Experimental glomerulopathy alters renal cortical cholesterol, SR-B1, ABCA1, and HMG CoA reductase expression

Abstract

Previous studies indicate that acute tubular injury causes free cholesterol (FC) and cholesteryl ester (CE) accumulation within renal cortex/proximal tubules. This study assessed whether similar changes occur with glomerulopathy/nephrotic syndrome, in which high-circulating/filtered lipoprotein levels increase renal cholesterol supply. Potential adaptive changes in cholesterol synthetic/transport proteins were also assessed. Nephrotoxic serum (NTS) or passive Heymann nephritis (PHN) was induced in Sprague-Dawley rats. Renal injury (blood urea nitrogen, proteinuria) was assessed 2 and 7 days (NTS), or 10 and 30 days (PHN) later. FC and CE levels in renal cortex, isolated glomeruli, and proximal tubule segments were determined. SR-B1 (a CE influx protein), ABCA1 (a FC exporter), and HMG CoA reductase protein/mRNA levels were also assessed. FC was minimally elevated in renal cortex (0 to 15%), the majority apparently localizing to proximal tubules. More dramatic CE elevations were found ( approximately 5 to 15x), correlating with the severity of proteinuria at any single time point (r >/= 0.85). Cholesterol increments were associated with decreased SR-B1, increased ABCA1, and increased HMG CoA reductase (HMGCR) protein and its mRNA. Tubule (HK-2) cell culture data indicated that SR-B1 and ABCA1 levels are responsive to cholesterol supply. Experimental nephropathy can increase renal FC, and particularly CE, levels, most notably in proximal tubules. These changes are associated with adaptations in SR-B1 and ABCA1 expression, which are physiologically appropriate changes for a cholesterol overload state. However, HMGCR protein/mRNA increments can also result. These seem to reflect a maladaptive response, potentially contributing to a cell cholesterol overload state.

Figure 1.

Renal injury after induction of NTS nephritis as assessed by BUN concentrations (left) and proteinuria (right). Marked BUN elevations were apparent at 2 days after NTS induction (**, P < 0.001), which completely resolved by the 7-day time point. Similarly, marked proteinuria, apparent at 2 days, in large part resolved by 7 days (ΔΔ = P < 0.025) (n = 6 experimental and control animals at each of the two time points).

Figure 2.

Renal injury after induction of PHN as assessed by BUN concentrations (left) and proteinuria (right). Significant azotemia (**, P < 0.0001) and marked proteinuria (*, P < 0.001) were seen 10 days after PHN induction. At 30 days, azotemia had primarily resolved (Δ, P < 0.05). However, proteinuria had almost doubled in amount, compared to the values observed at the 10-day time point (n = 6 to 8 animals for control and experimental groups at each time point).

Figure 3.

Renal cortical FC and CE levels at 2 and 7 days after induction of NTS nephritis. Only trivial elevations in FC levels occurred after NTS, with a statistically significant increase (*, P < 0.01) being observed only at the 7-day time point (c, control tissues). In contrast, substantial increases in CE levels were observed at both time points (**, P < 0.001; n = 6 for each group at each time point).

Figure 4.

Renal cortical FC and CE levels at 10 and 30 days after induction of PHN. Significant FC levels were observed at both time points, in comparison to control (c) tissue samples. Much more dramatic CE elevations were observed, particularly at the 30-day time point (**, P < 0.001; n = 6 to 8 animals for each group at each time point).

Figure 5.

FC and CE levels in glomeruli and proximal tubule segments (PTS) isolated from 30-day PHN rats. Isolated proximal tubules, but not isolated glomeruli, demonstrated significant elevations in FC levels (left: Δ, P < 0.05). Both glomeruli (*, P < 0.01) and tubules had elevated CE levels, but the increase was approximately twice as great in the tubule versus the glomerular preparations (n = 6 for control and experimental glomeruli; n = 4 for control and experimental tubules). C, controls.

Figure 6.

Western blotting for HMG CoA reductase (HMGCR; top lanes) and heat shock protein-72 (HSP-72; bottom lanes) in renal cortex harvested from PHN and NTS kidneys at the time points (d, days) shown, and in paired control (C) tissues. HMGCR appears as an ∼120-kd band (top) and as an ∼94-kd band (bottom) corresponding with inactive/uncleaved and active/cleaved bands, respectively. In both the NTS and PHN kidney samples, there was an increase in the active band. A trend toward a decrease in the inactive band was also apparent (see Figure 7 ▶ for quantitation). The diseased kidneys also showed an obvious increase in HSP-72 expression, consistent with a renal stress response (bottom lanes).

Figure 7.

Quantitation of HMGCR Western blots by densitometry in tissues harvested from control (c), PHN, and NTS nephritis rats. For these analyses, data obtained at each time point for the PHN (10 + 30 days) or NTS (2 + 7 days) rats were combined and compared versus their simultaneous controls. The left panel compares expression of the active 94-kd band, and the right panel compares the 120-kd inactive band. Significant increases in the active band were observed for both experimental groups (ΔΔ, P < 0.025). Although there was a trend toward a decrease in the inactive band in the PHN and NTS groups (suggesting increased cleavage), these differences did not achieve statistical significance (n = 4 to 5 at 10 days; n = 8 to 9 at 30 days, respectively).

Figure 8.

Western blots of SR-B1 and ABCA1 in renal cortical tissues. SR-B1 was detected at 82 kd. A modest decrease in expression was observed in NTS and PHN kidneys, compared to their controls (C). The most prominent reductions were at 30 days after PHN induction, consistent with the fact that the nephrotic syndrome and renal cortical cholesterol overload were most fully expressed in this group (see Figure 9 ▶ for quantitation). ABCA1 blotting of 30-day PHN kidneys are shown at the bottom. It is observed as a triplet at ∼220 kd (consistent with differing degrees of glycosylation; see text). In contrast, normal renal cortex had minimal ABCA1 expression (as is typical for unstimulated tissue; see text). When subjected to densitometric analysis, a greater than 2× increase in ABAC1 expression was observed in PHN tissue samples.

Figure 9.

SR-B1 densitometric analysis in NTS (left) and in PHN (right) tissue samples versus their simultaneous controls (c). Overall analysis of NTS versus control samples and PHN versus control samples demonstrated statistically significant reductions in SR-B1 expression in each disease model (*, P < 0.01; n = 4 to 5 determinations per group).

Figure 10.

SR-B1 (top) and ABCA1 (bottom) in cultured HK-2 cells: sensitivity to cell cholesterol manipulation. SR-B1 was reduced in HK-2 cells by incubation with serum (which increases cell cholesterol levels; S, serum; C, controls). This is consistent with cholesterol loading causing physiological suppression of SR-B1 protein mass. ABCA1 levels were suppressed by incubation with mevastatin, consistent with suppression of this cholesterol exporter by reductions in cell cholesterol levels (M, mevastatin; C, controls).