Germination of Soybean Embryonic Axes NUCLEOTIDE SUGAR METABOLISM AND INITIATION OF GROWTH 1

UDP-Sugars comprise the domnat cass of nucleot sugars in isolated soybeaa axes duri early m at While "dry" axes contain 1 nanomole per axis of UDP-sugars, further synthesis i initiated upon in_bbton such that the concentration of total UDP-sugars reaches 8 na_omoles per axis or roughly 1 after 10 hours, when the axes begin to elogate. The GDP-sugars are essentialy absent before inblbtio, acculate rapidy for 90 min to 173 p per axis, tben decrease somewhat, reattaig the earer palevel shortly before growth begins. MeanwhWie, the levd of ADP-sugrs is c T data indcate that the 10hOw lg period precedlug axis growth does not resudt from a dbinhed abflity to synthesize a u4or category of nucleotide sugars. Relatve rates of synthesis of individu UDPand GDP-sugars were determined by incorporatio of 'Hjuridlne or I3Hlguaose. The distributio of lab in the different casses of UDP-sugars and in the sigle chss of GDP-sugar was quantitatively similar whe analyze before, at the onset, or duing early growth It therefore seems unlkely that synthesis of a key nucleotide sugar controls the intiation of growtL The possible relevance of n -cotide sugars to growth is discussed and new methods for enymlc analysis of picomole levels of nucleotide sugars are described.

Growth of isolated soybean embryonic axes is preceded by a lag period of 9 to 10 h.Substantial synthesis of ATP and GTP occurs at the outset ofthis period while the rate ofprotein synthesis increases subsequently, but prior to the initiation ofcell elongation (22).Separation of growth from earlier changes in rates of ATP, GTP, and protein synthesis has also been shown for wheat em- bryos (1), where mobilization of poly(A)+-RNA into polysomes also occurs well before growth begins (23).Such observations suggest that although axis cells quickly recover basic metabolic activities, a longer period may be required until more complex processes reach optimal rates.Since growth in other plant tissues is accompanied by such changes in the cell wall matrix as a correlative increase in dry weight (24), a decrease of pectinase-extractable galactose (12), and a change in the wall hydroxyproline content (2), it seemed reason- able that the early growth of seed axes might be accompanied by an increased availability of cell wall precursors.Nucleotide sugars are prime participants in glycosidation reactions (6,14), including those of monosaccharide polymerization: UDP-Glc in cell wall f8-glucan synthesis (7,13,27), UDP-Ara and UDP-Xyl in the synthesis of other wall polymers (10,15), and GDP-sugars in Ii-glucan synthesis through GDP-Glc (9) and glucomannan biosyn- thesis through GDP-Man (8).Considering therefore that the timing ofthe appearance ofeither the UDP-sugars or ofthe GDP- sugars might be related to the initiation of growth by the seed axis, we undertook an analysis of the levels of these compounds at different times of germination.We also examined the rate of synthesis of several specific nucleotide sugars on the further possibility that growth would involve a sharp transition in the synthesis of one of these compounds.

GERMINATION AND EXTRACTION OF AXES
Embryonic axes were removed from whole, dry, soybean seeds and were stored at 4 C for periods of up to 3 weeks.Thirty embryonic axes, placed concave side upward on three layers of Whatman No. 1 filter paper in a 5.5-cm Petri dish, were incubated with 1.6 ml H20 containing 10 isg/ml chloramphenicol.Dishes were kept in the dark at 25 C in a humid chamber.For incubations of more than 15 h, additional H20 was added as required.At the end of the incubation, the axes and upper layer of filter paper were briefly blotted from below on paper towels to withdraw excess H20 and the axes were weighed, or trichloroacetic acid extracts were prepared.For the latter procedure, axes were ho- mogenized in 0.5 ml of cold 10% trichloroacetic acid, then in 2 x 2 ml of cold 5% trichloroacetic acid.The homogenate was centri- fuged at 24,000g for 30 min and the trichloroacetic acid was removed from the supernatant by partitioning three times into 10 ml of cold, H20-saturated ether.Residual ether was removed by vortexing briefly at 60 C. The extract was neutralized, frozen at -70 C, rethawed, and centrifuged for 15 min to remove a small precipitate, and stored at -70 C.

NUCLEOTIDE ASSAYS
Levels ofnucleotide sugars were measured using enzymic assays with 3P.The rationale for these procedures is presented under "Results."UDP-Sugars.To 100 ,ul of diluted sample was added a 100ylO mixture containing 20 p1 of buffer A (0.25 M Tris-HCI [pH 7.6], 0.35 M KCL 10 mm MgCl2), and 1.2 Atg of APase2, (Sigma, type VII, from calf intestine).This was incubated at 30 C for 20 min, then at 90 C for 2 min.Another 100-ilO mixture containing 40 p1 of buffer A and 0.032 unit of nucleotide pyrophosphatase (NPPase; Sigma, type III) was added and the incubations at 30 C and 90 C were repeated.Subsequently, 100 pl containing 20 I1 of buffer A, 500 pmol ATP, 500 pmol PEP, 4 jig AK, and 4 Ag pyruvate kinase (PK) were added.Again, the reactants were 2Abbreviations: APase: alkaline phosphatase; PEP: phosphoenolpyru- vate; AK: adenylate kinase; PK: pyruvate kinase; NMP: nucleoside mono- phosphate; TEA: triethylamine; PEI: polyethyleneimine; GlcUA: glucu- ronic acid; GalUA: galacturonic acid.975 incubated at 30 C and 90 C. Next a 50-pl solution was added which contained 10 pl of buffer A, 25,000 cpm of ["2P]PEP (18), 4 ,ug PK, and 0.1 unit NMP kinase.The mixture was incubated 20 min at 30 C, then 600 ,u of 0.2% acid-washed charcoal in cold 5% trichloroacetic acid was added.The charcoal mixture was diluted with 4 ml of 5% trichloroacetic acid, and collected on filters of Whatman GF/C.The filters were washed four times with cold 5% trichloroacetic acid, dried, and counted.The step with AK was used because the reactivity of NMP kinase to UMP, AMP, and GMP is 100:4:1, and this step removes AMP derived from NADlike compounds and ADP-sugars.Standards for the assay were either UDP-Glc or UMP, both being used with omission of the APase treatment.The reaction was linear up to 120 pmol/assay.GDP-Sugars.The first two incubations with APase and NPPase were as described above, then a 100 ,l-solution was added which included 50 p1 of buffer A, 1 nmol ATP, 500 pmol PEP, 25,000 cpm [mP]PEP, 4 ,ug GMP kinase, and 4 ,ug PK.The mixture was incubated for 20 min at 30 C and charcoal (in 5% trichloroacetic acid) was added as described above.The reaction, which used GMP as the standard, was linear up to 60 pmol of GMP per assay.
ADP-X.The sample was reacted with APase, NPPase, then with 100 pd containing 20 pl of buffer A, 500 pmol ATP, 500 pmol PEP, 4 ,ug AK, 4 ,tg PK, and 25,000 cpm of [2P]PEP.The reaction was linear for 100 pmol of AMP.ADP-X compounds included ADP- sugars and NAD-like compounds.

INCORPORATION OF 3H-NUCLEOSIDES AND |4C]GLC INTO
NUCLEOTIDE SUGARS Nucleotide sugars were synthesized in vivo from the appropriate 3H-nucleoside precursor.[5,6-3H]uridine (47 Ci/mmol, 1 mCi/ml, 2.1 x 10-5 M) and 60 axes were placed in a plastic scintillation vial (1 ml for 4-and 9-h samples or 1.2 ml for 15-h samples).Incubation was routinely for 60 min at 25 C on a reciprocating shaker. 8-3H]guanosine (15 Ci/mmoL 1 mCi/ml, 6.7 x 10 M) and [methyl-3Hlthymidine (57 Ci/mmoL 1 mCi/ml, 1.8 x 10-5 M) were used similarly.In the case of double labeling with [U-'4CjGlc (313 mCi/mmol, 150 ,uCi/1.2ml, 4 x l0-M) and .3H]uridine, the radioactive compounds were dried under N2 at 40 C and redissolved in H20.Each incubation medium contained chlor- amphenicoL 10 ,tg/ml.At the end of the labeling period, the medium was withdrawn and the axes quickly rinsed five times with 20 ml each of cold H20.The axes were separated into two lots and trichloroacetic acid extracts were prepared in the usual manner.The processed extracts (about 9 ml total) were diluted 5- fold, with buffer A being added to a final concentration of 20 mm Tris-Cl (pH 7.6), 28 mm KCL and 0.8 mm MgCl2.APase was added (5.25 units/ml of diluted extract) and incubation was carried out at 30 C for 20 min and terminated by heating 2.5 min at 90 C. The APase-treated extract was concentrated at 40 C using a Buchi Roto-vapor fitted with a dry ice trap.

SEPARATION METHODS
The concentrated, APase-treated extract was applied to a col- umn of Bio-Gel P2 (100-200 mesh, 2 x 37 cm) and was eluted at room temperature with 2.5 mm TEA-carbonate buffer (pH 7.8) at a rate of 20 ml/h.Two-ml fractions were collected.The nucleotide sugar fractions, which eluted close to the column void volume, were pooled, cooled to 4 C, and applied to a column of Whatman DE52 (0.9 x 15 cm).The column was washed with 2.5 mm TEA- carbonate buffer for 40 fractions (1.55 ml each, 20 ml/h) then the nucleotide sugars were eluted with a 120-ml gradient of 2.5 to 250 mM TEA-carbonate at a flow rate of 7.5 ml/h.When GDP-sugars were being analyzed, a 120-ml gradient of 5 to 500 mm TEA- carbonate was employed.The extracts were further eluted with 15 ml of the final (250 or 500 mM) buffer, then with 8 ml each of 1 and 2 M buffer to regenerate the column.Appropriate fractions were pooled and concentrated in vacuo at 40 C. The concentrate was repeatedly dissolved in methanol with further evaporation, until the volatile TEA-carbonate was removed.The final residue was taken up in 1 ml of H20.
TLC was carried out at room temperature in closed tanks with relevant marker compounds being mixed with samples before chromatography.System A used thin layer plates of PEI-cellulose with UV indicator (Brinkmann), which were developed for 6 h or more with sodium borate (pH 7.2) (19,20).Due to the low RF values of many of the nucleotide sugars, some plates were devel- oped for up to 40 h by continuous flow which was achieved by stapling filter paper (Whatman No. 1 or 3) "blotters" to the tops of the plates.Where noted in the text, 0.25 M LiCl was added to the borate buffer.System B used TLC plates of cellulose MN 300 with UV indicator which were developed with tert-butanoL meth- ylethyl ketone, H20, and formic acid, 44:44:11:0.26.System C used cellulose plates with UV indicator, with H20-saturated secbutanol as the solvent.System D was used to detect free sugars.Cellulose plates with filter paper blotters attached were employed as described earlier, but here, the blotters were also covered with plastic film to retard evaporation.The solvent system was pyridine-ethyl acetate-H20, 2:8:1 for 16 h.After chromatography the plates were dried with air then sprayed with a solution containing aniline (4 ml), diphenylamine (4 g), and H3PO4 (20 ml) in acetone (200 ml).The plates were air-dried for 5 to 10 min, then heated at 80 C for 2 min.Due to quenching effects, parallel samples were chromatographed with one lane sprayed to ascertain the coincidence of markers with the radioactivity.The second lane was used for actual quantitation.System E used plates of PEI-cellulose with UV indicator which were developed with 1 M ammoinum acetate titrated to pH 7 with acetic acid, and 95% ethanol, 7.5:3.Again, the plates were developed by continuous flow ascending chromatography.The thin layer plates were analyzed quantitatively for radioactivity by removing strips of the support material (5-mm sections) from the celluloid backing with a razor blade, and counting the strips in scintillation fluid.When required, compounds were recovered from PEI-cellulose plates by scraping the PEI-cellulose containing the sample from the developed plate, making a small column of the scrapings, washing with methanol, eluting with 0.5 M TEA-carbonate (pH 7.8) and evaporating under N2.

CHEMICAL AND ENZYMIC ANALYSES OF [3'HuDP SUGARS
Free nucleoside and free sugar moieties were generated from the nucleotide sugars in two ways.Samples made 2 N in trifluo- roacetic acid were sealed in ampoules and heated at 12 C for 60 min.Trifluoroacetic acid was removed by evaporation with a stream ofN2.GMP was destroyed by this procedure.Alternatively, the free nucleosides and free sugars were generated enzymically in a 20-pl mixture which included the sample, 0.25 units ofalkaline phosphatase, 0.032 unit nucleotide pyrophosphatase, 5 mm TEA- carbonate (pH 7.8), and 0.8 mM magnesium acetate.The mix was incubated 60 min at 30 C and 2 min at 90 C, then 200 pl of cold ethanol was added and the precipitate removed by centrifugation.The supernatant was dried down at 50 C using N2, dissolved in 50 pl ethanoL and dried twice more.One-half of the sample was used for chromatography.
UDP-Glc was converted to UDP-GlcUA by the enzyme UDPG-dehydrogenase.The 50-pl reaction mix contained sample, 84 mm TEA-carbonate buffer (pH 8.5), 50 nmol 8-NAD+, and 0.005 units of UDPG-dehydrogenase.After an incubation at 30 C for 20 min, 50 yt of methanol was added and the mixture was dried at 40 C with a stream of N2.The residue was dissolved in H20, then UDP-Glc and UDP-GlcUA were separated chromato- Eraphically in system A. In most instances authentic UDP- [ CJGlc (2 nmol) was completely metabolized to UDP-[14CJ- GlcUA; any radioactivity (less than 5%o) remainng as UDP- [1CJGlc could be totally converted by a second enzyme treatment.

SOURCES OF MATERIALS
Nucleotide sugars and monosaccharides were from Sigma Chemical Co.UDP-[q4CJAra (putative) was synthesized by the action of mung bean extract on UDP-[U-14CJXylose (3).The following enzymes were from Boehringer-Mannheim: NMP ki- nase (beef liver), AK, GMP kinase, and PK.All other enzymes were from Sigma.Radiochemicals were from New Engand Nu- clear.Triethylamine was redistilled prior to its use.Soybeans, Glycine max var.Scott (Marlboro, Md.1976), were supplied by Dr. Abdul-Baki of USDA, Beltsville, Md.

RESULTS
Germination of Isolated Axes.Three stages of germination of embryonic axes of soybean seeds were identified by measuring increases in the fresh weights of the axes (Fig. 1): an imbibition phase from 0 to 90 min of incubation, a quiescent period from 90 min to 10 h, and a subsequent growth period characterized by a constant rate of increase in fresh weight and by visible elongation of the axes.
Measurement of Nucleotide Sugar Levels.Procedures devised to measure accurately low levels of NDP-sugars in trichloracetic acid extracts of plant tissues are shown in Figure 2. Extracts were first treated with alkaline phosphatase to dephosphorylate all phosphate esters except those with internal phosphate linkages.NMPs were then generated from these latter compounds with nucleotide pyrophosphatase, and specific kina were used to phosphorylate the individual NMPs to specific NDPs at the expense of added ATP.In the case of the UDP-sugars, NMP kinase was used to phosphorylate the pyrimidine monophosphates to form pyrimidine diphosphates.Although CDP-X compounds (e.g.CDP-choline) are also active in this reaction, the predominance of UDP-sugars in plant tissues makes the reaction essentially specific for the uridine nucleotides.Finally, the stoichiometric amount of ADP generated in each monophosphate kinase reaction was quantitated using pyruvate kinase and radioactive 32PIPEP (18).
When the levels of UDP-sugars at various stages of early germination were measured (Fig. 3), the dry axes were found to contain 1.1 nmol of UDP-sugar per axis, and the embryonic axes were capable of synthesizing more UDP-sugars immediately upon incubation in H20.The level of UDP-sugars increased steadily from 0 to 16 h of incubation up to a level of about 13 amol/axis.The rate of accumulation of UDP-sugars did not correlate directly with the rate of increase in the fresh weight of the axes (Fig. 1) nor with the level of ATP (see ref. 22, showing that maximum net accumulation ofATP occurs during the first 40 min ofimbibition).
The levels of GDP-sugars were also determined (Fig. 4).The Germinotion time, h FIG. 3. UDP-Sugars in isolated axes during germination.Each data point represents a separate extract from 30 axes.dry axes contained very little GDP-sugar, but rapid synthesis of these compounds occurred during the imbibition phase.This was followed by a transient decrease in the level of GDP-sugars during the early quiescent phase, and a further accumulation of these compounds thereafter.Levels of ADP-X compounds (Fig. 5) showed little change during the period 0 to 16 h.Distribution of I'HiUridine in Individual UDP-Sugars.The sustained accumulation ofthe UDP-sugars indicates that this class of compounds is synthesized continuously during early germination.Are there significant changes in the levels of specfi'c UDP- sugars?This question was approached by labeling the nucleoside moiety of the UDP-sugars and ascertaining the relative rates of synthesis of different compounds within this group.Axes were incubated with [3H]uridine for 60 min after a preincubation of4, 9, or 15 h in H20, time points chosen to represent the middle of the lag period, the end of quiescence, and the early growth phase.Neutralized trichloroacetic acid extracts of labeled axes were treated with alkaline phosphatase and applied to a Bio-Gel P-2 column.Figure 6a shows a labeling profile representative of all labeling regimes.During germination, the total amount of radio- activity incorporated into UDP-sugars increased, the relative in- corporation at 4, 9, and 15 h of germination being in the ratio of 1:2.7:3.5.Of the I3H]uridine taken up, 54, 61, and 57% (4, 9, 15 h) were incorporated into the UDP-sugar fraction.
The UDP-sugar peak was applied to a column of Whatman Germination time, h FIG. 5. ADP derivatives in isolated axes during germination.Each data point is the average of similar values determined from two extracts of 30 axes.
DE52 and three 3H-labeled peaks were eluted with a linear gradient of TEA-carbonate (pH 7.8) (Fig. 6b).The Awo of the extract provided a marker for the positions of the peaks of radioactivity.As will be shown, the major peak, peak B, included all of the UDP-neutral sugars and UDP-N-acetylamino sugars.The UDP-glycuronic acids eluted in peak C. All of the radioactiv- ity in peaks B and C ftkom each of the 4-, 9-, and 15-h samples was recovered as [3H]UMP upon acid hydrolysis with 2 N trifluoroac- etic acid, while less than 1% of the radioactivity in the acid- hydrolyzed peak A could be recovered as UMP, TMP, or CMP.When samples from each ofthe three DE52 peaks were hydrolyzed by a mixture of nucleotide pyrophosphatase and alkaline phos- phatase, 3H-radioactivities from peaks B and C were recovered as [3H]uridine (solvent system C), while that from peak A was not converted to uridine, thymidine, or cytidine.Thus, all of the 3H radioactivity in peaks B and C is derived from UDP-sugars while the identity of H-labeled compound(s) of peak A is unknown.Chromatography of peak A (RF = 0.79) in solvent system B indicates that it is substantially unlike the known UDP-sugars or UMP (RF values = 0), nor is it the degradation product of uridine, fl-Ala (RF = 0.04).
The distribution of I3Hluridine in the three fractions of UDP- sugars was similar at all three labeling periods (Table I).The significance of the increase in relative labeling of peak A at 15 h is not clear, especially in view of the facts given above, and the observation (data not presented) that the relative labeling of this peak is substantially enhanced by the omission ofchloramphenicol during incubation.The peaks of radioactivity eluted from the DE52 column were pooled separately and subjected to further chromatography on thin layer plates of PEI-cellulose using solvent system A (Fig. 7).
By this ultimate fractionation procedure, a total of 10 classes of  UDP-sugars were delineated.Classes III, IV, VII, VIII, and IX co-migrated with markers of known UDP-sugars.Putative UDP- ['4CJAra, prepared enzymically from UDP-[ 4C]Xyl, chromato- graphed between UDP-Gal and UDP-Olc (data not shown).The data of Table H show the distribution of radioactivity in the 3H- UDP-sugar compounds labeled at 4, 9, and 15 h.The remarkable feature is the similarity ofthe labeling patterns at each germination period tested.
Since primary cell walls are rich in xyloglucan ( 25) it was of interest to know the proportion of radioactivity from the [13iuridine labeling found in UDP-Xyl relative to that in UDP-Glc.This was determined by treating the UDP-sugars in DE52 peak B with UDPG-dehydrogenase and separating the UDP-glcUA formed (position IV in Fig. 7) chromatographically from the residual UDP-sugar (putatively UDP-Xyl) in position VIII.When 4-h, 9-h, and 15-h samples were analyzed in this way, and a correction was made for a small amount of UDP-Glc standard which was not metabolized under parallel conditions, it was found that the formed 3H-UDP-GlcUA accounted for 94, 95, and 94% of the 3H radioactivity originally present at position VIII.Conse- quently, only 5 to 6% of the initial label of peak VIII could belong to UDP-XyL such that a maximum of 3.6% of the total 3H-UDP- sugar (see Table III) was present as 3H-UDP-Xyl.
We were unable to distinguish specifically the 3H-UDP-sugars with sugar moieties of -Ara, -Man, -Rha, or -Fuc from UDP-GaL or to separate UDP-mannuronic acid from UDP-GalUA.There- fore, we sought to label the sugar moieties of the UDP-sugars and to identify the sugar moieties after hydrolysis.The 15-h axes were labeled with both [3HJuridine and [14CJGlc and the three [3H]- uridine peaks were collected.The profiles of the DE52 columns showed coelution of 14C radioactivity with the 3H peaks.However, in the UDP-glycuronic acid fraction (peak C), the 3H and "C radioactivities eventually were separable (40-h chromatography in solvent system A, with 0.25 M LiCl) indicating that the UDP- glyCuronic acids were not labeled by [14CJGlc.Of the 14C radio- activity in peak B ,.3.4% co-chromatographed (in solvent system A) with UDP-Gal (Fig. 7, class VII), 81.6% with UDP-Glc (Fig. 7, class VIII) and 15.1% with UDP-GlcNAc (Fig. 7, class IX).The UDP-Gal peak was indeed labeled by ["4C]Glc, as indicated by the co-chromatography of the 14C radioactivity with Gal (in solvent system D) after treatment with nucleotide pyrophospha- tase and alkaline phosphatase (Table III).This analysis also ruled out UDP-['4C]Man (or any other UDP-14C-sugar) being present in the UDP-Gal peak.In a similar manner the '4C radioactivity at the position of UDP-GlcNAc was identified as N-acetyl hex- osamine.Of the 14C radioactivity in the UDP-Glc fraction, 14% chromatographed with Glc and only a trace with Xyl.Further- more, UDPG-dehydrogenase converted only 13% of the 14C ra- dioactivity to UDP-["4CJGlcUA.Thus, [14CjGlc metabolites other than the UDP-sugars had co-purified with UDP-Glc.We conclude that of the UDP-sugars, only UDP-GaL UDP-Glc and UDP-GlcNAc, are substantially labeled by Glc.
Distribution of [3Hguanosine in GDP-X.In experiments anal- ogous to those with [3Hluridine, axes were labeled for 60 min with [8-3Hlguanosine.Labeling of the GDP-sugars was substantially less efficient than was labeling of the UDP-sugars.Incubation of 60 axes in I mCi of [3H]guanosine between 3.5 and 4.5 h resulted in the uptake of 7.3 liCi.When the alkaline phosphatase-treated extract was applied to a Bio-Gel P2 column, 1.6% of the radioac- tivity eluted in the nucleotide sugar region, 20.5% with guanosine and 77.9% in a low mol wt fraction, the latter suggesting extensive degradation of the label by the axes.After DE52 chromatography, a single peak was obtained, over 90% of which co-chromato- graphed with GDP-Man by TLC (Fig. 8a).Based on theoretical considerations (based on ref. 11) we assume that GDP-Rha and Table III.'4C-Monosaccharides Released by Hydrolysis of UDP-Sugars Soybean axes were incubated for a total of 16 h.At 11 h [14C]Glc was added and at 15 h [3H]uridine was added.UDP-sugars were separated on Bio-Gel P2 and DE52, then DE52 peak B was fractionated by solvent system A, yielding peaks of 14C radioactivity associated with 3H-UDPsugars of classes VII, VIII, and IX.These UDP-sugar regions were individually eluted, treated with nucleotide pyrophosphatase and alkaline phosphatase or with 2 N trifluoroacetic acid to generate free sugars, and the free sugars were chromatographed in solvent system D. Radioactivity remaining at the origin was associated with charged metabolites such as organic acids, amino acids, etc.No radioactivity was found at the positions of Man, Ara, Rha, or Fuc.Numbers in parentheses refer to per cent distribution within the class.
['4C]-Component Ana- GDP-Fuc would also chromatograph to the position of GDP- Man.No label was associated with GDP-Glc.Labeling at 15 h improved the uptake of [3Hlguanosine with 30 ,uCi taken up out of 1.2 mCi.The nucleotide sugar region of the Bio-Gel P2 column held 7% of the radioactivity and again, the radioactivity chro- matographing as GDP-sugar on DE52 co-chromatographed with GDP-Man (Fig. 8b).The finding that all of the 3H-GDP-sugar is represented by this single class makes it probable that there is no general switch in the class of GDP-sugars being supplied to the germinating axes (e.g. from GDP-Man to GDP-Glc), at least between 4 and 16 h of ge tion.
Reports of plant enzyme systems which metabolize TDP-Glc to TDP-Rha ( 16) prompted labeling experiments with [methyl-31_]thymidine (1 mCi for 60 min) after 4, 9, and 15 h of incubation.Phosphatase-treated extracts were run directly on DE52, and three peaks of 3H radioactivity were eluted in the nucleotide sugar region.These three peaks could be further resolved into a total of seven radioactive compounds by solvent system A. None of the compounds, however, were acid-labile, ie.no [3H]TMP was re- leased by 2 N trifluoroacetic acid, although authentic TDP-Glc was easily hydrolyzed to TMP under similar conditions.In solvent system E, marker TDP-Glc was fully resolved from all of the radioactive compounds.Thus, TDP-sugars apparently are not synthesized from 13H]thymidine in germinating soybean axes.DISCUSSION Synthesis of UDP-sugars, as a group, is apparently not limiting to growth, since the steady-state level of UDP-sugar increases several hours before initiation of growth.The UDP-sugar level rises from I nmol/axis before incubation to 8 nmol/axis (approx- imately 1 mM) after 10 h (Fig. 3) when growth begins.This general observation still allows that a major change in the relative pro- portion of individual species of UDP-sugar could be essential for growth, e.g. the appearance of a key epimerase (4), pyrophospho- rylase (26), polysaccharide synthetase, etc.When the relative distribution of [3H]uridine into classes of UDP-sugars was ana- lyzed, it was found, however, that 4-h (early quiescent), 9-h (end of quiescence), and 15-h (growing) axes all had similar patterns for distribution of label.Thus, there is neither a sudden appear- ance ofa new UDP-sugar (as far as our separation methods allow), nor a switch in the preferred synthesis (or utilization) of a specific UDP-sugar.
Of the less abundant nucleotide sugars, the ADP-sugars, a group which includes NAD-type compounds, do not change sig- nificantly in amount before growth begins (Fig. 5), and the ADP- Distance, cm FIG. 8. TLC of GDP-4H-sugars on PEI-ellulose.After chromatography on DE52, the GDP-4H-sugar peak was analyzed in solvent system A (40 h of chromatography with 0.25 M LiCl being added). (a) Axes incubated 3.5 h before labeling for 60 min with [3H]guanosine; (b) axes incubated for 15 h before labeling.Quantitation of radioactivity from the 3.5-h sample required that more extract be spotted; the higher solute concentration caused bunching of the nucleotides in this region of the chromatogram.
sugars could not be labeled by [3H]adenosine.We found no evidence for the synthesis of TDP-sugars, key donors of Rha (21) in bacterial cell wall synthesis.The bimodal increase in GDP- sugars (Fig. 4) suggests that at least one system for metabolizing GDP-sugars is operative during the quiescent period.The GDP- sugars, which were labeled by I Hlguanosine, belong to one chro- matographic class which could only include GDP-Man, GDP- Rha, and GDP-Fuc.No 3H-GDP-Glc, a proposed precursor of a cellulose-like l-1,4-glucan (9), was observed.Albersheim and co-workers have reported that the primary cell walls of dicots are particularly rich in Ara (28%), Glc (25%), Gal (14%), GalUA (13%), and Xyl (10%) with lesser amounts of Rha, Fuc, and Man (25).The 60-min labeling pattern of 3H-UDP- sugars from the soybean axes is quite unlike that of the cell wall composition.In the soybean axis, 3H-UDP-Glc (55%) and 3H- UDP-Gal (20%) predominated, while forms for which there were no counterpart in the cell wall (UDP-N-acetylhexosamine, UDP- GlcUA, and class X) accounted for another 15% of the labeled pool of UDP-sugar.3H-UDP-Xyl and -Ara were poorly repre- sented (<4%) in the 3H-UDP-sugar pooL yet these compounds presumably supply the Ara and Xyl for cell wall synthesis.Two situations could explain this phenomenon.One possibility would be that UDP-sugars carrying monosaccharides specifically des- tined for cell wall synthesis turned over very rapidly, either with rapid incorporation of the monosaccharide into oligosaccharide or attachment to intermediate carriers which are more immediate substrates for cell wall synthesis.The result would be that only a small portion of the steady-state pool of UDP-sugar would be required for cell wall synthesis, while most of the UDP-sugar would contribute to other reactions such as the metabolism of sucrose (5,17) or the synthesis of glycoproteins not structurally associated with the cell wall.Alternatively, the first few hours of axis growth may occur without the necessity of oligosaccharide addition to the cell wall matrix.A fLrther variation of this idea would be that the axis cells may already possess sufficient levels of oligosaccharide precursors to provide for the initiation of growth.

FIG. 1 .FIG. 2 .
FIG. 1. Changes in fresh weight of isolated axes during germination.Relative fresh weight is the final over initial fresh weight, the latter being 128 to 145 mg/30 axes.
FIG.4.GDP-Sugars in isolated axes during germination.Each data point was determined from a separate extract made from 30 axes.

Table I .
Distribution of [3HlUridine by DE52 Column ChromatographySixty axes were incubated for 60 min in ['Hiuridine after preincubation for the times indicated.

Table II .
Relative Distribution of3H-UDP-Sugars during Early Germination Sixty axes were incubated for the times shown plus 60 min in [3HJuridine.The various nucleotide sugars were separated as in Figures6 and 7.The classes of UDP-sugars were numbered on the basis of their relative mobilities.