The preparation of ribonucleic acid from yeast, tobacco leaves and tobacco mosaic virus

Patron: Her Majesty The Queen Rothamsted Research Harpenden, Herts, AL5 2JQ Telephone: +44 (0)1582 763133 Web: http://www.rothamsted.ac.uk/

Rothamsted Research is a Company Limited by Guarantee Registered Office: as above. Registered in England No. 2393175. Registered Charity No. 802038. VAT No. 197 4201 51. Founded in 1843 by John Bennet Lawes.

Rothamsted Repository Download

A - Papers appearing in refereed journals

Holden, M. and Pirie, N. W. 1955. The preparation of ribonucleic acid

from yeast, tobacco leaves and tobacco mosaic virus. Biochemical

Journal. 60 (1), pp. 46-53.

The publisher's version can be accessed at:

•

https://dx.doi.org/10.1042/bj0600046

The output can be accessed at:

https://repository.rothamsted.ac.uk/item/8wyx0

.

© 1 May 1955, Portland Press Ltd.

46

M.

HOLDEN AND N.

W. PIRIE

I955

The action catalysed by extracts from leaves

(Pirie, 1950) and sprouted barley (Shuster & Kaplan, 1953) resembles that catalysed by the enzyme from spleen and some snake venoms in being moreextensivethan the actioncatalysedby PRNase but the nature of the end products has not yet beenestablished. Parker(1952), on the basis of

the solubility of the Ba salts, recognized

mono-nucleotides in LRNase hydrolysates but did not

characterise them, and Shuster & Kaplan (1953) mention preliminary evidence that 5'-nucleotides

are a product of the action of the barley enzyme, which suggests that this enzyme, unlike 1dRNase,

splits the molecule between the phosphate and

carbon-3 on the ribose. There is no published

evidence that mononucleotides are thesole,oreven theprincipal,products oftheaction andtheremay beoligonucleotides thatdifferfromthose remaining after PRNase action by being soluble in UrTCA.

Finallythere is no evidence that only one type of RNase is present in leaf extracts; it is only for con-venience that we havereferred to LRNase as if it were oneenzyme. All that is clear is that LRNase is not anunspecific phosphodiesterase because it is

onlycrudepreparations that carryactivity towards

diestersunrelated to RNA.

This fractionation was undertaken to ascertain the range of substratespecificityof LRNase and it hasonly been carried farenoughtosatisfy us that

phosphate esters and polynucleotides other than

those containing ribose are not attacked.

Inci-dentally, the activity of theenzyme per mg. of N has beenincreased230-foldandthefinalproductis as active on suitable substrates asare crystallized preparations of the pancreatic enzyme. At this stagethereis amarked diminution inthestabilityof LRNase so that, for our purpose, there is no ad-vantage incarrying thefractionation further.The final preparations areobviously inhomogeneousand none of the usual criteria ofpurityhaveyet been

appliedtothem. Further workonthe fractionation

is in progress.

SUMMARY

1. From peaseedlings ribonucleasepreparations have been made which attack P-containing sub-stratesother thanribonucleic acidso slowlyas to make itunlikelythat the enzyme has anunspecific

action.

2. The enzyme differs from pancreatic ribo-nuclease in that ithydrolysesnucleicacid so com-pletely that noacid-precipitable'core' isleft.

3. Less thoroughly fractionated preparations have been made from tobacco leaves and some properties of the enzyme in other leaves are described.

REFERENCES

Axelrod, B. (1947). J. biol. Chem.167, 57.

Bheemeswar, B. &Sreenivasaya,M. (1953). J.8ci. indudtr. Bes. 12B, 529.

Boroughs, H. (1954). Arch. Biochem. Biophys. 49, 30. Brown, D. M., Heppel, L. A. & Hilmoe, R. J. (1954). J. chem.

Soc. p. 40.

Durand, M. C. & Thomas, R. (1953). Biochim. biophy8. Acta, 12, 416.

Euler,H. &Josephson, K. (1927). Liebigs Ann.452, 158.

Folin,0. &Ciocalteu, V. (1927). J. biol. Chem. 73, 627. Frisch-Niggemeyer, W.,Keck, K., Kaljunen, H. &

Hofmann-Ostenhof,0. (1951). Mh. Chem. 82, 758.

Holden,M. (1952). Biochem.J.51, 433.

Holden,M. & Pirie, N. W.(1955a). Biochem. J. 60, 46. Holden, M. & Pirie, N. W. (1955b). Biochem. J. 60, 53.

Jono,Y. (1930). ActaSch.med.Univ.Kioto, 13,162, 182.

Keilin,D. & Hartree, E. F. (1951). Biochem. J. 49, 88.

Markham, R. & Smith, J. D. (1954). In The Proteins, vol. IIA,p. 1. Ed. H. Neurath& K.Bailey. Academic PressInc., N.Y.

Parker,G. (1952). Biochem. J. 51, 389. Pirie, N. W. (1950). Biochem.J. 47, 614.

Schlamowitz,M. &Garner, R.L.(1946). J.biol.Chem.163, 487.

Shuster,L. &Kaplan,N. 0.(1953). J. biol. Chem.201,535.

Singer,T.P.&Kearney,E.B. (1950). Arch. Biochem.29, 190.

Zamenhof, S.&Chargaff,E.(1950). J.biol. Chem.186, 207.

The

Preparation of Ribonucleic Acid from

Yeast,

Tobacco Leaves

and

Tobacco

Mosaic Virus

BY MARGARET HOLDEN AND N. W. PIRIE

Rothawmted Experimental Station, Harpenden,Herts

(Received 12October 1954)

There is no conclusive evidence thatnucleic acid everexists in vivo inthe free state. Markham (1953) has argued that in some situations, e.g. turnip yellow mosaic virus, it is free and simply held as aclathratecomplexinside a protein cagesothatit is

RIBONUCLEIC ACID FROM THREE SOURCES

linkage appears to be stronger. At one time,

apparently(cf. Sevag&Smolens, 1941;Greenstein, 1944),thelinkagewasthoughttobeelectrovalent. Ifthis viewwasindeedheld itwasbaseless because

nucleic acid had only been made fromtissuesafter

treatmentsthat denaturemanyproteins, andthere was already evidence from work on plant virus

preparationsthat in them,atanyrate,the linkwas

stableover awide pHrangeinmanydifferentionic environments.

If,therefore,weassumethatmost,evenifnotall, ofthe nucleic acid in acell is normally linked to

other materials, such as protein, by more than

electrovalencies, the free acid is largelyorentirely anartifact and thetypeof change it undergoeson

liberation depends on the at present undefined

natureofthe links involved. The position is

com-parableto that of haematin and globin in haemo-globin. Nucleic acid can be liberated in many

differentways, e.g.bytreatmentwith acid, alkali,

urea,guanidine, ethanoland byboiling; itmaywell

bethatthesecausedifferent changes in the nucleic

acid whichareincomparable. There is,therefore,no

basis for trying to arrange the treatments in a

sequencerepresenting theextentof alteration. The

common assumption that those preparations with

thelargest particle size are the least modified has nonecessaryfoundation; one cause ofapparently

large size is incomplete removal of protein andsome

treatmentsmaywellcauseaggregationrather than

depolymerization.

Theagentsthatsplitnucleoproteinsare also,to some extent, selective and for each particular nucleoprotein some aremoresuitable than others.

Thus, trichloroacetic acid works with the

micro-somes from normal tobacco leaves but not with tobaccomosaic virus (Pirie, 1950), whereas

stron-tiumnitrateworks with thelatter butnotthe former (Pirie, 1954). Similarly, detergents and urea

separate nucleic acid from some viruses but not

fromothers(Sreenivasaya & Pirie, 1938; Bawden&

Pirie, 1940). It is reasonable to suppose that a

starting material as complex as plant or animal

tissue or a yeast cell, will contain very many

differenttypesofnucleic acid held inmanydifferent

typesofcombination. Unlessallthe nucleic acid is

being isolated it is therefore reasonabletosuppose

that each method willbringoutthe nucleic acids in

adifferent ratio andthismaybethecauseofsomeof

theconflictintheliterature.

Whennucleicacid is isolateddirectly fromacell

without intermediate isolation ofanucleoprotein,

the cell wall must be broken or damaged before evenanuncombined nucleic acidcanbereleased.

Thispresentslittledifficulty withmostanimal and higher-plant tissues for the cellsareeasilydestroyed

by grinding. Yeast is more robust, but Chargaff

etal. (1950) haveground it in a mill designed for

bacteria. In other methods the yeast is dried and extracted with fat solvents or heated to 60-1000

(Clarke & Schryver, 1917; Markham & Smith,

1952a). The efficacy of these treatmentsprobably

depends notonly on the denaturation of the protein and consequent liberation of nucleicacid, but also onthe breaking of the cell wall. Yeast cells can be broken openbydigestionwith theenzyme mixture from snails' crops under the conditions used with leaves byHolden,Pirie &Tracey (1950) but there is ribonuclease in the crop fluid and we have not found this treatmentsatisfactory as aprelude to making nucleic acid. A more refined enzyme preparation

would, however, probably be admirable. The traditional method for opening yeast cells is ex-posure to 0-5 or 1ON-NaOH at 00;this treatment also denatures the protein sufficiently to make it

largely insolubleinthe presence of salts atpH6-7. There are disadvantages in working with nucleic acid that has been exposed to such an alkaline environment but there seem to be comparable disadvantagesinall the other treatments.

The result ofany of these processes is a solution

containingmore orlessmodified nucleic acid and a variable but small amount of protein as well as othermaterials, polysaccharides, metaphosphates,

etc., of varied molecular weight. During the last 15 years the protein has generally been removed

by shaking with chloroform under various condi-tions derived from the method developed by

Tsuchihashi (1923)andpopularized bySevag (e.g.

Sevag, Lackman & Smolens, 1938). The emulsion

thatisformed containsahigherratio ofproteinto nucleic acid than the aqueous layer, so that by

repetitionoftheprocess most ofthe proteincanbe removed without great loss of nucleic acid. It shouldbeemphasized,however, that thisis potenti-allyamethod offractionatingthe nucleic acidtoan extent that willdepend onthe amount ofprotein beingremoved. It has been condemnedby Jungner

&

Allgen

(1950) as

leading

to depolymerization of the nucleic acid.

These commentsapply

equally

totheribonucleic

acids and the deoxyribonucleic acids; in the

re-mainder of this paper theunqualifiedtermnucleic

acidwill be used for ribonucleic acid

only.

Most published work has been done with

com-mercial yeast nucleic acid (YNA), although the

authorsgenerallycomment onits poor andvariable quality. Abrief search in recent patent literature showsthat the acid mayhave beenexposedtosuch

savage treatment thatthese strictures are

amply

justified. Many methodsforpreparingYNA in the

laboratoryhave beenproposedand inthem treat-ments have been advocated and condemned on

grounds that do not always seem tobe based on

experiment. We have tried to

keep

conditions as

mildasiscompatiblewith

getting

a

high

yield

and

we show that ourproduct is as highlyaggregated as those of otherworkers and that a repetition of some of the treatments does not lead to further break-down. This suggests that the breakdown is not a progressive matter but it does not exclude the

possibilitythatvulnerable structureswere destroyed atthestartofoperations. As we have said,. we look onnucleic acid as an

artifact;

if wearecorrectthere isnopossibilityofseparatingunmodified material, and the best that can behopedfor is tominimize the

modification.

It issometimes convenient to start with commer-cialYNArather than with yeast; we therefore give amethod forseparating fromit arelatively satis-factory product. For comparison with YNA we have also made nucleic acid from normal tobacco leaf and from tobacco mosaic virus by methods similar to that used with yeast.

EXPERIMENTAL AND RESULTS

Preparation ofribonucleicacidfromyeast

Pressed baker's yeast is

su§pended

evenly in its own weight of water, cooled to00and anequal volume of 2N-NaOH, also at

00,

is added with vigorous

stirring. After 1 hr. at00the mixture is neutralized

bythe addition of5N-HClwith vigorous and con-tinuousstirring. Greatcareistakento prevent any regions from becoming acid and the final adjust-ment to pH 6-0-6-5 is made with more dilute acid. Most of the insoluble material is removed by

centrifuging at 1500g. The supernatant fluid is

adjustedtopH 3 with HCI, 70 g. NaClareadded to each11. offluid,andthesmallprecipitate removed

byfilteringthrougha layer of Hyflo Supercel (Johns Manville Co.). To the clear filtrate solid NaCl is added to full saturation; a turbidity develops at onceandprecipitationstarts in a few minutes and continues for several hours. After standing for a

day the precipitateiscentrifuged off. More material is precipitatedwhenthe pH of the sat. NaClsoln.is

adjustedto 2. Eachppt. isdissolved separately in 20-30 times its volume of water and the pH

ad-justedtoabout 4. The preparationsare deprotein-ized by stirring with 0-25 vol.

CHC18

and 0 1 vol.

amyl alcohol in an 'Ato-mix' (Measuring and ScientificEquipment Co.). The emulsion is

centri-fugedandthe lower layers discarded. At least three

cycles oftreatment are necessary. The solution is thendialysedfor several days at0°,against frequent

changesofdistilledwater, inacellophansacprotected

fromburstingby asleeve of stainless-steel

gauze.

The dialysedpreparations of Na nucleate contain

8'0-8

5% P and we have no evidence for any

P-containing component besides ribonucleic acid.

Thus, lipid solvents do not extract any P and on

precipitation with HCI, although 1-2

%

ofthe P remains in solution, the light absorption of this

solution at 260m,. is approximately that found with solutions ofpartly degraded nucleic acid with the same P content. We therefore conclude that

little phospholipid or metaphosphate is present. The questionofdeoxyribonucleic acid contamina-tion is considered later.

The Fe content of both thesepreparations is less than 0 01

%.

TheNaClprecipitateatpH3contains

0 1% of Ca and Mg but the precipitate at pH 2 contains less than 0-03

%.

For these metal determi-nations,the nucleic acid was incinerated withH2SO4

until charred and then cleared withHC104. After dilution, Fe was determinedcolorimetrically after adding NaCNS; Ca was precipitated at pH 3 as the oxalate and this, afterwashing, was titrated with KMnO4- Mg was determined colorimetrically on

the supernatant fluid from Ca oxalatebyamethod based on that ofLudwig&Johnson(1942)in which the lakegiven with Titan Yellow in alkaline solution isstabilized by starch. Control determinations in which 10mg. quantities of YNAwereincinerated after the addition of 1-10 ,ug.quantitiesof the three

metals showed that these methods were

satis-factory with this material. We do not therefore agree with those who have claimed thatmetals,and

especially Mg (cf. Jungner, 1951), are an integral

part of YNApreparationsmadeby gentlemethods of isolation. The small amounts that we find, in preparations in which the primary valencies are neutralizedby Na+,may well be derived from the reagents used in thepreparation.

Of the various methods used to detect small amounts of protein in materials such as nucleic acid,those based on the biuretreactionappeartobe

theleastunsatisfactory. Of themwe

prefer

those in which the colour isdevelopedinasmallvolume and

thencompared

qualitatively

withasetof standards rather than those in which it is developed in a

larger volume and measured in a colorimeter. Osborne'stechnique(cf. Markham,1955)is the most

sensitive, buttogetsatisfactoryresultsweuseless

ethanol than is recommended by Markham. To 1ml. of aqueous nucleic acid solution 0.05 ml. of 10g./l. CuSO4, 5H20 and 0-3ml. of ethanol are

added. The colour is developed and the ethanol forced out of solutionby theaddition of

approxi-mately0-9g. of KOH. The biuretcolour extracts

intothe ethanollayerwhereasmostof the

uncom-bined copper and many formsofcolour in theoriginal solution, which confuse other methods of doing

biuretreactions, remain inthe aqueous

layer.

By

thismethod60

pg.

of casein is

clearly

seenand 30,ug.

is perceptible. With 3-5 mg. of nucleic acid the presenceof 1% ofproteinis

readily

detected. The fraction of YNA thatprecipitatesat

pH

3 contains 1% of protein or a little less; the fraction

pre-cipitatingatpH2contains much less than1

%

and sometimes nocolour isdetectable.

RIBONUCLEIC ACID FROM

THREE SOURCES

Purification

ofcommercialnucleicacid

The contaminants that interfere most with

enzymemeasurements are degraded material and

metals. The first must be removed because it is incompletely precipitated by thereagents usedto

precipitate YNA; this is partly achieved by

re-precipitation with diluteHCIorconcentrated acetic

acid. Fe is not removed in this way, and Zittle

(1946) found that Cuwasnoteither.

Commercial YNA is suspended in water and

neutralized with NaOH to give a 60-100g./l. solution; this is dialysed in a sac protected by a

stainless-steel gauze sleeve. During 1-2 weeks at

0° the outside water is changed frequently. The

amountofmaterial diffusingawayvaries from batch

tobatchbutasmuchas80%maybe lost. Protein,

and alsosomenucleic acid, is removed by shaking

with CHC13 and amyl alcoholasintheothermethod.

NaCl is addedto the fluid at pH4and at a

con-centration of 100-150 g./l. precipitation begins; it finishes inafew hours and the solid iscentrifuged

off. NaCl is addedtosaturation and another

pre-cipitate is separated off;athirdcomesoutwhen the

pH isadjustedto2.The three fractionsaredialysed

separately; they contain approximately equal

amounts of nucleic acid and together account for about half ofthe Ppresentinthe initially indiffusible material.

Itwould obviously bemoreconvenienttoreverse

thesequence sothatthere is only onedialysis but

with the commercial products thatwehavehandled,

there islittleor noprecipitationonsaturationwith

NaClatpH 3-4 unless the diffusible material has first been removed. The fraction most readily precipi-tated by NaCl is the most highly coloured and containsthemostFe.Thusaproduct which, after

simple acid precipitation at the stage before the NaCl precipitation, contained 120 atoms of P for eachatomofFe,gaveNaClprecipitates with P/Fe

ratiosrising from 40to350. The latter ratioamounts to 0 025% Fe and this is more than we find in

preparations made from yeast directly. Purified commercial preparations have invariably been

more highly coloured than those made directly

fromyeast,buttheyareequallyfree fromprotein, CaandMg.

Precipitation by other 8alts. We have found no moreconvenient precipitant than NaCl.

Precipita-tionby NH4C1 is similartothatwithNaCl; KC1and

NaBrarelesseffectiveatthe samesaturation and

(NH4)2SO4and

Na2SO4

aremuchlessso. Mgand La

salts are well known precipitants for the nucleic

acidsbutappeartobe lessselective.

Preparations from tobacco-leafnucleoprotein Nucleoprotein (NP) was made from the sap of

young uninfected tobacco leaves by differential

4

ultracentrifuging (Pirie, 1950) and used while still

fresh. It can be dissociated into free nucleic acid and denatured protein in many ways. NaCl pre-cipitation of the nucleic acid is satisfactory after fission by pouring a solution containing

approxi-mately 10g./l.of NP at pH 8 into an equal volume ofboiling 0-1 M-NaClthat iskeptboilingduring the addition, orby exposing the NP solution to

0-5N-NaOH for 40 min.at00,orto 50g./l.trichloroacetic acid (TCA) at room temperature for 8-12 hr. (Pirie, 1950). Most of the denatured protein is easily filteredfrom the boiled preparation; from the alkaline one it can be removed by adjusting the pH to 7 orby addingone-fifth volume of saturated

(NH4)2SO4 solution; after TCA fission the nucleic acid is extracted from the protein precipitate at pH 8. The three types of extract now behave similarly. A small amount of protein, accompanied by only alittlenucleic acid, precipitates when the

pHisadjustedto 4. This iscentrifuged off andthe fluid half-saturated with NaCl. There is an immedi-ateopalescenceandprecipitationiscompletein afew hours. This nucleic acid precipitate is centrifuged off and the supernatant is saturated with NaCl;

again an opalescence is followed by precipitation.

Fromthesupernatantafurtherquantityof nucleic acid separatesslowlyontheadditionofHCI.

The three products are dissolvedseparately at

pH4and shaken with

CHC13

andamylalcohol; if

the removal of denatured protein after the first stage of the preparation was satisfactory, little

emulsion is formed. The process is repeateduntil there is none, and the supernatant fluids are

dialysed.

For reasons that are not understood but that maywell be connected with thephysiologicalstate

ofthe originaltobaccoplant, the propertiesof NP

are not constant in different preparations. The P contentvaries from 15 to 40

%.

Ingeneral,

pre-parationswith low P contents are accompaniedby chlorophyll-containing particles and contain up to 10

%

oflipid.With starting material of this type the

separationof denaturedproteinfrom nucleic acid is easierand morecompleteifthelipidisremovedby precipitating the nucleoprotein with 10 vol. of a mixture ofequalpartsof ethanol and etherbefore

fissionbyboilingorexposure to alkali. AfterTCA fission it is easier to remove thelipid byextracting theprecipitatewithethanol-etherbeforethe extrac-tion at pH8. There is no apparent advantage in

removing the small amount of lipid present in

preparations containing 2-7

%

P or more. This variation in the starting material may in part explain the lessregular distributionof thenucleic acid amongthe differentfractionsderivedfromNP

thanfromother nucleicacid sources, butthepoint

has not yet been studied in detail because these

fractionations aredifficulttoreplicate exactly. The

Bioch.1955, 60

result of arepresentative experimentin which the

three types of fission were compared on one

pre-parationof NPare set out inTable1.Three of the

differences that appear in this table have been found consistently in many comparisons; TCA

fision is least effective in separating nucleic acid fromprotein, NaOH fission leaves more of the P in a form that is not precipitable by either NaClor

HClthantheother two, andthebestyieldof

NaCl-precipitablenucleic acid isgiven by boiling.

Table1. Distributionofphosphorus between

fractionsafter

fission

ofNPby three methods

Descriptionofthefraction

ProteincoagulumatpH7 ProteinprecipitatedatpH4

Protein removed from the nucleic acidpreparations byCHC13+amyl alcohol Both NaCl precipitates

HCI precipitate Final supernatantfluid

Percentageof theP inthe original NPthatappears

finallyineachfraction after fission by

Exposure

TCA Heating toalkali

20 11 10

Itisclear from Tables1and 2 thattherearelosses

ofP inthe initialproteincoagulum, inthe

precipi-tates

separated

atpH 4,and in the emulsionthat is

made by shaking with CHC13 and amyl alcohol.

With NP theselossesareserious. A further

quantity

ofnucleic acid can be made by suspending these protein fractions in water atpH 9-10 and adding

NaCl or preferably

(NH4)2SO4

to 0-1 saturation.

Mostoftheprotein coagulatesandbrings outvery little

P,

moreis removed atpH4andthenucleic

acidcanthen beprecipitated by addingHC1; it is

purified

asbefore.

Table 2. Distributionofphosphorusbetween fractions after fission of TMVby twomethods

t]

a]

fi

4 3 7 Descriptionof the fraction

5 7 3 Initialproteincoagulumand precipitate atpH 4

Precipitateonsaturation withNaCl 27 32 20 PrecipitateonaddingHCI to lower

19 30 35 pHto2

18 15 23 Final supernatant fluid

'ercentage of the P in heoriginalTMV that

6ppears finallyineach raction after fission

by

Exposure Heating toalkali

2-3 1-5-2

70 28

14 43

14 27

Preparation from tobaccomosaicvirus(TM V)

Virus prepared by differential ultracentrifuging

only, contains variableamounts ofP that canbe

separatedby incubation with trypsin; preparations

made by precipitation with (NHI4)2SO4 and acid

(Bawden& Pirie, 1943)are free from thisandare,

therefore, preferableassourcesofvirusnucleic acid.

Nucleic acid is most conveniently separated from

theproteinbyheating (Bawden & Pirie, 1937), but when separatedby various othermethodsoffissionit isalsoprecipitable by NaCl. In Table2theresullts

aresummarized ofanexperimentinwhichhalfof a viruspreparation was denatured byheating for

3min. at 1000 in 0-1 M-NaCl and halfbyexposure

for 30 min. to 0-5N-NaOH. Thereafter, the treat-ment was similar to that used with NP. The differences are clear. Afterheating, alarger

pro-portion of the nucleic acid isprecipitated by NaCl thanafteralkalinefission, and, in agreement with

earlierobservations(Gr6goire, 1950),theproportion of the nucleic acid that remains unprecipitated

evenby HCl is large when NaOHasdiluteas0-5N

is used for thefimsion. If stronger alkali is usedthe proportionprecipitated by NaCl is however smaller. Only 2-3%ofthe nucleic acid is lost when the NaCl orHCIprecipitatesareshakenwith CHC13 andamyl alcohol to remove residual protein, but the final

productshavenotbeensofreefrom proteinasthose

made fromyeast.

Composition ofNP and TMV nucleic acid

pre-parations. Nucleic acid preparations made from NP and TMV are rarelysofree fromnprotein, deter-mined by thebiuret reaction, as those made from yeast. Thefractionsprecipitating withNaCl have contained 1-3

%

and thosesubsequentlyprecipitated

by HCI contain 1% or a little less. The dried Na salts contain7-5-8-0% P. Because ofshortageof

material,metaldeterminationshave notbeenmade onmany preparations. Furthermore, the nucleo-proteins used for making them were not purified

with any special precautions to avoid accidental

contaminationfromtraces ofmetal inthereagents. We have, therefore, no reason to think that the

0-10-0-16%

Ca,

0-01-0-03%

Fe, and less than 0-1

%

Mgfound in thepreparations precipitatedby

NaCl are anintegral part ofthenucleic acid. The metal contents of preparations made by HCI precipitationwerelower.

Thepresence of

deoxyribose

nuleic acid in ribose nucleic acid preparations. The diphenylamine re-action (Dische, 1930) gives acolour which canbe

readily seen, or measured, on a photoelectric colorimeter, with an amount of deoxyribonucleic

acid (DNA) which contains 1

jig.

of P. 3 mg.

RIBONUCLEIC

ACID FROM THREE

SOURCES

from NP have never given as littlecolour as those from yeast but the colour is stillsofaint as tobe uncertain. Preparations from TMV, on the other hand, have always given a distinct colour; the intensity varies irregularly in the different fractions

and corresponds to the presence of 0 5-2*0 % of DNA. It is notcertain that DNA isthe cause of this colour, but the possibility that it is due to the

presence of pectin or its breakdown products

(Holden, 1953) seems to be excluded by the fact that the intensity is not increased bypreliminary

acid hydrolysis. Furthermore, Hoff-J0rgensen (1952), by an entirely different method of assay, found DNA in a TMVpreparation. TMV prepara-tions arecommonly contaminated by a wide range of materials (cf. Pirie, 1949, 1953); we do not, therefore,claimthat DNA is anessentialpartof the structureof TMV, and work is in progre#s to find out

whether pretreatments thatdonotrobthevirusof

itsinfectivity will remove it.

O8motic pressure meaeuremente. The osmotic pressures exerted by these nucleic acid solutions weremeasured in simple osmometers with cellophan membranes at 00. The level of the meniscus was

observedregularly until equilibrium wasreached;

the osmometer was then disequilibrated and

equilibrium attained from the other direction. Each process took 3-4 days and from the final value the height of capillaryrise ofthe same fluid was

subtracted. The concentration of nucleic acid was

determined from theP contentofthefluidinside the

sac

atequilibrium. All measurements were made at

pH6 and,todiminish theosmotic pressure sot up

bythe uneven distribution of ions withthis highly chargedparticle, M-NaCl wasused as the outside fluid. Measurements made withthe same prepara-tionof nucleicacidagainstdifferentconcentrations ofNaCl showedthat there was no further diminution ofthe pressure when theNaCl concentration was

increasedabovethisvalue.

InTable 3 the resultsaresetout. For convenience values for the 'molecular weight' are included; these were calculated on theassumptionthat all the preparations contained 8-5

%

P andthatthe van't Hofflawapplies althoughthe applicability ofthe law to systemswhich,like this one, containhighly

charged, hydrated, and anisometric particles has

been abundantlycriticized.The'molecular

weight'

figureisonlyincluded to show that these

prepara-tions fall within the range found by others with

YNAthat has not been subjected to extreme condi-tions. Thevalues werecalculated by the

approxi-mation:

'molecularweight'

Pcontent ing./l. x

2.6

x 106. height of fluid column inmm.

YNA'core' is the part of aYNApreparationthat isresistant to attack by PRNase and was made by

the method outlined

ip

an accompanying paper

(Holden & Pirie,

d1955a).

Two preparations have been used and, although it isclearthat theparticles

are smaller than those of the parent YNA, these

measurements do not support the conclusion

reached by Markham & Smith (1952b) that the particlesof core have ameanchainlengthofonly

four residues. These osmotic pressure measurements obviously only give evidence about the degree of dispersion of thepreparationsinM-NaCland this is not the

environment

used in our enzyme

experi-ments or inthose ofMarkham& Smith.

Further-more, there may be anequilibrium between oligo-nucleotides and aggregates of them. The low value for the osmoticpressure should, however,betaken into account in any attempt to decide on the nature of all theinternucleotidelinks in YNA.

(hange8

in the properties of nucleic acid brought

about by thetreatment8usedin thei8olation.We have

already pointed out that there is no source of

'native' nucleic acid that can be used as astandard

onwhichtofollowthe effects ofthevarious

treat-ments used. The only

apparent

alternative is to

repeat, onisolatedproducts,thetreatmentsused in

the isolation. This procedure would clearly

only

recognizeaprogressivechangewhich hadnotgone tocompletion during theisolation.

It is

well

known

that the exposure of YNA to

alkalineconditions for a few hoursat room temper-atureorafewdaysat

00

diminishes its acid

precipit-ability.This treatment also diminishesits precipit-ability by NaCl. The pH of the yeast suspension duringthe treatment withNaOHis 11-3;wedonot

Table3. The

o8notic

pressureof variowus types of nucleic acid

Type ofpreparation

YNA, NaCl ppt.

YNA,HCIppt.

YNA, purifiedcommercial YNA, 'core'

Tobacco-leafnucleicacid,NaCl ppt.

TMV nucleicacid, NaClppt.

TMVnucleicacid,HCIppt.

P content ofsolution

(g./l.)

1-8 2-34 1-36 0-308

0-52

0-44 0-41

Increasein

height of

capillaryrise (mm.)

58 129 104

44

38 17 27

'Molecular weight'

81000 47000 34 000

18000 36 000 67 000 39500

find any change in the precipitability of YNA at pH 3.5byNaClafterexposurefor 60 min. at00 to thispH and a change is only perceptible after 3 hr.

Similarly, precipitation with HCl at pH 1-7 and exposureto thatpH for the 3-5 min. necessary to

centrifugedownthe precipitate had no effect onthe

osmotic pressure of several preparations that had not, untilthen, been exposed to such acid conditions. Thechangesthat we recognize during the course of preparing YNA are in the other direction, for the process ofprecipitation with NaCl makes reprecipi-tation more easy. Thus material that has precipi-tated atpH

3*5

and

0*5

saturationwith NaClwill

partly reprecipitate at the same pH and 0-33 saturation, while material that did not precipitate till full saturation will partly reprecipitate at 0 5 saturation. Preparations from which the protein has beenremoved by shaking withCHC13and amyl alcohol are largely precipitable by as little as 0-2 saturation, a concentration of NaCl with which little or nothing can be precipitated from the original

extract fromyeast, NP or TMV.

DISCUSSION

It issurprising that the apparently general

applic-ability to the preparation ofnucleic acids of

pre-cipitation byNaCl and some other salts in slightly acidsolutionhas notbeenmorewidely commented on. Passing remarks in a few papers suggest that it

has been noticed by others but, so far as we are aware, its use as a preparative procedure has not been advocated. The method shares with those

dependingonprecipitation with ethanol orMgSO4 themeritthatexposure to alow pH isavoidedand

this may well beimportant. Butthe cycleof

pre-cipitation with HCI and re-solution at neutrality need not occupy more than 2min. and can be carried out at

00;

there is no

published

evidence

thatsuch treatment degrades nucleicacid,and, as wehaveshown,it is aneffectivemethod of separat-ing CaandMgfromthe final product.The method, like that in which the initial fission of nucleoprotein was brought about by

Sr(NO3),,

may be useful because it is different from the existing methods without necessarilybeingbetter than them.

In addition to these possible advantages,

pre-cipitation byNaCl has the realadvantage that it is

gradual. In consequence the precipitate that separates under any particular circumstances is morelikelytobe in equilibrium with the environ-ment whereas the usual acid precipitation is so

nearly instantaneous that most of it probably takesplaceat a

pH

other than that of the bulk of

the fluid. Wehavenot yet shown that successive

fractions, separating withincreasing NaCl

concen-tration, differ qualitatively though they differ in theirprecipitability and osmotic pressure, but the

method holds somepromiseofbeingsuitable for the fractionation of bulkpreparations of nucleic acid.

We havealready pointed out that all methods of separating nucleic acid from other tissue com-ponents maymodify it and NaCl precipitation is no exception. Notonlyis theremodificationwhen the linkage withprotein is broken but material that has been precipitated once with NaCl is more readily precipitatedby it again. Changesinprecipitability are mostprobably the consequence of associations betweenhitherto independent particles but there is no obvious method of assessing the magnitude or the detailed nature of thechange. The nucleic acid intheextractsthat wehavestudied,andthese are made from yeast, leaf microsomes and virus particles, separates in a number of fractions of differing precipitability. We do not yet know whether this is a consequence of an initial diversity inthe source or ofpartialmodificationduringthe isolation. Notonlydo we not knowthe relationship between nucleic acid in cell extracts and nucleic acid in thecell, we do not even know therelationship

between nucleic acid in extracts and in purified

preparations. The osmotic pressure measurements

published here, and those already published by others, suggest that the meanparticle weight in

nucleic acid preparations can be large. This is

confirmedbytheir otherphysical properties.Thus YNAthat has beenprecipitated by NaCl gives a

solution that is very viscous especially at low temperatures; some neutral solutions containing only10 g.YNA/l.do not flow at 0 butwillstay inan

invertedtesttube.

In an accompanying paper (Holden & Pirie,

1955b) wedescribe the effect ofleafandpancreatic ribonucleases on these nucleic acid preparations

and find nostriking differences. Theirabsorption

spectraarealso similar(Holden&Pirie, 1955c)and aresimilarlyaffectedbytreatmentscausing hydro-lysis. Ourresultsdo not,therefore, throwanylight

onthenatureofanyspecificity thatmayreside in thenucleicacids.

SUMMARY

1. Most ofthenucleicacid insuitablyprepared

extracts from yeast, tobacco-leaf microsomes and

tobaccomosaic virus(TMV)canbeprecipitatedat

pH3-4byNaCl. A small part of the nucleic acid in

commercialproductscanalso be

precipitated.

2. Preparations ofnucleic acid made from yeast inthis way appear to have ahigh mean

particle

weightand tocontain lessthan1

%

ofprotein.

3. Their contentsofCa,Fe andMgarelow and

probably the result of contamination. Those from

TMV contain significant amounts ofdeoxyribose nucleicacid butthe othertwo aresubstantiallyfree from it.

Vol. 6o

RIBONUCLEIC ACID

FROM

THREE

SOURCES

53

REFERENCES

Bawden,F. C. & Pirie, N. W. (1937). Proc. Roy.Soc. B, 123, 274.

Bawden, F. C. & Pirie, N. W. (1940). Biochem. J. 34,1258. Bawden, F. C. & Pirie, N. W. (1943). Biochem. J. 37, 67.

Chargaff, E., Magasanik, B., Vischer,E.,Green, C.,Doniger, R. &Elson, D. (1950). J. biol. Chem. 186, 51.

Clarke,G. &Schryver,S. B. (1917). Biochem.J. 11, 319. Dische, Z. (1930). Mikrochemie,8, 4.

Greenstein, J.P. (1944). Advane.ProteinChem.1, 209.

Gr6goire,J. (1950). Bull.Soc.Chim.biol., Paris,32, 359.

Hoff-Jorgensen,E.(1952). Biochem. J. 50, 400. Holden, M. (1953). Analyst, 78, 542.

Holden,M. & Pirie, N. W. (1955a). Biochem.J. 60, 39. Holden, M. & Pirie, N. W. (1955b). Biochem. J. 60, 53.

Holden,M. &Pirie, N. W. (1955c). Biochim.biophys.Aeta, 16, 317.

Holden,M., Pirie, N. W. & Tracey, M. V. (1950). Biochem. J. 47, 399.

Jungner, G.(1951). Science, 113, 378.

Jungner,G. &Allgen,L. (1950). Actachem.scand. 4,1300. Ludwig, E. E. & Johnson, C. R. (1942). Indu8tr. Engng

Chem. (Anal. ed.), 14, 895.

Markham, R. (1953). Advances inVirusResearch, 1, 315. Markham, R. (1955). In Modern Methods of Plant Analysis,

vol.4(ed. K. Paech and M. V. Tracey). Springer-Verlag. Markham,R. & Smith, J. D. (1952a). Biochem. J. 52, 552. Markham, R. & Smith, J. D. (1952b). Biochem. J. 52,

565.

Pirie, N. W. (1949). Exp.Cell Res. (Suppl.), 1, 183. Pirie, N. W. (1950). Biochem. J. 47, 614.

Pirie, N. W. (1953). 6th nt.Congr. Microbiol. Symposium on 'Interactionof Viruses andCelLs',p. 11.

Pirie, N. W. (1954). Biochem. J. 56, 83.

Sevag,M.G., Lackman, D.B. &Smolens,J.(1938). J.biol.

Chem. 124, 425.

Sevag,M. &Smolens, J.(1941). J.biol. Chem.140, 833.

Sreenivasaya, M. & Pirie, N. W. (1938). Biochem. J. 32, 1707.

Tsuchihashi,M. (1923). Biochem.Z. 140, 63. Zittle, C. A. (1946). J. biol. Chem. 163, 111.

A

Comparison of Leaf and Pancreatic Ribonuclease

BY MARGARET HOLDEN ANm N. W. PIRIE

Rotham8ted ExperimentalStation,Harpenden, Herts (Received 12October1954)

The ribonucleases are relatively specific phospho-diesterases and their action can be followed by

methods that depend either ontheappearanceof an acidorhydroxylgroup when the ester link isbroken

or else by methods that detect changes in the

particlesizeof thenucleicacid. It maybe profitable

to discuss the merits ofthevarious methods that havebeenused,paying particular attention to their

applicability to nucleic acid and nucleoprotein

solutions in which the concentrations of nucleic acid ornucleoproteinfall in thephysiologicalrange, e.g. equivalenttoabout 10 mg.P/l. Methods that

dependmainlyonthefirst steps in the attack on the

macromolecule areof moreinterest than those that

place equalweight on all its stages.

The appearance of a new acid group has been followed by measuring the increase in buffering

power intheregion pH 6

8-7*9

(AIlen&Eiler,1941) or manometrically in bicarbonate buffer (Bain &

Rusch, 1944). The lattermethodhas been used in severallaboratoriesbut it calls fornucleicacid con-centrations ashigh as 5 g.

P/1.

Diminutioninthe molecularweightofthenucleic acidhasbeenfolloweddirectly by Carter &

Green-stein (1946),whocarriedoutthe reaction indialysis

sacsandmeasured

spectrophotometrically

thesplit products that diffused out. This method depends mainlyonthefirst stages of the reaction but it is

tediousif manydeterminationsarebeingmadeand,

as Markham & Smith (1952) have emphasized, variationsinthesalt concentration causevariation

in the diffusibility of partly hydrolysed nucleic

acid. For these reasons, diminution inmolecular weightisgenerally inferred fromachangeinthe acid

precipitability of the nucleic acid. Many different conditionshave been usedand,intheexperimental section of thispaper, we compare some of them.

On hydrolysis there is a diminution in the absorption oflightat 300m,u. (Kunitz, 1946), but theeffectisnotlargeand ismainlyaconsequenceof the later stagesoftheaction.Thereis anincreasein

absorption at 260m,., which is discussed more

fullyinanotherpaper(Holden&Pirie, 1955c);this increase depends on the extent ofhydrolysisbut is notproportional to it.The volumeofthe solution increases and then diminishes (Vandendriessche,

1951) and the

ability

to bind

trimethyl

p-(p-hydroxybenzeneazo)phenyl ammonium chloride decreases(Cavalieri, 1952). Inorganic phosphateis aproductoftheaction of some enzyme preparations but there is reason tothink that thisaction is due to contamination with a phosphatase so that any measurements depending on it deal with two enzymessimultaneously.

It is important to remember that the different methods of following the reaction may measure differentstages of it.ThusKunitz

(1940)

found that