Testing the hypothesis that amphiphilic antineoplastic lipid analogues act through reduction of membrane curvature elastic stress

Abstract

The alkyllysophospholipid (ALP) analogues Mitelfosine and Edelfosine are anticancer drugs whose mode of action is still the subject of debate. It is agreed that the primary interaction of these compounds is with cellular membranes. Furthermore, the membrane-associated protein CTP: phosphocholine cytidylyltransferase (CCT) has been proposed as the critical target. We present the evaluation of our hypothesis that ALP analogues disrupt membrane curvature elastic stress and inhibit membrane-associated protein activity (e.g. CCT), ultimately resulting in apoptosis. This hypothesis was tested by evaluating structure-activity relationships of ALPs from the literature. In addition we characterized the lipid typology, cytotoxicity and critical micelle concentration of novel ALP analogues that we synthesized. Overall we find the literature data and our experimental data provide excellent support for the hypothesis, which predicts that the most potent ALP analogues will be type I lipids.

Figure 1

The lateral stress profile of an amphiphilic molecule, where z is the distance from the interface, τ(z) is the stress at z, where stress is the negative of pressure. Annotations describe the ‘rules of thumb’, derived from the lateral pressure profile, which we have used to predict the structure activity relationships of ALP and APC analogues.

Figure 2

Lyotropic liquid crystal contact preparations of phosphoramide phospholipid analogues, where L₁, micellar solution; L₂, inverse micellar solution; I₁, micellar cubic phase; H₁, hexagonal phase; H_II, inverse hexagonal phase; Q₁/Q₂, cubic phases; L_α, fluid lamellar phase; L_β, solid lamellar phase; C′, hydrated crystal and C, crystal. Divisions show the phases present within the temperature range.

Figure 3

Lyotropic liquid crystal contact preparations of ALP analogues with heterocyclic headgroups. Definitions of terms are the same as given legend of figure 2.

Figure 4

Lyotropic liquid crystal phase diagrams of glycolipid analogues. Definitions of terms are the same as given in figure legend 2. (a) HDPC (2), (b) compound 47, (c) compound 48, (d) compound 49, (e) compound 50, (f) compound 51, (g) compound 52, (h) compound 53, (i) compound 54, (j) compound 55.

Figure 5

A comparison of ED₅₀ against lipid typology for phosphoramide compounds (14–27), lysophospholipids with heterocyclic headgroups (28–38), classic type I amphiphiles (39–46) and glycolipid analogues (47–55). Diamonds, type I; triangles, type 0; crosses, undetermined; squares, type II.

Figure 6

Plot of the ratio of the ED₅₀:CMC for phosphoramide compounds (14–27), lysophospholipids with heterocyclic headgroups (28–38), classic type I amphiphiles (39–46) and glycolipid analogues (47–55). Compounds with a ratio of greater than 1 probably have a lytic effect on cellular membranes. Diamonds, type I; triangles, type 0; crosses, undetermined; squares, type II.