Reconstituted membranes of tumour cells (proteoliposomes) induce specific protection to murine lymphoma cells.

Antigens presented on cell membranes or on liposomes are usually more immunogenic than antigens in soluble form, this being one of the reasons for the weak immunogenicity of extracted tumour-associated transplantation antigens (TATA). The main objective of this study is to solubilize TATA from tumour cells and to present them on a membrane-like structure to the immune system. Crude tumour cell membranes of SL2 lymphosarcoma cells (a spontaneously arising, weakly immunogenic tumour) were solubilized with octylglucoside or sodium deoxycholate, and reconstituted membranes (proteoliposomes) were prepared by detergent removal. Mice immunized s.c. with reconstituted membranes were protected against an i.p. challenge with tumour cells. Although octylglucoside solubilized only 41% of the membrane proteins, the reconstituted membranes were as immunoprotective as crude membranes. (Glyco)proteins were probably the major membrane components in the reconstituted membranes that induce immunoprotection, as mice immunized with preparations constituted of (glyco)lipids from SL2 cells could not reject SL2 cells. If Freund's complete adjuvant was used with the first immunization injection, no potentiation of the elicited immune responses was observed. Besides the membrane TATA, SL2 cells contained an apparently non-membrane-bound TATA, which was found in the cytoplasm. It is concluded that detergent solubilization of membranes and subsequent preparation of reconstituted membranes can be used to obtain membrane tumour-associated antigens that retain activity for induction of protective tumour immunity. The major advantage of this method is that membrane proteins are solubilized and are subsequently presented on a membrane-like structure that resembles the tumour cell membrane. On theoretical and practical grounds it provides a promising alternative for whole-cell vaccines.

31P magnetic resonance phospholipid profiles of neoplastic human breast tissues.

Phospholipids from malignant, benign and noninvolved human breast tissues were extracted by chloroform-methanol (2:1) and analysed by 31P MR spectroscopy at 202.4 MHz. Thirteen phospholipids were identified as constituents of the profiles obtained among the 55 tissue specimens analysed. Observed patterns in phospholipid tissues profiles were distinct, allowing qualitative characterisation of the three tissue groups. Multivariate analysis of lysophosphatidylcholine (LPC) and an uncharacterised phospholipid were shown to be independently significant in predicting benign tissue histology as either fibrocystic disease or fibroadenoma in 92% of cases. Univariate analysis of relative mole-percentage of phosphorus concentrations of individual phospholipids using the Scheffé comparison procedure revealed that in malignant tissues, phosphatidylethanolamine was significantly elevated compared to benign (+ 32%) and noninvolved tissues (+ 22%). Phosphatidylinositol (+ 33%) and phosphatidylcholine plasmalogen (PC plas) (+ 25%) were increased in malignant compared to benign and LPC was decreased (-44%) in malignant compared to noninvolved. LPC was significantly depressed (-39%) in benign tissue compared to normal. Phospholipid indices computed to further characterise the three tissue groups showed PC plas/PC elevated in malignant tissue compared to benign and PE plas/PE depressed in malignant tissue compared to noninvolved. These findings support previous investigations reporting that the alkyl-phospholipid analogues of phosphatidylcholine are released by malignant tissues and that levels of ethanolamine are elevated in malignant tissues. Indices describing the choline-containing phospholipids showed that these lipids are depressed significantly in malignant tissue relative to healthy tissue.