5, 6 As the genetic risk increases, IgA nephropathy presents at progressively younger ages.6 Collaborative studies over several decades at UAB with Mestecky, Galla, Susan Jackson, Zina Moldoveanu, Michael W. Russell, F. Bryson Waldo, Milan Tomana, Cecil Czerkinsky, Warmold van den Wall Bake (visiting scientist), Karel Matousovic (visiting scientist), Hitoshi Suzuki (visiting scientist), Kazuo Takahashi (visiting scientist), Matthew B. Renfrow and Jan Novak have shown that galactose-deficient IgA1 plays a pivotal role in the development of IgA nephropathy. Patients with IgA nephropathy have elevated serum levels of IgA1 with less galactose in its hinge region than does serum IgA1 of healthy persons.8-10 IgA1 usually has three to six glycans (carbohydrate side chains which include galactose if fully glycosylated) per hinge region8; however, the distribution of the sites of attachment is not random. Thus, galactose-deficient IgA1 consists of multiple glycoforms. The serum level of galactose-deficient IgA1 measured by a lectin-based assay developed at UAB is above the 90th percentile for healthy individuals in about three-fourths of patients with IgA nephropathy.7 A study with Gharavi and colleagues found that biological relatives of patients with IgA nephropathy often had high serum levels of galactose-deficient IgA1, although they usually lacked any clinical or laboratory manifestation of kidney disease.11 A high level was more common in first-degree relatives (28%) than in second-degree relatives (17%). The level in a biological relative was more likely to be high if the index case with IgA nephropathy had a level ≥95th percentile for healthy persons. Among parents, an elevated level was more frequently found in one parent than in both parents or neither parent. In a study with Na Zhao and colleagues, the serum level of galactose-deficient IgA1 at the time of biopsy diagnosis was a prognostic biomarker for long-term kidney outcome.12 Circulating galactose-deficient IgA1 is primarily within immune complexes bound by antibody that recognizes the galactose-deficient hinge region glycans.13 Studies with H. Suzuki, J. Novak and William J. Placzek at UAB showed that IgG is the predominant antibody isotype that increases binding to galactose-deficient IgA1.14 The heavy chains of this IgG have a somatically determined mutation; serine rather than the usual alanine occupies position 97 in the complementarity 3 determining region.15 When the IgG antibody was produced as a recombinant protein with either serine 97 or alanine 97, the IgG with serine bound more galactose-deficient IgA1.15 In a study with Francois Berthoux and colleagues, the level of antibody specific for galactose-deficient IgA1 at the time of biopsy diagnosis predicted long-term kidney outcome.16 Serum levels of IgG autoantibody but not IgA autoantibody correlate with serum galactose-deficient IgA1 levels.14 Many patients with IgA nephropathy who have normal serum galactose-deficient IgA1 levels have high serum IgG autoantibody levels.17 In vitro experiments at UAB showed that IgA1-containing immune complexes from the sera of patients with IgA nephropathy stimulate primary human mesangial cells in culture to proliferate and secrete matrix proteins.18 In 2011, the UAB IgA Nephropathy Research Team proposed a four-hit hypothesis that IgA nephropathy is an autoimmune disease: circulating galactose-deficient IgA1 (autoantigen; Hit #1) is recognized by circulating antibody specific for galactose-deficient IgA1 (autoantibody; Hit #2) to form immune complexes (Hit #3) that accumulate in the glomeruli to induce kidney injury (Hit #4).19 The nephritogenic effects can be modified by genetically determined and environmental factors. Since its proposal, other researchers in the field worldwide have used this hypothesis as a blueprint for developing disease-targeted biomarkers19 and therapy.20, 21 The first published in vivo evidence for the injury-inducing nature of galactose-deficient IgA1-IgG complexes was shown in studies by Moldoveanu, Suzuki, Colin Reily, Lea Novak, Todd J. Green, Stacy Hall, J. Novak and others at UAB.22 Intravenous injection of immune complexes formed from galactose-deficient IgA1 and IgG autoantibody into immunodeficient mice induced the mesangioproliferative injury characteristic of human disease. The observation that IgG is not detected by routine immunofluorescence microscopy of kidney-biopsy tissue of many patients with IgA nephropathy raised questions about the applicability of the UAB four-hit hypothesis for the pathogenesis of IgA nephropathy. A study at UAB with Dana V. Rizk, L. Novak, J. Novak and others addressed this question.23 IgG was extracted from remnant kidney-biopsy tissues from 34 patients with IgA nephropathy, even from the 20 biopsies that had no IgG detected by routine immunofluorescence microscopy. This IgG bound galactose-deficient IgA1. After staining the kidney tissues with a nanobody specific for IgG-Fc, confocal microscopy detected IgG in all biopsies. The single-optical-plane images indicated that the glomerular IgA and IgG colocalized, suggesting that the two immunoglobulins were within immune complexes. These findings indicated that the amount of glomerular IgG in many patients with IgA nephropathy is less than the minimum that can be detected by reagents generally used for immunofluorescence examination of kidney biopsies. The results strengthened the hypothesis that IgG autoantibodies play a central role in the pathogenesis of IgA nephropathy. As research has unravelled the disease mechanisms of IgA nephropathy, clinical trials to evaluate various new approaches to treatment have emerged.20, 21 Hopefully, disease-targeted treatment(s) will be found to slow the decline in kidney function that shortens life expectancy of many patients.24 More importantly, the goal of discovery of a cure for IgA nephropathy has become more realistic. In 1987, at the 2nd International Symposium on IgA Nephropathy in Bari, Italy, Giuseppe D'Amico (Italy), Hideto Sakai (Japan), Clarkson (Australia) and I organized an informal group, the IgA Nephropathy Club. The purpose was to increase awareness of the clinical importance of IgA nephropathy and to encourage national and international collaborations in clinical and basic science research. We helped to organize international meetings about every 2 years. In 2000, the IgA Nephropathy Club recognized the need to rejuvenate its format. A more encompassing name was selected: The International IgA Nephropathy Network. New efforts included developing a steering committee with additional members from more research centres. A website was started to better disseminate scientific information and educational materials to physicians, investigators, students and the general public. I participated with an international group of nephrologists and pathologists to develop the 2009 Oxford MEST classification of the kidney pathology of IgA nephropathy.25, 26 Four light-microscopy features predict long-term kidney outcomes, independent of clinical or laboratory findings such as proteinuria.26 These features are mesangial hypercellularity, endocapillary hypercellularity (in the absence of prior immunosuppression therapy), segmental glomerulosclerosis and tubular atrophy/interstitial fibrosis. Those histology-outcome findings have been confirmed in subsequent studies. In 2017, the presence of glomerular crescents was added to the classification because of its prognostic value.27 In 2013 (Figure 1), I retired from the care of hospitalized patients but shortly thereafter resumed my participation in clinical research on various aspects of IgA nephropathy. I fully retired from outpatient clinical activities in November 2022. I continue to work at Reliant Glycosciences, LLC, a start-up biotechnology company formed in 2015 at the recommendation of the Dean of the UAB School of Medicine by Renfrow, Placzek, Rizk, J. Novak and myself. The company is focused on developing diagnostic and prognostic tools for patients with IgA nephropathy. We are nearing our goal to market assays to measure serum levels of galactose-deficient IgA1 and IgG autoantibodies specific for galactose-deficient IgA1. For 45 years, I have been very fortunate to have collaborated with so many extraordinarily talented researchers. In 1975, my vision of my anticipated professional career was very far off target. I encourage investigators, wherever they are in their career, to be attentive to new opportunities to expand, re-direct or re-invigorate their research interests; they often arrive in very unexpected circumstances. I thank the many colleagues and collaborators with whom I have worked during my career, the participants in the research studies, the study coordinators and nurses who have shared in the care of the participants in these studies, and the National Institutes of Health, the IgA Nephropathy Foundation and the pharmaceutical companies who have financially supported my research. National Institutes of Health grants DK078244, AI149431 and DK082753 and a gift from IGA Nephropathy Foundation. I am a co-founder and co-owner of Reliant Glycosciences, LLC. I am a co-inventor on US patent application 14/318082 (assigned to the UAB Research Foundation [UABRF] and licensed by UABRF to Reliant Glycosciences, LLC). Reliant Glycosciences, LLC had no role in the preparation of this manuscript.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.