Biochemical Studies of Bacterial Sporulation and Germination PHOSPHATE METABOLISM DURING SPORULATION*

SUMMARY Acid-soluble phosphate compounds represent about 25 % of the total phosphate of Bacillus megaferium vegetative cells, but only 7% of the total phosphate in dormant spores. In vegetative cells, 12 compounds make up 90% of the acid-soluble phosphate. These same compounds predominate throughout growth and early sporulation, although their relative amounts change during the process. One hour before refractile spores appear within sporangia, 3-phospho-D-glyceric acid (PGA) becomes a major component of the sporangial pool and, with the appearance of refractile spores, PGA becomes more difficult to extract. PGA is the largest component of the acid-soluble phosphate pool of B. megaterium spores produced in rich as well as minimal media, and also predominates in Bacillus cereus and Bacillus subfilis spores. Spore PGA appears to be synthesized in a compartment which is inaccessible to inorganic phosphate in the medium, although, at the time of synthesis, Pi from the medium does enter the sporangium. PGA in dormant spores is very Crmly bound, and does not readily exchange with external

vegetative cells, but only 7% of the total phosphate in dormant spores. In vegetative cells, 12 compounds make up 90% of the acidsoluble phosphate. These same compounds predominate throughout growth and early sporulation, although their relative amounts change during the process. One hour before refractile spores appear within sporangia, 3-phospho-D-glyceric acid (PGA) becomes a major component of the sporangial pool and, with the appearance of refractile spores, PGA becomes more difficult to extract.
PGA is the largest component of the acid-soluble phosphate pool of B. megaterium spores produced in rich as well as minimal media, and also predominates in Bacillus cereus and Bacillus subfilis spores. Spore PGA appears to be synthesized in a compartment which is inaccessible to inorganic phosphate in the medium, although, at the time of synthesis, Pi from the medium does enter the sporangium. PGA in dormant spores is very Crmly bound, and does not readily exchange with external PGA.
During sporulation, major shifts in metabolism occur. Certain compounds, such as poly-P-hydroxybutyrate, accumulate and are later utilized (1) ; extensive turnover of macromolecules occurs (2,3), and some enzyme levels change dramatically (2); new and distinctive spore structures are synthesized, while previously existing structures are degraded (4). Finally, a metabolically inert spore is formed.
Are these metabolic changes reflected in the kinds and amounts of small molecules in sporulating cells? Do dormant spores contain ATP, nucleotides, coenzymes, and other intermediates typically found in growing cells? * This investigation was supported in part by grants from the National Institutes of Health (United States Public Health Service) and the National Science Foundation.
This work was taken in part from a thesis submitted by David L. Nelson  The pivotal role of phosphorylated compounds in biosynthetic and energy-yielding reactions led us to compare the pools of low molecular weight phosphate compounds in growing cells with those in spores and to examine the variations in phosphate pools during sporulation.
The earlier studies of Fitz-James (5) and of Yamakawa, Aida, and Uemura (6) indicated that spores contained relatively small amounts of acid-soluble phosphate.
We report here significant differences in the kinds of phosphate compounds found in vegetative cells and spores. In an accompanying paper (7), phosphate metabolism during germination is considered.
We have also examined pools of free amino acids (8) and sulfur compounds (9) in spores.

Methods
Growth of Cells and Spores-Cells were grown in flasks with vigorous aeration either at 37" in a supplemented nutrient broth (10) or at 30" in a synthetic medium (11) containing sucrose as sole carbon source. The SNBI medium, containing 2.0 to 2.5 mM Pi and about 0.2 mM organic phosphate, was used in all experiments except those in Table IV and Figs. 2 and 3. To improve the uptake of 32Pi in these experiments, the Pi content of the nutrient broth was reduced as follows.
Eighty grams of Difco nutrient broth were dissolved in water to give 300 ml of a viscous solution, to which were added slowly (at 45") first 25 ml of 25% MgC12.7Hz0 and then 1.8 ml of concentrated ammonium hydroxide.
A heavy precipitate formed immediately and stirring was continued for 2 hours at 4". The suspension was centrifuged for 30 min at 10,000 x g to remove the bulk of the precipitate, and the supernatant fluid was filtered, under vacuum, through previously washed filter paper. The clear filtrate was diluted with Hz0 to 500 ml. This 20.fold concentrated nutrient broth, when diluted to make SNB, contained 0.21 mM Pi. The amount of NH3 added in this procedure was not enough to change the final pH of the diluted broth, which was 6.8. Cells of Ba- The time of appearance of refractile spores within sporangia was determined by phase contrast microscopy of unstained culture samples. Cells and spores were counted directly in a Petroff-Hauser chamber.
To make septa of dividing cells more clearly visible, vegetative cells were diluted with 1 volume of 5'y0 trichloracetic acid before counting. For determinations of dry weight, cells were harvested by centrifugation, washed once with cold water, and then suspended in water.
Aliquots in tared metal planchets were dried to constant weight under an infrared lamp.
Cleaned spores were weighed in the same way.
Washing of Cells and Spores--In order to harvest and wash labeled cells rapidly, samples (0.2 to 0.5 ml) of the culture were immediately filtered through 0.45 p Millipore filters and then washed twice on the filters with 2.0 ml of unlabeled SNB at 15". When the filters had been previously soaked in 1 M Pi, very little background adsorption of s2Pi occurred.
The washed cells on the filter were immediately transferred to a small beaker containing either 2.0 ml of boiling Hz0 or 2.0 ml of cold 5% trichloracetic acid. Less than 30 set was required for this washing procedure.
When larger volumes of culture were to be harvested, samples of 5 to 100 ml were chilled 3 min in an ice bath and then centrifuged 3 min at 8000 x g. The supernatant fluid was decanted and discarded.
The cell pellet was drained and then quickly suspended in cold 5% trichloracetic acid. A comparison of these two methods of harvesting showed that cells washed on a Millipore filter contained 5 to 10cyO less Pi than those harvested by centrifugation.
Spores were harvested after maximal sporangial lysis, about 20 hours after the end of growth, and then cleaned in one of three ways. When sporangial lysis was complete and cultures contained no sporangial debris visible in the light microscope, spores were simply washed eight times with cold HzO. After each centrifugation (10 min, 10,000 X g), the light layer at the top of pellet was rinsed off and discarded.
Alternatively, if sporangial lysis was not complete, spores were cleaned with lysozyme and sodium dodecyl sulfate, as previously described (12). In some experiments with Bacillus subtilis, spores were freed of sporangia and debris by partitioning the culture pellet in the two-phase System Y of Sacks and Alderton (13). The methods all yielded spores uncontaminated by sporangia or detectable debris, and the phosphate content of cleaned spores was the same regardless of the method of cleaning.
Extraction of Cells and Spores-To extract acid-soluble eontents, vegetative cells were suspended in 5% trichloracetic acid and subjected to four 1-min periods of sonic oscillation, with continuous cooling in an ice bath. The broken cell suspension was held for 30 min at 0" and then centrifuged for 15 min at 12,000 x g. The supernatant fluid was extracted four times with 2 volumes of ether to remove trichloraeetic acid. A second cold trichloracetic acid extraction without sonic treatment released essentially no additional phosphate.
The residue was extracted twice with 5% trichloracetic acid at 80" for 15 min to release nucleic acids. Fitz-James (5) showed that disruption of spores was essential for the proper fractionation of their phosphate compounds.
Several methods were therefore tested for their effectiveness in releasing phosphate compounds from dormant spores. Disruption in the Gifford-Wood Mini-mill (14) released a maximum of 25 to 307, of the phosphate from spores of B. subtilis suspended in neutral buffer. This same limit was also reached when disruption was by sonic oscillation with glass beads (8) or by dry rupture (15) or by treatment with lysozyme after exposure to mercaptoethanol and urea (16). These methods all proved unsatisfactory for the analysis of acid-soluble pools, since nuclease activity in the spore extracts released enough nucleotides during the extraction to obscure the true dormant spore pools.
When B. subtilis spores were suspended in 57, trichloracetic acid and then thoroughly disrupted in the Mini-mill (14) at O-lo", 5% of their total phosphate was released.
However, after extraction of the trichloracetic acid with ether, neutralization with NaOH resulted in the precipitation of 807; of the acidsoluble phosphate, leaving only 1 o/0 of the total spore phosphate in the acid-soluble, pH 'i-soluble fraction.
The material which precipitated at pH 7 was easily dissolved in 5% trichloracetic acid, and, upon removal of the trichloracetic acid with ether, 100% of the 32P was again precipitated.
As a control, 14C-ATP was added to intact spores before their disruption in trichloracetic acid. The ATP also showed anomalous solubility properties in that all of the added ilTP was recovered in the cold trichloracetic acid extract, but 95% of it was precipitated at pH 7. Passage of the original cold acid extract through a short column of Dowex 50-Na+ succeeded in removing a factor, perhaps a metal ion, responsible for this precipitation at neutral pH. -4fter such a Dowex treatment, less than 30/, of the added .4TP and less than 10% of the extracted spore g2P was precipitated at pH 7.
We were concerned that the factors which caused the insolubility at pH 7 of 32P and ATP in the cold acid extracts might also prevent the extraction of normally acid-soluble phosphorus compounds of the spore. To examine this possibility, spores labeled with 32P were extracted by disruption in 5y0 trichloracetic acid, and the insoluble debris was then suspended and stirred with Dowex 50-Na+ beads. The cold acid extract contained 5% of the total spore phosphate, and subsequent stirring with Dowex 50 released less than 0.2rr/, of the remaining phosphate.
Thus it seems unlikely that there is a significant quantity of small, phosphorus-containing molecules sequestered in the spore as, for example, in metal complexes.
No anomalous solubility properties were observed when spores of B. megaterium were suspended in cold 5% trichloracetic acid and disrupted by sonic oscillation in the presence of glass beads, as previously described (8). This method was therefore used, unless otherwise noted, for the extraction of B. megaterium dormant spore pools.
Treatment of B. subtilis or B. megaterium spores at 100" for 10 min in either water or neutral buffer caused the complete release of their acid-soluble phosphate pools, as shown under "Results." It was therefore convenient to use this method to release all PGA from both cells and developing spores in our studies of PGA accumulation during sporulation. After treatment for 30 min in water at loo", cell and spore debris was separated from the extract by centrifugation at 12,000 x g for 15 min.

Fractionation
and Detection of Acid-soluble Phosphate Compound-High voltage paper electrophoresis was performed in '20 mM sodium citrate buffer, pH 3.5, on Whatman No. 3MM paper. With the paper immersed in an organic coolant, at 20-25", a potential of 125 volts per cm was applied for 25 min. Under these conditions, Pi migrated 30 cm toward the anode.  (Table I).
bility to charcoal was tested by mixing a labeled compound with Fractionation of the acid-soluble extracts of vegetative cells acid-washed Norit A in the presence of carrier Pi and CMP, at revealed 12 major components (Fig. 1). These included Pi, acid pH. Compounds adsorbed to Norit were eluted with 50% ATP, CTP, GTP, UTP, DPN, CMP, AMP, UMP, ADP, FDP, ethanol containing 3 ml of concentrated NH,OH per 100 ml. and cr-glycero-P (Table II). The ratio of ATP to AMP and ADP The susceptibility of compounds to hydrolysis by 5'-nucleo-was lower than expected at myokinase equilibrium, suggesting tidase or bacterial alkaline phosphatase was determined as that some ATP breakdown occurred during extraction. Pi was previously described (22). For the determination of acid lability, the predominant component of the acid-soluble fraction, and the a 32P-labeled compound was treated for 15 min with 1 M HCl at other compounds occurred in amounts ranging from 0.2 to 20 loo", and liberated 32Pi was assayed with the reagent of Sugino and Miyoshi (23), which specifically precipitates Pi.  Fig. 6) and spores were pro-3-l' dehydrogenase under previously described conditions (24). duced in SNB. Acid-soluble phosphate and nucleic acids were The stoichiometric conversion of PGA to glyceraldehyde-3-P extracted as described under "Methods." Phospholipids were was obtained by trapping the product with hydrazine, and DPN extracted as previously described (26). Residual phosphate is that fraction not released by any of these treatments (cf. Fig. 4). production was measured by the decrease in absorbance at 340 I mk. This assay is specific for the 3-P isomer of PGA; 2-P. and B. subtilis contained 1.0 to 1.6% of their dry weight as phosphorus, but the distribution was quite different from that found in vegetative cells. Nucleic acids made up 45% of the total phosphate, but the acid-soluble fraction was only 3 to 7% in spores (Table I). A fraction resistant to extraction with hot trichloracetic acid (the "residual fraction") made up 40 to 50% of the total phosphate in spores of both species (Table  I).

Acid-soluble Phosphate Compounds
The acid-soluble phosphate extracted from B. megaterium spores was resolved into at least nine components by paper electrophoresis (Fig. 2). Two of these components made up more than 90% of the extracted phosphate. One, containing 20% of the acid-soluble phosphate, was identified as Pi. The second major species, slightly more acidic than Pi at pH .3.5, represented about 75% of the extracted phosphate.

PGA as Predominant
Acid-soluble Phosphate Compound in Spores-The most abundant acid-soluble phosphate compound in spores was purified by chromatography on Dowex 1, from which a peak of constant, specific radioactivity was obtained. The purified compound had the same electrophoretic mobility at pH 3.5 or 5.0 as phosphoglyceric acid. It reacted with 4,5-dihydroxy-2,7-naphthalene disulfonic acid (chromotropic acid) under defined conditions (25), yielding a product with the same absorption spectrum and extinction coefficient as glyceric acid and its phosphomonoesters.
The intact compound was insensitive to periodate but, after the removal of phosphate with bac- . Neutralized were grown in SNB containing a2P,, harvested by Millipore filtra-extracts were subjected to electrophoresis at pH 3.5. The pattion, and immediately extracted with cold trichloracetic acid. The terns above were obtained by scanning the electropherogram for neutralized extract was subjected to chromatography in two di-radioactivity at two levels of instrument sensitivity. Radioacmensions as described under "Methods" and radioactive regions tivity is represented on the ordinate scale, and the positions of were visualized by autoradiography. unlabeled markers are indicated by arrows.
terial alkaline phosphatase, periodate liberated 1 mole of formaldehyde per mole of phosphate originally present. The infrared spectrum of the compound from spores was identical with that of authentic 3-P-glyceric acid.
Glyceraldehyde-3-P dehydrogenase and PGA kinase together catalyzed the oxidation of DPNH in the presence of the spore compound, ATP, and Mg++. When hydrazine was used to trap the glyceraldehyde-P and drive the reaction to completion, 1.0 eq of DPN was produced per phosphate.
Under exactly these assay conditions only 3-phospho-n-glyceric acid, and not the 2 isomer, served as substrate.
These analyses, summarized in Table III, establish the identity of the spore compound as  3. Thin layer chromatographic separation of acid-soluble phosphate compounds from spores. Extracts of labeled spores were prepared as in Fig. 2 and fractionated as in Fig. 1. 3-P-n-glyceric acid. Spores of B. subtilis, B. cereus, and Bacillus thuringiensis, produced in SNB medium, also contained PGA as the major component of their acid-soluble pools (26), as did B. megaterium spores produced in SNB containing Pi at a growthlimiting concentration (0.2 mM) or in a synthetic (11) medium, Less Abundant Acid-soluble Phosphate Compounds in Spores---Minor components of B. megaterium spore extracts (Fig. 3) have been tentatively identified, on the basis of criteria summarized in Table IV, as AMP, ADP, CMP, GMP, UMP, glycero-P, and DPN. Each of these components represents less than 5% of the acid-soluble phosphate of spores.
The levels of ATP and other nucleoside triphosphates in spores of B. megatwium and B. subtilis were too low to detect by the methods used here. When "C-ATP was added to dormant spores before the extraction of their acid-soluble pools, more than 90% of the l*C was recovered in the extracts as ATP. Yet, when extracts of 82P-labeled spores were fractionated by paper electrophoresis or anion exchange column chromatography, essentially no "P was associated with the l*C-ATP. Furthermore, when spore extracts were fractionated by thin layer chromatography as in Fig. 3  Spores were produced in "low Pi SNB" supplemented with a*P (final Pi concentration was 1 mM). Acid-soluble components of spores, extracted and separated as in Fig. 3, were eluted and characterized as in Table II PGA was released from spores at 93' (in water) after 20 to 30 min (26).
Similarly, spores took up less than 6% as much external '*Pi as the Pi in the spores. No germination occurred in these experiments. Release of spore phosphate at 100" in neutral, acid, and alkaline solutions.
Cleaned spores labeled with SzP were held at 100" in either 0.1 M potassium phosphate buffer (pH 7.0), 5% trichoracetic acid (TCA), or 0.1 M NaOH. At the indicated times, aliquots were chilled and then centrifuged for 10 min at 15,000 X g, and the fraction of the radioactivity in the supernatant fluid was determined.  II  I  I  I  I  I  I  I  II  11  11  I  I  I I  I I  I I  I  I I  1  PGA to exchange with endogenous pools of these compounds points to their relative inaccessibility in the dormant spore. Free glutamate in spores was also nonexchangeable (8).

Acid-insoluble
Phosphate Compounds in SpwesDNA and RNA made up about 45% of the total spore phosphate in B. subtilis and B. megaterium (Table I). The amount of nucleic acid per cell or spore and the ratios of RNA to DNA were in good agreement with other published values (5,12,28).
Spores of both species contained about half of their phosphate in a form not extractable under conditions normally used to extract nucleic acids. Spores held at 100" in neutral buffer released a maximum of 7 to 12% of their total phosphate (Fig. 4), including PGA. Treatment of intact spores with 5% trichloracetic acid at 100' quickly released 50 to 60% of their total phosphate, but further treatment caused little or no additional phosphate release (Fig. 4). Similarly, 0.1 M NaOH at 100" extracted 55 to 60% of the spore phosphate within 5 min, but longer treatment gave no more release (Fig. 4). This "resistant residue" fraction was described by Fita-James (5) and appears to be associated with spore coats. Although the exact composition of residue phosphate is not known, there is a report (29) that, in B. megaterium spores, phosphomuramic acid is present.
When B. subtilis spores were mechanically disrupted by Minimill treatment, almost all of the phosphate in the water extract was insoluble in cold 5% trichloracetic acid. The acid-insoluble fraction of this extract consisted chiefly of nucleic acid, but it also contained a non-nucleic acid component, representing about 8% of the total spore phosphate.
This fraction did not dissolve m neutral buffer after precipitation with trichloracetic acid, whereas the nucleic acid dissolved readily.
The fraction did dissolve in 1 mM NaOH, giving a viscous yellow solution. This component has not been identified although some of its properties suggest that it may be a cell wall polymer, derived, perhaps, from the spore coat.
Variations in Phosphate Content of Cells during Growth and Sporulation-The uptake of Pi from the medium paralleled growth in both B. megaterium and B. s&i&, reaching a maximum at the end of exponential growth (Fig. 5). In the following 2 hours, there was a rapid loss of about '10% of the total phosphate, after which the phosphate content remained nearly constant for 2 hours. At about the time that refractile spores appeared, but before any sporangial lysis was apparent, another rapid decline in phosphate content began. Twenty-five hours after inoculation, both cultures contained more than 9O70 free spores, with a phosphate content of about 40% the amount in cells at the end of growth.
Washing the spores reduced this value to about 25y0 of the amount in vegetative cells.
Acid-soluble phosphate represented about 20% of total phosphate throughout the growth and sporulation of B. megaterium. The amount of acid-soluble phosphate reached a maximum after the end of exponential growth, and then remained nearly constant for 4 hours before decreasing along with the total phosphate.
Electrophoretic fractionation of these acid-soluble fractions revealed no major qualitative changes in the patterns of these compounds during growth and sporulation. There were, however, significant changes in the relative 'amounts of these components during sporulation, including an increase in the level of PGA.

PGA Synthesis during Sporulation
Kinetics of PGA Accumulation-PGA accumulation began about 1 hour before the appearance of refractile spores (Fig. 6) not extracted by suspending cells in 5% trichloracetic acid without sonic disruption; its release required treatment at 100" in water.
This "inaccessible" fraction of PGA accumulated at the same time as refractile spores appeared (Fig. 6). At about the time of sporangial lysis, the level of PGA decreased by about 50%, leaving only the "inaccessible" fraction, which was localized in the spores. PGA in the culture medium was not determined. The low and constant level of PGA present in hot water extracts of growing cells and during early sporulation may be an artifact of the hot water extraction.
Essentially no PGA was found in Forespore as Probable Site of PGrl Synthesis and Impermeability of Forespore-We attempted to chase the 32P-label from previously labeled P-glycerate by adding unlabeled Pi at intervals during PGA accumulation (Fig. 7). A 25.fold excess of unlabeled Pi, added to the medium 1 hour before the first refractile spores appeared, when PGA synthesis had not yet begun, caused a dilution of only 37% in the specific radioactivity of PGA isolated from mature spores. Similar chases at later times, when 7, 25, and 75"/c of the cells contained refractile spores, resulted in less than a 57, decrease of the PGA specific radioactivity in mature spores, compared with that in a culture with no chase. The Pi isolated from spores chased at the same times showed a 5-fold decrease in specific radioactivity with the earliest chase, but progressively less dilution in specific radioactivity occurred when the chase came later (Fig. 7). These results suggest that the forespore compartment at this stage is relatively inaccessible to Pi in the medium.
Nonetheless, Pi does cuter the sporangium, and is incorporated into the outer layers of the spore, when added to the medium even at these later stages of sporulation.
Insolubility in hot acid was used to define the coat phosphate fraction and to determine the time of its synthesis.
At least part of the coat phosphate fraction was synthesized quite lat.e, about 2 hours after PGA synthesis was completed (Fig. 7). In contrast t'o the results with PGA, the phosphate in this coat fraction was diluted by Pi added to the medium late in sporulation (Fig. 7). DISCUSSIOX Spcra Pool of :lcid-soluble Phosphate Is Unique-The absence of metabolic activity in dormant spores is reflected in their acidsoluble phosphate compounds, which differ strikingly both in amount. and in kind from those of vegetative cells. Spores contain only 3 to 7% of their phosphate as acid-soluble compounds, a fraction appreciably lower than the 20 to 307, present in vegetative cells and sporangia.
Even more significant are the differences in the compositions of the acid-soluble phosphate pools. Spores contain very low levels of ATP, whereas ATP is a major component of vegetative cells. On the other hand, 3-P-glyceric acid constitutes as much as 75y0 of the acid-soluble phosphate in the spore, but no more than 57, of the acid-soluble phosphate of the vegetative cell. PGA is present at these high levels in B. megaterium spores produced in a rich or minimal medium, and is also the predominant component in the acidsoluble phosphate pools in spores of three other species, B. cereus, B. subtilis, and B. thuringiensis (26).
Spores contain, in addition to PGA and Pi, much smaller amounts of other acid-soluble phosphate compounds, including DPN, the nucleoside 5'-monophosphates, ADP, and glycero-P. Some of these less abundant spore components are similar to or identical with the major acid-soluble phosphate compounds of vegetative cells. Moreover, the levels of these components (e.g. I)PN) are comparable in vegetative cells and spores, when expressed in terms of cell or spore dry weight; a more detailed analysis of adenine, pyridine, and flavin nucleotides in spores and vegetative cells will be preseuted elsewhere.2 PG=1 May Be Made in Forespore-Our observations of a slight increase in acid-solul le phosphate after the end of growth and a 2 P. Setlow and A. Kornberg, manuscript in preparation.
sharp decrease during sporangial lysis agree with those of Yamakawa et al. (6). Their "sugar-l I" fraction, which should have included PGA, increased about 2-fold during the period of spore formation, and then decreased with sporangial lysis. We have shown that the amount of PGA eventually found in spores is about half of the total amount accumulated in sporangia, the remainder being lost during sporangial lysis, perhaps by release to the medium.
PGA became inaccessible to direct extraction with trichloracetic acid at the time that refractile spores appeared.
The large dilution of 32P in the spore coat phosphate, by Pi added to the medium indicates access of exogenous Pi to the mother cell compartment.
In contrast, the relatively slight dilution of a2P in PGA of the spore, when a large excess of exogenous Pi was added at the onset of PGA synthesis, suggests that PGA is synthesized from a precursor pool distinct from that of the mother cell, perhaps within the forespore.
Alternatively, the size of the pool of phosphorylated precursors for PGA synthesis may be large compared with the pool of spore coat precursors. The fact that the internal Pi pool was diluted less than coat phosphate makes this alternative less attractive to us.
It is very difficult physically to separate the forespore from its mother cell and thus to compare the metabolism in the two compartments.
The preferential extraction of small metabolites from one of the two compartmeuts and the different accessibilities of the compartments to the external medium may possibly offer means of exploring metabolism in the forespore compartment even in intact sporangia.
Role of dcid-soluble Pool in Spore-The role of small phosphate compounds, and particularly of PGA in spores, is not clear. These compounds could be metabolic byproducts of sporulation, without significance for the maintenance of dormancy or the mechanics of germination.
However, the similarity in distribution of phosphate compounds in spores of different species, produced in different media, suggests that t.his particular collection of small molecules is in some way essential to the spore. Clearly, coenzymes such as pyridine and adenine nucleotides are needed during germination, when metabolic activity sets in rapidly. Of particular interest to us is the possibility that PGA serves as a ready source of energy in the earliest stages of germination.
It is far more stable chemically than BTP, and can yield ATP via phosphoenolpyruvate without the expenditure of energy. The amount of PGA in spores, 10 to 15 pmoles per g, dry weight, is actually 3 to 5 times greater than the amount of ATP in vegetative cells2 The possibility of PGA serving as an "ATP reservoir" in the spore is explored further in the following paper (7).