Plant Physiol. (1994) 106: 1389-1 394

Genotypes of the Common Bean (Phaseolus vulgaris 1.) Lacking the Nodule-Enhanced lsoform of

GI utami ne Synthetase'

Fei Cao and Peter P. Wong*

Division of Biology, Ackert Hall, Kansas State University, Manhattan, Kansas 66506

Glutamine synthetase (CS) is an octameric enzyme. l h e nodule cytosol of the common bean (Phaseolus vulgaris 1.) has two major types of CS subunit polypeptides (B and y). As a result, nine different isozymes containing varied proportions of B and y can be generated. l h e isozymes are resolvable by native polyacrylamide gel electrophoresis. Staining the gel for GS activity reveals two isoforms, CS.,, which is nodule enhanced and is composed of the eight y polypeptide-containing isozymes, and CS.,, which i s the isozyme &. We screened 104 cultivars and genotypes of common beans for variations in isozyme formation and found two, PI317350 and P1326054, that had no CS.,. l h e PI beans appeared to nodulate normally and had cytosolic protein concentrations and total CS activities similar to those of the cultivar Ul-111, which has CS.,. They accumulated the y polypeptide, which had the same molec- ular weight (46,000) and isoelectric point (6.3) as they polypeptide of UI-111. Experiments with extracts prepared by mixing UI-11 1 and the PI bean nodules suggested that the PI bean nodule extracts did not have an inhibitor or a proteolytic system that specifically inhibited or degraded CS.,. Nodules from UI-111 and the PI beans were dissected into cortex and central infection zone tissue frac- tions. CS.? was found in the cortex and the central infection zone tissue of all beans. Our results suggested that the reason we were unable to detect CS., from the PI beans was not because their CS., and CS.? had an identical electrophoretic mobility, nor was it due to an inhibited or unstable CS.,. Our results suggested that either their polypeptide to assemble or the assembly of CS may require a chaperone. In the two PI beans, the chaperone accumulated to a lower level than it did in Ul-111. This lower amount limited the assembly of the y polypeptide into CS.

gene had mutated in the region that is essential for the

GS (EC 6.3.1.2) is the first enzyme in the ammonia assim- ilation pathway of most higher plants (Miflin and Lea, 1980). In legume root nodule cytosols, ammonia produced by the Nz-fixing bacteroids is assimilated into glutamate by GS and glutamate synthase (Ohyama and Kumazawa, 1980). Culli- more et al. (1983) reported that common bean (Phaseolus vulgaris L.) nodule cytosol has two isoforms of GS (designated GS,l and GSnz), which can be separated from each other by ion-exchange chromatography or native PAGE. The holoen-

Supported by United States Department of Agriculture, National Research Initiative Competitive Grants Program grant No. 92- 37305-7734, and National Aeronautics and Space Administration contract No. NAGW-1197. This is contribution No. 95-13-J from the Kansas Agricultural Experiment Station.

* Corresponding author; fax 1-913-532-6653.

zymes of both isoforms have an identical mol wt (380,000) and comprise eight subunit polypeptides. The subunit poly- peptides are of two major types (designated /3 and y) (Culli- more et al., 1983; Lara et al., 1984). The /3 and 7 polypeptides have a similar mol wt (46,000) but different isoelectric points, 6.0 and 6.3, respectively (Cai and Wong, 1989).

The presence of /3 and y subunit polypeptides in the nodule cytosol results in the generation of nine isozymes, each having a different proportion of /3 and y (i.e. 7 8 , 77/31, 76/32,

%P3, y4P4, Y ~ P s , 72/36, 71/37, and P d (Robert and Wong, 1986; Cai and Wong, 1989). Cai and Wong (1989) also demon- strated that GSnz is the isozyme and that GSn1 is a com- posite of the other eight isozymes. Although the isozymes can be easily detected, the molecular mechanism for the assembly of subunit polypeptides into different isozymes is unknown. In this paper we report two common bean geno- types that could accumulate the y polypeptide, but in which the polypeptide could not assemble into holoenzymes. As a result, the two genotypes showed no detectable GSnI, but they had active GSn2. These genotypes may be useful for elucidating the molecular mechanism of GS assembly and the physiological significance of GSn1.

MATERIALS AND METHODS

Plant Culture and Nodule Harvest

Seeds of common bean (Phaseolus vulgaris L. cv UI-111) were purchased from Idaho Seed Bean Co. (Twin Falls, ID). Seeds of common bean cultivars from other countries and seeds of common bean land race genotypes and wild ances- tors (P. vulgaris L. aborigineus) were obtained from Dr. Richard Hannan of the United States Department of Agri- culture Regional Plant Introduction Station (Pullman, WA). Rhizobium leguminosarum biovar phaseoli strain 127K14 was obtained from Nitragin Co. (Milwaukee, WI). Conditions for plant growth and procedures for inoculating plants with Rhizobium strain 127K14 were described earlier (Manhart and Wong, 1979; Cai and Wong, 1989).

Nodules were detached from roots of bean plants 30 d after sowing. Nodules were frozen in liquid NZ and stored at -7OOC until use.

Abbreviations: GS, Gln synthetase; GSnl, nodule-enhanced Gln synthetase; GSn2, second form of nodule Gln synthetase.

1389 www.plantphysiol.orgon July 22, 2020 - Published by Downloaded from

Copyright © 1994 American Society of Plant Biologists. All rights reserved.

1390 Gao and Wong Plant Physiol. Vol. 106, 1994

2 3

GSm jGSn2

Figure 1. Native PACE (7.5% gel) of common bean nodule extracts.Lane 1, Extract of UI-111; lane 2, extract of PIS 17350; lane 3, extractof PI326054. Each lane received 90 ^g of protein. Electrophoresiswas conducted for 6 h at 4°C and at a constant power of 8 W/plate. After electrophoresis, the gel was stained for CS transferaseactivity. UI-111 showed two CS isoforms, CSn, and GSn2. PI317350and PI326054 showed only one CS isoform, CSn2. The experimentwas repeated four times. Similar results were obtained each time.

Preparation of Nodule Crude Extracts

All procedures were carried out at 0 to 4°C. Root nodulesfrom each cultivar or genotype of common bean were mac-erated in a mortar with a pestle in the presence of anextraction buffer (0.05 M Tris-HCl, pH 7.5, containing 1.0rrtM DTT) at a ratio of 2 mL of buffer to 1 g of nodules aspreviously described (Robert and Wong, 1986). The maceratewas filtered through two layers of Miracloth (Calbiochem),and the filtrate was centrifuged at 27,000g for 30 rnin. Thesupernatant was the nodule crude extract.

In one experiment, nodule extracts were prepared by mix-ing two sources of nodules. Nodules (1 g) from cv UI-111were mixed with 2 g of nodules from either PIS 17350 orPI326054. The mixed nodules were macerated in 6 mL ofextraction buffer as described above. Protein contents of allextracts were measured colorimetrically by the method ofBradford (1976).

Separation of Central Infection Zone andCortex of Nodules

Root nodules (3 g) from UI-111, PI317350, and PI326054were halved with a razor blade. The central infection zonetissue of individual halved nodules was dissected from thecortex with a sharp scalpel. The cortex tissue was washedwith gentle agitation four times with the extraction buffer.Crude extracts of the two tissue fractions were prepared asdescribed above. All procedures were conducted at 0 to 4°C.

PACE

Three PAGE systems were used: (a) a discontinuous nativegel system conducted according to Davis (1964); (b) an SDS-PAGE system carried out according to Laemmli (1970); and(c) an IEF in slab gel system conducted by the method ofKung et al. (1974). The necessary modifications of the exper-imental conditions for each PAGE system were describedpreviously (Robert and Wong, 1986; Cai and Wong, 1989).

The native gel system was used to separate GSnl from GSn2and to reveal their isozyme compositions. The SDS-PAGEwas employed to separate subunit polypeptide ft from yaccording to their mol wts. The mol wts of the polypeptideswere determined by comparing them with those of the stand-ard marker proteins. The IEF system was used to separatethe two polypeptides according to their isoelectric points. Theisoelectric points of the polypeptides were determined bydirectly measuring the pH of the gel after electrophoresis at2-mm intervals with a microelectrode.

After electrophoresis, the gels were either stained for GStransferase activity (Barratt, 1980) or the proteins in the gelswere electrophoretically transferred onto nitrocellulose sheets(Towbin et al., 1979). GS on the sheets was subsequentlyvisualized by ELISA using anti-GS antiserum as the primaryantibody and horseradish peroxidase-conjugated goat anti-rabbit IgG antiserum as the secondary antibody (Cai andWong, 1989).

RESULTS

Robert and Wong (1986) analyzed by native PAGE thenodule extracts of 62 cultivars of common bean that arecommercially grown in the United States and found that allof them have distinct GSnl and GSn2 isoforms similar to thatof cv UI-111 as shown in Figure 1. Because the genetic baseof the American commercial common beans may be small,26 additional commercial cultivars from many different coun-tries were analyzed for this report (Table I). The resultsshowed that they also had two distinct GS isoforms. How-ever, when 13 land race genotypes and three wild ancestorswere studied, one from each group showed no GSn] (Fig. 1).Land race genotype PI317350 came from Mexico. The wildancestor PI326054 was collected in Venezuela.

Table I. Cultivars and genotypes of P. vulgaris that have CSn, andCSn2 isoforms of GS in their nodules

w, Wild ancestor. Ir, Land race genotype. All others are com-mercial cultivars.

Country of Origin PI No.

Argentina 266910(w), 326090(w)Canada 136704Chile 151020Colombia 151385, 207165, 207431(lr)Costa Rica 209483, 209484, 209491El Salvador 201013(lr)Guatemala 163583, 181996, 201010(lr), 201018(lr),

343950(lr)Iran 140305Kenya 209803Lebanon 181785,181786Mexico 150941, 165419, 165442, 203934,

309792, 417662, 417655, 325683(lr),417625(lr), 417626(lr), 417662(lr),430198(lr)

Pakistan 219701Peru 209259,390770(lr)Spain 226910(lr)Turkey 165038, 165078, 169890Venezuela 109860

www.plantphysiol.orgon July 22, 2020 - Published by Downloaded from Copyright © 1994 American Society of Plant Biologists. All rights reserved.

Common Beans That Lack an Isoform of Gin Synthetase 1391

Table II. Protein concentration and CS activity of nodule extracts of UI-117, PI317350, and PI326054beans

Each value is the average ± so of four determinations. CS was assayed by its transferase activity(Farnden and Robertson, 1980). One unit of activity catalyzed the formation of 1 jimol of 7-glutamylhydroxamate/min at 25°C.

Nodule Extract

UI-111PI317350PI326054

Protein Concentrationmg/mL

3.18 ±0.14 (100%)2.69 ±0.1 4 (85%)3. 62 ±0.22 (114%)

Activityunits/mi, extract

13.32 ±0.08(100%)11.89 ±0.10(89%)12.19 ±0.09(92%)

Specific Activityunits/mg protein4.19(100%)4.42 (106%)3.36 (80%)

Both PI beans were nodulated by Rhizobiutn strain 127K14.The nodules were smaller than those of UI-111. Other aspectsof nodule morphology appeared to be normal, althoughdetailed studies had not yet been conducted.

Soluble protein concentration of PIS 17350 and PI326054nodule extract was 85 and 114%, respectively, that of UI-111(Table II). The total GS activity of PI317350 and PI326054was 89 and 92%, respectively, compared to UI-111 (Table II).Because the total GS activity of the two PI beans was con-tributed solely by GSn2, the PI beans seemed to compensa-torily accumulate higher levels of GSn2 than did UI-111 (TableII; Fig. 1).

The possibility that the PI beans may have inactive GSniwas investigated by electrophoretically transferring proteinsin the gel from native PAGE to a nitrocellulose sheet. GSbands on the sheet were visualized by ELISA. The resultsshowed that UI-111 had eight isozyme bands and the two PIbeans had none in the GSni region (Fig. 2). However, a lower

1 2 3 _

Figure 2. Western blot of native PACE of common bean noduleextracts. The conditions for this experiment were identical to thoseof Figure 1 except that each lane received 60 ng of protein and thatafter electrophoresis, the proteins were eletrophoretically trans-ferred to a nitrocellulose sheet at 4°C for 1 h at 500 mA (westernblotting). The sheet was probed with anti-CS antiserum. The CSbands were then visualized by ELISA. In UI-111 (lane 1), the GSn1

isoform showed eight individual CS isozyme bands; its CSn2 hadonly one band. PI317350 (lane 2) and PI326054 (lane 3) showedone band at the CSn2 region and a light lower mol wt band (doublearrowhead) near the bottom of the sheet. The experiment wasrepeated four times. Similar results were obtained each time.

mol wt band was detected in both PI beans (Fig. 2). Thisband had no GS activity, as shown in Figure 1. This bandwas probably not a proteolytic degradation product. In aseparate experiment, the extracts were incubated at roomtemperature for 1 h prior to electrophoresis. This treatmentdid not increase the intensity of the band.

The mol wt of the GS subunit polypeptides from the twoPI beans were examined by SDS-PAGE. As shown in Figure3, both PI beans had the /? and y subunit polypeptides, asdid UI-111. The mol wts of the two polypeptides were similarat about 46,000. The isoelectric points of the subunit poly-peptides were determined by IEF-PAGE. The isoelectricpoints of the /3 and y polypeptides of the PI beans were 6.0and 6.3, respectively (Fig. 4). The isoelectric points of the twopolypeptides of UI-111 were also 6.0 and 6.3, respectively(Fig- 4).

The PI bean nodule extracts may contain a proteolyticsystem that degrades specifically the 7-containing GS iso-zymes. To explore this possibility, extracts were prepared bymacerating one part of UI-111 nodule with two parts ofPI317350 or PI326054 nodules. The extracts were subjectedto native PAGE. When the gel was stained for GS activity,GSni of UI-111 was detected in both mixed nodule extracts(Fig. 5A). When the proteins in the gel were electrophoreti-cally transferred to a nitrocellulose sheet and the GS bandsvisualized by ELISA, the 7-containing GS isozymes of UI-111 were shown to be present in the two mixed noduleextracts (Fig. 5B).

1 2 3

Figure 3. SDS-PAGE of common bean nodule extracts. Lane 1,Extract of UI-111; lane 2, extract of PI317350; lane 3, extract ofPI326054. Each lane received 45 ^g of protein. Electrophoresis wasconducted at 4°C for 5 h at a constant power of 4 W/plate. Afterelectrophoresis, the proteins were blotted onto nitrocellulose, andthe CS subunit polypeptides (|8 and 7) were subsequently visualizedby ELISA as described in Figure 2. The experiment was repeatedthree times. Similar results were obtained each time. www.plantphysiol.orgon July 22, 2020 - Published by Downloaded from

Copyright © 1994 American Society of Plant Biologists. All rights reserved.

1392 Cao and Wong Plant Physiol. Vol. 106, 1994

pp12426779006 20x10.11 31211 2 3 1 2 3 1 2 3 .

Base

AcidFigure 4. IEF-PAGE of common bean nodule extracts. Lane 1,Extract of UI-111; lane 2, extract of PI317350; lane 3, extract ofPI326054. Each lane received 60 Mg of protein. IEF-PAGE wasconducted at room temperature for 19 h at a constant voltage of400 V. After electrophoresis, the proteins were blotted onto nitro-cellulose and the GS subunit polypeptides (0 and y) were subse-quently visualized by ELISA as described in Figure 2. The experi-ment was repeated three times. Similar results were obtained eachtime.

The 7 polypeptide may be located only in the centralinfection zone of the nodule, whereas the /? polypeptide maybe present in the central infection zone as well as the nodulecortex (Cai and Wong, 1989; Forde et al., 1989). Anotherreason for the lack of GSni in the two PI beans could be thatthe |8 and y polypeptides in the central infection zone some-how cannot assemble to form the octameric holoenzymes.To test this possibility, nodules from UI-111 and the two PIbeans were dissected and separated into the cortex and thecentral infection zone. The cortical, central infection zone,and whole nodule extracts were analyzed by native PAGE.The activity stain showed that in UI-111 the cortex had verylittle GSni activity but had substantial GSn2 activity (Fig. 6A).Both isoforms that were detected in the whole nodule extractwere also detected in the central infection zone extract (Fig.6A). Only GSn2 was detected in the cortical, central infectionzone, and whole nodule extracts of the two PI beans (Fig. 6,

Figure 6. Native PAGE of common bean cortical, central infectionzone, and whole nodule extracts. A, Extracts of UI-111; B, extractsof PI317350; C, extracts of PI326054. Lanes 1, Whole noduleextract; lanes 2, cortical extract; lanes 3, central infection zoneextract. Each lane received 90 /ig of protein. After electrophoresis,the gel was stained for GS transferase activity. The conditions forelectrophoresis were the same as described in Figure 1. The exper-iment was repeated twice. Similar results were obtained each time.

B and C). Analysis by ELISA also showed that UI-111 corticalextract had a slight amount of GSn] but an amount of GSn2similar to those of the whole nodule extract and the centralinfection zone extract (Fig. 7A). GSn2, but no GSni, wasdetected in all three extracts of the PI beans (Fig. 7, Band C).

DISCUSSION

The possible mechanisms that can explain why PIS 17350and PI326054 beans had no GSni are as follows:

(a) GSni and GSnz of the two PI beans have an identicalelectrophoretic mobility; as a result, the two GS isoforms

1 2 3

GSniGSn2

2 3 _

aid1 2 3 1 2 3 1 2 3 _

A BFigure 5. Native PAGE of common bean mixed nodule extracts.Conditions for native PAGE were the same as described in Figure1. In A, the gel was stained for GS transferase activity, and eachlane received 90 ^g of protein. In B, each lane received 60 jig ofprotein; after native PAGE, the proteins were blotted onto nitrocel-lulose and the GS isoforms were subsequently visualized by ELISAas described in Figure 2. In both A and B, nodule extract of UI-111(lane 1) showed two GS isoforms, GSni and GSn2. Extract preparedby mixing UI-111 and PI317350 nodules (lane 2) similarly showedtwo GS isoforms. Extract prepared by mixing UI-111 and PI326054nodules (lane 3) also showed two GS isoforms. The experiment wasrepeated twice. Similar results were obtained each time.

GSniGSn2:

B

Figure 7. Western blot of native PAGE of common bean cortical,central infection zone, and whole nodule extracts. A, Extracts ofUI-111; B, extracts of PI317350; C, extracts of PI326054. Lanes 1,Whole nodule extract; lanes 2, cortical extract; lanes 3, infectionzone extract. Each lane received 60 n% of protein. The conditionsof native PAGE were the same as described in Figure 1. After nativePAGE, the proteins were blotted onto nitrocellulose and the GSisoforms were subsequently visualized by ELISA as described inFigure 2. Double arrowhead, Low mol wt band. www.plantphysiol.orgon July 22, 2020 - Published by Downloaded from

Copyright © 1994 American Society of Plant Biologists. All rights reserved.

Common Beans That Lack an lsoform of Gln Synthetase 1393

appeared as a single band as shown in Figures 1 and 2. This mechanism is incorrect because Figure 4 clearly shows that the isoelectric points of the /3 and y polypeptides of the two PI beans were different, as were the isoelectric points of the two polypeptides of cv UI- 11 1. So if the y polypeptide of the two PI beans could assemble to generate GSnl, their GSnI should be electrophoretically separable from their GSn2, as are the two isoforms of UI-111.

(b) GSnI of the two PI beans is catalytically inactive either because of inhibition or because it is unstable. This mecha- nism may also be incorrect because, although the inactive form of GSnI could not be detected by activity stain (Fig. l), it should be detected by ELISA. Figure 2 clearly shows that the two PI beans had no inactive GSnI. The presence of an inhibitor or a proteolytic system in the PI bean nodule extracts that specifically inhibited or degraded GSnI also seems un- likely because using mixed nodule extracts of the two PI beans, GSnI of UI-111 was neither inhibited nor degraded (Fig. 5, A and B). However, Figure 2 does show the presence of a low mol wt inactive GS band. This inactive form may be due to a mutation in the y gene that affected its assembly site or a change in the system that assists the assembly of polypeptide. These two possibilities are discussed below.

GS gene of the two PI beans may have mutations in the region that is essential for assembly. This “mutated assembly site” mechanism may be correct because it is rea- sonable to assume that a certain sequence domain of the y polypeptide is essential for its assembly into octameric holo- GS isozymes. If mutations occurred in the domain, then the y polypeptide’s ability to assemble into an intact enzyme molecule may be affected. To prove this mechanism is correct, the gene of the two PI beans will have to be cloned and sequenced. Comparing the sequences of the y gene from the PI beans with that of cv Bush Blue Lake, which has GSnI (Bennett et al., 1989), may reveal the mutations in the assem- bly site.

(d) The assembly of GS subunit polypeptides into the octameric holoenzymes may require a chaperone. In the two PI beans, the chaperone may accumulate to a lower level than it does in UI-111. This ”chaperone” mechanism seems likely because there are reports in the literature that indicate that GS assembly in higher plants requires a chaperone. Lubben et al. (1989) have shown that GroEL-related chloro- plast molecular chaperone (cpn60) forms stable associations with GS and six other proteins imported into the chloroplasts under in vitro conditions. Their results suggest that cpn60 may be involved in the assembly or folding of GS in pea chloroplasts. Pushkin et al. (1982) reported the structure and some properties of a protein from pea leaves that is similar to GroEL protein of Escherichia coli. They found that this protein can be co-purified with GS. More recently, the same workers (Tsuprun et al., 1992) have used EM and image analysis to demonstrate the formation of a complex between the GroEL-like protein and GS.

For the chaperone mechanism to be correct, an assumption is needed. In the PI beans, either the chaperone level is lower because it is not found in the central infection zone or the y polypeptide requires a higher amount of the chaperone for assembly than does the /3 polypeptide, because the /3 poly- peptide could assemble into GSn2. Arguing against the first

(c) The

part of the assumption are the data shown in Figures 6 and 7, which show that GSn2 was present not only in the cortex but also in the central infection zone of the two PI bean nodules. This indicates that the chaperone accumulated in both tissue fractions, where it could assist in the assembly of /3 polypeptide.

There are reports that lend some support to the second part of this assumption. Temple et al. (1993) transferred an alfalfa GS gene into tobacco and found that a factor was limiting the assembly of the alfalfa GS. Hemon et al. (1990) genetically engineered transgenic tobacco plants to contain chimeric genes encoding the y polypeptide of the common bean. They found in the leaves of mature transgenic plants that some amount of the y polypeptide failed to assemble into active GS, suggesting that a factor may be limiting the assembly of the y polypeptide into GS holoenzyme. More interestingly, Swamp et al. (1990) demonstrated that the a and GS genes of common bean are expressed in its coty- ledons during germination. They found that the y mRNA represented about 20% of the total GS mRNA and that the y polypeptide was about as abundant as the CY polypeptide. They could detect the activity of GS isozymes containing CY

polypeptide, but they found almost negligible activity of GS isozymes containing the y polypeptide. Their results sug- gested that a factor may be specifically limiting the assembly of the y polypeptide in the cotyledons.

Based on these reports, it is reasonable to suggest that the PI beans might accumulate a lower amount of the chaperone than does UI-111. The lower amount limits the assembly of y polypeptide but has no effect on the assembly of the /3 polypeptide.

Received April 22, 1994; accepted August 13, 1994. Copyright Clearance Center: 0032-0889/94/106/1389/06.

LITERATURE CITED

Barratt DHP (1980) Method for the detection of glutamine synthe- tase activity on starch gels. Plant Sci Lett 1 8 249-254

Bennett MJ, Lightfoot DA, Cullimore JV (1989) cDNA sequence and differential expression of the gene encoding the glutamine synthetase y polypeptide of Phaseolus vulgaris L. Plant Mol Biol

Bradford MM (1976) A rapid and sensitive method for the quanti- tation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 7 2 248-254

Cai X, Wong PP (1989) Subunit composition of glutamine synthetase isozymes from root nodules of bean (Phaseolus vulgaris L.). Plant Physiol91: 1056-1062

Cullimore JV, Lara M, Lea, PJ, Miflin BJ (1983) Purification and properties of the two forms of glutamine synthetase from the plant fraction of Phaseolus root nodules. Planta 157: 245-253

Davis BJ (1964) Disc electrophoresis. 11. Method and application to human serum proteins. Ann NY Acad Sci 121: 404-427

Farnden KJF, Robertson JG (1980) Methods for studying enzymes involved in metabolism related to nitrogenase. In FJ Bergersen, ed, Methods for Evaluating Biological Nitrogen Fixation. John Wdey, New York, pp 265-314

Forde BG, Day HM, Turton JF, Shen W-J, Cullimore JV, Oliver JE (1989) Two glutamine synthetase genes from Phaseolus vulgaris L. display contrasting developmental and spatial patterns of expression in transgenic Lotus comiculafus plants. Plant Cell 1:

Hemon P, Robbins MP, Cullimore JV (1990) Targeting of glutamine synthetase to the mitochondria of transgenic tobacco. Plant Mol Bioll5 895-904

12 553-565

391-401

www.plantphysiol.orgon July 22, 2020 - Published by Downloaded from Copyright © 1994 American Society of Plant Biologists. All rights reserved.

1394 Cao and Wong Plant Physiol. Vol. 106, 1994

Kmg SD, Sakano K, Wildman SG (1974) Multiple peptide compo- sition of the large and small subunits of Nicotiana tabacum fraction I protein ascertained by finger-printing and electrofocusing. 13iochem Biophys Acta 365: 138-147

Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature New Biol 227:

Lara M, Porta H, Folch J, Sanchez F (1984) Heterogeneity of glutamine synthetase polypeptides in Phaseolus vulgaris L. Plant I’hysiol76 1019-1023

Lubben TH, Donaldson, GK, Vitanen PV, Gatenby AA (1989) Several proteins imported into chloroplasts form stable complex with the GroEL-related chloroplast molecular chaperone. Plant Cell 1: 1223-1230

Manhart JR, Wong PP (1979) Nitrate reductase activities of rhizobia and the correlation between nitrate reduction and nitrogen fixation.

680-685

- Can J h4icrobiol25 1169-1174

Miflin B1, Lea PI (1980) Ammonia assimilation. In BI h4iflin. ed. The Biochekistry Lf Plants, Vol 5. Academic Press,-New York, pp

Ohyama T, Kumazawa K (1980) Nitrogen assimilation in soybean nodules I. The role of GS/GOGAT system in the assimilation of ammonia produced by nitrogen fixation. Soil Sci Plant Nutr 2 6

169-202

109-1 15

Pushkin AV, Tsuprun VL, Solovjeva NA, Shubin W, Evstigneeva ZG, Kretovich WL (1982) High molecular weight pea kaf protein similar to the GroE protein of Escherichia coli. Biochirn Biophys Acta 704 379-384

Robert F, Wong PP (1986) Isozymes of glutamine synthetase in Phaseolus vulgaris L. and Phaseolus lunatus L. root nodiiles. Plant Physiol81: 142-148

Swarup R, Bennett MJ, Cullimore JV (1990) ExpressiorL of gluta- mine synthetase genes in cotyledons of germinating Phaseolus vulgaris L. Planta 183 51-56

Temple SJ, Knight TJ, Unkefer PJ, Sengupta-Gopalan (1993) Mod- ulation of glutamine synthetase gene expression in tobacco by the introduction of an alfalfa glutamine synthetase gene in sense and antisense orientation: molecular and biochemical analysis. Mol Gen Genet 236 315-325

Towbin H, Staehelin T, Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: proce- dure and some applications. Proc Natl Acad Sci USA 7 6

Tsuprun VL, Boekema EL, Pushkin AV, Tagunova IV (1992) Electron microscopy and image analysis of the GroEL-like protein and its complexes with glutamine synthetase from F’ea leaves. Biochim Biophys Acta 1099 67-73

4350-4354

www.plantphysiol.orgon July 22, 2020 - Published by Downloaded from Copyright © 1994 American Society of Plant Biologists. All rights reserved.