Main

Bilirubin is known to have a variety of neurotoxic effects, which are of particular concern in neonates due to the high incidence of unconjugated hyperbilirubinemia in that age group. Despite this, the molecular mechanism(s) responsible for these effects remain(s) incompletely understood. In vivo, bilirubin is transported in serum in the unconjugated state largely bound to lysine residues on albumin, whereas in the hepatocyte it is bound to lysine residues on ligandin, a glutathione S-transferase(1–4). Polylysine has been used to bind bilirubin in artificial organ systems(5, 6). Bilirubin also binds to lysine on myelin basic protein as well as lysine-rich histones(7).

The functional importance of the bilirubin-lysine interaction is not clear. In particular, it is not known whether binding to lysine prevents or attenuates the ability of bilirubin to interfere with biologic functions. Recent studies have demonstrated that bilirubin has a potent and widespread inhibitory effect on protein phosphorylation systems both in vitro(8, 9) and in vivo(10). We have previously shown that bilirubin inhibits the in situ phosphorylation of neuronal phosphoproteins presumed to be involved in neurotransmitter release, suggesting a mechanism whereby bilirubin might affect inter-neuronal signaling mechanisms(11). Other studies have shown that bilirubin inhibits a variety of protein kinases in vitro(8, 9, 12, 13). Bilirubin inhibits the catalytic subunit of PKA by a noncompetitive mechanism, suggesting that bilirubin interacted with a noncatalytic domain on the enzyme(13). Interestingly, the protein kinase family contains an invariant lysine on the ATP-binding subdomain II as a conserved feature(14), indicating that this might represent a possible target for bilirubin.

In the present study we have examined whether synthetic peptides, both with and without lysine, have the ability to modulate the toxic effects of bilirubin on protein phosphorylation in vitro. We demonstrate that the inhibitory effects of bilirubin on phosphorylation catalyzed by PKA are largely prevented by polylysine. A synthetic decapeptide, which copied part of the amino acid sequence of the kinase ATP-binding site subdomain II, attenuated the inhibitory effect of bilirubin. Such effects were not observed in the presence of peptides that did not contain lysine.

METHODS

[32P]ATP was obtained from ICN Biomedicals, Inc. (Irvine, CA). The model substrate was a synthetic peptide derived from the protein phospholemman, comprising the extreme 15 carboxy-terminal amino acids (residues 58-72 of the canine protein, GTFRSSIRRLSTRRR in single letter code)(15). A synthetic decapeptide (Gly-Asp-His-Tyr-Ala-Met-Lys-Ile-Leu-Glu) was derived from the sequence of subdomain II of the kinase family. Two other decapeptides were identical except for the substitution of alanine or arginine, respectively, for lysine. All of these peptides were made by the Peptide Synthesis Facility, Forskningsparken, University of Oslo, Norway. Bilirubin, BSA, the catalytic subunit of PKA, poly-L-lysine, poly-D-lysine, polyarginine, and polyglutamate were from Sigma Chemical Co. (St. Louis, MO). Other reagents of analytical grade or better were from standard commercial suppliers. SpinZyme separation units were from Pierce (Rockford, IL).

Phosphorylation of phospholemman peptide. Phosphorylation of the synthetic phospholemman peptide was analyzed by incubations (in triplicate) in microtiter plates in a medium containing 20 mM Tris-Cl (pH 7.4) and 10 mM MgCl2, in the presence of bilirubin (final concentration, 50 μM, solubilized in 0.1 N NaOH), one of the peptides (final concentration 100-300 μM), and BSA (0.1 μM). The assays were performed in a reaction volume of 100 μL with [ATP] kept constant (0.2 mM final concentration, with 1-4 μCi of [γ-32P]ATP per assay) and peptide substrate concentrations being varied between 10 and 50 μM. Reactions were initiated by addition of the catalytic subunit of PKA (final concentration, 2.5 nM). Samples were incubated in a water bath at 30 °C for 3 min, and the reactions were terminated by addition of 0.2 M EDTA. The solution was transferred to SpinZyme separation units and washed three times with 75 mM phosphoric acid. Radioactivity in the filters was measured as Cerenkov radiation.

Statistical methods. Statistical analysis was by analysis of variance or Mann-Whitney U tests as appropriate. p < 0.05 was chosen as the level of significance.

RESULTS

The lysine-containing decapeptide (300 μM) from the kinase family subdomain II (Gly-Asp-His-Tyr-Ala-Met-Lys-Ile-Leu-Glu) had a significant, albeit moderate, ability to prevent the inhibitory effect of bilirubin (50 μM) on phospholemman peptide (30 μM) phosphorylation by PKA (q = 8.3, p < 0.01; Fig. 1). Essentially identical results were observed when the substrate concentration was 10 or 50 μM, respectively (results not shown). The decapeptides which did not contain lysine did not have this abilty (F = 2.31, p = 0.097; results not illustrated).

Figure 1
figure 1

A lysine-containing decapeptide that mimics subdomain II on the protein kinase family reduces the inhibitory effect of bilirubin on phospholemman phosphorylation by PKA. Phospholemman peptide (30 μM) was phosphorylated by PKA (2.5 nM) in the absence or presence of bilirubin (50 μM) and the lysine-containing decapeptide (Gly-Asp-His-Tyr-Ala-Met-Lys-Ile-Leu-Glu) (300 μM) at 30 °C for 3 min. Reactions were stopped by adding 0.2 M EDTA, the reaction solution was filtered and washed, and the radioactivity in the filters was counted as Cerenkov radiation. □, control; ▪, bilirubin; [square with lower left to upper right fill], bilirubin + lysine-containing decapeptide; F = 450, p < 0.0001. Contrast of bilirubin effects vs effects of bilirubin plus decapeptide (Tukey's multiple comparison test): q = 8.3, p < 0.01.

Full size image

Poly-L-lysine (100 μM) completely blocked the inhibitory effect of bilirubin (50 μM) on phosphorylation of the phospholemman peptide (30 μM) by PKA (F = 13.01, p = 0.0066; Fig. 2). Essentially the same results were found when the substrate concentration was 10 or 50 μM, respectively (data not shown). When the stereochemical specificity of this effect was examined in a separate set of experiments, both poly-L-lysine and poly-D-lysine (at 300 μM concentration) significantly reduced (Mann-Whitney U' = 49, p = 0.0006 for both) the inhibitory effects of bilirubin (50 μM) on phosphorylation of the phospholemman peptide (50 μM) (Fig. 3), with no differences observed in the ability of the two stereoisomers to inhibit the bilirubin effect (Mann-Whitney U = 17, p = 0.38). In contrast, neither the acidic polymer polyglutamate nor the basic polymer polyarginine modulated the effects of bilirubin on this reaction (results not shown).

Figure 2
figure 2

Poly-L-lysine blocks the inhibitory effect of bilirubin on phospholemman phosphorylation by PKA. Phospholemman peptide (30 μM) was phosphorylated by PKA (2.5 nM) in the absence or presence of bilirubin (50 μM) and poly-L-lysine (100 μM) at 30 °C for 3 min. Reactions were stopped by adding 0.2 M EDTA, the reaction solution was filtered and washed, and the radioactivity in the filters was counted as Cerenkov radiation. □, control (poly-L-lysine alone); ▪, bilirubin; [square with lower left to upper right fill], bilirubin + poly-L-lysine; F = 13.01, p = 0.0066.

Full size image
Figure 3
figure 3

Both poly-L-lysine and poly-D-lysine block the inhibitory effect of bilirubin on phospholemman phosphorylation by PKA. Phospholemman peptide (50 μM) was phosphorylated by PKA (2.5 nM) in the absence or presence of bilirubin (50 μM) and poly-L-lysine or poly-D-lysine (300 μM) at 30 °C for 3 min. Reactions were stopped by adding 0.2 M EDTA, the reaction solution was filtered and washed, and the radioactivity in the filters was counted as Cerenkov radiation. □, control; ▪, bilirubin; [square with lower left to upper right fill], bilirubin + poly-L-lysine; [square with upper left to lower right fill], bilirubin + poly-D-lysine. Contrast of bilirubin vs bilirubin + poly-L-lysine or bilirubin + poly-D-lysine: Mann-Whitney U' = 49, p = 0.0006 for both contrasts. Contrast of poly-L-lysine vs poly-D-lysine: Mann-Whitney U = 17, p = 0.38.

Full size image

DISCUSSION

Our results demonstrate the ability of lysine-containing peptides to modulate the inhibitory effects of bilirubin on protein phosphorylation reactions. This is in agreement with the hypothesis that binding to lysine may be important in the mediation and/or modulation of bilirubin toxicity.

According to this hypothesis, the lysine-containing subdomain II of the catalytic subunit A might be a target for bilirubin effects. Interestingly, the synthetic subdomain II peptide that contained the invariant lysine, although not able to completely prevent bilirubin effects, did induce a partial reversal of the bilirubin effects (Fig. 1). Hence, lysine as opposed to arginine or alanine appeared to have a special effect on bilirubin-PKA interactions. In further studies we therefore characterized the effects of several synthetic poly-amino acid peptides on bilirubin inhibition of PKA. Addition of poly-L-lysine prevented most of the bilirubin effect on PKA activity (Fig. 2). The steric conformation of polylysine (L versus D) did not appear to be important (Fig. 3). Other polypeptides not containing lysine, including both the basic poly-Arg and the acidic poly-Glu were unable to interfere with the actions of bilirubin on peptide phosphorylation.

Inhibitory effects of bilirubin have been demonstrated in a number of different biologic systems and reactions [see Hansen and Bratlid(16) for a review]. Lysine is present in many molecules and structures that participate in reactions or biologic phenomena that bilirubin has been shown to influence or inhibit, although so far a direct connection had not been shown. Our results strongly indicate that bilirubin bound to lysine-containing peptides may no longer have access to the inhibitory sites on substrates and/or kinases. The lysine-binding region of bilirubin may therefore be important relative to its toxicity.

In summary, bilirubin affects many biologic processes, protein phosphorylation is involved in the regulation of many of these processes, and the presence of lysine at or near the active site has been demonstrated for several of the involved enzymes and substrates. The present observations regarding the ability of lysine to modulate and/or mediate the inhibitory effects of bilirubin on protein phosphorylation indicate that binding of bilirubin to lysine may play a role in toxicity.