Frank C. Bancroft

A simple technique for the simultaneous measurement of relative levels of a specific mRNA in numerous small samples of animal cells or tissue is described. The technique involves denaturation of cytoplasmic preparations, followed by dotting of up to 96 samples onto a single sheet of nitrocellulose, hybridization with a 32P-labeled cDNA plasmid, autoradiography, and scanning. By analyzing cytoplasmic preparations instead of purified RNA, manipulations of multiple samples prior to analysis is minimized. Experiments with a clonal line of rat pituitary tumor (GH3) cells showed that this technique can be employed to follow the induction by Ca2+ of prolactin mRNA sequences, employing cytoplasm prepared from as little as 2.5 x 10(4) cells. The specificity of the technique for prolactin mRNA was shown by employing GC cells, a GH3 cell variant lacking detectable prolactin mRNA sequences. Experiments with cultured rat hemipituitaries showed that the prolactin mRNA present in cytoplasm corresponding to as little as 1/100 of a pituitary could be readily detected. This technique is quite simple, requires very small amounts of cells or tissue, and permits the simultaneous analysis of multiple samples. Hence, it should be quite useful for studies with various experimental systems of the regulation of specific mRNA levels.

Several established clonal strains of rat pituitary cells which produce growth hormone in culture have been shown to secrete a second protein hormone, prolactin. Prolactin was measured immunologically in culture medium and within cells by complement fixation. Rates of prolactin production varied from 6.6 to 12 microg/mg cell protein per 24 hr in four different cell strains. In these cultures ratios of production of prolactin to growth hormone varied from 1.0 to 4.1. A fifth clonal strain produced growth hormone but no detectable prolactin. Intracellular prolactin was equivalent to the amount secreted into medium in a period of about 1-2 hr. Both cycloheximide and puromycin suppressed prolactin production by at least 94%. Hydrocortisone (3 x 10(-6)M), which stimulated the production of growth hormone 4- to 8-fold in most of the cell strains, reduced the rate of prolactin production to less than 25% of that in control cultures. Conversely, addition of simple acid extracts of several tissues, including hypothalamus, to the medium of all strains increased the rate of production of prolactin six to nine times and decreased growth hormone production by about 50%. We conclude that multifunctional rat pituitary cells in culture show unusual promise for further studies of the control of expression of organ-specific activities in mammalian cells.

Thyroid or glucocorticoid hormone increases the synthesis of growth hormone (GH) by clonal lines of rat pituitary tumor cells. To investigate whether these increases arise from increased accumulation of GH-specific RNA sequences in the cytoplasm and nuclei of these cells, we adapted two existing procedures so that a 32P-labeled hybrid plasmid containing a cDNA sequence could be used to quantitate relative concentrations of the corresponding mRNA. One method (RNA gel blot hybridization) used electrophoresis of RNA, transfer to nitrocellulose paper, and hybridization to 32P-labeled plasmid. The other (RNA dot hybridization) used covalent attachment of RNA to activated cellulose paper squares and hybridization to 32P-labeled plasmid. As probe, we used a hybrid plasmid (pBR322-GH1) which we show by restriction analysis to contain a DNA sequence coding for rat GH. The results were comparable from both techniques and showed that incubation of GH3 cells with a thyroid hormone (triiodothyronine), a glucocorticoid hormone (dexamethasone), or both hormones caused an increase of cytoplasmic pre-GH mRNA sequences of about 4-, 22-, and 13-fold, respectively. Results obtained with the RNA gel blot hybridization method showed that hormonal stimulation leads to the induction of a single 1.0-kilobase species of pre-GH mRNA in the cytoplasm and of 2.7- and 1.0- kilobase species of GH-specific RNA in the nucleus.

Previous determination of the chromosomal location of unique genes have required that the chromosomes of interest be fractionated, either by species-specific chromosome loss from interspecies hybrids, or by physical fractionation procedures. We have developed a general technique for the localization of a unique gene, which requires no prior chromosome fractionation. The technique involves the use of a labeled hybrid cDNA plasmid for direct hybridization in situ to metaphase cells from the organism under investigation. As a model system for development of this technique, we have employed a human alpha-globin cDNA plasmid (JW101) to localize the corresponding gene cluster. To obtain a sufficiently large autoradiographic signal, we have both labeled this plasmid with 125I to a high specific activity (10(9) dpm/micrograms) and taken advantage of the ability of a double-stranded probe to form networks. To obtain sufficient hybridization specificity, various experimental procedures were used, the most important of which was the choice from among a range of probe concentrations of the highest that does not yield excessive background hybridization. We have shown that, with an autoradiographic exposure time of only 12 days, use of this technique correctly localized the human alpha-globin gene cluster to chromosome 16. This technique should be generally applicable to the localization of any gene for which a corresponding cDNA hybrid plasmid is available.

Abstract This account describes a rapid and accurate fusion technique of X‐ray fluorescence analysis that is based on the method of Norrish and Hutton, 1 but modified from the latter in a number of respects to enable more rapid preparation and processing of samples without any loss of accuracy. The technique allows a wide range of materials to be analysed including silicate rocks and minerals, cements, bricks, carbonates, evaporites, soils and some ores.