Effects of sphingomyelin headgroup size on interactions with ceramide

Abstract

Sphingomyelins (SMs) and ceramides are known to interact favorably in bilayer membranes. Because ceramide lacks a headgroup that could shield its hydrophobic body from unfavorable interactions with water, accommodation of ceramide under the larger phosphocholine headgroup of SM could contribute to their favorable interactions. To elucidate the role of SM headgroup for SM/ceramide interactions, we explored the effects of reducing the size of the phosphocholine headgroup (removing one, two, or three methyls on the choline moiety, or the choline moiety itself). Using differential scanning calorimetry and fluorescence spectroscopy, we found that the size of the SM headgroup had no marked effect on the thermal stability of ordered domains formed by SM analog/palmitoyl ceramide (PCer) interactions. In more complex bilayers composed of a fluid glycerophospholipid, SM analog, and PCer, the thermal stability and molecular order of the laterally segregated gel domains were roughly identical despite variation in SM headgroup size. We suggest that that the association between PCer and SM analogs was stabilized by ceramide's aversion for disordered phospholipids, by interfacial hydrogen bonding between PCer and the SM analogs, and by attractive van der Waals' forces between saturated chains of PCer and SM analogs.

Figure 1

Thermotropic properties of bilayers prepared from SM analog and PCer (1:1, by mol). Representative DSC-thermograms of the last heating (A) and cooling (B) scans of 12 consecutive cycles (1°C/min) of two independently repeated experiments are shown. For clarity, only the temperature range around the gel-to-fluid transition is shown omitting long stretches of uneventful baseline.

Figure 2

Formation of ceramide-rich domains in laterally heterogeneous bilayers. The ability of the SM analogs to interact with PCer to form ordered gel phase domains was determined by 7SLPC-induced fluorescence quenching of tPA-Cer as a function of temperature. The F₀ samples had the following compositions: 70 nmol POPC, 15 nmol SM analog, 15 nmol PCer, and 1 nmol tPA-Cer. One system consisted only of 70 nmol POPC and 15 nmol PCer and 1 nmol tPA-Cer. In the composition which also included the quencher (F-samples), 30 nmol 7SLPC replaced an equal amount of POPC. The curves are representative of at least two separate experiments for each composition.

Figure 3

Thermotropic properties ordered domains formed in POPC/SM analog/PCer bilayers (70:15:15, by mol). Representative DSC-thermograms of the last heating (A) and cooling (B) scans (1°C/min) of two independently repeated experiments are shown. For clarity, the high temperature region of the uneventful thermogram baseline has been omitted.

Figure 4

tPA-Cer fluorescence lifetimes in bilayers containing the SM analogs. The longest lifetime component (solid circles) and average lifetime (open circles) of tPA-Cer (1 mol %) were measured at 23°C in bilayers composed of (A) POPC/PCer or POPC/SM analog (85:15 by mol), and (B) POPC/SM analog/PCer (70:15:15). Each value is the average from three separate experiments ± SD.

Figure 5

Detection of domain melting in the presence of PC-analogs or DPG. The thermal stability of laterally segregated domains formed by the PC-analogs and PCer, or by the SM analogs and DPG, was determined by 7SLPC-induced fluorescence quenching of tPA-Cer as a function of temperature. The bilayers were composed of (A) POPC/PC-analog/PCer (70:15:15 mol %) and (B) POPC/SM analog/DPG (70:15:15 mol %). In quenched curves, the composition was (A) POPC/7SLPC/PC-analog/PCer (40:30:15:15 mol %) and (B) POPC/7SLPC/SM analog/DPG (40:30:15:15 mol %). Representative quenching curves from two reproducible experiments are shown.