Lipoprotein glomerulopathy

INTRODUCTION — Lipoprotein glomerulopathy (LPG) is a unique and rare disorder of renal lipidosis that was first reported in a Japanese patient in 1989 [1]. It is characterized by lipoprotein thrombi in glomeruli, an abnormal plasma lipoprotein profile that resembles type III hyperlipoproteinemia, and a marked increase in serum apolipoprotein E (apo E) concentrations. LPG was subsequently found to be due to mutations of the APOE gene [2]. However, intrinsic renal factors are also involved in the pathogenesis of LPG because of the incomplete penetrance based upon many asymptomatic carriers of APOE variants [3-5].

A review of the clinical features of LPG is presented in this topic review.

PATHOGENESIS — Initial reports of familial cases as well as recurrence in transplanted kidneys suggested that a systemically acting genetic factor may underlie lipoprotein glomerulopathy (LPG). It was also observed that the serum apolipoprotein E (apo E) isoform was inconsistent with the APOE genotype among affected individuals.

These findings predicted that an uncharacterized variant(s) of the APOE gene may be responsible for the pathogenesis of this disorder [6]. This was subsequently confirmed as mutations in the APOE gene have been found in all reported cases of LPG in which DNA sequencing has been performed (algorithm 1).

APOE variants — Apo E, an essential apolipoprotein, is a major constituent of various plasma lipoproteins. Plasma lipoprotein levels are determined in part by the binding activities of apo E to the low-density lipoprotein (LDL) receptor, LDL receptor-related protein, and very-low-density lipoprotein (VLDL) receptor. Apo E is composed of 299 amino acids and has a relative molecular mass of 34 kDa.

A number of serum isoforms of apo E have been described. They differ based upon their migration properties as determined by isoelectric focusing polyacrylamide gel electrophoresis (IEF). The most common serum apo E isoform is apo E3.

Observations in patients with type III hyperlipoproteinemia provided insight into the pathogenesis of LPG. Type III hyperlipoproteinemia is a multifactorial disorder that is characterized by the presence of two apo E2 alleles. It is also occasionally associated with renal lipidosis showing sclerosis and foam cells in the glomerulus [7-9].

Among affected individuals with LPG (which is directly associated with renal lipidosis), the serum apo E isoform was noted to be inconsistent with the APOE genotype. This suggested that an uncharacterized variant(s) of the APOE gene may underlie the pathogenesis of this disorder [6].

Subsequently, mutations of the APOE gene have been described in all patients with LPG in whom DNA sequencing has been performed [2,3,10], except for the cases in one report from China [11]:

●In most cases from the eastern part of Japan, DNA sequence analysis of the APOE gene reveals a nucleotide substitution of G to C at codon 145 [6]. This missense mutation results in an amino acid substitution of the proline residue for arginine residue at position 145 of apo E and was termed apo E Sendai (figure 1).

●Another major mutation called apo E Kyoto is found worldwide in various ethnicities, including Japanese, Chinese, and European Americans [12-17]. This mutation changes the amino acid at position 25 of apo E from arginine to cysteine. This position is not directly related to the central portion (position 136 to 158) of apo E, which is involved in binding to the LDL receptor. However, the removal of arginine at 25 could break the tertiary structure of apo E, as could the change at 145 in apo E Sendai [12].

●Several other apo E variants have been found in LPG patients (figure 1). Some are missense mutations [13,16,18-24], while others are in-frame deletions [25-27]. As a whole, 18 different causative APOE mutations associated with LPG have been reported [23,24,28,29].

Most mutations detected in patients with LPG are found in and/or around the LDL receptor binding domain (figure 1). In particular, the substitutions of proline (eg, apo E Sendai [6], apo E Chicago [30], apo E Guangzhou [20], and apo E Osaka/Kurashiki [31,32]) are important mutations because they may produce severe structural changes in the middle of the alpha-helix, thereby altering the three-dimensional conformation of the protein [33]. The incompatibility of proline in helical regions may induce LPG by thermodynamic destabilization, hydrophobic surface exposure, and aggregation of apo E [34]. On the other hand, the nonproline-substituted apo E mutants with LPG, including apo E Kyoto, may also contribute to protein aggregation in glomerular capillaries [35]. Thus, it is hypothesized that the abnormal lipoproteins aggregate and then deposit in the glomerular capillary walls and/or around the mesangium, with subsequent transformation into lipoprotein thrombi.

In vitro studies and murine models of LPG support the view that apo E variants are directly responsible for these glomerular lesions [10]:

●LPG results after transduction of the apo E Sendai variant into mice with genetic deletion of the wild type APOE gene [36,37].

●In vitro studies reveal that the apo E Sendai variant lipoprotein has reduced affinity for the LDL receptor [38]. This decreased binding ability is shared with the apo E2 allele found in patients with familial type III hyperlipoproteinemia [39-41], which is also sometimes associated with renal lipidosis [7-9].

Role of other factors — Despite the evidence in favor of a causative role for apo E variants in LPG, several observations suggest that additional factors are involved in the induction of LPG. As examples:

●There are asymptomatic carriers of apo E variants in families with LPG.

●LPG-like lesions do not develop in all apo E Sendai transfected mice.

●LPG-like lesions have been observed in mice with hyperlipidemia and genetic deletion of the APOE genes [42,43].

●Lesions are confined to the glomeruli of LPG, suggesting that unique intrinsic glomerular factors found in the affected host interact with apo E variants.

●Cases of LPG sometimes occur in association with other glomerular diseases (eg, immunoglobulin A [IgA] nephropathy, membranous glomerulopathy, lupus nephritis [2,3]).

Macrophage impairment represented by Fc receptors (FcRs) may be one such non-apo E factor. FcRs mediate the uptake of immune complex- or C-reactive protein (CRP)-binding LDL into macrophages, thereby inducing phagocytosis and atherosclerosis with foam cells [44-46]. In mice, FcR dysfunction can be induced by eliminating the FcR gamma chain, which is required to allow the surface expression and signal transduction of the receptor. In FcR gamma chain (FcR gamma) knockout mice, LPG-like lesions characterized by lipoprotein thrombi are found in glomeruli [47]. The lack of scavenger receptors induced by FcR dysfunction may suppress the uptake of lipids by macrophages and contribute to form lipoprotein thrombi [48]. In addition, several studies in mice show the interaction between abnormal macrophage function and apo E in the pathogenesis of LPG [49,50]. In one study, LPG-like changes are generated by injecting various apo E vectors in APOE and FcR gamma double knockout mice [49].

EPIDEMIOLOGY — Since the first published case in 1989, more than 270 cases of lipoprotein glomerulopathy (LPG) have been reported worldwide, and approximately 200 cases have been assayed by DNA sequencing (figure 2) [51-53]. Most cases have been found in Japan and China [2,51]. However, cases outside of Asia, such as the United States and Europe, have been identified [13,21,22,30,54-61]. In the seven cases from the United States, four are of European ancestry [13,21,57], two are of Asian ancestry [54,58], and one is Mexican [30]. Three Italian and two French cases are White individuals [22,55,56]. Cases from Russia, Brazil, and Canada have also been reported [53,62,63].

CLINICAL FEATURES — Most patients with lipoprotein glomerulopathy (LPG) are asymptomatic. The average age is 32 years old, although the age at onset varies widely (ranging from 4 to 69 years [2]). The male to female ratio is approximately 3:2.

Affected patients usually come to medical attention after the detection of proteinuria on routine screening examination. Some patients present with symptoms and signs of the nephrotic syndrome.

Inheritance of LPG appears to be autosomal dominant with incomplete penetrance [3-5]. Apolipoprotein E (apo E) Sendai [4] and apo E Kyoto [5] may have descended from a single founder.

Kidney findings — At presentation, all patients with LPG have proteinuria of variable severity, ranging from mild to severe. Most eventually develop nephrotic syndrome or nephrotic-range proteinuria. The average urinary protein excretion of reported patients at time of kidney biopsy is 4.8 grams per day (range of 0.3 to 18 grams/day). Microscopic hematuria is usually not observed.

Average creatinine clearance of patients is 90 mL/min per 1.73 m² (range of 35 to 143 mL/min per 1.73 m²) at the time of kidney biopsy. One-half of patients eventually develop end-stage kidney disease (ESKD) at 1 to 27 years after disease onset [2]. On average, this occurs within approximately 7.5 years from time of diagnosis, as calculated by Kaplan-Meier method.

In several cases, LPG has been associated with other glomerulopathies. These include IgA nephropathy, membranous nephropathy, and lupus nephritis [3].

Non-kidney manifestations — Characteristic non-kidney symptoms appear to be limited:

●In most cases, systemic manifestations, such as hypertension, atherosclerosis, cardiovascular change, and hepatic dysfunction, are mild. Among those with significant proteinuria, edema and decreased albumin levels are found.

●Clinical symptoms characteristic of lipidosis, such as corneal arcus, xanthoma, and Achilles tendon thickening, are rarely observed.

●Cases with psoriasis vulgaris have also been reported [64].

Despite comparatively high serum triglyceride levels, no cases of pancreatitis have been reported.

Lipid and lipoprotein profiles — Most patients with LPG have hyperlipidemia with a predominance of triglycerides. When isolated by ultracentrifugation, high cholesterol levels are observed in very-low-density lipoprotein (VLDL) and intermediate-density lipoprotein (IDL) fractions [65]. Agarose gel electrophoresis for lipoprotein shows a broad beta band. These findings are consistent with type III hyperlipoproteinemia [66,67]. (See "Hypertriglyceridemia in adults: Management".)

However, the increased levels of triglycerides and cholesterol in plasma are frequently less pronounced than patients with familial type III hyperlipoproteinemia and are even occasionally within normal ranges [25,68]. Among reported patients, average plasma triglyceride and total cholesterol levels are 313 mg/dL (range of 74 to 1019 mg/dL) and 272 mg/dL (range of 107 to 720 mg/dL), respectively [2].

A high serum concentration of apo E is another characteristic finding in LPG. The average plasma apo E level is 17.1 mg/dL (range of 3.9 to 71.0 mg/dL), which is approximately twofold higher than the upper normal range [2].

In most cases, the APOE phenotype determined by isoelectric focusing polyacrylamide gel electrophoresis (IEF) shows heterozygosity for the E2 allele (E2/3 or E2/4) [3]; this differs from that of familial type III hyperlipoproteinemia, which is homozygous for E2. In some cases, however, the protein band by IEF is recognized at the position of E1 [25,26] or between E2 and E3 [14,56].

On the other hand, the APOE genotype routinely determined by restriction fragment length polymorphism analysis (RFLP) with restriction enzyme HhaI shows E3/3 or E3/4, which is inconsistent with the APOE phenotype as determined by IEF [6].

Histopathologic findings — Findings on kidney biopsy include the following.

Light microscopy — Light microscopic examination of kidney biopsy material shows marked dilatation of the capillary lumen in the glomeruli by a pale-stained substance (picture 1) [1]. Foam cells, which are characteristic of lipidosis, are rarely seen in either the glomeruli or the interstitium.

Electron microscopy — Electron microscopy shows that the thrombus-like substances in the glomerular capillaries are composed of granules and vacuoles of various sizes forming concentric lamellate like a fingerprint (picture 2) [3,69].

Confirmation of lipoprotein deposits — Using snap-frozen kidney sections, immunofluorescence studies reveal deposition of apolipoprotein B (apo B) and apo E (picture 3), and Sudan or oil red O staining reveals lipid droplets in the thrombus-like substances (picture 4) [2].

EVALUATION — Lipoprotein glomerulopathy (LPG) should be suspected in the patient who presents with proteinuria and dyslipidemia with a predominance of triglycerides. Initial evaluation includes lipoprotein and apolipoprotein E (apo E) tests and agarose gel electrophoresis. Findings consistent with LPG include increased plasma very-low-density lipoprotein (VLDL) and intermediate-density lipoprotein (IDL) levels, a broad beta-band on agarose gel electrophoresis, and a high serum apo E level.

If this initial evaluation is consistent with LPG, we always obtain a kidney biopsy and perform analysis of APOE phenotype and genotype. DNA sequence analysis is useful for the confirmation of diagnosis.

DIAGNOSIS — The clinical elements most consistent with lipoprotein glomerulopathy (LPG) include the following:

●Proteinuria.

●Type III hyperlipoproteinemia with high levels of very-low-density lipoprotein (VLDL) and intermediate-density lipoprotein (IDL) plus increased apolipoprotein E (apo E) concentration [65-67]. This is usually associated with a heterozygous APOE phenotype, E2/3 or E2/4, by immunoelectrophoresis but sometimes with an uncommon type (eg, E1/3 or others).

●Kidney biopsy showing, on light microscopy, dilatation of glomerular capillary lumina with pale-stained substances and, on electron microscopy, stone- or sand-like granules occupying the capillary lumina.

To establish the diagnosis, DNA sequence analysis of APOE should be conducted, if possible. APOE variants have been detected in all LPG patients in whom DNA sequence was determined. At present, DNA sequence analysis of APOE can be assayed in various commercial laboratories using blood samples.

DIFFERENTIAL DIAGNOSIS — A significant number of primary kidney diseases also present with significant proteinuria and kidney dysfunction. Although analysis of lipoproteins and apolipoproteins may distinguish lipoprotein glomerulopathy (LPG) from these diseases clinically, the diagnosis is definitively obtained by kidney biopsy. (See "Glomerular disease: Evaluation and differential diagnosis in adults".)

LPG is associated with segmental sclerosis of the glomerulus and mesangial interposition, findings also observed in focal segmental glomerulosclerosis (FSGS) and membranoproliferative glomerulonephritis, respectively. These disorders can be distinguished from LPG after DNA sequencing.

In several reports, familial type III hyperlipoproteinemia due to APOE2 homozygosity (apolipoprotein E [apo E]2 homozygote glomerulopathy) has been associated with glomerulosclerosis, including foam cells [7-9,70], and occasionally with LPG-like lesions such as lipoprotein thrombi [71,72]. One case, with the combination of APOE2 and APOE2-Tokyo/Maebashi genes, showed both foam cells and lipoprotein thrombi, providing the relationship between histologic characteristics and APOE gene abnormalities [73].

Moreover, a new form of apo E-related glomerular disease, which is different from LPG and apo E2 homozygote glomerulopathy, has been reported [74-77]. This disease is similar to membranous nephropathy but has no immunologic findings. All patients have a novel variant apo E Toyonaka (Ser197Cys) associated with APOE2 homozygosity [74-77]. Since apo E Toyonaka is a point mutation in the hinge region between N-terminal and C-terminal domains and is involved in the dysfunction of the C-terminal domain that connects with lipids, it is suggested that altered apo E molecules without lipids may accumulate in the glomerulus.

As mentioned above, glomerular lesions due to various APOE mutations have been reported in addition to LPG, and these have been categorized as APOE-related glomerular disorders [78].

THERAPY — There are no high-quality data that have established an effective therapeutic regimen for lipoprotein glomerulopathy (LPG). However, severe hypertriglyceridemia aggravates the features of LPG in both human cases and animal models. Accordingly, preventing hypertriglyceridemia may be an important aspect of treatment.

Specific therapies — Lipid-lowering therapy consisting of different agents, either administered alone or in combination, have been examined:

●In some case reports, intensive therapy using lipid-lowering agents including fibrates resulted in clinical remission with histologic resolution [19,79-81]. These cases reported reductions in serum cholesterol, triglyceride, and apolipoprotein E (apo E) levels and complete disappearance of lipoprotein thrombi in serial kidney biopsies.

●In one controlled study, the three-year patient and kidney survival rates were significantly higher in the fibrate-administered group than in control [5].

●The effects of hydroxymethylglutaryl (HMG)-CoA reductase inhibitors (eg, statins) are controversial. Statins do not seem to lower urinary protein excretion, even in those with remission of dyslipidemia [2,3].

The effects of plasmapheresis or low-density lipoprotein (LDL) apheresis are uncertain [3], but beneficial treatment with staphylococcal protein A immunoadsorption was reported as a pilot study from China [82]. Protein A binds to the Fc receptor gamma chain (FcR gamma) of immunoglobulin G (IgG) [83]. As noted above, deficiency of FcR gamma produces LPG-like lesions in mice independently of the apo E abnormality [47]. Actually, the lack of LDL receptors [49] and scavenger receptors [48] due to FcR gamma deficiency may suppress the uptake of lipoproteins and contribute to lipoprotein thrombi in the glomerulus. Thus, protein A immunoadsorption may be effective because dysfunction of FcR gamma may be important in the development of human LPG.

Another report showed that LDL apheresis using a heparin-induced extracorporeal lipoprotein precipitation system was successful because of lowering LDL cholesterol and triglycerides [84].

The administration of glucocorticoids, immunosuppressant agents, and/or anticoagulants are ineffective [2,3].

Despite the absence of published evidence, we suggest the use of lipid-lowering therapy. Our initial regimen consists of the administration of lipid-lowering agents, including a fibrate (eg, fenofibrate, bezafibrate, and pemafibrate). We subsequently monitor serum lipid levels and serum creatinine concentrations as well as urinary protein excretion. We aim for normalization of lipid levels, reduction of urinary protein excretion, and stabilization of the serum creatinine concentration.

If initial therapy is ineffective in sufficiently lowering proteinuria to goal levels by six months, we add angiotensin-converting enzyme (ACE) inhibitors or angiotensin II receptor blockers (ARBs). (See 'Nonspecific therapy' below.)

Nonspecific therapy — There is no evidence that ACE inhibitors and/or ARBs are effective in LPG. However, based upon their effectiveness in other forms of proteinuric chronic kidney disease, we recommend the administration of ACE inhibitors or ARBs both for blood pressure control and to slow progression of the kidney disease among those with persistent proteinuria (which is defined as inadequate lowering of proteinuria to goal levels by six months from initiation of specific therapy).

The treatment goals with angiotensin inhibition are the same as those in other forms of proteinuric chronic kidney disease as described in the Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines (see "Antihypertensive therapy and progression of nondiabetic chronic kidney disease in adults"):

●We recommend a minimum reduction in protein excretion of at least 60 percent from baseline values, with goal protein excretion being less than 500 to 1000 mg/day. Given simplicity and cost considerations, we initiate therapy with a single agent (either an ACE inhibitor or an ARB).

●The goal blood pressure is the same as it is in other patients with proteinuric chronic kidney disease. (See "Antihypertensive therapy and progression of nondiabetic chronic kidney disease in adults".)

Transplantation — To date, six kidney transplantations have been performed, and LPG has recurred in five transplanted allografts [15,56,68,85,86]:

●In one case report, kidney function was lost within one year [68].

●Serum creatinine levels reached approximately 2 mg/dL (176.8 micromol/liter) within four years in two cases [15,56].

●One case revealed histologic recurrence one year after surgery [85].

●In one case, the recipient's kidney failure was erroneously attributed to membranoproliferative glomerulonephritis, but LPG was correctly diagnosed in a protocol kidney allograft biopsy [86].

●One case revealed recurrence of LPG after 20 years follow-up [87].

Dialysis — There is no special selection between hemodialysis and peritoneal dialysis. However, hemodialysis appears to ameliorate the dyslipidemia in most cases.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Glomerular disease in adults".)

SUMMARY AND RECOMMENDATIONS

●Epidemiology – Lipoprotein glomerulopathy (LPG) is a rare and unique disorder of renal lipidosis. More than 270 cases of LPG have been reported worldwide, with most cases being found in Japan and China. However, cases outside of Asia, such as the United States and Europe, have been identified. (See 'Introduction' above and 'Epidemiology' above.)

●Pathogenesis – LPG results from mutations in the APOE gene, with most mutations detected being found in and/or around the low-density lipoprotein (LDL) receptor binding domain. Although unproven, it is hypothesized that the abnormal lipoproteins aggregate and then deposit in the glomerular capillary walls and/or around the mesangium, with subsequent transformation into lipoprotein thrombi. (See 'Pathogenesis' above.)

●Clinical features – At presentation, most patients are asymptomatic and come to medical attention after the detection of proteinuria on routine screening examination. All patients with LPG have proteinuria of variable severity, ranging from mild to severe proteinuria. Most eventually develop nephrotic syndrome or nephrotic-range proteinuria. One-half of patients eventually develop end-stage kidney disease (ESKD). (See 'Kidney findings' above.)

Most patients also have hyperlipidemia with a predominance of triglycerides. Findings with ultracentrifugation and agarose gel electrophoresis are consistent with type III hyperlipoproteinemia. (See 'Lipid and lipoprotein profiles' above.)

●Pathology – Kidney biopsy reveals marked dilatation of the capillary lumen in the glomeruli by a pale-stained substance on light microscopy and, on electron microscopy, thrombus-like substances in the glomerular capillaries that are composed of granules and vacuoles of various sizes. These substances form concentric lamellate, like a fingerprint. (See 'Histopathologic findings' above.)

●Diagnosis – LPG is diagnosed in the patient with proteinuria, type III hyperlipoproteinemia with high levels of very-low-density lipoprotein (VLDL) and intermediate-density lipoprotein (IDL) plus increased apolipoprotein E (apo E) concentration and characteristic findings on kidney biopsy. The diagnosis should be confirmed, if possible, with DNA sequence analysis of the APOE gene. (See 'Diagnosis' above.)

●Treatment – No effective therapeutic regimen has been established for LPG. We suggest the use of lipid-lowering therapy including a fibrate (Grade 2C). We subsequently monitor serum lipid levels and creatinine concentrations, as well as urinary protein excretion. We aim for normalization of lipid levels, reduction of urinary protein excretion, and stabilization of serum creatinine concentration. (See 'Specific therapies' above.)

For patients with persistent proteinuria (>500 to 1000 mg/day), we recommend angiotensin inhibition with an angiotensin-converting enzyme (ACE) inhibitor or angiotensin II receptor blocker (ARB) (Grade 1A). We initiate monotherapy and target a minimum reduction in protein excretion of at least 60 percent from baseline values, with goal protein excretion being less than 500 to 1000 mg/day. (See 'Nonspecific therapy' above.)