A single aldehyde group can serve as a structural element for recognition by transmembrane protein CD36

Abbreviations: fl, fluorescently labeled; LCFAs, long-chain fatty acids; oxLDL, oxidized low-density lipoprotein; oxPC, oxidized phosphatidylcholine; PC, phosphatidylcholine.

Abstract

Transmembrane protein CD36 is considered to bind its distinct ligands such as long-chain fatty acids primarily by recognizing their terminal carboxyl moiety. In this study, we provide evidence that long-chain fatty aldehydes, such as oleic aldehyde, can be recognized by CD36. We suggest that a single aldehyde group may also serve as one of the structural elements recognizable by CD36.

Key words:

• aldehyde functional group
• CD36 ligands
• fatty acid-related molecules
• oleic aldehyde
• receptor-binding moiety

CD36 is a member of the class B scavenger receptors characterized by two transmembrane domains, a broad expression pattern, and an ability to recognize a variety of ligands.¹⁾ The primary function of CD36 is considered to participate in the capture (i.e. clearance) of modified lipoproteins and advanced glycation products that would cause damage to tissues.¹⁾ One of the most well-studied CD36 ligands is oxidized low-density lipoprotein (oxLDL).^1–3) In an earlier study, the particles were found to undergo capture by the receptor protein because of the surface localization of distinct forms of oxidized phosphatidylcholine (oxPC) (referred to as CD36-specific oxPC ligands or oxPC_CD36), which have a terminal γ-hydroxyl(or oxo)-α,β-unsaturated carbonyl on the sn-2 acyl group.²⁾ The terminal moiety was postulated to be essential for recognition by CD36, because non-oxidized counterparts of oxPC_CD36 exhibited no detectable interaction with the receptor.²⁾ Later, by a study with a series of artificial constructs of oxPC, a terminal negatively charged carboxylate at sn-2 position was demonstrated to serve as an important requisite for binding with high affinity to CD36.³⁾

CD36 is also believed to participate in the cellular uptake, trafficking, and metabolism of long-chain fatty acids (LCFAs).⁴⁾ It would not be surprising if LCFAs should be directly recognized by CD36, because the lipids have a terminal carboxyl group. We found recently that (i) fluorescently labeled oxLDL (fl-oxLDL) bound specifically and saturably to a chemically synthesized peptide representing the oxLDL-binding site on mouse CD36 (amino acid 150 to 168) (hereinafter CD36 mimic) and (ii) the binding of fl-oxLDL to the mimic was inhibited by unsaturated LCFAs, such as the oleic and linoleic acids, as well as by an oxPC_CD36 with a terminal carboxyl group at sn-2 position.⁵⁾ These data provided one piece of molecular evidence that distinct LCFAs could indeed be recognized by CD36. On the other hand, fatty acid derivatives, oleic alcohol, and methyl oleate had no inhibitory effects.⁵⁾ Based on others’ and our findings,^2,3,5) one can propose that a terminal carboxyl group linked to (or concomitant with) distinct hydrocarbon chains could serve as a structural element recognized by CD36.

Our aim was to identify potential structural elements recognizable by CD36, which correspond to the carboxyl group in LCFAs. For this purpose, we attempted to examine the ability of lipids with only one distinct species of hydrocarbon chain to inhibit the binding of fl-oxLDL to a CD36 mimic, named herein CD36_150–168 (a 22-amino acid peptide with an N-terminal biotin⁵⁾). If certain of test lipids exhibited an inhibitory effect, the acceptability of the non-hydrocarbon chain moieties by CD36 would be indicated. We have previously shown that the degree of inhibition of fl-oxLDL binding to CD36_150–168 varies among unsaturated LCFAs.⁵⁾ Thus, the hydrocarbon chain of test lipids should be identical. In this study, we selected oleic acid as a reference standard and several lipids with oleyl group(s) [CH₃(CH₂)₇CH=CH(CH₂)₇] as test lipids, based on the criteria that they are present in the diets of human (or possibly those of other distinct mammals) or occur in their gastrointestinal tracts as digests.^6–8)

The purity of test lipids obtained from the manufacturers was at least more than 95%. Cholesteryl oleate, cis-11-hexadecenal, 1,2-dioleoyl-rac-glycerol (1,2-diolein), oleamide, oleic acid, oleyl (oleic) alcohol were from Sigma-Aldrich Japan (Tokyo). Trioctadecenoin (triolein) and 1,3-diolein were from Nu-Chek Prep (Elysian, MN). Monoacylglycerols (1-monoolein and 2-monoolein) were from Olbracht Serdary Research Laboratories (Toronto, Canada). Octadecanal (stearic aldehyde) was from Tokyo Chemical Industry (Tokyo). 4-Hydroxy nonenal (4-hydroxy-trans-2-nonenal) dissolved in ethanol at a concentration of 1% w/v was from Cayman Chemical (Ann Arbor, MI). Oleic aldehyde (also referred to as cis-9-octadecenal) was prepared by the chemical conversion of oleic acid. Briefly, oleic acid was once converted into oleic alcohol by treatment with lithium aluminum hydride. The alcohol obtained was purified by silica gel column chromatography and then converted into oleic aldehyde by treatment with 2-iodoxybenzoic acid. The product was purified by silica gel column chromatography and subjected to ¹H- and ¹³C-nuclear magnetic resonance spectrometries and gas chromatograph–mass spectrometry for quality assessment. The purity of oleic aldehyde we obtained was more than 98%. Each of the test lipids (except cholesteryl oleate) was dissolved in ethanol (99.5% v/v, Nacalai Tesque, Kyoto, Japan) at a concentration of less than 3% w/v and stored at −20°C until use. Five milligrams of cholesteryl oleate was suspended in 0.77 mL of ethanol and stored at −20°C until use. The procedures for preparation of fl-oxLDL and CD36_150–168 and the protocol for fl-oxLDL binding and competition assay using the CD36 mimic were exactly the same as described previously.⁵⁾ Briefly described for the assay protocol, the wells of a streptavidin-coated 96-well plate were prewashed with phosphate-buffered saline (PBS, 8 mM Na₂HPO₄, 1.5 mM KH₂PO₄, 136 mM NaCl, and 2.7 mM KCl, pH 7.4) with 0.4% w/v bovine serum albumin (BSA) (PBS–BSA). PBS–BSA alone or PBS–BSA with CD36_150–168 at a concentration of 5 µM was added to each of the wells. Then, the plate was incubated for 1 h at ambient temperature. After washing with PBS with 0.05% w/v BSA, each well received assay mixture (PBS–BSA with 8 µg/mL of fl-oxLDL, 20 µM butylated hydroxytoluene, and 100 µM diethylenetriamine penta-acetic acid) in the presence and the absence of test lipids. Again, the plate was incubated for 1 h at ambient temperature. After washing with PBS, the fluorescence value in each well was measured with a plate reader at an excitation wavelength of 485 nm and an emission wavelength of 528 nm. The statistical analysis (one-way analysis of variance and post hoc Dunnett’s test) and the determination of concentration required for 50% inhibition of fl-oxLDL binding (the IC₅₀ concentration) were performed with the aid of a commercially available software (Prism® version 5.0f, GraphPad, San Diego, CA). A p-value less than 0.05 was considered to be statistically significant.

In this study, we aimed at identifying other potential structural elements than carboxyl group recognizable by CD36, by comparing the effect on the fl-oxLDL binding to CD36_150–168 of oleic acid with those of lipids with one or more oleyl groups, cholesteryl oleate, triolein, 1,2-diolein 1,3-diolein, 1-monoolein, 2-monoolein, oleamide, and oleic aldehyde. These lipids are known to occur in the diets of human (or of other distinct mammals) or in their bodies as digests^6–8) and thus could often encounter CD36. We added the test lipids to the assay mixture at a fixed concentration (1 mM), referring to that oleic acid caused a potent inhibition around the concentration.⁵⁾ We measured the effect of oleic alcohol (a negative control lipid⁵⁾) under the same conditions.

The results of the assay are presented in Fig. 1. The data for oleic acid and alcohol were similar to those shown previously.⁵⁾ Either of the mean fluorescence values obtained using the wells incubated with fl-oxLDL and cholesteryl oleate, triolein, 1,2-diolein, 1,3-diolein, 1-monoolein, 2-monoolein, or oleamide was not substantially lower (even higher in a few of these lipids) than that obtained using the wells incubated with the labeled ligand alone. Indeed, statistical analysis revealed no significant differences in either of the two mean values. Based on the results, we conclude that these lipids exhibit no or little inhibitory effects on the fl-oxLDL binding to CD36_150–168 and suggest that cholesteryl, glycerol, and amide (or amino) groups do not primarily serve as elements recognizable by CD36.

Fig. 1. Analysis of the inhibition of fl-oxLDL binding to CD36_150–168 by lipids with oleyl group(s).

Notes: The wells of an assay plate were pretreated with PBS–BSA alone or PBS–BSA containing CD36_150–168. At least three each of the CD36 mimic-untreated and mimic-treated wells were incubated with an assay mixture containing fl-oxLDL alone (None) or those containing the labeled ligand and any one of the test lipids at a concentration of 1 mM. Note that any of the assay mixtures contain ethanol at a final concentration of 10% v/v, because they received the solvent or test lipids dissolved or suspended in the solvent. We evaluated the effects of these lipids by subtracting the mean fluorescence value for the CD36 mimic-untreated wells from the fluorescence values for the mimic-treated wells. The mean fluorescence value for None was evaluated in the same manner and was set at 1.0. Data are expressed as mean ± SD for at least three independent wells treated with CD36_150–168. Data were analyzed by one-way analysis of variance followed by post hoc Dunnett’s test. *p < 0.05 and **p < 0.01 vs. None.

Oleic aldehyde is a long-chain fatty aldehyde that is reportedly present in smoked fish as an odor component.⁸⁾ The aldehyde is also known to exist in the female gland of Diatraea grandiosella Dyar (a species of moth), although the biological significance in the insect is unclear.⁹⁾ The mean fluorescence value obtained using the wells incubated with fl-oxLDL and oleic aldehyde was about 65% of and statistically significantly lower than that obtained using the wells incubated with fl-oxLDL alone (Fig. 1). We then examined the effects on the binding of fl-oxLDL to CD36_150–168 of oleic aldehyde at various concentrations (Fig. 2). As references, we presented the data for oleic acid and alcohol (Fig. 2). The IC₅₀ concentrations were determined to be 1.1 mM (R² = 0.93) for oleic aldehyde and 0.47 mM (R² = 0.99) for oleic acid (not given for oleic alcohol). These results indicated that oleic aldehyde exhibited an inhibitory effect on fl-oxLDL binding to CD36_150–168, albeit at lower efficiency than oleic acid. Here, we suggest that a single aldehyde is acceptable as a structural element for recognition by CD36.

Fig. 2. Analysis of inhibition of fl-oxLDL binding to CD36_150–168 in the presence of various concentrations of oleic aldehyde.

Notes: The wells were pretreated with PBS–BSA alone or PBS–BSA containing CD36_150–168 was incubated with the assay mixtures containing fl-oxLDL alone or the labeled ligand and oleic aldehyde (CHO) at the concentrations indicated. As references, we determined the concentration-dependent effects of oleic acid (COOH) and oleic alcohol (CH₂OH). Note that any of the assay mixtures contain ethanol at a final concentration of 10% v/v. The fluorescence values for the wells incubated with fl-oxLDL alone and those incubated with the labeled ligand and the lipids at concentrations indicated were determined by subtracting the mean fluorescence value for the CD36 mimic-untreated wells from those for the mimic-treated wells. We were given the plots (i.e. the values relative to that obtained by the wells incubated with fl-oxLDL alone assigned as value 100) and the curves for plots of the data on oleic acid and oleic aldehyde (but not that on oleic alcohol) as outputs through a program of Prism® software for the determination of IC₅₀ concentrations. Data are expressed as mean ± SD for three independent wells treated with CD36_150–168.

To validate our prediction regarding the effect of an aldehyde group, we finally addressed whether there are additional lipids with a single aldehyde group that inhibit fl-oxLDL binding to CD36_150–168. cis-11-Hexadecenal is a long-chain fatty aldehyde that acts as a pheromone in distinct species of insects (Fig. 3(A)).⁹⁾ In Drosophila, detection of the aldehyde has been demonstrated to occur via a mechanism that requires sensory neuron membrane protein-1, a homolog of CD36 in the insect.¹⁰⁾ As shown in Fig. 3(B), cis-11-hexadecenal inhibited the binding of fl-oxLDL in a concentration-dependent manner. The IC₅₀ value was determined to be 0.97 mM (R² = 0.72). This result reinforces our notion about the effect of a single aldehyde. On the other hand, another long-chain fatty aldehyde, stearic aldehyde (known to be used by certain species of insects as a pheromone) (Fig. 3(A)),¹¹⁾ had no inhibitory effect at any concentrations tested (Fig. 3(B)). Note that the data for oleic and stearic aldehydes correspond to our previous observations that oleic acid but not stearic acid inhibited the binding of fl-oxLDL to CD36_150–168.⁵⁾ It has been shown that 4-hydroxy-trans-2-nonenal (Fig. 3(A)), a lipid peroxidation product, displayed no detectable inhibitory effects on the binding of oxLDL to CD36.²⁾ We also had no evidence for inhibition by this hydroxyl fatty aldehyde of fl-oxLDL binding to CD36 mimic (Fig. 3(B)). Altogether, we propose that an aldehyde group as well as a carboxyl one when linked to distinct hydrocarbon chains could serve as a structural element recognized by CD36. Also, we predict that the acceptability of an aldehyde group by CD36 might associate with the electron-withdrawing ability that depends substantially on the adjacent structure. To define such neighboring structures, we are planning to examine the inhibitory effects of several other lipids with a terminal aldehyde group on the binding of fl-oxLDL to CD36 mimic. At present, we cannot exclude the possibility that higher concentrations of candidate CD36 ligands, including oleic aldehyde, only caused a disruption of fl-oxLDL to inhibit the binding to CD36_150–168. Therefore, direct interaction between the candidates and CD36 mimic must also be evaluated.

Fig. 3. Analysis of the inhibition of fl-oxLDL binding to CD36_150–168 by the other lipids with a single aldehyde group.

Notes: (A) Chemical structures of cis-11-hexadecenal, stearic aldehyde, and 4-hydroxy-trans-2-nonenal. For reference, we herein present the chemical structure of oleic aldehyde. (B) Analysis of inhibition of fl-oxLDL binding to CD36_150–168 in the presence of various concentrations of cis-11-hexadecenal, stearic aldehyde, or 4-hydroxy-trans-2-nonenal. The wells were pretreated with PBS–BSA alone or PBS–BSA containing CD36_150–168 was incubated with the assay mixtures containing fl-oxLDL alone or the labeled ligand and lipids with a single aldehyde group at concentrations of 0.5, 1, and 2 mM. As a reference, we simultaneously determined the effect of oleic aldehyde at a concentration of 2 mM. Note that any of the assay mixtures contains ethanol at a final concentration of 10% v/v. The fluorescence values for the wells incubated with fl-oxLDL alone and those incubated with the labeled ligand and the fatty aldehydes at concentrations indicated were determined by subtracting the mean fluorescence value for the CD36 mimic-untreated wells from those for the mimic-treated wells. Plots (i.e. the values relative to that obtained by the wells incubated with fl-oxLDL alone assigned as value 100) and the curves for plots of the data on cis-11-hexadecenal (but not those on stearic aldehyde and 4-hydroxy-trans-2-nonenal) were given as outputs through a program of Prism® software for the determination of IC₅₀ concentrations. Data are expressed as mean ± SD for three independent wells treated with CD36_150–168.

Author contributions

S.T., T.A., and T.F. designed the research plan; S.T. and T.O. performed the fluorescently labeled oxidized low-density lipoprotein binding and competition assay; Y.K. prepared the fluorescently labeled oxidized low-density lipoprotein; T.O. had charge of data processing and computational analysis; S.T. wrote the manuscript with assistance from T.A. and T.O.; T.A. served to procure oleic aldehyde; S.T., T.A., T.O., S.M., K.I., and T.F. analyzed the data; and all authors read and approved the final manuscript.

Funding

This work was supported by the Science and Technology Research Promotion Program for agriculture, forestry, fish- eries and food industry, and JSPS KAKENHI [grant (B) number 25292071] to Tohru Fushiki.

Acknowledgment

We would like to express our sincere gratitude to Mr Yasutaka Ohkubo, R&D Center, T. Hasegawa Co., Ltd., for the preparation of oleic aldehyde.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

Abbreviations: fl, fluorescently labeled; LCFAs, long-chain fatty acids; oxLDL, oxidized low-density lipoprotein; oxPC, oxidized phosphatidylcholine; PC, phosphatidylcholine.