Proc. NatL Acad. Sci. USA Vol. 78, No. 9, PP. 5470-5474, September 1981 Biochemistry Spermidine, an intrinsic component of turnip yellow mosaic virus (polyamines/RNA/norspermidine/spermine) SEYMOUR S. COHEN AND MICHAEL L. GREENBERG Department of Pharmacological Sciences, State University of New York at Stony Brook, New York 11794 Contributed by Seymour S. Cohen, June, 1, 1981 ABSTRACT The major polyamine of turnip yellow mosaic MATERIALS AND METHODS virus has been identified as spermidine by gas chromatography Growth and Purification of TYMV. Chinese cabbage seeds and mass spectrometry ofthe trifluoroacetamido derivative. Very (Brassica pekinensis, var. Pak Choy) were obtained from Ni- small amounts of putrescine and cadaverine, but not norspermi- chols Garden Nursery (Albany, OR). Plants were grown in a dine, have been detected in the virus. The spermidine contents growth chamber (Scientific Systems, Baton Rouge, LA) set for of numerous virus preparations were in the range 200-700 mol- with lux from incandescent and ecules per virion and were considerably in excess of those of 18-hr days at 280C 20,000 spermine. The RNA and spermidine contents of small amounts of fluorescent lighting and 6-hr dark periods at 220C. At about 5 the virus were determined after serological precipitation in pu- weeks, the rosettes of each plant were removed, leaving two rified preparations and in thejuice of infected plants. Under con- large leaves to be mechanically inoculated with carborundum ditions of the precipitation or purification from juice by differ- and a solution ofTYMV (0.1 mg ofvirus per ml in 0.02 M phos- ential centrifugation, only small amounts ofexogenous radioactive phate buffer, pH 7.0). Three weeks after infection, newly spermidine and spermine became bound to virus. Although ad- emergedderibbed leaves were frozen and stored at -200C. sorbed radioactive spermidine could be removed almost quanti- Frozen leaves were homogenized in a Waring Blendor with tatively by washing the virus.in dilute buffers, only a small part 0.05 M pH 4.8 potassium acetate buffer at 1 ml/g. The juice ofadsorbed.spermine was removed by.such treatment. However, was expressed through eight layers of cheese cloth and centri- >95% of the newly attached spermine was separated from the fuged at 3000 X g for 30 min. The supernatant fluid was cen- virus without loss of the original spermidine by sedimentation in trifuged at 78,500 X g for 150 min and the drained pellets were buffers containing 0.5 M NaCI or 0.06 M MgCl2. Crystallization soaked in cold acetate buffer overnight. The pellets were trit- of virus in 40% saturated MgS04 or 7.5% heparn or dialysis did urated with buffer and' after another cycle of low- and high- not decrease viral spermidine. Although the virus coat may adsorb speed'centrifugation, the pellets were redissolved in the acetate small amounts ofthe polyamines reversibly, the virus is imperme- buffer and centrifuged at 3000 x g. Such solutions contain both able toexogenous spermidine and spermine and does notexchange empty, capsids and intact virions (7). Empty capsids, separated or leak. internal spermidine. The spermidine present in purified by sucrose density gradient centrifugation, are essentially de- virus is. associated.with viral RNA at the time. of packaging and void of both RNA and polyamine (unpublished data). These formation of intact virions. particles are significantly less stable than intact virions, and denatured virus protein precipitates on storage at 40C. After Turnip yellow mosaic virus (TYMV), a RNA virus known to con- 2-3 months the empty capsids had essentially disappeared from tain polyamine (1, 2), is believed to multiply in chloroplast ag- the preparation. Virus concentrations were estimated from the gregates in Chinese cabbage (3). The virus was reported to con- extinction coefficient El% = 8.4 cm2/mg (8). Virus preparations tain far more triamine than the tetramine, spermine (1, 2), at A2 of 100-300 were stored at 40C in acetate buffer pH 4.8 although the latter is present in the plant and binds to RNA far containing 1 mM sodium azide. more tightly than does spermidine (4). The accumulation of Crystallization ofTYMV. Crystallization ofTYMV with salts, putrescine, spermidine, and spermine is exaggerated in the in- ethanol, or heparin most commonly yields octahedra (9). In the fected plant (5), and we have been investigating the ability of current study, we obtained tetragonal bipyramids with salts and chloroplasts to synthesize the viral triamine. We have reinves- fragile rectangular plates with heparin. Virus preparations con- tigated the identity of the viral triamine because several re- taining 10 and 26 mg/ml could be crystallized at 40C from 33% cently discovered natural triamines (6), including norspermi- saturated ammonium sulfate by mixing or by free interphase dine, originally thought to be the viral triamine (1), are not easily diffusion. Crystals obtained by the latter method were larger separated from spermidine. and more regular in shape than those obtained by mixing or by To determine if the spermidine in the virus was an artifact dehydration. TYMV crystals obtained from ammonium sulfate ofpreparation, as a result of the sopping up ofspermidine into are not easily analyzed for amines by the dansyl method. Crys- RNA contained within a possibly sponge-like virion, we have tals were obtained from 40-46% saturated magnesium sulfate studied the adsorption ofpolyamines to virus as well'as the sta- orfrom 5-10% (wt/vol) heparin. Afterremoval ofmotherliquors bility ofpolyamine content. The virus does contain spermidine by fine capillary pipettes, the crystals were dissolved in acetate as the sole triamine, which is essentially nonexchangeable, in- buffer, and these solutions were extracted in 3% perchloric acid dicatingthat the triamine is associated with theviral RNA before for polyamine analysis. packaging into virions. Abbreviations: TYMV, turnip yellow mosaic virus; norspermidine, 1,7- The publication costsofthis article were defrayed in partby page charge diamino4 azaheptane; GC/MS, gas chromatography/mass spectrom- payment. This article must therefore be hereby marked "advertise- etry; dansyl, dimethylaminonaphthylsulfonyl; F3CCONH, trifluora- ment" in accordance with 18 U. S. C. §1734 solely to indicate this fact. acetamide. 5470 Downloaded by guest on September 24, 2021 Biochemistry: Cohen and Greenberg Proc. Nati Acad. Sci. USA 78 (1981) 5471 Polyamine Analyses. Hydrochlorides ofspermidine and nor- (wt/vol) sucrose were prepared in cellulose nitrate tubes with spermidine were recrystallized from mixtures ofethanol, water, an Isco gradient mixer. A virus sample, 1 ml containing 10-11 and isopropanol. The hydrochloride of N-3'-aminopropyl-1,5- mg ofTYMV in 0.05 M KOAc at pH 4.8, was layered on 37 ml diaminopentane was recrystallized from absolute ethanol. ofa 5-20% gradient ofsucrose in this buffer. The tube was cen- These substances as well as hydrochlorides of spermine, pu- trifuged at 26,000 rpm for 150 min at 4°C in an SW 27 rotor in trescine, and cadaverine were >99% pure according to chro- a Spinco model L5-50 ultracentrifuge. A 30% (wt/vol) sucrose matography as dansyl derivatives. solution displaced the fractions upward in an Isco density gra- Extracts ofvirus preparations or plantjuice were made in 3% dient fractionator; 0.9-ml fractions were collected and absor- perchloric acid, chilled, and centrifuged. The cold acid extracts bances were determined. Aliquots (0.1 ml) ofthe fractions were were neutralized with KOH, chilled, and centrifuged to remove added to 10 ml of Aquasol, and radioactivity was estimated in KC104. A portion ofthe supernatantfluid was dansylated. Dan- a Packard scintillation counter. syl polyamines were extracted into benzene and separated by Radioactive Substances. Radioactive polyamines were ob- thin-layer chromatography on silica gel G plates with devel- tained from New England Nuclear. [1,4-'4C]Putrescine dihy- opment with cyclohexane/ethyl acetate, 2:1 (vol/vol), two to drochloride, [tetramethylene-1,4'4C]spermidine trihydrochlo- four times. The fluorescent dansyl amines were quantified with ride, and [tetramethylene-1,4'4C]spermine tetrahydrochloride a Farrand VIS-UV chromatogram analyzer. were available at 6-9 x 107 cpm/pmol. Spermine tetrahydro- Serological Method. Rabbit antisera to TYMVwere obtained chloride was also obtained as [3-aminopropyl-3-3H(N)]2sper- from the American Type Culture Collection and were used to mine'4HCl at 15-30 Ci/mmol. precipitate virus specifically under conditions of equivalence or of most rapid flocculation (5, 10). In estimates of the poly- RESULTS amine or RNA content of virus in purified preparations or in Diamines and Triamines in TYMV. Fig. 1 compares the gas plantjuice, samples (0.2 ml) containing virus in phosphate buf- chromatographic patterns ofelutions ofF3CCONH derivatives fer (0.05 M, pH 7.1) were incubated with diluted antiserum or ofspermidine, norspermidine, and amines extracted from sam- buffer (0.2 ml) at 370C for 2 hr and stored overnight at 40C. The ples ofTYMV. These patterns were monitored in the mass spec- tubes were centrifuged at 40C for 1 hr at 850 X g, and the pre- trometer at m/e 126, defining the ion CF3CONHCH2-found cipitates were drained and washed similarly with phosphate in the analysis of all the derivatized diamines and triamines buffer. The polyamines were extracted with 0.4 ml of 3% per- studied. Peaks from F3CCONH norspermidine ofm/e 126 and chloric acid at 40C. Perchloric acid (1.2 ml of3%) was added to 350 appeared simultaneously at 6.3-6.5 min, whereas peaks the drained precipitate and heated to 90°C for 15 min. After from F3CCONH spermidine of m/e 126 and 364 appeared at cooling and centrifugation, an ultraviolet absorption spectrum 7.2 min. The peak of the major F3CCONH amine extracted was determined and compared with that of a mixture of nu- from TYMV monitored at m/e 126 and 364 eluted at 7.2 min cleotides prepared with the ratio ofbases estimated to be that and had the additional methylene ofspermidine, distinguishing of group I type TYMV (11).