Familial IgA nephropathy in southeastern Kentucky

Abstract

Background

Two decades ago, pedigrees of patients with IgA nephropathy (IgAN) from Pike County, KY, USA, provided evidence for a role of genetic factors in the pathogenesis of this disorder. Subsequently additional pedigrees were described for several communities from northern Italy. Recently, we found another cluster of patients in the Clay County, KY area, about 100 miles southwest of Pike County.

Aim

The purpose of this study was to evaluate and expand the pedigrees of patients with IgAN from Clay County, KY to provide additional insight into the mechanisms of inheritance of IgAN and assess the possible influence of a founder effect on the prevalence of IgAN in the region.

Method

Since 1980, most patients with IgAN and their relatives in eastern KY have provided personal genealogic data. These data were used to construct pedigrees that included the patients born in Clay County. Nine of 11 patients with IgAN born in Clay County, KY, USA were members of 1 or more of 5 pedigrees, each with 3 – 11 patients with IgAN.

Conclusion

Our findings suggest the possibility of a low-penetrance ancestral mutation in the IgAN kindreds from Clay County.

Introduction

In 1985, we described a pedigree with 6 individuals having IgAN and descending over 8 generations from a common ancestor who was one of the original settlers of Pike County, KY, USA [1]. Prior to that report, the evidence for genetic influences in IgAN was limited to HLA associations [2, 3] and the occurrence of IgAN in identical twins [4] and first-degree relatives [5, 6, 7, 8, 9]. Subsequently, we examined genealogic data for 80 patients with IgAN and identified a group of 29 patients descending from 28 pairs of ancestors [10]. The birthplaces of those patients, their parents, grandparents, and great-grandparents clustered in the extreme eastern portion of the state that included Pike County and adjacent counties to the north and west. Our hypothesis was that these occurrences could be attributed to a founder effect, whereby a gene(s) conveying susceptibility to IgAN was carried into the region by 1 or more of the early settlers.

In 1992, 3 pedigrees from the Valsavoire Valley in the Brescia district of northern Italy were described that appeared very similar to the Kentucky pedigrees [11]. Subsequently, several more pedigrees were discovered in the Valsavoire valley [12] and, more recently, additional kindreds from the Valtrompia valley just to the south of the Valsavoire valley have been described [13]. Using genome-wide analysis of 30 of these multiplex kindreds, (24 Italian and 6 US; 5 from Kentucky), linkage with chromosome 6q22-23 (IgAN1) was shown for 60% of the families [14]. Subsequently, two additional linkage studies have reported IgAN susceptibility loci distinct from IgAN1 [15, 16].

Recently, we found another cluster of patients in Clay County, KY, an area about 100 miles southwest of Pike County. In 2000, the population of Clay County was about 24,500 persons. Because only a few of the patients from Clay County had been included in our published pedigrees [10, 14], we initiated this study to extend the study of familial IgAN to patients from Clay County. Our goal was to provide additional insight into the possible mode of the inheritance of IgAN and to assess whether a founder effect was operational for this cluster of patients.

Methods

Since 1980, most patients from eastern Kentucky with IgAN have participated in studies investigating the pathogenesis and mechanism of inheritance of IgAN. For these studies, patients and their relatives were asked to provide personal genealogic data. Starting in 2001, a National Institutes of Health-supported grant (DK061525) enrolled all living patients with IgAN from Clay County. The protocol for this study was approved by the Institutional Review Boards of the University of Tennessee Health Science Center and the University of Alabama at Birmingham. Institutional review boards approved previous studies at one or both of these institutions or at the University of Kentucky Medical Center, for work done prior to 1984.

A brief history and physical examination were performed and blood and urine samples were collected on one occasion from patients and their family members for determination of serum creatinine concentration, urinalysis and random urinary protein/creatinine (U P/C) ratio. Microscopic hematuria was defined as ≥ 1+ (small) blood on dipstick (Multistix, Bayer) and pathologic proteinuria was denoted by a U P/C ratio > 0.20. End-stage renal disease (ESRD) was defined as Stage 5 CKD with estimated glomerular filtration rate (GFR) less than 15 ml/min/1.73 m² by the MDRD equation, corrected for race and gender [17] or the need for dialytic therapy or renal transplantation. The diagnosis of IgAN required a renal biopsy with IgA as the dominant immunoglobulin in a typical mesangial distribution and the absence of clinical and laboratory evidence for systemic disease [18]. Henoch-Schönlein purpura nephritis (HSPN) was defined by typical clinical features and immunohistological findings indistinguishable from those of IgAN [19]. These clinical laboratory findings and the features of renal biopsy were used to assign clinical phenotype: 1) biopsy-proven IgAN, 2) HSPN, with or without biopsy, 3) ESRD without definitive diagnosis, and 4) clinical nephritis (history of macroscopic hematuria, isolated microscopic hematuria, isolated proteinuria, or hematuria and proteinuria, but no renal biopsy for confirmation).

Genealogic investigation was performed by review of published transcripts of the Clay County census dating from 1850 to 1930. Matching of ancestors was determined by several factors, including birth and death dates, children listed in the census and locality of residence within the county. Spousal confirmation was ascertained using census information (when listed) and a transcribed index of marriage records and permission notes for marriage called “Clay County’s Missing Marriage Permission Slips” published by the Clay County Genealogical and Historical Society. Data from these investigations and previously collected pedigrees of patients diagnosed with IgAN or HSP were entered into a Progeny 5 database (Progeny Software, LLC, South Bend, IN, USA). Genealogic data were entered into the pedigree database and compiled according to the most distant relative known at the beginning of the study. After all known family members from individual pedigrees were compiled, the subsequent step was to identify common ancestors. The names of these ancestors were confirmed by examining all available information, including birth/death dates, marriage history and offspring, in the published transcripts of the Clay County Census dating from each decade between 1850 and 1930. Expansion of pedigrees was accomplished by methodical examination of surnames in the aforementioned census transcripts in consideration of the same data that confirmed the common ancestors. In any instance in which a relative had undergone kidney transplantation, we diligently attempted to confirm the nature of the underlying kidney disease. However, this information was not always available, and some patients were simply designated as having undergone transplantation without a specific diagnosis.

Results

Eleven patients with IgAN were born in Clay County and were members of 5 pedigrees. Clinical and demographic data for these 11 IgAN patients are shown in Table 1. The surnames of the founding ancestors for all 5 pedigrees indicate that they originated from the British Isles. No individual of African or Italian ancestry was introduced by marriage into the pedigrees. The clinical spectrum of disease for these patients was quite variable, with 6 having progressed to ESRD.

Table 1.

Demographic and clinical data for patients with IgA nephropathy who were born in Clay County, KY, USA.

Patient	Gender	Year of diagnosis	Age at diagnosis	Age at ESRD or last follow-up	Renal function at last follow-up*	Proteinuria last follow-up (Urine protein to creatinine ratio)	Pedigree(s)
C1	M	1993	39.7	40.1	ESRD		1, 2, 3
C3	F	1983	17.8	39.1	99	0.09	1, 2, 5
C4	M	1986	44.3	50.9	ESRD		1, 2
C5	F	1977	11.6	38.6	85	0.09	1
C16	M	1983	16.2	31.8	ESRD		2, 6
C17	F	1997	32.4	40.8	38	19.3	2, 4, 5
C18	M	1992	50.5	50.5	ESRD		3
C19	M	1988	23.1	27.7	ESRD		3
C22	F	1996	38.7	47.6	43	1.89	4
C24	M	1988	39.5	39.5	ESRD		none, related to A4
C25	M	2000	16.8	22.2	67	2.35	none

^*Estimated GFR, ml/min/1.73 m².

Pedigree 1 contains the most patients from Clay County and contiguous counties; 6 had IgAN and 2 had HSPN. Four of those with IgAN and 1 with HSPN were born in Clay County. The founders of the pedigree had resided in the extreme southeastern corner of the county. Of the patients not born in Clay County, 1 with IgAN (A1) was born in a county adjacent to Clay but had the same surname as the founder of the pedigree, the father of another patient with IgAN (D2) was born in Clay County, and the mother of another patient with adult-onset biopsy-documented HSPN (D1) was born in Clay County. This pedigree also included 2 individuals born in Clay County who had progressed to ESRD without a biopsy diagnosis: A2 was presumed to have diabetic nephropathy and C2 had IgAN documented by biopsy in a renal allograft. Subject C5 had 5 siblings, 2 nieces and 2 nephews with clinical evidence of glomerulonephritis and another niece with childhood HSPN who did not undergo renal biopsy. Four of these affected siblings had microscopic hematuria or a history of macroscopic hematuria in 1999. C10 was described as unaffected in our earlier publication [14] but later exhibited three episodes of macroscopic hematuria. When screened again in 2004, subject C5 and her affected siblings, including C10, all had normal urinalyses. Four individuals shown in this portion of the pedigree (U1, U2, U3, U4) were normal when screened in 2004. The portion of the pedigree that included C5 and her first- and second-degree relatives showed linkage to the IgAN1 locus [14]. A larger portion of Pedigree 1, having three patients with IgAN (C3, C4, D2), was included in Group 2 of a previous study [10].

Pedigrees 1 – 5.

Square symbols denote males and round symbols, females. A line through the symbol indicates that the person is deceased. Solid black symbols indicate that the person has biopsy-proven IgAN. Solid light-blue symbols indicate that the person had ESRD without a native-kidney biopsy for diagnosis. Checked black symbols indicate that the person had HSPN. Red half-filled symbols indicate that the person had clinical glomerulonephritis (history of macroscopic hematuria, microscopic hematuria, and/or proteinuria but no renal biopsy for confirmation). Subjects designated by C (e.g., C1) were born in Clay County, those designated by A were born in counties adjacent to Clay County, and those designated by D were born in more distant counties. Subjects designated by U in Pedigree 1 were classified as unaffected. Unaffected relatives are not shown except in the portion of Pedigree 1 that was previously reported [14].

Pedigree 2 contains 11 patients with IgAN, 5 of whom were born in Clay County. Two of these patients (A1, A3) were born in adjacent counties and the mother of another (A3) was born in Clay County. One patient (D6) was born in central Kentucky more than 100 miles from Clay County and underwent renal biopsy in Birmingham, AL, USA, about 350 miles distant. The members of this pedigree descend from the founder over a span of 6 – 9 generations. This pedigree contains the largest number of patients with IgAN known to descend from a single ancestral pair. One portion of this pedigree, with 3 patients with IgAN (C16, D4, D5), was included in Group 1 of a previous study [10].

Pedigree 3 includes 3 patients with IgAN who were born in Clay County. A fourth patient with IgAN was born in an adjacent county, but both of her parents were born in Clay County. The pedigree also contains 2 patients with ESRD but without a biopsy-proven diagnosis (C2, C20). Another individual (C21) had clinical evidence of glomerulonephritis. This pedigree has not been previously reported. Three members of Pedigree 3 with IgAN were not members of Pedigrees 1 or 2.

Pedigree 4 contains only 3 patients. 2 are first cousins born in Clay County, while the third was born in an adjacent county, but the mother was born in Clay County. This pedigree has not been previously reported. C17 was also a member of Pedigree 2. The 2 first cousins were included in one of the pedigrees in the previous linkage study [14].

Pedigree 5 has 4 patients with IgAN and 1 patient with HSPN. Two of the patients with IgAN were born in Clay County. One with IgAN and another with HSPN were born in adjacent counties, with the IgAN patient’s mother having been born in Clay County. This pedigree has not been previously reported. All of the patients in this pedigree, except for A6, were members of previous pedigrees.

Several patients are members of more than one pedigree. Patient C1 is in three pedigrees (1, 2, 3); C3, in 3 (1, 2, 5); C16, in 3 (2, 4, 5); D3 in 2 (2, 5); C4, in 2 (1, 2); A3, in 2 (2, 5); and A1, in 2 (1, 2).

Urine screening consisting of urinalysis and random U P/C ratio was performed for 36 first-degree relatives of the subjects with IgAN or HSPN included in the 5 pedigrees. The only case of clinical glomerulonephritis detected by screening was C23, the 51-year-old brother of C22, who a U P/C ratio of 0.38 and an estimated GFR of 61 ml/min/1.73 m², but dipstick analysis negative for blood.

Discussion

Most members of the early Kentucky pedigree that we described in 1985 [1] were from Pike County, located on the border with West Virginia. The founder of this pedigree settled in Pike County in the 1790’s. Our extensive analysis of genealogic data for patients with IgAN from eastern and central KY showed that the ancestors of patients from many pedigrees clustered in Pike and surrounding counties [10]. We concluded that the patients descended from only a few ancestors, supportive of the hypothesis that the eastern Kentucky patients with IgAN shared a common gene pool that confers susceptibility to the development of the disease. Based upon these observations, we suggested that an individual could carry the trait(s) for susceptibility to IgAN despite the absence of clinically apparent disease and that the geographic clustering of the ancestors was due to a founder effect.

The first reported Italian pedigree contained 7 patients with IgAN [11]. Six patients were from the village of Cevo, population 1,200, in Valsaviore, a small, isolated valley in the Brescia district of northern Italy. The patients descended over 7 – 9 generations from the founders who lived in the region in 1690. Two additional pedigrees from the neighboring villages of Savoire and Valle contained 2 and 3 patients with IgAN. All three pedigrees also included numerous individuals with clinical glomerulonephritis. However, in contrast to our experience with fewer patients from Clay County, there was a marked male predominance.

A subsequent report described additional pedigrees from the Brescia region, such that 26 patients with IgAN belonged to 10 pedigrees [20]. However, among the 159 patients with IgAN from the region, there was no indication that any patient with IgAN was known to be related to another patient. More recently, 3 additional Italian pedigrees were reported from northern Valtrompia. Urine samples were screened for 5,642 subjects from 3 villages, representing 91% of the inhabitants. The largest pedigree from San Columbano had 6 patients with biopsy-proven IgAN who descended from the founding ancestors over 11 – 13 generations since 1570. Pedigrees from Collio and Bovegno contained 2 and 4 patients with IgAN, respectively. All 3 pedigrees contained additional patients with mesangial proliferative glomerulonephritis or clinical nephritis. Thus, a wide spectrum of clinical features was observed in these pedigrees.

Pedigree 1 of our current report dates to 1750. The founders of this pedigree moved into the Clay County region from Virginia. This was a very common sequence of settlement, with each new area further west than the previous area. One custom in this early era of settlement in the region was the relatively common change in county designations. This practice renders assignment of the ancestry of persons to a modern-day county more challenging. In the course of developing all of the pedigrees, any ambiguity as to the individual’s birth county resulted in the county being listed as “unknown”. No assumptions were made in this regard.

Until recent years, the population of Pike and Clay Counties could be considered to be a genetic isolate. Most of the residents had been born there and only a few individuals moved into the county from the outside. While this pattern of settlement has clearly changed over the past three decades, for this and our previous reports, the ancestors of virtually all of the patients identified with IgAN and HSPN had lived in the region for many generations.

Excluding the 2 patients not fitting into any of the pedigrees in this report, 5 of 9 Clay County patients have reached ESRD. This feature would seem to support the opinion that familial IgAN is associated with an increased risk of ESRD [20], although this view was recently challenged in a larger study of Italian patients [22]. In any event, the high frequency of progressive disease in the Clay County cohort may represent a selection bias, as most of the patients were diagnosed either after presenting at, or near, ESRD, or in childhood with normal renal function. Thus, there appears to be an under-representation of adults presenting with normal or mildly impaired renal function in the KY cohort.

At least a third of patients with IgAN diagnosed by biopsy in childhood will eventually have a normal urinalysis [23]. Adults identified with IgAN may also clinically improve over time to exhibit a normal urinalysis, as we found for 17% of 153 subjects enrolled in one of our recent studies [24]. This feature underscores the importance of repeated screening over time for a disease such as IgAN that may spontaneously remit and relapse. Correct assignment of affectation status (clinical nephritis, unaffected, unknown) for the purpose of conducting molecular genetic studies such as linkage analysis will be enhanced by repeat screenings.

One should not assume that our documentation of familial disease in eastern KY implies a high frequency of IgAN in that region. In fact, for the period 1990 – 1994, the incidence was 12.2 cases per million population per year (MPPY) outside of Fayette County (Lexington) as compared to 13.2 cases per MPPY in Fayette County where no case of familial disease had been documented [25]. In fact, in Clay County the detection of 11 cases of IgAN over a 23-year span represents an incidence of about 20 cases per MPPY.

The structure of the eastern Kentucky pedigrees is similar to that of pedigrees from Iceland or Sardinia which have proven to be very powerful for identification of genes for complex traits. In population isolates that are of recent origin, linkage disequilibrium (i.e., the length of the shared chromosomal segments) surrounding the susceptibility alleles is expected to be stronger, thereby reducing the requirements for sample size and marker density while simultaneously increasing the chances of detecting genes by association approaches. Hence, these populations have facilitated identification of susceptibility genes for diabetes, high uric acid levels or eye pigmentation [26, 27, 28].

In conclusion, the data presented in this report suggest the possibility of a low-penetrance ancestral susceptibility allele in the IgAN kindreds from Clay County, KY. This hypothesis can now be tested by genome-wide association studies, which have recently become feasible and can identify chromosomal segments shared by descent among affected individuals.