A role for geranylgeranylation in interleukin-1β secretion

Authors


Abstract

Objective

Mevalonate kinase deficiency (MKD) is an autosomal-recessive disorder characterized by recurring episodes of inflammation. MK catalyzes the phosphorylation of mevalonic acid, which is an early step in isoprenoid biosynthesis. The goal of our study was to determine whether a temporary shortage of certain isoprenoid end products and/or the accumulation of mevalonic acid is the cause of interleukin-1β (IL-1β) secretion in MKD.

Methods

We studied the effect of the addition of intermediate metabolites and inhibitors of the isoprenoid biosynthesis pathway on IL-1β secretion by peripheral blood mononuclear cells (PBMCs) of patients with MKD and healthy controls.

Results

Inhibition of enzymes involved in geranylgeranyl pyrophosphate (GGPP) synthesis or geranylgeranylation of proteins led to a marked increase of lipopolysaccharide-stimulated IL-1β secretion in PBMCs of control subjects. Furthermore, the increased IL-1β secretion by PBMCs of patients with MKD was reversed by supplementation with GGPP as well as with mevalonic acid. IL-1β secretion was increased only when control PBMCs were incubated with excessive amounts of mevalonic acid. Finally, a reduction in IL-1β secretion by MKD PBMCs was also observed when sterol biosynthesis was inhibited, favoring nonsterol isoprenoid biosynthesis.

Conclusion

Our results indicate that a shortage of geranylgeranylated proteins, rather than an excess of mevalonate, is likely to cause increased IL-1β secretion by PBMCs of patients with MKD.

Mevalonate kinase deficiency (MKD) is an autosomal-recessive autoinflammatory disorder characterized by recurring episodes of high fever associated with headache, arthritis, nausea, abdominal pain, diarrhea, and skin rash (1, 2). Originally, 2 distinct syndromes had been defined, classic mevalonic aciduria (MA) (3) and hyperimmunoglobulinemia D with periodic fever syndrome (HIDS) (4), but after the discovery that both disorders are caused by a deficiency of the enzyme MK (5, 6), they are now recognized as the severe and mild presentations of MKD. Patients with HIDS typically have recurrent episodes of fever with associated inflammatory symptoms (1), whereas patients with MA, in addition to these episodes, show developmental delay, dysmorphic features, ataxia, cerebellar atrophy, and psychomotor retardation and may die in early childhood (2). Cells of patients with HIDS show residual MK enzyme activity of 1–8% (6–8), but in cells of patients with MA the enzyme activity is below detection level (2, 9, 10). This difference in residual enzyme activity is also reflected in the occurrence of high levels of mevalonic acid in the plasma and urine of patients with MA and low to moderate levels of mevalonic acid in patients with HIDS.

Blood analyses during the episodes of fever indicate an acute inflammatory state, with a marked rise in serum levels of proinflammatory cytokines, such as interleukin-6 (IL-6) and interferon-γ (11, 12). Also, between attacks, isolated peripheral blood mononuclear cells (PBMCs) of patients with MKD secrete increased amounts of proinflammatory cytokines, such as IL-1β (13, 14). IL-1β is the prototypical proinflammatory cytokine and appears to play an important role in the pathogenesis of several autoinflammatory diseases, including MKD (15). This is supported by the beneficial effect of a recombinant form of IL-1 receptor antagonist, anakinra, used in the treatment of those autoinflammatory diseases (16–22).

MK catalyzes the ATP-dependent phosphorylation of mevalonate to produce 5-phosphomevalonate and is the first enzyme to follow the rate-limiting and highly regulated enzyme hydroxymethylglutaryl-coenzyme A (HMG-CoA) reductase in the isoprenoid biosynthesis pathway (Figure 1) (23). The isoprenoid biosynthesis pathway provides cells with several bioactive molecules, including isoprenyl groups, the polyprenyl chain of heme A, dolichol, and sterols. The isoprenyl groups come in 2 forms: the farnesyl groups (from farnesyl pyrophosphate [FPP]) and the geranylgeranyl groups (from geranylgeranyl pyrophosphate [GGPP]). Both can be attached to proteins of the Ras superfamily.

Figure 1.

Schematic overview of the isoprenoid biosynthesis pathway, indicating the various enzyme steps that were inhibited in this study. Acetyl-CoA = acetyl-coenzyme A; HMG-CoA = hydroxymethylglutaryl-coenzyme A; MKD = mevalonate kinase deficiency; P = phosphate; PP = pyrophosphate; tRNA = transfer RNA; FTI = farnesyltransferase inhibitor; FTase = farnesyltransferase; GGTase = geranylgeranyltransferase; ZAA = zaragozic acid A; GGTI = geranylgeranyltransferase inhibitor.

The precise molecular mechanism by which the depressed activity of MK leads to increased IL-1β secretion and fever episodes is still unknown. However, there are indications that it is due to a temporary shortage of certain isoprenylated proteins (24).

Recently, Simon et al (25) reported the outcome of simvastatin treatment of 6 patients with the HIDS phenotype, which led to shorter periods of fever in most patients. The rationale for testing simvastatin in these patients was based on the assumption that the elevated mevalonic acid levels were causing the inflammation. Since statins, such as simvastatin, are competitive inhibitors of HMG-CoA reductase, this treatment would lead to a lowering of mevalonate levels and thus was predicted to reduce inflammation.

In contrast, we previously reported that it was not elevated mevalonic acid levels, but a shortage of isoprenoid end products, that contributed to the inflammation in MKD (14, 24, 26). In order to resolve this apparent discrepancy, we studied the inflammatory response of both MKD and control PBMCs by measuring IL-1β secretion upon stimulation with lipopolysaccharide (LPS) after exposing the cells to a range of concentrations of mevalonate. Furthermore, we studied the effect of various enzyme inhibitors and intermediate metabolites of the isoprenoid biosynthesis pathway on IL-1β secretion by PBMCs of patients with MKD and controls. Our results indicated that increased IL-1β secretion is correlated with a shortage of certain nonsterol isoprenoids rather than elevated mevalonic acid levels.

PATIENTS AND METHODS

Patients and samples.

After approval by the ethics review board of the University Medical Center and after written informed consent was obtained from their parents, blood was drawn from 4 patients with MKD ranging in age from 3 to 14 years, by venipuncture using sterile, pyrogen-free heparinized tubes (Vacuette; Greiner, Alphen aan den Rijn, The Netherlands). Healthy volunteers served as controls. PBMCs were isolated by density-gradient centrifugation using Lymphoprep according to the manufacturer's protocol (Axis-Shield, Oslo, Norway). Cells (1 × 105/per well) were seeded in 96-well flat-bottomed microtiter plates in RPMI 1640 (Gibco Invitrogen, Carlsbad, CA) supplemented with 10% heat-inactivated fetal calf serum (Gibco Invitrogen), 10 mM HEPES (Gibco Invitrogen), and 1% penicillin/streptomycin. Next, PBMCs were incubated with culture medium that contained the indicated compound, at 37°C in a humidified atmosphere containing 5% CO2 in air. After 18 hours of incubation, Escherichia coli O55:B5 LPS (final concentration of 100 ng/ml; Sigma, St. Louis, MO) or medium was added to the cultures, followed by 24 hours of additional incubation. After this, supernatants were collected and stored at −20°C until analysis.

Mevalonate was prepared by hydrolyzation of mevalonic acid lactone with 0.1M NaOH, followed by neutralization and stabilization at pH 7.4 with 1M HEPES and 0.1M HCl. Simvastatin and zaragozic acid A (ZAA) were prepared as previously described (27). Farnesyltransferase (FTase) inhibitor and geranylgeranyl transferase (GGT) inhibitor (FTI-277 and GGTI-298; Calbiochem, La Jolla, CA) were dissolved in DMSO (20 mM). Pamidronate (a gift from Novartis) was dissolved in distilled water (10 mM).

Cytokine measurements.

IL-1β concentrations were measured in duplicate in thawed supernatant samples, using enzyme-linked immunosorbent assays (ELISAs) (PeliKine-compact human IL-1β ELISA kit; Sanquin Reagents, Amsterdam, The Netherlands) according to the manufacturer's instructions.

Statistical analysis.

All results are expressed as the mean ± SEM. Statistical significance was calculated using Friedman's paired nonparametric analysis of variance followed by Dunn's multiple comparison test or a 2-tailed Wilcoxon's matched pairs signed rank test. P values less than 0.05 were considered significant.

RESULTS

Effect of upstream inhibition of MK and of mevalonic acid supplementation.

The effect of inhibiting HMG-CoA reductase, which is the enzyme preceding MK, was studied by incubating PBMCs from patients with MKD and control subjects with or without 5 μM simvastatin, after which they were stimulated with LPS. In the absence of simvastatin, stimulation of IL-1β secretion by LPS was greater in MKD PBMCs than in control PBMCs, whereas LPS-stimulated IL-1β secretion was markedly increased after incubation with simvastatin both in control and MKD PBMCs (Figure 2A). These results showed that lowering the levels of mevalonate and/or downstream isoprenoids led to an increase in IL-1β secretion.

Figure 2.

Effect of upstream inhibition of mevalonate kinase (MK) and mevalonic acid supplementation. A, Interleukin-1β (IL-1β) secretion by peripheral blood mononuclear cells (PBMCs) from 4 controls and 4 patients with MK deficiency (MKD). Cells were incubated in the absence or presence of simvastatin for 18 hours and stimulated for 24 hours with lipopolysaccharide (LPS). B, The effect of mevalonate on IL-1β secretion by PBMCs from controls and patients with MKD. The IL-1β concentrations determined in independent incubations using PBMCs obtained from 7 different controls and 3 different patients with MKD are shown. Incubations with 10 mM mevalonate were performed in PBMCs from 3 controls and 2 patients with MKD. Values are the mean and SEM.

To investigate whether elevated mevalonate levels also can be proinflammatory, we next determined the LPS-induced IL-1β secretion by PBMCs from control subjects and patients with MKD when they were exposed to increasing concentrations of mevalonate. Incubation of freshly isolated PBMCs from control subjects with mevalonate concentrations ranging from 0.05 to 10 mM resulted in a noticeable increase of LPS-stimulated IL-1β secretion only at 5 mM and 10 mM (Figure 2B). When PBMCs of patients with MKD were incubated with the same concentrations of mevalonate, there was a marked decrease in IL-1β secretion with increasing mevalonate concentrations (Figure 2B).

Effect of inhibition of enzymes located downstream of MK.

The reduced IL-1β secretion observed after mevalonate supplementation supports the hypothesis that LPS-induced IL-1β secretion by PBMCs of patients with MKD is due to a shortage of one of the isoprenoid end products and not to elevated mevalonate levels. To further substantiate this, we studied the effect of inhibiting different enzymes located downstream of MK in the isoprenoid biosynthesis pathway. To this end, we incubated control PBMCs with increasing concentrations of pamidronate, which inhibits FPP synthase, the enzyme that catalyzes the formation of geranyl pyrophosphate and FPP (Figure 1). This resulted in a significant elevation in LPS-stimulated IL-1β secretion (Figure 3A), which confirmed that a shortage of isoprenoid end products plays a role in elevated LPS-stimulated IL-1β secretion levels.

Figure 3.

Effect of inhibiting enzymes downstream of MK. A, IL-1β secretion by LPS-stimulated PBMCs in 4 controls in the absence or presence of pamidronate. ∗ = P < 0.05 versus concentrations determined in PBMCs stimulated in the absence of pamidronate. B, IL-1β secretion by PBMCs in 4 controls and 4 patients with MKD, stimulated with LPS in the absence or presence of zaragozic acid A (ZAA). ∗ = P < 0.05 versus concentrations determined in PBMCs incubated in the absence of ZAA. Values are the mean and SEM. See Figure 2 for other definitions.

Both simvastatin and pamidronate inhibited the synthesis of sterol as well as nonsterol isoprenoid end products. Because our previous results indicated that the shortage underlying the symptoms in MK deficiency was probably related to nonsterol isoprenoid end products (14, 24, 26), we next studied the effect of increasing concentrations of ZAA on IL-1β secretion by control and MKD PBMCs. ZAA is a specific inhibitor of squalene synthase, the first enzyme committed exclusively to sterol isoprenoid biosynthesis. Thus, ZAA inhibits the synthesis of sterol isoprenoids and, by doing so, promotes the synthesis of nonsterol isoprenoids (Figure 1). Moreover, the reduction in the synthesis of sterol end products leads to increased transcription of early isoprenoid biosynthetic genes (23). Consistent with this, the incubation of MKD PBMCs with ZAA resulted in reduced LPS-stimulated IL-1β secretion (Figure 3B).

Role of isoprenylated proteins in IL-1β secretion.

The previous experiment pointed to a role for nonsterol isoprenoid end products in IL-1β secretion. The most prominent nonsterol isoprenoids with a known biologic function are the farnesyl and geranylgeranyl moieties, derived from FPP and GGPP, respectively. Both can be covalently attached to proteins in a process known as protein isoprenylation. To determine whether increased IL-1β secretion is due to a shortage of farnesyl or geranylgeranyl moieties, we tested the effect of incubating control PBMCs with FTase inhibitor or GGT inhibitor on LPS-stimulated IL-1β secretion. Results showed that the addition of GGTI led to a marked increase in LPS-stimulated IL-1β secretion, whereas the addition of FTase inhibitor did not have a noticeable effect on the secretion of IL-1β (Figure 4A).

Figure 4.

Role of isoprenylated proteins in IL-1β secretion. Shown are levels of IL-1β secretion by PBMCs from controls (shaded bars) and patients with MKD (solid bars), incubated in the absence or presence of farnesyltransferase inhibitor (FTI), geranylgeranyl transferase inhibitor (GGTI), simvastatin, and geranylgeranyl pyrophosphate (GGPP) for 18 hours, followed by stimulation with LPS. A, Control PBMCs (n = 4) incubated with FTI and GGTI. ∗ = P < 0.05. B, Control PBMCs (n = 3) stimulated with LPS after incubation with simvastatin and/or GGPP. ∗ = P < 0.05. C, MKD PBMCs (n = 4) stimulated with LPS in the absence or presence of GGPP. D, Findings in paired samples (with and without GGPP) from individual patients, showing similar results in all 4. Values in A–C are the mean and SEM. See Figure 2 for other definitions.

Consistent with these findings, we also found that the increased LPS-stimulated IL-1β secretion observed when control PBMCs were incubated with simvastatin could be completely reversed when the cells were incubated with GGPP (Figure 4B). Also, the increase in LPS-stimulated IL-1β secretion in PBMCs of patients with MKD could be reduced to control levels with GGPP (Figures 4C and D).

DISCUSSION

Patients with the HIDS and MA phenotypes experience similar inflammatory episodes despite marked differences in both residual MK activity and accumulating levels of mevalonic acid. This suggests that mevalonate itself has no major effect on the inflammatory response. This was confirmed in our study, in which only a small increase in LPS-stimulated IL-1β secretion was observed when control PBMCs were incubated with 5 mM and 10 mM mevalonate but not at lower concentrations. Because 5 mM mevalonate is 10 times higher than the mevalonate levels found in the blood of MA patients and 1,000 times higher than those found in the blood of HIDS patients, it seems very unlikely that mevalonic acid is the main cause of the inflammatory response. In fact, the increase in LPS-stimulated IL-1β secretion by MKD PBMCs is even reversed by the addition of mevalonate.

These data are consistent with previously reported results (26), which showed that MA and HIDS cells compensate for the reduced MK activity by elevating their intracellular mevalonate levels by increasing HMG-CoA reductase activity. This increased HMG-CoA reductase activity was also down-regulated when MKD cells were incubated with the isoprenoid precursors farnesol, geranylgeraniol, or mevalonate (26). Thus, in order to maintain the flux through the isoprenoid biosynthesis pathway, it seems important to increase the levels of intracellular mevalonic acid. Failure of this compensatory mechanism might even lead to increased IL-1β secretion and inflammation, as is suggested by the severe inflammatory crises provoked by treating 2 MA patients with lovastatin in an attempt to lower the mevalonic avid levels (2).

Other studies showed that incubation of control PBMCs with lovastatin or fluvastatin increased IL-1β secretion (14, 28, 29) and that lovastatin was able to further increase the elevated IL-1β secretion in MKD PBMCs (14). This was also observed in the present study, when simvastatin was added to PBMCs from patients with MKD and controls. These observations confirm that mevalonate, by itself, is not the cause of the LPS-stimulated hypersecretion of IL-1β by PBMCs of patients with MKD.

The results of our study strongly indicate that it probably is a shortage of nonsterol isoprenoid end products, notably geranylgeranyl groups, that leads to increased IL-1β secretion in MKD. This is due to the fact that the inhibition of enzymes involved in GGPP synthesis or geranylgeranylation of proteins, i.e., HMG-CoA reductase, FPP synthase, and GGT, all led to a marked increase in LPS-stimulated IL-1β secretion by control PBMCs. Moreover, the increased IL-1β secretion by PBMCs of patients with MKD could be reversed by supplementation with GGPP, whereas the inhibition of sterol synthesis with ZAA, which promotes the synthesis of nonsterol isoprenoids, also resulted in a reversal of LPS-stimulated IL-1β secretion by MKD PBMCs. Together, these findings clearly indicate an important role for nonsterol isoprenoids, notably geranylgeranyl groups, in the regulation of the inflammatory response.

There are currently no established treatments for MKD, although treatment with the recombinant form of the IL-1 receptor antagonist anakinra was shown to be effective in relieving the symptoms (19). Our results indicate that raising the availability of geranylgeranyl moieties in MKD, for example by increasing the pathway flux toward nonsterol isoprenoid biosynthesis, may provide another option in the treatment of MKD.

Acknowledgements

The authors thank Novartis for kindly providing pamidronate, Merck, Sharpe, and Dohme for kindly providing simvastatin, and Merck for kindly providing ZAA.

Ancillary