S.L. Bonting, E. Drenthe and F.J.M. Daemen,     Dept. of Biochemistry, University of Nijmegen, Nijmegen, The Netherlands

ABSTRACT

A study has been made of the distribution of the three major phospholipids of the bovine photoreceptor membrane over the two faces of the lipid bilayer.Three different approaches have been used: a. Treatment of isolated intact outer segments and disk vesicles (lysed outer segments) with three different phospholipases and determination of the degradation pattern. As randomized control preparations a retinal lipid suspension and detergent solubilized disks were used, b. Treatment of intact and lysed outer segments with the amino group reagent trinitrobenzenesulfonic acid (with and without treatment by phospholipase D) and determination of modification of phosphatidylethanolamine and phosphatidylserine. c. Determination of the fatty acid composition of the membrane phospholipids during treatment by phospholipase D or TNBS.The results consistently indicate a nearly symmetric distribution of phosphatidylcholine, phosphatidylethanolamine and phosphatidylserine and their fatty acids over the two membrane faces, with a slight preponderance of the first phospholipid on the inner side and possibly a slight excess of the second one on the outer side.

KEYWORDS

Asymmetry

fatty acids

phospholipases

phospholipids

photoreceptor membrane

trinitrobenzenesulfonate

INTRODUCTION

The importance of the phospholipid environment for proper functioning of membrane proteins is by now well documented (see e.g. Sanderman, 1978). Membrane bound enzymes commonly require phospholipids for activity, in some cases like Ca2+-Mg2+ activated ATPase an annulus of about 30 moles of phospholipids per mole of enzyme (Warren and others, 1975), in others like Na+-K+ activated ATPase a larger cloud of at least 90 moles of phospholipids per mole of enzyme (De Pont and others, 1978). In the case of the photoreceptor membrane removal of 90 % or more of the 65 moles of phospholipid normally present per mole rhodopsin severely lowers the thermal stability and the regeneration capacity of the visual pigment and blocks its photolytic sequence after the formation of metarhodopsin I (Van Breugel and others, 1978) in parallel with lateral aggregation of the rhodopsin molecules (Olive and others, 1978), which changes are completely reversible upon detergent-mediated reconstitution with phospholipids. There appears to be little specificity for a particular phospholipid in these three cases (see e.g. De Pont and others, 1978).

It is generally agreed that the phospholipids in cell membranes are arranged in large part as a bilayer (Singer, 1974). For the photoreceptor membrane this has recently been confirmed by means of 31P-NMR studies (De Grip and others, 1979b). This membrane has as its three major phospholipids phosphatidylcholine, phosphatidylethanolamine and phosphatidylserine, comprising 36, 45 and 16 % of total phospholipid, respectively (Anderson and others, 1975).

In recent years studies have been made to determine how the various phospholipids are distributed between the two faces of the bilayer, since this could obviously be of significance for our understanding of the role of phospholipids in membrane function. It has been claimed that asymmetric distribution of phospholipids occurs in various biological membranes (Rothman and Lenard, 1977), although there is controversy in a number of cases especially with regard to intracellular membranes (see e.g. Van den Besselaar and others, 1978). It is clear that different independent methods need to be used with appropriate caution before reliable conclusions can be drawn.

In this paper we describe the methods used by us and the results obtained for the rod photoreceptor membrane, which can be considered to be an intracellular membrane.

EXPERIMENTAL APPROACHES

The methods that can be used include the application of phospholipases, of group specific reagents, of phospholipid exchange proteins and of NMR spectroscopy (Bergelson and Barsukov, 1977). The first two methods have been applied by us so far.

Three phospholipases, phospholipase A2, C and D, have been employed to determine the distribution of the three major phospholipids of the photoreceptor membrane. Phospholipase A2 removes the fatty acid from the glycerol-C-2 position with the formation of a lysophospholipid. Phospholipase C removes the phosphate ester, leaving a diglyceride. Phospholipase D removes the base group, leaving phosphatidic acid. In all cases two-dimensional thin layer chromatography with phosphate determinations of the spots is used to determine the phospholipids and their water-insoluble hydrolysis products (Broekhuyse, 1968). In addition, phosphate analysis of the aqueous layers is performed to allow determining a complete balance of the phospholipid breakdown.

This approach is based on the assumption that these enzymes will not penetrate an intact membrane, and thus will only attack the phospholipids in the outer membrane face. This requires the use of intact outer segments with stacked disks to ensure right-side-out photoreceptor membrane orientation, and the observation of the early phase of enzymatic degradation to minimize the risk of membrane and phospholipid inversion. Fresh preparations must be used, since freezing and thawing may invert the photoreceptor membrane (Adams and others, 1979). In addition, the specificity of the phospholipases toward different phospholipids should be taken into account. Randomized control preparations are used for this purpose.

The other approach involves the use of trinitrobenzenesulfonate (TNBS), which reacts with the free amino groups of phosphatidylethanolamine and phosphatidylserine yielding their trinitrophenyl derivatives. These derivatives appear as discrete yellow spots on the two-dimensional thin layer chromatogram (Gordesky and others, 1975). The approach is again based on the assumption that the reagent will not penetrate the membrane or reverse the membrane orientation or cause phospholipid inversion. This requires selection of suitable reaction conditions and comparison of the effects fects with those on randomized control preparations. In addition to reaction with TNBS alone, the treatment can also be combined with phospholipase treatment before or after reaction with TNBS, which allows a direct comparison of the effects of the two agents.

Finally, the distribution of fatty acids in the phospholipids on either side of the membrane can be determined by gaschromatographic analysis of the methylated fatty acids (Morrison, 1964) present in the TLC spots before and after treatment of the membrane preparation with phospholipase D or TNBS.

EXPERIMENTAL DETAILS