388 B. MAGASANIK 



diately homogenizing the tissue in concentrated guanidine hydrochloride. 

 The latter reagent is an effective protein denaturant. These procedures 

 were found to be applicable to a number of mammalian tissues; details of 

 the methods of preparation follow. 



Fresh or frozen tissue was cut in small pieces and blended for 6 to 8 minutes 

 with 3 vol. per gram of tissue of a 0.15 M sodium chloride-0.02 M phosphate buffer, 

 pH 6.8. A few drops of octyl alcohol were added to reduce foaming. The homogen- 

 ate was then centrifuged at 3000 g for 30 minutes. Essentially all the deoxyribo- 

 nucleic acid was removed in the form of an insoluble nucleic acid-protein complex 

 as described by Mirsky and Pollister.i^ All operations were carried out between 2 

 and 5°. 



To the supernatant solution enough solid guanidine hydrochloride was added, 

 with rapid stirring, to make the solution 2 M with respect to guanidine hydrochloride. 

 The solution was placed in a 38° bath and allowed to stand at this temperature for 

 30 minutes, then chilled at 0° for 1 hour. Under these conditions most of the protein 

 of the tissue extract remained soluble, while a gelatinous precipitate formed which 

 contained ribonucleic acid and a small amount of protein. The precipitate was washed 

 twice with a cold solution of 2 M guanidine hydrochloride (1 vol. per gram of original 

 tissue) and extracted with chloroform-octyl alcohol (5:1). The suspension of nucleic 

 acid in guanidine hydrochloride was added to an equal volume of the chloroform- 

 octyl alcohol mixture, warmed to 40°, then shaken mechanically for 30 minutes. 

 The mixture was centrifuged and the upper aqueous layer containing the nucleic acid 

 removed. The extraction of the aqueous solution at 40° was repeated twice with fresh 

 chloroform-octyl alcohol. Extractions in the cold, or in saline or water solutions, 

 resulted in incomplete separation of the nucleic acid from protein. Nucleic acid was 

 precipitated in the cold from the guanidine solution by adjusting the acidity to pH 

 4.2-4.5 with acetic acid and adding 2 vol. of cold ethanol. The white, flocculent ribo- 

 nucleic acid precipitate was centrifuged and washed twice with cold 70% alcohol. The 

 precipitate was then dissolved in water, carefully adjusted to pH 6.8 with dilute 

 sodium hydroxide, and any insoluble material (denatured protein) centrifuged off. 

 The ribonucleic acid was purified by adding enough 1 M sodium chloride to bring the 

 final concentration to 0.05 M sodium chloride and precipitating the sodium ribonu- 

 cleate with 2 vol. of cold ethanol. The product was washed twice with cold 70% 

 ethanol. 



In the second method the tissue was immediately homogenized with 3 vol. per 

 gram of tissue of cold 2.5 M guanidine hydrochloride solution. The rest of the pro- 

 cedure followed that of the first method, except that the ribonucleic acid-protein 

 complex was washed at least three times with cold 2 M guanidine hydrochloride to 

 ensure complete removal of any contaminating DNA. Excess foaming, which occurred 

 during the blending in the presence of guanidine hydrochloride, was alleviated by 

 adding a* few drops of octyl alcohol after the solution had warmed a few minutes in 

 the 38° bath. 



Dup icate liver ribonucleic acid preparations made bj^ the two methods had an 

 essentially identical analytical composition, indicating that in liver little or no en- 

 zymatic hydrolysis occurred in the first procedure. 



The mammalian ribonucleic acids readily dissolved in water to give clear, color- 

 less solutions. Preparations to be stored were lyophilized from water solutions. 

 Concentrations as high as 20 mg. per cubic centimeter failed to give a reaction with 

 diphenylamine reagent, indicating that all the nucleic acid was of the ribose type. 



