SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

The hepatic expression of Niemann-Pick C1-like 1 (NPC1L1), which is a key molecule in intestinal cholesterol absorption, is high in humans. In addition to NPC1L1, Niemann-Pick C2 (NPC2), a secretory cholesterol-binding protein involved in intracellular cholesterol trafficking and the stimulation of biliary cholesterol secretion, is also expressed in the liver. In this study, we examined the molecular interaction and functional association between NPC1L1 and NPC2. In vitro studies with adenovirus-based or plasmid-mediated gene transfer systems revealed that NPC1L1 negatively regulated the protein expression and secretion of NPC2 without affecting the level of NPC2 messenger RNA. Experiments with small interfering RNA against NPC1L1 confirmed the endogenous association of these proteins. In addition, endocytosed NPC2 could compensate for the reduction of NPC2 in NPC1L1-overexpressing cells, and this demonstrated that the posttranscriptional regulation of NPC2 was dependent on a novel ability of NPC1L1 to inhibit the maturation of NPC2 and accelerate the degradation of NPC2 during its maturation. Furthermore, to confirm the physiological relevance of NPC1L1-mediated regulation, we analyzed human liver specimens and found a negative correlation between the protein levels of hepatic NPC1L1 and hepatic NPC2. Conclusion: NPC1L1 down-regulates the expression and secretion of NPC2 by inhibiting its maturation and accelerating its degradation. NPC2 functions as a regulator of intracellular cholesterol trafficking and biliary cholesterol secretion; therefore, in addition to its role in cholesterol re-uptake from the bile by hepatocytes, hepatic NPC1L1 may control cholesterol homeostasis via the down-regulation of NPC2. (HEPATOLOGY 2011)

Niemann-Pick C1-like 1 (NPC1L1) is a key protein involved in intestinal cholesterol absorption.1, 2 In most animal species, NPC1L1 is highly expressed in the proximal intestine (where dietary cholesterol is absorbed).1, 3-5 In humans, it has been reported that NPC1L1 is highly expressed in the liver in addition to the intestine.1, 3In vivo studies of mice expressing human NPC1L1 from a liver-specific promoter have revealed that hepatic NPC1L1 may be involved in cholesterol reabsorption from the bile by hepatocytes.6 On the basis of these findings, NPC1L1 is believed to play critical roles in cholesterol uptake in the intestine and in re-uptake in the liver.

NPC1L1 was originally identified as a homolog of Niemann-Pick C1 (NPC1)7; mutations of the latter result in the acquisition of NPC disease, which is a neurovisceral disorder characterized by an accumulation of free cholesterol within endosomes and lysosomes.8, 9 In addition to mutations in the NPC1 gene, mutations in the NPC2 gene also cause an accumulation of cholesterol in late endosomes and lysosomes and can result in the development of NPC disease in some patients.10 NPC2 is a small secretory protein that is widely expressed in the body and specifically binds unesterified sterols with nanomolar affinity.11 Recent studies have revealed that NPC2 can transfer its bound cholesterol to NPC1 in late endosomes and lysosomes to facilitate intracellular cholesterol trafficking.12, 13

In addition to its role as a regulator of intracellular cholesterol trafficking, NPC2 may contribute to whole-body cholesterol homeostasis. Klein et al.14 found that NPC2 is expressed in the liver and secreted into the bile in both mice and humans. Moreover, in a recent study,15 we found that biliary NPC2 positively regulates biliary cholesterol secretion by stimulating cholesterol efflux, which is mediated by a heterodimer of adenosine triphosphate–binding cassette G5 (ABCG5) and ABCG8, a cholesterol exporter expressed in the liver.16 Physiologically, the amount of cholesterol secreted into the bile each day is similar to the amounts synthesized in the liver and absorbed from the intestine,17 and this suggests the importance of biliary cholesterol in cholesterol homeostasis. The biliary secretion of NPC2 is, therefore, thought to be important in the maintenance of the whole-body cholesterol level.

Although it has been shown that NPC2 cooperates with NPC112, 13 and ABCG5/ABCG8,15 the functional interaction between NPC1L1 and NPC2 has not yet been clarified. Although it has been demonstrated that secreted NPC2 has little effect on NPC1L1-mediated cholesterol uptake,15, 18 considering the fact that NPC1L1 is expressed in intracellular compartments besides the plasma membrane,3 we have hypothesized that there may be an intracellular interaction between NPC1L1 and NPC2.

In this article, we show that NPC1L1 interacts with NPC2 during the maturation of NPC2. In addition, the results of in vitro assays with NPC1L1-overexpressing cells or cells in which NPC1L1 was knocked down with small interfering RNA as well as analyses with human liver specimens indicate that NPC1L1 down-regulates the protein expression and secretion of NPC2 by inhibiting its maturation and by accelerating its degradation during the maturation process. These findings demonstrate a novel function of NPC1L1 as a negative regulator of NPC2 in addition to its role as a cholesterol (re-)uptake transporter.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Construction of Expression Vectors and Recombinant Adenoviruses.

The expression vector for c-terminally HA-tagged human NPC1L1 complementary DNA (NPC1L1-HA) was constructed as reported previously.(2,19,20) The expression vectors for c-terminally c-myc- and 6 x histidine-tagged human NPC2 complementary DNA (NPC2-Myc-His) and human GM2 ganglioside activator protein complementary DNA (GM2AP-Myc-His) were constructed as described in the supporting information.

Recombinant adenoviruses expressing each complementary DNA [Niemann-Pick C1-like 1–expressing adenovirus (Ad-NPC1L1), Niemann-Pick C1-like 1/HA tag–expressing adenovirus (Ad-NPC1L1-HA) and Niemann-Pick C2/c-myc tag/6 x histidine tag–expressing adenovirus (Ad-NPC2-Myc-His)] were prepared with the Adeno-X Tet-Off 1 expressing system (Takara Bio, Inc., Shiga, Japan) according to the manufacturer's instructions and were purified by cesium chloride gradient centrifugation. A tetracycline-responsive transcriptional activator–expressing adenovirus and a green fluorescent protein–expressing adenovirus (Ad-GFP)21 were purified with the same method. The titer of each purified virus (plaque-forming units per milliliter) was determined with the Adeno-X rapid titer kit (Takara Bio), and the multiplicity of infection (MOI) was determined by the normalization of the virus titer to the cell count in each experiment.

Immunoblot Analyses.

Immunoblot analyses were performed as described in the supporting information.

Metabolic Labeling of NPC2.

Chinese hamster ovary K1 (CHO-K1) cells infected with the indicated adenoviruses were first incubated in methionine/cysteine-free minimal essential medium (Invitrogen Life Technologies, Carlsbad, CA) for 30 minutes. The cells were then incubated in a labeling medium containing a 100 mCi/mL [35S]methionine/cysteine cell labeling mix (PerkinElmer, Waltham, MA) and were collected at the indicated times. The cells were lysed with a radio immunoprecipitation assay buffer (0.1% sodium dodecyl sulfate, 0.5% deoxycholate, and 1% Nonidet P-40) and immunoprecipitated with 1 μg of a mouse anti-Myc antibody (Roche Applied Science, Indianapolis, IN), as described in the supporting information. The immunoprecipitates were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and exposed to a super-resolution phosphor imager (PerkinElmer). The radioactivity was detected with a Cyclone phosphor imager (Packard, Haverhill, MA).

Quantitative Real-Time Polymerase Chain Reaction (PCR).

To determine the messenger RNA (mRNA) levels of NPC2 and NPC1L1, quantitative real-time PCR was performed as described in the supporting information.

Immunohistochemical Staining.

Immunohistochemical staining with HepG2 cells was performed as described in the supporting information.

Cholesterol Staining.

For the detection of free intracellular cholesterol, cells were fixed with 4% paraformaldehyde and stained with filipin according to the manufacturer's instructions (Cayman Chemicals). The relative intensity of filipin staining in intracellular compartments was quantified by the division of the intensity above the low threshold by the number of total pixels.22, 23

Collection of Human Liver Specimens.

All experiments involving human specimens were conducted according to a study protocol approved by the institutional review board of the University of Tokyo and Tsukuba University after informed consent was obtained from all subjects. Tumor tissue and surrounding tissue appearing to be grossly normal were obtained from nine liver cancer patients upon surgical resection. The tumor-adjacent normal tissue specimens were used for immunoblot analyses to determine the protein levels of NPC1L1 and NPC2 and for quantitative real-time PCR to determine the mRNA levels of these genes.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

NPC2 Coimmunoprecipitates With NPC1L1.

To examine the interaction between NPC1L1 and NPC2, we performed a coimmunoprecipitation assay. As shown in Fig. 1, NPC2 in the total cell lysate was detected at approximately 20 to 26 kDa, as reported previously.24 The lower molecular weight (LMW) form of NPC2 (20 kDa) coimmunoprecipitated with NPC1L1 (HA, Fig. 1). Conversely, NPC1L1 coimmunoprecipitated with NPC2 (Myc, Fig. 1). These results suggest that NPC1L1 can interact with LMW NPC2.

thumbnail image

Figure 1. Molecular association between NPC1L1 and NPC2. CHO-K1 cells were infected with Ad-NPC2-Myc-His, Ad-NPC1L1-HA, and Ad-GFP (the control) at 5 MOI. Twenty-four hours after infection, the cells were harvested, and a coimmunoprecipitation assay was performed as described in the supporting information. Anti-HA IPs, anti-Myc IPs, and the total lysate (the input) were subjected to IB analysis with an anti-HA antibody (to detect NPC1L1-HA) and an anti-His antibody (to detect NPC2-Myc-His). Abbreviations: IB, immunoblot; IP, immunoprecipitate.

Download figure to PowerPoint

NPC1L1 Down-Regulates the Protein Expression of NPC2.

The expression of NPC2 and particularly its higher molecular weight (HMW) forms, which are detected at approximately 26 kDa,24 was markedly reduced by the coexpression of NPC1L1 in adenovirus-based experiments (input, Fig. 1). This reduction in the NPC2 protein level was also observed in plasmid-based experiments (lanes 2 and 3, Fig. 2A), although the expression level of NPC2 mRNA was not altered by the coexpression of NPC1L1 (Fig. 2B). Because the expression of endogenous NPC1, endogenous cathepsin D (Fig. 2A), and exogenous GM2AP (Supporting Fig. 1), which are also lysosomal proteins, was hardly affected by the coexpression of NPC1L1, the reduction in the NPC2 protein level did not likely result from a nonspecific effect of NPC1L1 on lysosomal protein expression.

thumbnail image

Figure 2. Effect of NPC1L1 on the expression of NPC2. (A) CHO-K1 cells were transfected with the indicated vectors or an empty control vector. Twenty-four hours after transfection, the cells were harvested and subjected to an immunoblot analysis. (B) The mRNA levels of NPC2 in CHO-K1 cells that were cultured for 24 hours after transient transfection with the NPC2-Myc-His vector and either the NPC1L1-HA vector or an empty control vector were determined with quantitative real-time PCR. The relative mRNA level of NPC2 in each cell was normalized to the level of β-actin mRNA. The columns and vertical bars represent the means and standard deviations of three determinations. (C) CHO-K1 cells were infected with Ad-NPC1L1 at the indicated MOI together with Ad-NPC2-Myc-His at 5 MOI. Twenty-four hours after infection, the total cell lysate and the concentrated media were subjected to an immunoblot analysis. Ad-GFP was used to equalize the total amount of the infected adenovirus. (D) The aforementioned total cell lysate and concentrated media were deglycosylated with either Endo H or PNGase F (see the supporting information). The deglycosylated and undigested proteins were subjected to immunoblot analyses. (E) CHO-K1 cells were infected with Ad-NPC1L1 or Ad-GFP (the control) at 10 MOI together with Ad-NPC2-Myc-His at 5 MOI. Twenty-four hours after infection, 35S-labeling was performed for the indicated times. The 35S-labeled NPC2 protein was detected with a phosphor imager. The lower panels show the ratio of the 35S-labeled HMW NPC2 radioactivity to the total 35S-labeled NPC2 radioactivity. The data points and bars represent the means and standard deviations of three specimens on different images. Abbreviation: NS, not significant.

Download figure to PowerPoint

Maturation and Secretion of NPC2 Protein Are Inhibited by the Coexpression of NPC1L1.

A detailed analysis of the molecular association between NPC1L1 and NPC2 was performed. As shown in Fig. 2C, the expression of intracellular HMW NPC2 was dramatically reduced as the expression of NPC1L1 increased. In addition, the ratio of cellular LMW NPC2 to cellular HMW NPC2 was elevated by the increase in the expression of NPC1L1. Furthermore, because it had been reported that NPC2 is secreted extracellularly,24, 25 we also analyzed the effect of NPC1L1 on NPC2 secretion. As shown in Fig. 2C, the amount of NPC2 secreted into media was decreased by NPC1L1 overexpression, and this was consistent with the decrease in the intracellular NPC2 protein level.

A glycosidase digestion assay was performed next. After digestion with endoglycosidase H (Endo H; upper panels, Fig. 2D) or peptide N-glycosidase F (PNGase F; lower panels, Fig. 2D), intracellular NPC2 was detected at 18 kDa, which corresponded to the nonglycosylated form of the protein. This result suggests that HMW NPC2 and LMW NPC2 are glycosylation variants. In addition, the secretion of maturely glycosylated HMW NPC2, which corresponded to a 23-kDa band after digestion with Endo H, was concomitantly reduced with the increase in NPC1L1 expression (upper panels, Fig. 2D). These results raise the possibility that NPC1L1 inhibits the maturation of NPC2.

To test this hypothesis, we characterized NPC2 maturation over time with a [35S]methionine/cysteine pulse-chase experiment. In control cells, the amounts of both cellular and secreted 35S-labeled HMW NPC2 increased in a time-dependent manner (upper panels, Fig. 2E). However, in NPC1L1-overexpressing cells, very little cellular or secreted 35S-labeled HMW NPC2 was detected, whereas the amount of 35S-labeled LMW NPC2 increased in a time-dependent manner (upper panels, Fig. 2E). Furthermore, the ratio of the cellular amount of 35S-labeled HMW NPC2 to the amount of total 35S-labeled NPC2 (HMW + LMW) increased time-dependently in control cells, whereas this ratio did not change in NPC1L1-overexpressing cells (lower panels, Fig. 2E). These results support the hypothesis that NPC1L1 inhibits the maturation of NPC2 from LMW forms to HMW forms.

NPC1L1 Interacts With NPC2 in Prelysosomal Compartments.

To clarify the cellular compartment in which the interaction between NPC1L1 and NPC2 occurs, we performed immunohistochemical staining for markers of various organelles. As shown in Fig. 3, NPC2 was localized in vesicle-like structures that were partially costained for the lysosomal marker cathepsin D (Fig. 3A) but not for the endoplasmic reticulum (ER) marker calnexin (Fig. 3B). However, in agreement with the results of the immunoblot analyses (Figs. 1 and 2A), when NPC1L1 was overexpressed, the expression of NPC2 protein was reduced, and minimal staining of NPC2 was observed (Fig. 3C).

thumbnail image

Figure 3. Intracellular colocalization of NPC1L1 with NPC2 in HepG2 cells. The cellular localization of NPC1L1-HA and NPC2 was examined with immunohistochemical staining. HepG2 cells were transfected with (A,B) the control vector or (C) the NPC1L1-HA vector. Twenty-four hours after transfection, the cells were stained for an immunohistochemical analysis. NPC1L1-expressing cells are indicated by circles. (D-F) HepG2 cells were transfected with the NPC1L1-HA vector. Twenty-four hours after transfection, the cells were treated with MG132 (10 μM) for 6 hours, and they were then subjected to immunohistochemical staining. Representative images are shown. In each panel, colocalization appears yellow in the merged image.

Download figure to PowerPoint

Because treatment with N-(benzyloxycarbonyl)leucinylleucinylleucinal (MG132), a proteasome inhibitor, could inhibit the degradation of NPC2 and increase LMW NPC2 even when NPC1L1 was overexpressed (Supporting Fig. 2), immunohistochemical staining was performed in the presence of MG132. Figure 3D shows that MG132-sensitive NPC2 was colocalized with NPC1L1 in intracellular compartments that were partially costained for calnexin (Fig. 3E) but not cathepsin D (Fig. 3F). The observation that NPC1L1 could be coimmunoprecipitated with LMW NPC2 (Fig. 1) and the observation that most of the MG132-sensitive NPC2 was LMW NPC2 (Supporting Fig. 2) suggest that NPC1L1 interacts with LMW NPC2 in prelysosomal compartments, including the ER.

Degradation of NPC2 Is Accelerated in the Presence of NPC1L1.

Because NPC1L1 reduces the expression of NPC2 protein, it was hypothesized that the degradation of NPC2 protein is accelerated by NPC1L1. To test this hypothesis, we analyzed the degradation rate of NPC2 protein. As shown in Fig. 4, the degradation of LMW NPC2 was more rapid in NPC1L1-overexpressing cells (half-life = 2.1 ± 0.1 hours) versus control cells (half-life = 7.1 ± 2.5 hours). In contrast, the half-life of HMW NPC2 was hardly affected by the overexpression of NPC1L1 (3.2 ± 0.8 hours in control cells and 2.3 ± 0.1 hours in NPC1L1-overexpressing cells). These results suggest that the ability of NPC1L1 to accelerate the degradation of LMW NPC2 may contribute to the lower expression of NPC2 protein in NPC1L1-overexpressing cells and that the reduced expression of HMW NPC2 in NPC1L1-overexpressing cells (Figs. 1 and 2A,C) could be explained by the inhibition of NPC2 maturation rather than the difference in the degradation speed of HMW NPC2.

thumbnail image

Figure 4. Effect of NPC1L1 on the degradation rate of NPC2 protein. CHO-K1 cells were infected with Ad-NPC1L1 or Ad-GFP (the control) at 10 MOI together with Ad-NPC2-Myc-His at 5 MOI. Twenty-four hours after infection, the cells were incubated with cycloheximide (100 μM). The total cell lysate, which was prepared at the indicated time points, was examined with an immunoblot analysis. The lower panels show quantitative comparisons of the levels of NPC2 protein normalized to the level of α-tubulin. The data points and bars represent the means and standard deviations of three specimens on different immunoblots. *Significantly different from control cells according to the Student t test (P < 0.05). **Significantly different from control cells according to the Student t test (P < 0.01).

Download figure to PowerPoint

Endogenous NPC1L1 Regulates the Protein Expression and Secretion of NPC2.

Although the overexpression of NPC1L1 caused a reduction in the protein expression and secretion of NPC2, an overexpression model is fraught with potential artifacts. To eliminate this possibility, we investigated the ability of endogenous NPC1L1 to regulate the expression and secretion of endogenous NPC2. For this purpose, HepG2 cells in which both NPC1L1 and NPC2 were expressed endogenously were transfected with a small interfering RNA targeted against Niemann-Pick C1-like 1 (siNPC1L1), and changes in NPC2 expression and secretion were analyzed. As shown in Fig. 5, after the transfection of siNPC1L1, the mRNA levels of endogenous NPC1L1 (Fig. 5A) and its protein levels (Fig. 5B) were reduced to approximately 40% and 60%, respectively, of the levels in cells transfected with the control small interfering RNA (siControl). Under these conditions, although the expression of NPC2 mRNA was unaltered (Fig. 5A), the level of NPC2 protein increased more than 1.9-fold (cellular NPC2, Fig. 5B). In addition, the amount of NPC2 secreted into the media also increased more than 2.5-fold after the suppression of NPC1L1 expression (secreted NPC2, Fig. 5B). However, transfection with siNPC1L1 had no effect on NPC1 protein expression (Fig. 5B). These results suggest that the expression and secretion of endogenous NPC2 are negatively regulated by endogenous NPC1L1.

thumbnail image

Figure 5. Effect of siNPC1L1 on the endogenous expression of NPC2. HepG2 cells were transfected with siNPC1L1 or siControl (see the supporting information) and cultured for 2 days. The extracted RNA and the total cell lysate were then analyzed with (A) quantitative real-time PCR and (B) an immunoblot analysis, respectively. (A) The relative mRNA levels of NPC1L1 and NPC2 were normalized to the β-actin mRNA levels. (B) The relative levels of the NPC1L1 and NPC2 proteins were calculated as the band densities of the proteins and were normalized to the α-tubulin protein level in each specimen. The columns and vertical bars represent the means and standard deviations of the three specimens in the immunoblot. **Significantly different according to the Student t test (P < 0.01). Abbreviation: NS, not significant.

Download figure to PowerPoint

Suppression of Intracellular Cholesterol Trafficking by NPC1L1 Overexpression Can Be Rescued by the Addition of Purified NPC2 Protein.

Because the loss of intracellular NPC2 suppresses intracellular cholesterol trafficking and causes cholesterol accumulation within lysosomes,10 the cholesterol distribution was analyzed in NPC1L1-overexpressing cells. In agreement with the previous results, the amount of endogenous NPC2 protein in HepG2 cells was reduced along with the increase in NPC1L1 expression, whereas the level of endogenous NPC1 protein was hardly affected (Fig. 6A). Because extracellular (secreted) NPC2 can be taken up into lysosomes via receptor-mediated endocytosis,26 NPC1L1-overexpressing cells were cultured in media containing exogenous NPC2 protein, and the amount of cellular NPC2 was examined. Exogenous NPC2 protein was purified from cell culture media containing secreted HMW NPC2. Figure 6B shows that the reduction of NPC2 in NPC1L1-overexpressing cells could be recovered by culturing with purified NPC2 protein, and this suggests that NPC1L1 may only minimally affect the protein stability of endocytosed NPC2. The endocytosed NPC2 protein in NPC1L1-overexpressing cells was colocalized with the lysosomal marker cathepsin D by immunohistochemistry (data not shown). These results are consistent with the observation that NPC1L1 interacts with NPC2 in prelysosomal compartments (Fig. 3E).

thumbnail image

Figure 6. Effect of NPC1L1 on intracellular cholesterol trafficking. (A) HepG2 cells were infected with Ad-NPC1L1 or Ad-GFP at the indicated MOI. Twenty-four hours after infection, the cells were collected and subjected to an immunoblot analysis. (B) HepG2 cells were infected with Ad-NPC1L1 or Ad-GFP (the control) at 3 MOI and were cultured together with purified NPC2 protein (10 nM). Twenty-four hours after infection, the cells were collected and subjected to an immunoblot analysis. (C) HepG2 cells were infected with Ad-NPC1L1 or Ad-GFP (the control) at 3 MOI and were cultured together with the indicated type of purified NPC2 protein (10 nM) for 24 hours or with U18666A (1.25 μM) for 8 hours. Twenty-four hours after infection, the cells were stained with filipin. The upper panels show the results of filipin staining. Representative images are shown. The lower panel shows a quantitative comparison of filipin staining in intracellular compartments. The columns and vertical bars represent means and standard deviations of three different images. **Significantly different according to an analysis of variance followed by Dunnett's test (P < 0.01). Abbreviation: NS, not significant.

Download figure to PowerPoint

Filipin staining was next performed in order to analyze the cellular distribution of cholesterol. As shown in Fig. 6C, in comparison with control cells, NPC1L1-overexpressing cells exhibited increased staining for intracellular cholesterol, whereas staining for cholesterol in the plasma membrane was reduced. A similar pattern was observed in cells treated with 3-beta-(2-diethylaminoethoxy)androst-5-en-17-one hydrochloride (U18666A), an inhibitor of intracellular cholesterol trafficking. When NPC1L1-overexpressing cells were incubated with purified wild-type (WT) NPC2, intracellular NPC2 levels were restored to the same levels found in control cells (Fig. 6B); intracellular cholesterol accumulation was significantly decreased, and in turn, the distribution of cholesterol in the plasma membrane was restored (Fig. 6C). This rescue effect was not observed when cells were cultured with NPC2 D72A, a loss-of-function mutant.27 These results suggest that the altered distribution of cholesterol in NPC1L1-overexpressing cells is mostly caused by the reduced expression of lysosomal NPC2 and not by NPC1L1 itself. In addition, NPC1L1 only slightly affects the function of lysosomal NPC2.

NPC2 Expression Is Negatively Correlated With NPC1L1 in Human Liver Specimens.

Finally, the correlation between the levels of NPC2 and NPC1L1 in human liver specimens was determined with immunoblot analyses and quantitative real-time PCR. In agreement with the in vitro results, the protein levels of NPC2 were negatively correlated with NPC1L1 levels in human liver specimens (Fig. 7A), although there was no significant correlation between the mRNA levels (Fig. 7B). Furthermore, the translational efficiency of NPC2, which is expressed as the ratio of the NPC2 protein level to the NPC2 mRNA level, was also negatively correlated with the protein level of NPC1L1 (Fig. 7C). These results support the hypothesis that the protein expression of NPC2 is posttranscriptionally regulated by NPC1L1 in the human liver.

thumbnail image

Figure 7. Correlation between the levels of NPC1L1 and NPC2 in human liver specimens. Homogenates and extracted RNA from human liver specimens were subjected to (A) an immunoblot analysis and (B) quantitative real-time PCR, respectively. (A) The relative protein levels of NPC1L1 and NPC2 were calculated as the band densities of the proteins and were normalized to the α-tubulin protein level in each specimen. (B) The relative mRNA levels of NPC1L1 and NPC2 were normalized to the β-actin mRNA level in each specimen. (C) The translational efficiency of NPC2 was calculated as the ratio of the NPC2 protein level to the NPC2 mRNA level. A correlation analysis was performed with Spearman's rank method. Abbreviation: Rs, Spearman's rank correlation coefficient.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

According to this study, in addition to its well-known role as a cholesterol importer, NPC1L1 has a novel function as a negative regulator of the expression and secretion of NPC2, which is based on the ability of NPC1L1 to inhibit the maturation of NPC2 protein and stimulate the degradation of LMW NPC2 protein. Because treatment with ezetimibe, an inhibitor of NPC1L1-mediated cholesterol import, does not affect the binding of NPC1L1 and NPC2 and cannot reverse the reduction in the NPC2 protein level in NPC1L1-coexpressing cells (Supporting Fig. 3), this novel function of NPC1L1 as a negative regulator of NPC2 protein is independent of its well-known role as a cholesterol importer.

Previously, it has been reported that NPC2 is degraded by the proteasome system.22 Therefore, we investigated whether treatment with MG132, a proteasome inhibitor, could inhibit the NPC1L1-mediated down-regulation of NPC2. In agreement with the previous report,22 in the absence of NPC1L1, the levels of both HMW NPC2 and LMW NPC2 were increased by MG132 treatment (Supporting Fig. 2). However, when NPC1L1 was coexpressed, MG132 treatment hardly increased the expression of HMW NPC2, although the expression of LMW NPC2 clearly increased (Supporting Fig. 2). This observation is in agreement with the finding that NPC1L1 inhibits the maturation of NPC2 from LMW forms to HMW forms (Fig. 2C,E). Furthermore, the stability of LMW NPC2 was altered by NPC1L1 coexpression (Fig. 4), and LMW NPC2 coimmunoprecipitated with NPC1L1 (Fig. 1). These data imply that NPC1L1 interacts with LMW NPC2, inhibits its maturation, and simultaneously promotes its degradation. As a result, intracellular NPC2 is reduced, and this in turn leads to a decrease in NPC2 secretion (Fig. 8).

thumbnail image

Figure 8. Suggested function of hepatic NPC1L1 in the regulation of biliary cholesterol secretion. (A) NPC1L1 inhibits the maturation and secretion of NPC2. (B) NPC1L1 accelerates the degradation of NPC2. (C) By reducing NPC2 secretion, NPC1L1 indirectly suppresses ABCG5/ABCG8-mediated cholesterol efflux. (D) In addition to its well-known function as a cholesterol importer, NPC1L1 contributes to the regulation of biliary cholesterol secretion by negatively regulating NPC2 secretion.

Download figure to PowerPoint

In contrast to LMW NPC2, a molecular association between HMW NPC2 and NPC1L1 was not detected by coimmunoprecipitation (Fig. 1). Because HMW NPC2 and LMW NPC2 are glycosylation variants (Fig. 2D), it is possible that the complex glycosylation of HMW NPC2 may inhibit its binding to NPC1L1, whereas core-glycosylated NPC2 can interact with NPC1L1. This idea is consistent with the observation that in addition to the expression of WT NPC2, the expression of NPC2 mutants (in which one or both of the glycosylation sites of NPC2 are mutated24) was reduced by the coexpression of NPC1L1 (Supporting Fig. 4). Because NPC1L1 colocalizes with NPC2 in prelysosomal compartments (Fig. 3) and the reduced expression of NPC2 can be rescued by exogenous secreted NPC2 (Fig. 6B), it seems that NPC1L1 interacts with NPC2 during the process of complex glycosylation.

The RNA interference studies using HepG2 cells (Fig. 5) and the correlation analysis of the expression levels of NPC1L1 and NPC2 in human liver specimens (Fig. 7) have revealed that endogenous NPC1L1 negatively regulates the expression of NPC2 posttranscriptionally. It has recently been reported that in addition to the NPC1L1-mediated regulation of NPC2, Nogo-B receptor (NgBR) interacts with NPC2 at the ER and enhances NPC2 protein stability by inhibiting its proteasomal degradation.22 Because NgBR is known to be expressed in the liver,28 NgBR may function as a positive regulator of NPC2 protein expression, whereas NPC1L1 acts as a negative regulator. The balance of the expression levels of NPC1L1 and NgBR may, therefore, determine the hepatic expression of NPC2 protein.

In addition to posttranscriptional modification, several groups have studied the transcriptional regulation of the NPC2 gene. For instance, Rigamonti et al.29 reported that in human macrophages, NPC2 mRNA is induced by activators of liver X receptor (LXR). Because LXR is activated by cellular cholesterol-related compounds, the expression of NPC2 mRNA may be positively regulated by cellular cholesterol levels. On the other hand, cholesterol has been shown to down-regulate the expression of NPC1L1 by transcriptional regulation via sterol regulatory element binding protein 2 and hepatocyte nuclear factor 4α.30, 31 Taken together, these data suggest that when the cellular cholesterol level increases, the expression of NPC2 protein is effectively elevated by a combination of positive transcriptional regulation via the LXR pathway and reduced posttranscriptional regulation via the interaction with NPC1L1. Because NPC2 is a crucial protein for intracellular cholesterol trafficking, which affects the regulation of cholesterol synthesis and uptake by delivering cholesterol to the sterol-sensing machinery in the ER,32 the expression level of NPC2 protein must be tightly regulated by various steps.

NPC1L1 affects the secretion of NPC2 protein by inhibiting the maturation and expression of intracellular NPC2 (Figs. 2C and 5). This regulatory mechanism of NPC2 secretion would be relevant in specific tissues such as the liver and intestine because NPC1L1 is predominantly expressed in these tissues in humans. In fact, because of the negative correlation between the protein levels of NPC1L1 and NPC2 in human liver specimens (Fig. 7), it is possible that hepatic NPC1L1 negatively regulates the biliary secretion of NPC2. Our recent study revealed the physiological function of biliary NPC2 as a positive regulator of biliary cholesterol secretion mediated by ABCG5/ABCG8 on the bile canalicular membrane of hepatocytes.15 Together with the results from this study, the data suggest that hepatic NPC1L1 may also indirectly affect ABCG5/ABCG8-mediated transport by decreasing biliary NPC2. In addition to its direct role in cholesterol re-uptake from the bile by hepatocytes,6 NPC1L1 may down-regulate the biliary secretion of NPC2 and, consequently, reduce NPC2-mediated cholesterol efflux by ABCG5/ABCG8 from hepatocytes into the bile15 (Fig. 8). When we consider that ABCG5 and ABCG8 are predominantly expressed in the liver and intestine and that ABCG5/ABCG8-mediated cholesterol excretion is an important process in cholesterol homeostasis, it makes sense that the regulatory mechanism for the secretion of NPC2 is working in these tissues to maintain adequate cholesterol levels in response to the dynamic cholesterol fluctuations in the body.

Collectively, the results of the present study suggest that NPC1L1 down-regulates the expression and secretion of NPC2 by interacting with NPC2 during its maturation process. Through this regulatory function, hepatic NPC1L1 is suggested to suppress the hepatic expression and biliary secretion of NPC2. In addition to its direct role in cholesterol re-uptake, hepatic NPC1L1 may effectively control biliary cholesterol secretion by negatively regulating NPC2 secretion because biliary NPC2 stimulates ABCG5/ABCG8-mediated cholesterol efflux.15 This is the first report suggesting a function for NPC1L1 besides its activity as a (re-)uptake transporter.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. References
  7. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_24772_sm_SuppFig1.tif751KSupporting Information Figure 1.
HEP_24772_sm_SuppFig2.tif915KSupporting Information Figure 2.
HEP_24772_sm_SuppFig3.tif1427KSupporting Information Figure 3.
HEP_24772_sm_SuppFig4.tif943KSupporting Information Figure 4.
HEP_24772_sm_SuppInfo.doc59KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.