Lipid biochemists often need to analyze total cellular lipids as well as cellular messenger RNA (mRNA) levels. Conventionally, different cell samples are used for extraction of each class of biomolecule. Here, we describe a procedure for the simultaneous isolation of both total cellular lipids and total RNA from the same sample. The method is based on the fact that mild organic solvents efficiently extract lipids but do not disrupt/degrade cellular RNA. This procedure not only reduces the time and expense of analysis, but also allows a direct investigation of any correlation between lipid and transcript levels. While this method has only been applied in fibro-blasts, prior delipidation of samples may be useful for extraction of nucleic acids from lipid-rich cells such as adipocytes. However, its application to other eukaryotic cell types needs to be tested. The method may not be useful in plant cells or bacterial cells, which are structurally quite different from eukaryotic cells.

Total lipids can be efficiently extracted from cultured cells using an organic solvent consisting of a 3:2 mixture of hexanes:isopropanol (1). While this method has been shown to completely extract various classes of lipids including fatty acids, steroids, phospholipids and triglycerides, it is not efficient for the extraction of gangliosides (2). Treatment of adherent cells with the solvent leaves behind a cell skeleton of nonlipid components from which RNA can be isolated using standard protocols. Cells that grow in suspension can be collected by centrifugation prior to extraction of lipids. This paper demonstrates the validity and feasibility of the technique for the simultaneous extraction of lipids and RNA from the same cell sample.

Normal human foreskin fibroblast (FSF) cells (GM04390) were obtained from the National Institute of General Medical Sciences Human Genetic Mutant Cell Repository (NIGMS). They were maintained and plated as described (3). Confluent FSF cells from each well of a 12-well plate were extracted twice sequentially with 500 L 3:2 hexanes:isopropanol for 13 min each on ice. The resulting lipid extract was evaporated under N₂ gas and resus-pended in 100 L dichloromethane (CH₂Cl₂). Analysis by thin layer chromatography revealed the presence of significant levels of cholesterol esters (CE) and phospholipids (PL) (see Figure 2A). The spot for free cholesterol (FC) was detected only when cells were preloaded with cholesterol. This protocol of lipid extraction is routinely used to quantitate the efflux of cellular cholesterol and phospholipids using cells loaded with radioactive lipids (4,5).

The remaining delipidated cellular material, which remained adherent, was solubilized in TRI reagent (Sigma, St. Louis, MO, USA), and total cellular RNA was isolated according to TRI reagent-specific instructions (6). TRI reagent (400 L) was added to each well of a 12-well plate, and the cellular material from 3 wells was combined for RNA isolation. The total yield of RNA from each delipidated sample was 5.78 ± 1.5 g (n = 34), whereas it was 5.92 ± 1.1 g (n = 38) when the direct RNA isolation procedure was used (Table 1). The quality of isolated RNA was estimated by obtaining a ratio of absorbance at 260 to 280 nm. The average ratio was identical for both methods indicating no compromise on yield or quality of RNA due to prior extraction of cellular lipids. The RNA was also evaluated by agarose gel electrophoresis and ethidium bromide staining. Identical results were obtained in each of multiple experiments. Figure 1A shows a representative gel. Normally isolated RNA (Figure 1A, lane 1) and RNA isolated after lipid extraction (Figure 1A, lane 2), both show typical bands for 28S and 18S ribosomal RNA (rRNA) as expected, with the 28S rRNA band being twice as intense as the one for 18S rRNA (7). Additionally, a higher molecular weight band was detected in the RNA sample prepared from delipidated cells. This band could represent the presence of aggregated RNA due to treatment with organic solvents. However, treatment of normally isolated RNA with 3:2 hexanes:isopropanol did not cause any obvious aggregation or appearance of a high molecular weight band (Figure 1A, lane 3). Another possibility was that delipidation of cells caused co-precipitation of some genomic DNA with the RNA. To investigate this possibility, the RNA sample prepared from delipidated cells was treated with DNase to digest any DNA present in the sample. Treatment with RQ1 DNase seemed to eliminate the high molecular mass band (Figure 1A, lane 4), indicating that RNA isolated from delipidated cells may be contaminated with DNA. However, a significant amount of RNA was lost during the reprecipitation step; thus, the disappearance of the high molecular weight band may be a nonspecific result of lower yield, independent of DNase treatment.