Biphasic Kinetic Behavior of Nitrate Reductase from ... - NCBI

Received for publication December 27, 1991 Accepted April 6, 1992

Plant Physiol. (1992) 100, 157-163 0032-0889/92/1 00/0157/07/$01 .00/0

Biphasic Kinetic Behavior of Nitrate Reductase from Heterocystous, Nitrogen-Fixing Cyanobacteria' lose Martin-Nieto2, Enrique Flores*, and Antonia Herrero Instituto de Bioquimica Vegetal y Fotosintesis, Universidad de Sevilla-Consejo Superior de Investigaciones Cientificas, Facultad de Biologia, Apartado 1113, E-41080 Sevilla, Spain ABSTRACT

the nature of the redox centers active in cyanobacterial NR, the existence in the enzyme of both iron-sulfur center(s) (24) and of a molybdenum cofactor has been inferred (2, 22). Routine assays of NR activity in cyanobacteria make use of dithionite-reduced MV as the electron donor (11, 16, 20). Reported Km values of the enzyme for nitrate determined with this assay are in the range of 0.7 to 2.1 mm in Synechococcus spp. (5, 8, 19) and P. boryanum (16). Catalytic or biochemical properties of NR from heterocyst-forming cyanobacteria are unknown. In this paper, we report a biphasic kinetic behavior detected in NR activity from N2-fixing, heterocystous cyanobacteria. The distribution of this phenomenon among different groups of cyanobacteria is also addressed. Possible kinetic and structural models for biphasic NRs are analyzed and discussed.

Nitrate reductase activity from filamentous, heterocyst-forming cyanobacteria showed a biphasic kinetic behavior with respect to nitrate as the variable substrate. Two kinetic components were detected, the first showing a higher affinity for nitrate (K., 0.050.25 mM) and a lower catalytic activity and the second showing a lower affinity for nitrate (Km, 5-25 mM) and a higher (3- to 5-fold) catalytic activity. In contrast, among unicellular cyanobacteria, most representatives studied exhibited a monophasic, MichaelisMenten kinetic pattern for nitrate reductase activity. Biphasic kinetics remained unchanged with the use of different assay conditions (i.e. cell disruption or permeabilization, two different electron donors) or throughout partial purification of the enzyme.

MATERIALS AND METHODS

Cyanobacteria are unique among microorganisms in being able to carry out an oxygenic, plant-type photosynthesis within a prokaryotic cell structure. These organisms can use nitrate or ammonium as the nitrogen source for growth (8). Many of them, however, have the ability to fix N2. N2 fixation in filamentous strains generally takes place inside specialized cells called heterocysts (30). Nitrate assimilation involves entrance of nitrate into the cell and its subsequent reduction to ammonia. In cyanobacteria, nitrate reduction takes place in two successive steps catalyzed by the enzymes NR3 and nitrite reductase, which make use of reduced Fd as the electron donor (8). NR, the enzyme responsible for the two-electron reduction of nitrate to nitrite, has been shown to be associated with thylakoid membranes in both unicellular (20) and filamentous cyanobacteria (10, 16, 25). The enzyme has been purified to homogeneity and characterized as a molybdo-iron protein consisting of a single polypeptide chain with a molecular mass of about 75 kD in the unicellular cyanobacterium Synechococcus sp. (5, 9), and of 83 to 85 kD in the filamentous, nonheterocystous cyanobacterium Plectonema boryanum (24). Although no definitive conclusion has been established about

Strains and Culture Conditions

Filamentous and unicellular cyanobacterial strains used in this work are listed in Table I. All the filamentous strains used belong to sections IV (Anabaena, Calothrix, and Nostoc) or V (Fischerella) as described by Rippka et al. (26), and these strains are capable of aerobic N2 fixation, with the exception of strains M-131 (3) and PCC 7601 (17), which are unable to develop heterocysts. Unicellular strains of section 1 (26) (Synechocystis, Synechococcus, and Gloeocapsa) were also studied, including strains CS501 and CS502, which have been recently isolated from nature. All organisms were grown photoautotrophically at 300C, with the exception of Fischerella and Synechococcus strains, which were grown at 400C. Strains PCC 7601, ATCC 29105, PCC 6301, PCC 7942, and CS501 were grown in the synthetic medium previously described (11), supplemented with 20 mm KNO3 and sparged with 5% CO2 in air. Strain ATCC 29413 (P9 derivative [13]), when cultured in large scale, was grown under the same conditions, except that concentrations of phosphate and bicarbonate in the medium were reduced by 50%. The other strains were grown in synthetic BG11 medium, which contained 17.6 mm NaNO3 (26), and sparged with air.

1 Supported by grant No. BI089-0527 from Comisi6n Interministerial de Ciencia y Tecnologia, Spain. 2 Present address: Division of Biology and Center for Basic Cancer Research, Kansas State University, 348 Ackert Hall, Manhattan, Kansas 66506. 3Abbreviations: NR, nitrate reductase; MV, methyl viologen; BPB, bromphenol blue; MTA, mixed alkyltrimethylammonium bromide;

Enzyme Assays Unless otherwise stated, NR activity was determined with dithionite-reduced MV as the electron donor (11). Cells permeabilized by treatment with toluene or with the detergent MTA provided a measurement of in situ activity. Crude

Tricine, N-tris(hydroxymethyl)methylglycine. 157

MARTIN-NIETO ET AL.

158

Plant Physiol. Vol. 100, 1992

Table I. Apparent Values of Kinetic Parameters Km (for Nitrate) and Vmax of Nitrate Reductase from Several Cyanobacterial Strains All strains were grown on nitrate-containing media, under the conditions described in 'Materials and Methods." Nitrate reductase activity was assayed under the conditions specified in the legend of Figure 1A, in cells permeabilized by treatment with toluene (Tol) or MTA or in crude extracts (Ce) or partially-purified fractions of the enzyme (P1, 42-fold purification and P2, 185-fold purification). Protein amounts assayed were 0.2-0.5 mg, except for P1, 20 ,ug, and P2, 5 ,ug. Apparent values of Km (mM) and Vmax (nmol NO2/min -mg of protein) were estimated from lines fitted to the limits of the corresponding Hofstee plots. Km (I) and Vmax (I) are estimates for high affinity components and Km (II) and Vmax (II) are low affinity components of NR. Strains

Cell Material Assayed

Km

(I)

(11)

(1)

Vmax

(11)

V

max

nmol N02-/ mM

Filamentous, N2-fixing Anabaena variabilis ATCC 29413-P9 Anabaena sp. PCC 7119

Anabaena sp. PCC 7120 Anabaena sp. M-131

Calothrix sp. PCC 7101 Calothrix sp. PCC 7601 Fischerella muscicola UTEX 1829 Nostoc sp. ATCC 29105 Nostoc sp. ATCC 29150 Unicellular Synechocystis sp. PCC 6803

Synechococcus sp. PCC 6301 Synechococcus sp. PCC 7942 Gloeocapsa sp. CS501 Gloeocapsa sp. CS502

Tol Ce P1

min mg protein

P2 Tol Ce Tol Tol MTA Ce MTA Tol

0.15 0.25 0.20 0.19 0.11 0.19 0.13 0.13 0.18 0.13 0.17 0.06

11 12 9 12 22 9 8 12 10 8 7 9

7 11 57 192 8 8 9 10 24 9 17 10

25 31 163 681 46 22 41 26 69 27 45 37

0.22 0.25 0.26 0.22 0.15 0.27 0.18 0.27 0.26 0.25 0.27 0.21

Tol Tol Tol

0.14 0.16 0.07

16 11 8

2 10 3

9 32 11

0.18 0.24 0.21

Ce Tol Tol Tol Tol Tol

0.55 6 0.55 9 1.7 2.0 9 1.3

14 30

25 52

0.37 0.37

extracts and partially purified fractions of the enzyme were used for measurements of in vitro activity. For in situ assays, cells were collected, washed by centrifugation, and resuspended in 5 mim Tricine-NaOH (pH 8.1) buffer. Toluene treatment was achieved before the assay by mixing the cell suspension with toluene (0.2-0.4 ,ul/,g of Chl) by vigorous vortexing for 2 min. MTA permeabilization was performed simply by including 50 Ag of the detergent (MTA-5; Sigma) per mL in the reaction mixture. Crude extracts were obtained after resuspending the cells in 10 mm Tris-HCl buffer (pH 7.5) at 50 to 150 gg of Chl per mL and disrupting the cells by ultrasonic treatment at 60 W for 2 to 5 min. Throughout extraction, solution temperature was maintained at 0 to 40C. After the solution was centrifuged at 2500g for 10 mmn, the activity was assayed in the supernatant, which constituted the crude extract. NR activity was measured at pH 10.5 and 300C for 3 min in the reaction mixture previously described (11), although modified to contain 100 mm KNO3. After the reaction was completed, cell debris was precipitated upon addition of 0.1 M NaOH and 0.1 M ZnSO4, and the amount of nitrite produced was determined (29). One unit of enzyme

47 36 49 78

activity corresponds to the amount of enzyme that catalyzes the formation of 1 gumol of nitrite per min. An altemative NR assay used dithionite-reduced BPB instead of MV as the electron donor (6, 14). This assay was optimized in this work to be used with Anabaena crude extracts. The reaction mixture contained, in a final volume of 1 mL, 100 ,Amol of Tris-HCl (pH 9.0), 100 ,umol of KNO3, 0.5 ,umol of BPB, 1 ,mol of EDTA, 5 umol of Na2S204, and crude extract, obtained as described above. The reaction was started by the addition of 100 uL of dithionite (dissolved in 50 mi NaHCO3), and the reaction mixture was incubated at 300C for 2 min. After the reaction was stopped by vigorous agitation, cell debris was precipitated by the addition of 0.2 mL of 1 M NaOH and 0.2 mL of 1 M ZnSO4, and nitrite formed was determined (29) in an aliquot of the supernatant obtained after centrifugation at 2500g for 10 min. Partial Purification of NR

Methods used were based on procedures previously used for purification of NR from other cyanobacteria (5, 16, 20,

NITRATE REDUCTASE FROM HETEROCYSTOUS CYANOBACTERIA

24). The buffer used throughout the purification was T25, consisting of 25 mi Tris-HCl, 0.5 mm EDTA (pH 7.5). All operations were performed at 0 to 40C. Eight grams (wet weight) of Anabaena variabilis 29413 (P9 derivative) cells grown on nitrate (large-scale cultures) were resuspended in 50 mL of T25 buffer and disrupted by ultrasonic treatment (90 W, 10 min) in the presence of 1 mm PMSF (Sigma). The supernatant obtained after centrifugation at 3300g for 15 min retained about 80% of total cell NR activity and represented the crude extract. This preparation was treated for 2 h with deoxyribonuclease I (Sigma) and ribonuclease I (Sigma) at 50 Ag/mL each, and then solid NaNO3 was added to a final concentration of 1 M to extract NR from membranes. After incubation in the presence of nitrate for 2 h with gentle, occasional mixing, the preparation was centrifuged at 104,000g for 90 min, resulting in most of the NR activity being recovered in the supematant. This was then subjected to fractionated precipitation with ethanol (50-70%). After addition of one volume of cold ethanol, the mixture was incubated on ice for 10 min and then centrifuged at 7900g for 15 min. Ethanol was then added to the supernatant to give a 70% (v/v) ratio, and, after incubation on ice and centrifugation, the pellet was air dried and resuspended in 10 mL of T25 containing 0.15 M NaCl. This fraction was dialyzed three times against 4 L of the same buffer and centrifuged at 43,500g for 20 min. The supernatant obtained was diluted 2-fold with T25 and applied to a DEAE-cellulose DE-52 (Whatman) column (0.8- x 6.8-cm bed) equilibrated with T25. The column was washed with 2.5 bed volumes of T25 containing 0.1 M NaCl, and NR was eluted by a linear gradient of NaCl (0.1-0.4 M; total volume, 140 mL) in T25 at a flow rate of 20 mL/h. Eluted fractions with NR activity were concentrated by dialysis against 40% (w/v) PEG 4000 in 0.1 M NaCl-supplemented T25 buffer and further equilibrated in the same buffer without PEG. Glycerol (3%, v/v) was added to the concentrated enzyme solution, and the mixture was applied to a Sephacryl S-300 Superfine (Pharmacia) column (0.8- x 31-cm bed) equilibrated with 25 mi Tris-HCl, 0.1 mm EDTA, 0.1 M NaCl (pH 7.5). Elution was performed at a flow rate of 29 mL/h.

159

procedure (21) and in cell-free preparations by the method of Bradford (4) with the Bio-Rad protein assay reagent. Egg albumin was the standard for both assays. RESULTS

Kinetic pattems that deviated from the hyperbolic pattem of Michaelis-Menten were obtained for NR from the cyanobacterium A. variabilis ATCC 29413-P9 when assayed with MV as the electron donor in the presence of different concentrations of nitrate. The Hofstee plot shown in Figure 1A illustrates this phenomenon for NR assayed in a crude extract from nitrate-grown cells. Apparent values of the kinetic parameters calculated from extrapolation of the straight segments of the Hofstee plot were: Km(I) = 0.25 mm and Vmax (I) = 11 milliunits/mg of protein for the high-affinity component; Km(II) = 12 mM and Vmax,(II) = 31 milliunits/mg of protein for the low-affinity component. The ratio between the two apparent Vmax values obtained from the Hofstee plot,

A 40 30 0_b

20 c

.....II..

10

E

0

0.

nv ,

10

0

20

30

40

0

E

c

Molecular Mass Estimation The molecular mass of NR was estimated by gel filtration chromatography (1) as described above. Marker proteins (Sigma) were rabbit muscle aldolase (158 kD), BSA (67 kD), and chicken egg albumin (43 kD). Four milligrams of each were dissolved in 0.4 mL of the equilibration buffer and applied to the column, and their elution volumes (Ve) were determined by the absorbance at 280 nm. Void (V0) and total (V,) column volumes were estimated from the elution volumes of dextran blue 2000 and potassium ferricyanide, respectively. Results were plotted as logarithm of molecular mass versus Kav, where Kay

=

(Ve -

Vo)/(Vt - Vo).

Analytical Methods

Chl was estimated from methanol extracts of the cells (18). Protein in whole cells was determined by a modified Lowry

o0'

2

0

v/

[N03-1

4

6

8

10

12

(nmol N02-/mg protein-min.mlMV) Figure 1. Hofstee plots of the effect of nitrate concentration on NR activity determined in crude extracts of A. variabilis 2941 3-P9 with MV (A) or BPB (B) as the electron donor. NR activity was assayed under the conditions indicated in "Materials and Methods," except that different concentrations of KNO3, between 50 ,M and 100 mm, were included. As controls, assays in which KNO3 was omitted from the reaction mixtures were run in parallel. Reaction times were 3 min for the MV-dependent activity and 2 min for the BPBdependent activity. Protein amounts in the assays were 0.33 (A) and 1.9 mg (B).

160

MARTIN-NIETO ET AL.

i.e. the Vmax corresponding to the high-affinity component [Vmax(I)] and the total Vmax [Vmax(t) = Vmax(I) + Vmax(II)] was 0.25. This biphasic kinetic behavior was not a specific characteristic of NR from the cyanobacterium A. variabilis ATCC 29413-P9 but was observed for all heterocystous strains studied in this work (Table I). Biphasic kinetic patterns were evident both in strains belonging to section IV (26) genera (i.e. Anabaena, Nostoc, and Calothrix), which form nonbranched filaments, and in Fischerella, which develops true branched filaments and thus belongs to section V (26). It should be noted that remarkably similar values of Km(I) and Km(II) and of the Vmax(I)/Vmax(t) ratio were obtained for a particular filamentous strain when NR was assayed either in crude extracts or in cells permeabilized with toluene or MTA (Table I). Similar values were also obtained with different cultures of the same strain. NR hyperbolic kinetics for the substrate, nitrate, have been reported for the unicellular cyanobacteria Synechococcus sp. strains PCC 6301 (5, 8) and PCC 7942 (19). For the sake of comparison, the kinetic behavior of NR was studied in these cyanobacteria as well as in other unicellular strains belonging to Gloeocapsa and Synechocystis. The same assay conditions that yielded two kinetic components in filamentous, heterocystous strains were used. Hyperbolic kinetic patterns were found for the Synechococcus and Gloeocapsa representatives, single values of Km and Vmax being obtained (Table I). Km values obtained for strains 6301 and 7942 were very similar to those previously reported (0.7-2.1 mM). For strain Synechocystis sp. PCC 6803, however, a biphasic pattem similar to that observed in heterocystous cyanobacteria was obtained. Similar values of Km(I), Km(II), and Vmax(I)/Vmax(t) were obtained for this strain in the in vitro and in situ assays (Table I). The electron donor BPB was substituted for MV to determine whether the biphasic kinetic pattem of NR was specific for MV. Figure 1B shows the Hofstee plot for the BPBdependent NR activity as a function of nitrate concentration, assayed in a crude extract of A. variabilis. Again, two components were evident in the enzyme system. Km values obtained with BPB were 0.08 and 8.6 ms for the high- and low-affinity components, respectively, and the Vmax(I)/ Vmax(t) ratio was 0.22. These values were reasonably similar to those obtained with the MV-dependent NR assay, indicating that the biphasic patterns are independent of two different electron donors that probably bind to different active sites of the enzyme (6, 14). Because the biphasic kinetic behavior observed could be explained by the existence in the cell of two different enzyme species able to catalyze the reduction of nitrate to nitrite, NR from A. variabilis ATCC 29413-P9 was partially purified. At different steps of the purification, measurements of kinetic parameters of NR with nitrate as the variable substrate were

performed. A crude extract was obtained from nitrate-grown cells of A. variabilis, NR was solubilized from membranes with NaNO3 as a chaotropic agent, and proteins in the supernatant fraction resulting from subsequent ultracentrifugation were fractionally precipitated with ethanol (50-70%). Following dialysis, a fraction was obtained that represented a 42-fold purification of NR. The determination of kinetic parameters


of NR in this fraction (fraction P1, Table I) revealed that the two kinetic components were still present and in a proportion similar to that in a crude extract of the cyanobacterium. The partially purified NR P1 fraction was further purified on an ion-exchange, DEAE-cellulose column. The elution profile showed a unique peak centered at about 0.15 M NaCl (Fig. 2A). The fraction displaying the highest NR activity, which represented a 185-fold purification of the enzyme, was assayed (fraction P2, Table I) for kinetic parameters of NR, and values of Km(I), Km(II), and Vmax(I)/Vmax(t) similar to those characteristic of strain 29413-P9 were again obtained. Similar values of these three parameters were also obtained for other fractions from the ascending and the descending portions of the NR activity peak (see Fig. 2A), indicating the absence of separation of two possible forms of NR. When the DEAEcellulose pool was further applied to a Sephacryl S-300 gel filtration column, a single activity peak was obtained (Fig. 2B), again revealing a lack of resolution of two possible NR species. From calculations based on the elution volume of marker proteins, the molecular mass of NR from A. variabilis could be inferred to correspond to that of a globular protein of approximately 76 kD (Fig. 3). This value was very close to those previously reported for purified NRs from other cyanobacteria (8, 24, 28). DISCUSSION

The possibility can be raised that the biphasic kinetic behavior obtained for NR from a number of cyanobacterial strains arises from artifacts derived from the assay procedures used. However, the available data do not support that possibility. Under the same assay procedures and conditions, the enzyme from some unicellular cyanobacteria exhibited hyperbolic kinetics (see Table I). Other considerations supporting the lack of artifacts in the assay are as follows. (a) The reaction times and conditions used for both the MV- and the BPB-dependent NR assays have been verified to ensure a linear appearance over time of the product nitrite (data not shown). (b) Nitrite was observed not to be reduced under the conditions of the NR assay, probably as a consequence of the difference between the optimal pH value of cyanobacterial nitrite reductase, about pH 7.5 (12, 23), and the pH of the NR assays (see 'Materials and Methods'). (c) Similar biphasic kinetics are obtained in assays carried out with crude extracts, in whole permeabilized cells, and throughout partial purification of the enzyme from A. variabilis. This rules out possible artifacts derived from methods used for cell disruption or permeabilization. (d) Biphasic kinetics was obtained in A. variabilis when BPB, which has been reported to donate electrons uniquely and directly to the molybdenum cofactor (6), substituted for MV as the artificial electron donor in the assay. (e) Biphasic kinetic behavior is not likely to result from a direct interaction of dithionite with NR, because nitrite formation in the absence of MV or BPB was not observed (results not shown), and the biphasic kinetic behavior remained unchanged after a 5-min incubation of a crude extract of A. variabilis with a dithionite concentration 5-fold higher than that in the assay (data not shown). The experimental kinetic data obtained for NRs exhibiting biphasic patterns were analyzed thlrough the consideration of a wide number of enzyme systems that determine multi-


9 Z

02

0.1 E Ul 1-

20

01 m

1'

z

161

Figure 2. A, Elution profile from DEAE-cellulose of NR from A. variabilis 29413-P9. After solubilization of NR and precipitation with ethanol, the fraction (8 mg of protein) was applied to a DE-52 DEAE-cellulose column (see "Materials and Methods"). The enzyme was eluted with a linear (0.1-0.4 M) NaCI gradient ), and 1.5-mL fractions were collected in which NR (0) and absorbance at 280 nm (-* -) were measured. The arrowhead indicates the maximum activity fraction in which kinetic parameters of NR were assayed. The bottom bar indicates the fractions that were pooled. The protein peak eluting at about 0.4 M NaCI was brown colored and contained Fd. B, Elution profile from Sephacryl S-300 gel filtration of NR from A. variabilis 29413-P9. The DEAE-cellulose pooled fractions were applied to a Sephacryl S-300 column in two successive chromatographic steps carried out with identical results (0.4 mL each time, containing 0.43 mg of protein). Fractions (0.4 mL) were collected and NR (0) and protein (----) were measured.

E

a

z I~-

0 A.

ELUTION VOLUME (ml) phasic kinetics deviating from the hyperbolic, MichaelisMenten behavior (27). For each system, values of the kinetic parameters governing the corresponding equations were sought, by means of computer simulation, that would determine theoretical curves fitting the experimental ones. Two kinetic models were found to predict kinetics that constituted good fits for the experimental data: 1. Two different, independent kinetic components. This system involves the existence of, alternatively, two NR isoenzymes, two different (e.g. modified plus unmodified) forms of NR, or an enzyme having two independent (noncooperative), distinct catalytic sites. Both components would separately obey Michaelis-Menten kinetics, the observed values being the sum of their individual contributions, according to the equation V

=

VI + Vii =

[S] [S+ Vmax(II)* Vmax(I)* Km(I) + [S] Km(II) + [S]J

By the application of the successive-approximation method of Spears et al., as described in ref. 27, individual Km and

Vmrax values for each separate component were calculated and were close to the apparent values obtained from the extrapolation of the straight segments from the Hofstee plot. Those values, once substituted in the equation above, defined a theoretical global velocity curve that represented an excellent fit for the experimental values. This was true for all the biphasic enzyme systems reported in this work. An illustrative example is shown in Figure 4A; both the global velocity curve and individual contributions of the two components are depicted. 2. An enzyme bearing two cooperative, equivalent sites. This kinetic model considers the possibility of interaction between two intrinsically identical catalytic sites when both are occupied by the substrate. As a result of this interaction, both the dissociation constant (Ks) and the catalytic rate constant of each site would become modified by factors a and b, respectively, in the substrate-enzyme-substrate complex. The velocity equation describing this model is vmaxV

[S]/bKs + +[S]2/aK/2 2[S]/K, [S]2/aK,2

1 +

162

MARTIN-NIETO ET AL.


to represent the phylogenetic ancestor of heterocystous cyanobacteria (7). Other unrelated unicellular cyanobacteria studied in this work would bear a monosite NR, thus exhibiting a Michaelis-Menten kinetic behavior. NR from A. variabilis showed an estimated molecular mass of 76 kD, a value very close to those reported for purified

5.3 I a

uu5 3r :3

1.5

4.9

0 g

1.0

4.7

0.5

4.5 1_

0.2

0.3

0.4 0.0

Kav Figure 3. Estimation of mol wt of NR from A. variabilis 29413-P9 by gel filtration chromatography. A fraction of the pooled fraction from DEAE-cellulose chromatography (0.4 mL, containing 0.04 mg of protein) was applied to a Sephacryl S-300 column. Fractions of 0.4 mL were collected, and NR activity was measured. Standards used were: A, aldolase; B, serum albumin; and C, egg albumin. The arrow indicates elution of NR.

vL

-1.5

-0.5

0.5

1.5

2.5

-0.5

0.5

1.5

2.5

0) 0

1.5 1.0

By means of computer simulation, values of the four kinetic parameters (Vma,x, K5, a, and b) could be found for every biphasic enzyme system studied in this work, which determined a global curve representing a good fit for the experimental data. An example is shown in Figure 4B. Values obtained for a were always in the range of 30 to 60 and for b in the range of 2 to 4. The values >1 obtained for both parameters indicate, respectively, negative cooperativity with respect to substrate binding and positive cooperativity with respect to catalysis. The consistency found in values obtained for kinetic parameters Km(I) and Km(II) and for the Vmax(I)/Vmax(t) ratio for NR from a particular strain through different cultures and assay conditions and, for A. variabilis, throughout different states of purification of the enzyme argues against models based on the existence of two NR enzymes. The absence of separation of two enzyme species indicates that, if this were the case, the two forms would bear very similar biochemical properties (i.e. surface charge and mol wt). The models considering a single enzyme species bearing two catalytic sites are more likely. A two-site NR could be a general feature of heterocyst-forming, N2-fixing cyanobacteria, because no exception to the reported biphasic kinetic behavior has been found in this work within this group of microorganisms. This property would be shared by the cyanobacterium Synechocystis sp. PCC 6803, which, interestingly, belongs to a subgroup of unicellular cyanobacteria that has been proposed

0.5

0.0_ -1 .5

log

[N03-1

Figure 4. Adjustment of kinetic experimental data obtained for NR of A. variabilis 29413-P9 to (A) model of multiple independent, distinct kinetic components (model 1, see text); and (B) model of simultaneous cooperative binding and catalysis (model 2, see text). Logarithmic plots of the experiment shown in Figure 1A are depicted. Equations and adjust procedures are indicated in the text. Values of parameters calculated were: Km(l), 0.21 mM; Vmax(l) 8.7 nmol of NO2j/min-mg of protein; Km (II), 26 mM; and V,5(ll), 38 nmol of N02/min * mg of protein, for model 1 (A), and K5, 0.21 mM; Vmax, 47 nmol of N02-/min.mg of protein; a, 45; and b, 3.7 for model 2 (B). Continuous lines are global theoretical curves obtained from the substitution of these values in the corresponding equations. Broken lines in A correspond to the contributions of separate kinetic components. Data points (0) are from experimental data. Nitrate concentration is expressed as mm and velocity as nmol of NO2j/min -mg of protein.


NRs from nonheterocystous cyanobacteria, such as the unicellular Synechococcus sp. (75 kD) (5) and the filamentous P. boryanum (83-85 kD) (24) and Phormidium laminosum (80 kD) (28). Because these NRs are composed of a single polypeptide chain, it seems reasonable to assume that NR from A. variabilis would also have a monomeric structure. Monomeric enzymes bearing multiple sites for the same substrate, although observed infrequently, have been previously reported. A well-known example is ribonucleotide reductase from Lactobacillus leichmannii (15).

13.

14. 15.

16.

ACKNOWLEDGMENTS

17.

We thank A. M&eida and S. Chavez for useful advice, P. Candau and A. Serrano for helpful discussions, and A.M. Muro-Pastor for Gloeocapsa strains. We are also indebted to L.P. Solomonson and I.H. Segel for helpful suggestions and criticism.

18.

19.

LITERATURE CITED

1. Andrews P (1964) Estimation of the molecular weight of proteins by Sephadex gel filtration. Biochem J 91: 222-233 2. Bagchi SN, Sherman L, Funkhouser EA (1987) Biochemical characterization of molybdenum cofactor in a cyanobacterium, Nostoc muscorum. Plant Cell Physiol 28: 1411-1419 3. Bancroft I, Wolk CP (1989) Characterization of an insertion sequence (IS891) of novel structure from the cyanobacterium

20.

4. Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of proteins utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254 5. Candau P (1979) Purificaci6n y propiedades de la ferredoxinanitrato reductasa de la cianobacteria Anacystis nidulans. PhD thesis. Universidad de Sevilla, Sevilla, Spain 6. Cherel I, Gonneau M, Meyer C, Pelsy F, Caboche M (1990) Biochemical and immunological characterization of nitrate reductase deficient nia mutants of Nicotiana plumbaginifolia. Plant Physiol 92: 659-665 7. Doolittle WF (1982) Molecular evolution. In NG Caff, BA Whitton, eds, The Biology of Cyanobacteria. Blackwell Scientific Publications, Oxford, UK, pp 307-331 8. Flores E, Ramos JL, Herrero A, Guerrero MG (1983) Nitrate assimilation by cyanobacteria. In GC Papageorgiou, L Packer, eds, Photosynthetic Prokaryotes: Cell Differentiation and Function. Elsevier, Amsterdam, The Netherlands, pp 363-387 9. Guerrero MG, Vega JM, Losada M (1981) The assimilatory nitrate-reducing system and its regulation. Annu Rev Plant Physiol 32: 169-204 10. Hattori A, Myers J (1967) Reduction of nitrate and nitrite by subcellular preparations of Anabaena cylindrica. II. Reduction of nitrate to nitrite. Plant Cell Physiol 8: 327-337 11. Herrero A, Flores E, Guerrero MG (1981) Regulation of nitrate reductase levels in the cyanobacteria Anacystis nidulans, Anabaena sp. strain 7119, and Nostoc sp. strain 6719. J Bacteriol 145: 175-180 12. Herrero A, Guerrero MG (1986) Regulation of nitrite reductase

22.

Anabaena sp. strain M-131. J Bacteriol 171: 5949-5954

21.

23. 24. 25.

26. 27. 28.

29.

30.

163

in the cyanobacterium Anacystis nidulans. J Gen Microbiol 132: 2463-2468 Herrero A, Wolk CP (1986) Genetic mapping of the chromosome of the cyanobacterium, Anabaena variabilis. Proximity of the structural genes for nitrogenase and ribulose-bisphosphate carboxylase. J Biol Chem 261: 7748-7754 Hoarau J, Hirel B, Nato A (1986) New artificial electron donors for in vitro assay of nitrate reductase from cultured tobacco and other organisms. Plant Physiol 80: 946-949 Holmgren A (1981) Regulation of ribonucleotide reductase. Curr Top Cell Regul 19: 47-76 Ida S, Mikami B (1983) Purification and characterization of assimilatory nitrate reductase from the cyanobacterium Plectonema boryanum. Plant Cell Physiol 24: 649-658 Kallas T, Coursin T, Rippka R (1985) Different organization of nif genes in nonheterocystous and heterocystous cyanobacteria. Plant Mol Biol 5: 321-329 MacKinney G (1941) Absorption of light by chlorophyll solutions. J Biol Chem 140: 315-322 Madueiio F, Borrias WE, Van Arkel GA, Guerrero MG (1988) Isolation and characterization of Anacystis nidulans R2 mutants affected in nitrate assimilation: establishment of two new mutant types. Mol Gen Genet 213: 223-228 Manzano C, Candau P, G6mez-Moreno C, Relimpio AM, Losada M (1976) Ferredoxin-dependent photosynthetic reduction of nitrate and nitrite by particles of Anacystis nidulans. Mol Cell Biochem 10: 161-169 Markwell MAK, Haas SM, Bieber LL, Tolbert NE (1978) A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal Biochem 87: 206-210 Martin-Nieto J, Flores E, Herrero A (1990) Mutants of Anabaena variabilis requiring high levels of molybdate for nitrate reductase and nitrogenase activities. FEMS Microbiol Lett 67: 1-4 Mendez JM, Herrero A, Vega JM (1981) Characterization and catalytic properties of nitrite reductase from Anabaena sp. 7119. Z Pflanzenphysiol 103: 305-315 Mikami M, Ida S (1984) Purification and properties of ferredoxin-nitrate reductase from the cyanobacterium Plectonema boryanum. Biochim Biophys Acta 791: 294-304 Ortega T, Castillo F, Cardenas J (1976) Photolysis of water coupled to nitrate reduction by Nostoc muscorum subcellular particles. Biochem Biophys Res Commun 71: 885-891 Rippka R, Deruelles J, Waterbury JB, Herdman M, Stanier RY (1979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria. J Gen Microbiol 111: 1-61 Segel IH (1975) Enzyme Kinetics. Behavior and Analysis of Rapid Equilibrium and Steady-State Enzyme Systems. John Wiley and Sons, New York Serra JL, Arizmendi JM, Blanco F, Martinez-Bilbao M, Alaiia A, Fresnedo 0, Urkijo I, Llama MJ (1990) Nitrate assimilation in the non-N2-fixing cyanobacterium Phormidium laminosum. In Ullrich WR, Rigano C, Fuggi A, Aparicio PJ, eds, Inorganic Nitrogen in Plants and Microorganisms. Uptake and Metabohism. Springer-Verlag, Berlin, Germany, pp 196-202 Snell FD, Sneli CT (1949) Colorimetric Methods of Analysis, Vol 3. Van Nostrand, New York, pp 804-805 Wolk CP (1982) Heterocysts. In Carr NG, Whitton BA, eds, The Biology of Cyanobacteria. Blackwell Scientific Publications, Oxford, UK, pp 359-386