Phosphate starvation-inducible synthesis of the alpha-subunit of the pyrophosphate-dependent phosphofructokinase in black mustard suspension cells.

PP(i)-dependent phosphofructokinase (PFP) activity, measured in the forward direction, increased approximately 19-fold when suspension cell cultures of black mustard (Brassica nigra) were subjected to 18 days of P(i) deprivation. Fructose 2,6-bisphosphate (2 microM) elicited a 10-fold activation of PFP from P(i)-deficient cells, compared to only a 2-fold activation of the enzyme from nutrient-sufficient cells. Also, PFP from P(i)-starved cells exhibited a greater affinity for the activator (Ka = 0.09 microM) than the enzyme from nutrient-sufficient cells (Ka = 0.32 microM). Western blots of extracts from P(i)-deficient cells were probed with rabbit anti-(potato tuber PFP) immune serum and revealed equal intensity staining immunoreactive polypeptides of M(r) 66,000 (alpha-subunit) and 60,000 (beta-subunit) that co-migrated with the alpha- and beta-subunits of homogeneous potato tuber PFP. By contrast, only the M(r) 60,000 beta-subunit was observed on immunoblots of extracts prepared from nutrient-sufficient cells. Quantification of immunoblots indicated that in black mustard cells experiencing transition from P(i) sufficiency to deficiency or vice versa, the relative amount of immunoreactive alpha-subunit correlated with the degree of activation of PFP by fructose 2,6-bisphosphate. These observations provide additional evidence that (i) plant PFP is an adaptive enzyme that may function in glycolysis during P(i) deprivation, and (ii) the alpha-subunit acts as a regulatory protein in controlling the catalytic activity of the beta-subunit and its regulation by fructose 2,6-bisphosphate.

tracts from Pi-deficient cells were probed with rabbit anti-(potato tuber PFP) immune serum and revealed equal intensity staining immunoreactive polypeptides of M , 66,000 (a-subunit) and 60,000 (&subunit) that co-migrated with the a-and &subunits of homogeneous potato tuber PFP. By contrast, only the M, 60,000 8subunit was observed on immunoblots of extracts prepared from nutrient-sufficient cells. Quantification of immunoblots indicated that in black mustard cells experiencing transition from Pi sufficiency to deficiency or vice versa, the relative amount of immunoreactive a-subunit correlated with the degree of activation of PFP by fructose 2,6-bisphosphate. These observations provide additional evidence that (i) plant PFP is an adaptive enzyme that may function in glycolysis during Pi deprivation, and (ii) the a-subunit acts as a regulatory protein in controlling the catalytic activity of the &subunit and its regulation by fructose 2,6-bisphosphate.
In animals, the direction of carbon flow through Fru-6-Pl and Fru-1,6-Pz is coordinated by the opposing activities of ATP:~-fructose-6-phosphate 1-phosphotransferase (EC 2.7.1.11) and ~-fructose-1,6-bisphosphate l-phosphohydrolase (EC 3.1.3.11). These enzymes catalyze thermodynamically irreversible reactions in vivo and are allosterically regulated by Fru-2,6-Pz, which activates ATP:~-fructose-B-phosphate 1-phosphotransferase and inhibits D-fructose-1,6-* This work was supported by the Natural Sciences and Engineering Research Council of Canada. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. $ Present address: Dept. of Biochemistry, University of Nebraska, Lincoln, NE 68583-0718.
The PFP of many plants consists of two pairs of different subunits of approximately M, 66,000 (a-subunit) and 60,000 (/?-subunit) (4-7). Several recent reports, however, have clearly demonstrated that a variety of molecular forms of PFP can exist within and between different plant tissues (8)(9)(10). Although there is an approximate 60% homology between the deduced amino acid sequence of the a-and @-subunits of PFP from potato tubers and castor oil seeds (ll), the two subunits are immunologically distinct (6,7). The functions of the aand /?-subunits of PFP have not been fully resolved. However, various evidence indicates that the @-subunit contains the catalytic site, while the a-subunit is involved in the regulation of catalytic activity by Fru-2,6-Pz (7,9,11,12).
We have previously shown that Pi starvation of heterotrophic suspension cells of Brassica nigra (black mustard) results in a large elevation in the activity of PFP, as well as dramatic reductions in the intracellular concentrations of adenylates and Pi (29). By contrast, PPi levels remain selectively high in Pi-deficient cells of B. nigra (29). These findings led us to postulate that the PFP of P,-deprived cells of B. nigra plays a pivotal role in the catalysis of one step of a series of Pi starvation-inducible glycolytic "bypasses." It was suggested (29) that these alternative glycolytic reactions circumvent the adenylate-and Pi-dependent reactions of glycolysis, thus allowing glycolysis to proceed in Pi-deprived cells. Two of these bypass reactions, PFP and phosphoenolpyruvate phosphatase, may also fulfill an additional role as a Pi-recycling system that converts esterified phosphate to Pi that would be rapidly reassimilated by the Pi-deficient cells. Potent feedback inhibition of both PFP (forward direction) and 21901 phosphoenolpyruvate phosphatase by their product Pi (14,301 is consistent with the proposal that these enzymes are active during Pi starvation.
The present study was undertaken to gain insight into the mechanism of regulation of PFP activity with respect to the Pi status of suspension cells of B. nigra. Our results indicate that de novo synthesis of the a-subunit of PFP occurs in response to Pi starvation and that this event coincides with a significant enhancement in the enzyme's sensitivity to Fru-2,6-P2.

EXPERIMENTAL PROCEDURES
Materials-Coupling enzymes, NADH, and EGTA were obtained from Boehringer Mannheim. Tris base and SDS were from Schwarz/ Mann. Dithiothreitol and 5-bromo-4-chloro-3-indolyl phosphate were supplied by Research Organics. Bradford protein assay reagent and ammonium persulfate were from Bio-Rad. Polyvinylidene difluoride membrane (Immobilon transfer; 0.45-pm pore size) was from Millipore. Sephadex G-25 (medium) was obtained from Pharmacia LKB Biotechnology Inc. Other biochemicals, acrylamide, N,N'-methylenebis-acrylamide, goat anti-rabbit immunoglobulin antibody-alkaline phosphatase conjugate, nitro blue tetrazolium, and polyvinylpolypyrrolidone were purchased from Sigma. All other reagents were of analytical grade and were obtained from British Drug House. Homogeneous potato tuber PFP and rabbit anti-(potato tuber PFP) immune serum were obtained as described previously (31).
Heterotrophic suspension cells of B. nigra were obtained as previously reported (30). Packed suspension cells (about 6 ml) were inoculated into 44 ml of Murashige-Skoog media (32) containing 6% (w/ v) sucrose and 0 or 10 mM Pi. Incubation was on a rotational shaker (120 rpm) at 25 "C for 18 days, with subculturing every 7th day. Cells for use in Pi-refeeding experiments were prepared by transfer of packed Pideficient cells (6 ml) into 44 ml of Murashige-Skoog media containing 10 mM P,, with subsequent culturing as described above. Cells were harvested and stored as described previously (30). All analytical procedures were performed on at least two extracts of suspension cells experiencing the transition from P, sufficiency to deficiency, or vice versa.
Enzyme Extraction-All procedures were carried out at 4 "C. Frozen cells (0.5-1.5 g) were ground (l:l, w/v) for 5 min using a pestle and a mortar containing a small scoop of acid-washed sand. The homogenate was centrifuged for 15 min at 16,000 X g using an Eppendorf microcentrifuge. With the exception of pH (7.5 instead of 6.9), the extraction buffer was as previously described (29). For enzyme assays, 1.5-ml aliquots of clarified homogenates were desalted a t 0.5 ml.min" on a column (1.0 X 14 cm) of Sephadex G-25 that had been pre-equilibrated in extraction buffer minus polyvinylpolypyrrolidone.
Enzyme Assays and Kinetic Studies-All cuvettes and glassware used for enzyme assays were treated overnight with 6 N HC1 and rinsed with deionized water prior to use. Fru-6-P and Fru-l,6-Pp were also acid treated (titrated to pH 3 with HC1, incubated for 15 min, and then neutralized with NaOH) to hydrolyze any contaminating traces of Fru-2,6-P2. Enzymes were assayed at 25 "C in a 1.0-ml final volume by following the reduction of NADP+ or oxidation of NADH a t 340 nm using a Gilford recording spectrophotometer. PFP was assayed in the forward direction in 50 mM Tris-HC1 (pH 7.5) containing 5 mM Fru-6-P, 0.4 mM PPi, 0.15 mM NADH, 5 mM MgCL, 1 unit of aldolase, 10 units of triose-phosphate isomerase, and 1 unit of glycerol-3-phosphate dehydrogenase. Assays were initiated by the addition of PPI, and PFP activity was monitored for approximately 3 min. Fru-2,6-P2 (2 p~ unless otherwise indicated) was then added, and the Fru-2,6-P2-stimulated PFP activity was recorded.
In a separate cuvette, desalted extract was added to a reaction mixture containing 10 p~ Fru-2,6-P2 but lacking Pi, and ~-fructose-1,6-bisphosphate 1-phosphohydrolase activity was recorded as C. Napi Coupling enzymes were desalted prior to use. All assays were conducted in duplicate; optimized with respect to pH, substrate, and cofactor concentrations; and corrected for any contaminating NADH oxidase or NADP+ reductase activity.
Activity in all assays was proportional to the amount of extract added and remained linear with respect to time. One unit of enzyme activity is defined as the amount of enzyme required to catalyze the formation of 1 pmol of product. Other Methods-Immunotitration of PFP activity was performed as described by Moorhead and Plaxton (31) with the exception that 0.088 unit of homogeneous potato tuber PFP and 0.068 unit of PFP from extracts of Pi-deficient B. nigra were incubated with various amounts of immune or pre-immune sera. Protein concentrations were determined by the method of Bradford (35) using bovine y-globulin as standard.

RESULTS
When suspension cells of B. nigra were subjected to 18 days of Pi deprivation, the extractable activity of the PFP was increased approximately 4-and 19-fold when assayed in the forward direction in the absence and presence, respectively, of 2 pM Fru-2,6-Pz (Fig. 1). Moreover, the enzyme from Pideprived cells exhibited a more than 3-fold greater affinity for the activator than the PFP from nutrient-sufficient cells (Fig. 1) did. Thus, a significant portion of the Pi starvationdependent induction of PFP activity can be attributed to a marked increase in sensitivity of the enzyme to Fru-2,6-P2.
The extractable activity of the Fru-2,6-P~-stimulated PFP assayed in the reverse direction increased from 0.025 to 0.038 unit.mg" when suspension cells of B. nigra were subjected to 18 days of Pi deprivation. This change could be attributed to a small increase in sensitivity of the enzyme's reverse activity to Fru-2,6-P2 since activities in the reverse direction measured in the absence of the activator were 0.016 and 0.017 unit mg" for the PFP from Pi-fed and Pi-starved cells, respectively. As Pi starvation caused a much greater elevation of PFP activity in the forward direction, the ratio of Fru-2,6-Pz-stimulated forward-reverse activities was increased about %fold following Pi deprivation. Interestingly, the activity of ~-fructose-1,6-bisphosphate 1- Correlation between PFP subunit ratio (CY/@) and relative PFP activity and -fold activation by Fru-2,6-P2. PFP activity was determined in the forward direction with saturating FN-2,6-P~ and is expressed relative to the maximal specific activity (0.062 unit.mg protein") that was observed in Pi-starved cells. -Fold activation by Fru-2,6-P2 was calculated as the -fold increase in PFP activity elicited by the addition of 2 p~ Fru-2,6-P2. Data were fitted using a computer statistics program (Statview Graphics), and in both cases the line of best fit was a third-order polynomial. combined activities of PFP in the forward direction and Dfructose-1,6-bisphosphate l-phosphohydrolase could function as a cytosolic pyrophosphatase, which makes both phosphates of PPi available to the Pi-deprived cells.
Immunological studies using rabbit anti-(potato tuber PFP) immune serum (31) were initiated to examine the molecular properties of the PFP from Pi-fed and -starved B. nigra cells.
Increasing amounts of the anti-(PFP) immune serum immunoprecipitated 100% of the activities of PFP from potato tubers or Pi-starved suspension cells of B. nigra (Fig. 2). The amount of antiserum required for 50% immunoprecipitation was about 15 and 45 p1. unit-' for the PFP from potato tubers and B. nigra, respectively.
Western blots of extracts prepared from Pi-starved B. nigra revealed immunoreactive polypeptides of 66 (a-subunit) and 60 kDa (@-subunit) staining in a 1:l ratio that co-migrated with the a-and @-subunits of homogeneous potato tuber PFP (Figs. 3 and 5). However, only the 60-kDa polypeptide was observed on immunoblots of extracts prepared from Pi-sufficient cells (Figs. 3 and 5). To ascertain whether the various immunoreactive polypeptides that were observed arose following tissue extraction via proteolytic degradation of PFP by endogenous proteases, Pi-fed and -starved cells of B. nigra were extracted under totally denaturing conditions in the presence of 10% (v/v) trichloroacetic acid (36). Trichloroacetic acid-precipitated proteins were solubilized and analyzed by immunoblotting; the pattern and intensity of the immunoreactive band(s) did not differ from those observed on immunoblots of the respective extracts shown in Figs. 3 and 5. Fig. 3 shows the immunoblot as well as the subunit ratio (a:@) and activity profile of the PFP in suspension cells during the transition from Pi sufficiency to deficiency. Induction of PFP activity and the appearance of the a-subunit began approximately 6 days after the transfer of Pi-fed cells to media lacking Pi and was complete by day 18 when an approximate 1:1 ratio of a-subunit:@-subunit was achieved (Fig. 3). Hence, the apparent Pi starvation-induced de novo synthesis of the a-subunit was coincident with an increase in the sensitivity of the enzyme to Fru-2,6-P2. No further increase in enzyme activity or the amount of the a-subunit occurred when the time course was extended to 22 days (data not shown). The subunit ratio (a:@) correlated well with Fru-2,6-P2-stimulated PFP activity and -fold activation of the enzyme by Fru-2,6-P 2 (Fig. 4). Fig. 5 presents immunoblots of extracts from Pi-starved cells that became Pi-sufficient when supplied with 10 mM Pi. Within 1 day of Pi addition, the a-subunit was no longer detectable on immunoblots, and the forward activities of the Fru-2,6-P2-stimulated and unstimulated PFP were reduced to the corresponding values originally observed for Pi-sufficient cells (see Figs. 1 and 3C).

DISCUSSION
The findings of the present investigation indicate that the amount of the a-subunit of PFP is tightly regulated in suspension cells of B. nigra and that this regulation is dependent on cellular Pi status. In contrast, the @-subunit of PFP from B. nigra is constitutively expressed under all nutrient regimes. The induction of PFP activity by Pi starvation of B. nigra appears to be based upon de novo synthesis of the enzyme's a-subunit, leading to a significant enhancement in sensitivity of the enzyme to its activator Fru-2,6-P2 (Fig. 3). Similarly, the large reduction in PFP activity that occurs when Pideprived cells become Pi-sufficient arises from an inhibition of synthesis and/or enhanced degradation of the a-subunit (Fig. 5). It is evident that some form of proteolytic specificity toward the a-subunit must exist to facilitate the selective disappearance of this polypeptide upon Pi refeeding. Overall, these results provide additional evidence that the a-subunit may function as a regulatory protein in controlling the catalytic activity of the @-subunit and its regulation by Fru-2,6-P2. The concentration of Fru-2,6-P2 in the cytosol of Pideprived B. nigra was previously estimated to be about 0.5 p M (29), a level that almost fully saturates the PFP from the Pistarved cells (Fig. 1).
Recent gel filtration chromatography studies have revealed that comparable to the PFP of many plants (4-7), the native enzyme from Pi-deprived B. nigra probably exists as an a& heterotetramer having a molecular mass of approximately 260 kDa.3 Further investigations involving the purified enzyme from nutrient-sufficient and Pi-deficient cells are required to fully resolve the structure-function relationships of the PFP isoforms of B. nigra. The results outlined above, however, are in accord with our previous suggestion (29) that the PFP of B. nigra is an adaptive enzyme that functions as a Pi starvation-inducible glycolytic bypass to ATP:~-fructose-6-phosphate 1-phosphotransferase when intracellular pools of ATP and Pi, but not PPi, are greatly depleted. It is of interest to note that the PFP of several heterotrophic plant tissues has also been proposed to operate as a glycolytic bypass to ATP:Dfructose-6-phosphate 1-phosphotransferase during periods of anaerobiosis (37). The use of PPi rather than ATP could confer a significant energetic advantage to plants subjected to environmental stresses such as Pi deprivation or anoxia.
Phosphate starvation-inducible synthesis of the 66-kDa asubunit of PFP in suspension cells of B. nigra is coincident with de m u 0 synthesis of cytosolic phosphoenolpyruvate car-boxyla~e,~ vacuolar phosphoenolpyruvate phosphatase, and a cell wall-localized nonspecific acid phosphatase (38). Parallel induction of these enzymes with a simultaneous enhancement in cellular Pi absorptive capacity (39) points to the existence of a plant "Pi stimulon" (i.e. a set of genes that are coregulated by Pi) as has been demonstrated in a variety of microorganisms (40).