than in uninfected roots (Sangwan and O'Brian, 1992), and hence, each symbiont ... 1; reviewed by Beale, 1990; Beale and Weinstein, 1991;. Jahn et al., 1992), ...
Plant Physiol. (1993) 102: 829-834
Expression of the Soybean (C/ycine max) Glutamate 1-Semialdehyde Aminotransferase Gene in symbiotic Root Nodules’ lndu Sangwan and Mark
R. O’Brian*
Department of Biochemistry and Center for Advanced Molecular Biology and Immunology, State University of New York at Buffalo, Buffalo, New York 14214
processes essential for heme formation in nodules can be inferred by observations that the Bradyrhizobium japonicum hemA gene is not essential for nodule development and symbiosis (Guerinot and Chelm, 1986), whereas expression of the hemH gene is required (Frustaci and OBrian, 1992). Soybean (Glycine max) synthesizes the tetrapyrrole precursor ALA in nodules (Sangwan and OBrian, 1991), and evidence supports the hypothesis that B. japonicum heme can be formed from that ALA (Sangwan and O’Brian, 1991), rendering the bacterial hemA gene nonessential. In tum, the soybean ALA formation activity is much greater in nodules than in uninfected roots (Sangwan and O’Brian, 1992), and hence, each symbiont affects the metabolism of the other with respect to heme formation. Some evidence supports the hypothesis that leghemoglobin heme is a bacterial product, but that possibility remains an open question. Plant involvement cannot be ruled out because soybean synthesizes ALA in nodules (Sangwan and O’Brian, 1991). Nodule heme formation is not only interesting with respect to symbiosis, it also allows the study of tetrapyrrole synthesis in a nonphotosynthetic plant organ that requires a high leve1 of expression. Tetrapyrrole synthesis in plants has been studied primarily in photosynthetic tissue because of the large quantity of Chl expressed there, and much less is known about its formation in other plant tissues. ALA is synthesized in chloroplasts from glutamate by the C5 pathway in leaves (Fig. 1; reviewed by Beale, 1990; Beale and Weinstein, 1991; Jahn et al., 1992), and ALA made in this way is incorporated into hemes as well as Chl (Schneegurt and Beale, 1986). The Cs pathway enzyme GSA aminotransferase has been purified from leaves of barley and pea (Grimm et al., 1989; Pugh et al., 1992), and a cDNA encoding the barley enzyme has been cloned (Grimm, 1990). The C5 pathway has also been described biochemically and genetically in bacteria and algae that express those enzymes (reviewed by Beale, 1990; Beale and Weinstein, 1991; Jahn et al., 1992). It is possible that ALA formation in roots and nodules is catalyzed by the Cs pathway enzymes as well, but this possibility has not been established and is addressed in the present work. Glutamate-dependent ALA formation activity was discemed in plant fractions of soybean root nodules (Sangwan and O’Brian, 1991), but expression of the C5 pathway could
Extracts of soybean (Glycine max) root nodules and greening etiolated leaves catalyzed radiolabeled 6-aminolevulinic acid (ALA) formation from 3,4-[3H]glutamatebut not from 1-[‘4C]glutamate. Nevertheless, those tissue extracts expressed the activity of glutamate 1-semialdehyde (CSA) aminotransferase, the C5 pathway enzyme that catalyzes ALA synthesis from CSA for tetrapyrrole formation. A soybean nodule cDNA clone that conferred ALA prototrophy, CSA aminotransferase activity, and glutamatedependent ALA formation activity on an Escherichia coli CSA aminotransferasemutant was isolated. The deduced product of the nodule cDNA shared 79% identity with the CSA aminotransferase expressed in barley leaves, providing, along with the complementation data, strong evidence that the cDNA encodes CSA aminotransferase. CSA aminotransferase mRNA and enzyme activity were expressed in nodules but not in uninfected roots, indicating that the Csa gene is induced in the symbiotic tissue. The Csa gene was strongly expressed in leaves of etiolated plantlets independently of light treatment and, to a much lesser extent, in leaves of mature plants. We conclude that CSA aminotransferase,and possibly the C5 pathway, is expressed in a nonphotosynthetic plant organ for nodule heme synthesis and that Csa is a regulated gene in soybean.
The nitrogen-fixing root nodule is a specialized plant organ elicited by rhizobia on certain legumes, and the relationship between the two organisms is symbiotic (reviewed by Long, 1989; Caetano-Annoles and Gresshoff, 1991; Verma, 1992). An increase in cellular hemes in nodules is one of numerous processes that occur as plant and bacterial cells differentiate from the asymbiotic to the symbiotic state (reviewed by Appleby, 1984; O’Brian and Maier, 1989). Rhizobial Cyt’s change both qualitatively and quantitatively in nodules to allow oxidative phosphorylation within the low oxygen milieu of the nodule. The legume host expresses a nodulespecific hemoglobin (leghemoglobin)to facilitate oxygen diffusion to the respiring bacteria. The copious amount of leghemoglobin found in nodules allows the heme protein to have a large oxygen-buffering capacity, and its high binding affinity results in a low free oxygen tension, thereby protecting the oxygen-labile nitrogenase. Evidence for interactive ‘ R i s work was supported by the Cooperative State Research Service, U.S.Department of Agriculture, under agreement No. 9137305-6750. * Corresponding author; fax 1-716-829-2725.
Abbreviations: ALA, 8-aminolevuliic acid; GSA, glutamate 1semialdehyde. 829
830
COOH
Sangwan and O'Brian
FH2
CHNH2
COOH Glutamate
COOH
FOOH
COOH
I
Glutamvl-IRNA
FH2 Redudase CHNH2 CO-tRNAG'" Glutamyl-tRNA
yH2
GSAAmim c H 2 translerase
CH2 CHNH2 CHO
Glutamate-7 Semialdehyde
CH2 +O CH2NH2 AIA
Figure 1. The Cs pathway for ALA synthesis from glutamate. Clutamyl-tRNA reductase and GSA aminotransferase are committed to ALA synthesis. Cofactors used in the reactions are not shown.
not be demonstrated (Sangwan and O'Brian, 1992). The ALA formation activity is significantly greater in nodules than in uninfected roots (Sangwan and O'Brian, 1992), and this increase is consistent with elevated levels of nodule hemes relative to asymbiotic cells. Thus, whereas ALA formation is reported to be regulated by light in leaves (Beale and Castelfranco, 1974a; Fluhr et al., 1975; Masoner and Kasemir, 1975; Klein et al., 1977), synthesis in soybean roots is affected by its symbiotic interaction with a bacterium. ALA formation by soybean in nodules is characterized biochemically and genetically herein toward the understanding of legume-rhizobia interactions essential for heme synthesis and of tetrapyrrole formation in nonphotosynthetic plant tissue. MATERIALS AND METHODS Bacteria and Plants
Escherichia coli strain HBlOl was used as an ALA prototroph in the present work and was grown in Luria broth (Ausubel et al., 1987) for enzyme assays. E. coli strain GE1377 (provided by Dr. D. 5611) is a mutant derivative of strain AB354 defective in hemL, the gene encoding GSA aminotransferase (Ilag et al., 1991); it is a leaky ALA auxotroph, and thus it was grown in Luria broth supplemented with ALA. Ampicillin (50 p g mL-I) was added to media for growth of strains harboring pUC18 or pBluescript derivatives. Bradyrhizobium japonicum strain I110 was the soybean endosymbiont used in the present work and was grown in glycerolsalts-yeast extract medium (Frustaci et al., 1991). Soybeans (Glycine max cv Essex), either inoculated with 8. japonicum or uninoculated, were grown in a growth chamber under a 16-h light/B-h dark regimen at 25OC. Nodules, leaves, and roots were harvested from 23- or 24-d-old plants for enzyme assays or RNA extraction. Etiolated soybean plants were grown in total darkness for 9 d and either left in the dark or exposed to direct light for the final 24 h before the leaves and roots were harvested for enzyme assays or RNA extraction. The light-exposed plants were green in color and had larger leaves than those that were grown in total darkness. ALA Formation from Labeled Glutamate
Enzymic incorporation of radioactivity into ALA from 10 pCi of 3,4-[3H]glutamateor 1 pCi of l-['4C]glutamate was measured in extracts of soybean leaves, nodule cytosol, and E. coli as previously described (Sangwan and O'Brian, 1992). The pH optima were previously determined to be 7.0 for nodule cytosol and 7.9 for the other samples (Sangwan and
Plant Physiol. Vol. 102, 1993
O'Brian, 1992). Reactions were camed out at their optimal pH values in the present work. ALA was isolated from reaction samples by ion-exchange chromatography antl solvent extraction as previously described (Sangwan and O'Brian, 1992), and incorporated radioactivity was measured with a scintillation spectrometer. GSA Aminotransferase Activity
Enzymic forrnation of ALA from GSA (provided by Dr. C. G. Kannangarai) by soybean nodule, root, and leaf extracts and by E. coli extracts was camed out in Mops buffer at pH 6.8 and 3OoC for 20 min as previously described (Hoober et al., 1988). Controls contained heat-inactivated protein samples. Reactions were terminated, and ALA was isolated as described for glutamate-dependent ALA formation (Sangwan and O'Brian, 1992). ALA was quantified spectrophotonietrically as the ALA-pyrrole as previously described (Frustaci et al., 1991). Activities in crude extracts of etiolated leaves were approximately 3% of that found in partially purified GSA aminotransferase from barley (Hoober et al., 1988). lsolation and Analysis of Soybean Nodule cDNA Encoiding GSA Aminotransferase
A soybean nodule cDNA expression vector library in pUC18 was a gift from Dr. M.L. Kahn and was constnicted as described by Udvardi et al. (1991).Library DNA was used to transform E. coli mutant strain GE1377 en masscr by electroporation, and cells were subsequently plated onto M9 minimal media (Ausubel et al., 1987) in the absence of ALA. The media contained the nonfermentable substrate succinate as the carbon source and 0.2 mg mL-' ampicillin, 40 fig mL-' Thr, Leu, and thiamine, and 1 m~ isopropyl &D-thiogalactopyranoside. Eight ampicillin-resistant colonies arose af ter 4 d, and only one of them contained an insert-containing plasmid. The other seven transformants were presumably ALA prototrophs because of spontaneous reversion of the hemL mutation, and they were not studied further. The cDNA insert of pKN4 was cloned into pBluescript IISK+ to construct pSKN4. Deletions were made in pSKN4 using an Exo-Mung deletion kit (Stratagene) according to the manufacturer's instructions, and the nucleotide sequence of both strands of the insert DNA was determined using a Sequenase kit (United States Biochemical).Sequence analysis was camed out using software of the Genetics Computer Group (Devereaux et al., 1984). RNA lsolation and Analysis
I
Nodules, leaves, and roots were excised, frozen in licpid nitrogen, and homogenized in a blender with homogenization buffer and phenol ( 2 2 3 , w/v/v). The homogenizaiion buffer contained 500 mM Tris (pH 8), 10 m~ MgC12, 1 mM EDTA, 100 m~ NaCI, 0.5% (w/v) deoxycholate, and 1 mM P-mercaptoethanol. Total RNA was isolated from the homogenate as described previously (Ausubel et al., 1987), m d poly(A)+ RNA was isolated using oligo(dT)-cellulose columns. Northem blot analysis of poly(A)+ RNA was canied out as previously described under high-stringency conditions (Ausubel et al., 1987). Ubiquitin cDNA was provided by Dr. D.P.S. Verma (Fortin et al., 1988).
83 1
Glutamate 1-Semialdehyde Aminotransferase in Soybean Root Nodules
Table 1. Clutamate-dependent ALA formation and CSA aminotransferase activities in extracts of
soybean nodules and greening etiolated leaves and of E. coli Activities are the average of three trials. ALA Formed from:
Source of Extract
10 pCi of
3,4-['H]Glutamate
1 pCi of 1-['4C]Clutamate
GSA
cpm h-' mg-' of protein cpm h-' mg-' of protein
Leaves Nodu les E. coli strain H B l O l
79,700 19,658 14,360
RESULTS ALA Formation from Clutamate and GSA
The carbon skeleton of glutamate is incorporated, intact, into ALA by the C5 pathway enzymes (Fig. l), and thus the ability of soybean nodule extracts to synthesize radiolabeled ALA from U-['4C]glutamate and 3,4-[3H]glutamate, but not from l-['4C]glutamate (Sangwan and O'Brian, 1992), is not predicted by that mechanism. In the present work, neither soybean leaf extract nor nodule extract was able to catalyze the incorporation of radiolabel from l-['4C]glutamate into ALA, although labeled ALA was synthesized from 3,4-t3H]glutamate (Table I). Extracts of E. coli, which contain the C5 pathway (Avissar and Beale, 1989; Grimm et al., 1989; Li et al., 1989; Ilag et al., 1991), catalyzed the formation of radiolabeled ALA from either glutamate isotope (Table I). The results show that the inability of the I-carbon of glutamate to be incorporated into ALA in soybean is not specific to root nodules because it was also observed in extracts of greening etiolated leaves, where the C5 pathway is presumed to operate and have maximal activity. Despite the anomalous observations for glutamate-dependent ALA formation in soybean tissue, both leaf and nodule extracts expressed a GSA aminotransferase activity (Table I), the final step of the C5 pathway (Fig. 1).The activity in nodules was approximately one-half of that found in greening etiolated leaves, and thus, GSA aminotransferase was expressed in nonphotosynthetic plant tissue. Glutamyl-tRNA reductase activity, which is the first committed enzymic step of the C5 pathway, was not measured because of the inability to purify or purchase soybean glutamyl-tRNA synthetase needed to synthesize the substrate. Complementation of a E. coli CSA Aminotransferase Mutant with a Soybean Nodule cDNA Clone
The E. coli hemL mutant GE1377 is defective in the gene encoding GSA aminotransferase and behaves as a leaky ALA auxotroph (Ilag et al., 1991). Strain GE1377 was transformed with a soybean nodule cDNA expression library, and of 5 X 106transformants that were screened, eight colonies grew on agar media in the absence of exogenous ALA. Seven of the transformants harbored plasmids with no insert DNA and presumably arose from spontaneous reversion of the ALA auxotroph phenotype; those transformants were not studied further. One transformant harbored a plasmid containing a 1.8-kb DNA insert, and this plasmid, designated pKN4, con-
nmol20 min-' mg-' of protein
2.5 1.2 5.4
O O 3,812
ferred ALA prototrophy on mutant strain GE1377 (Fig. 2). The ALA-independent growth of strain GE1377[pKN4] was due to the plasmid, rather than a chromosomal reversion, because pKN4 was able to restore ALA prototrophy of the mutant with 100% frequency. Strain GE1377[pKN4] expressed 10-fold more GSA aminotransferase activity than did GE1377[pUC18] and that activity was more than one-half of that found in the hemL+ strain HBlOl (Table 11). In addition, extracts of GE1377[pKN4] catalyzed glutamate-dependent ALA synthesis, and thus, C5 pathway activity was restored. The data show that the isolated soybean nodule cDNA conferred on strain GE1377 ALA prototrophy, GSA aminotransferase activity, and C5 pathway activity. Nucleotide Sequence and Putative Produd of the Cloned Soybean Nodule cDNA
The nucleotide sequence of the soybean cDNA insert of pKN4 was determined and found to be 1863 bp in length, excluding the linkers added during library construction (Fig. 3). An open reading frame encoding a 467-amino acid protein was identified beginning with Met and terminating with a TGA codon (Fig. 3). In addition, a termination codon was identified upstream of the open reading frame and in the same reading frame, showing that the entire coding region was present in the cloned cDNA. The putative protein shared 79% identity with the GSA aminotransferase expressed in barley leaves (Grimm, 1990; Fig. 4), and 55 to 71% identity
1.o
--C-
0.8
V
5:
-8 v)
C
0.6
ea
0.4
n
0.2
s
4
pCIC18,noAIA pUC18. with A I A pKN4,noAlA pKN4, with A I A
0-0
O
L
1
2
3
4
5
6
7
Time (hours) Figure 2. Crowth curve of E. coli CE1377 transformants bearing
pUC18 (squares) or pKN4 (circles).Cells were grown in the pres(filled symbols) or absence (open symbols) of ALA.
ente
Sangwan and O'Brian
832
Plant Physiol. Vol. 102, 1993 ~~
Table II. Complementation of the E. coli CSA aminotransferase mutant GEJ377 soybean nodule cDNA borne on pKN4 Activities are the average of three trials. ALA Formed from:
Strain CSA
1O pCi of 3,4-[3H]Clutamate
nmol20 min-' mg-' of protein
cpm h-' mg-' ofprotein
cpm h-' mg-' ofprotein
6.5 0.4 3.8
13,457 512 4,772
1672 41 557
HB101 GE1377 (pUC18) GE1377 (pKN4)
with bacterial aminotransferases (Elliott et al., 1990; Grimm et al., 1991). This identity, along with the observed complementation of the GSA aminotransferase structural gene mutant of E. coli (Table 11), strongly suggests that the cloned cDNA encodes a soybean GSA aminotransferase. We designate the gene Gsa as recommended for the barley gene (Grimm, 1990). The RNA from which the cloned cDNA was synthesized was polyadenylated, indicating that the Gsa message was processed in the nucleus. The N-terminal sequence of the GSA aminotransferase is characteristic of a peptide targeted to plastids with respect to the three structural domains and positions of severa1amino acid residues that typify such transit peptides (Gavel and von Heijne, 1990). Barley GSA aminotransferase contains a transit peptide (Grimm, 1990), and the high homology between the barley and soybean enzymes is delimited at one end by the known or predicted N termini of the mature proteins (Grimm, 1990; Figs. 3 and 4). Expression of the Gsa Gene in Soybean
The GSA aminotransferase cDNA was used to detect Gsa mRNA in various soybean tissues by northem blot analysis (Fig. 5 ) . Gsa mRNA was expressed in nodules from 23-d-old plants but was almost completely absent in uninfected roots of the same age or from those of etiolated plants (Fig. 5 ) . Likewise, enzyme activity of GSA aminotransferase was detected in nodules but not in uninfected roots (Fig. 5 ) . It was shown previously (Sangwan and O'Brian, 1992) that glutamate-dependent ALA formation activity was induced in nodules, and the present work indicates that the induction is due, at least in part, to activation of the Gsa gene. Because soybean nodules are subterranean organs, and ALA formation is reported to be induced by light in leaves and cotyledons (Beale and Castelfranco, 1974a; Fluhr et al., 1975; Masoner and Kasemir, 1975; Klein et al., 1977), it was germane to ask whether expression of the Gsa gene in leaves requires light. Gsa mRNA was strongly expressed in leaves from dark-grown etiolated plants, and this level was somewhat higher in lightexposed, greening etiolated plants (Fig. 5 ) . In addition, GSA aminotransferase activity in leaves from etiolated plants was independent of light exposure (Fig. 5). These data show that neither Gsa message levels nor posttranscriptional processes required light in those tissues. DISCUSSION
We have isolated a soybean nodule cDNA encoding the C5 pathway enzyme GSA aminotransferase and have shown
1 pC1 of 1-['4C]Clutamate
that the Gsa gene is expressed in nodules. Therefore, soybean GSA aminotransferase and possibly the C5 pathway are not confined to pho'tosynthetic tissue. The GSA aminotransfiirase mRNA and enzyme activity present in nodules was almost completely absent in uninfected roots (Fig. 5 ) , showing that expression of the Gsa gene was induced in the symbiotic tissue. Activation of the Gsa gene is likely to be at least partially resporrsible for the inducible glutamate-depenident ALA formation activity observed in nodules (Sangwan and O'Brian, 1992), although the other enzymes involved in ALA formation may be regulated as well. Gsa mRNA was strcingly expressed in leaves of etiolated plantlets and to a lesser e:utent in those of older plants (Fig. 5 ) . Hence, Gsa is not a constitutively expressed gene in leaf or root tissues. The inability of the I-carbon to become incorporated into ALA remains an unexplained phenomenon, but the present work shows that it is not a nodule-specific phenomenon (Table I). Beale and Castelfranco (1974b) observed that kidney bean ( P h a s d u s vulgaris) cotyledons incorporated the 1-carbon of glutainate into ALA less efficiently than did cucumber cotyledons. It is likely that the difference between the legumes under di.;ccussion and other plant systems is not in the mechanism of the C5 pathway enzymes but, rather, in the immediate fate of radiolabeled glutamate. In soybean and kidney bean, some or a11 of the radiolabeled glutamate may be first converted to other compounds, of which some of the label is retumed to the glutamate pool. The proposed role of light for induction of ALA formation activity in greening plant tissues (Beale and Castelfranco, 1974a; Fluhr et al., 1975; Masoner and Kasemir, 1975; Klein et al., 1977) and its unlikely role as a positive regulator in root nodules prompted us to address whether light was necessary for activation of the Gsa gene in leaves. Light was not essential for full expression of the Gsa gene in leaves of etiolated plantlets as seen by the high transcript level and enzyme activity in the dark-grown soybeans (Fig. 5 ) . Gsa mRNA is also expressed in dark-grown barley shoots (Griinm, 1990), but posttransciptional processes were not reported. It is possible that light plays a regulatory role in leaves of plants grown under a diurna1 light-dark cycle, and it may regulate other steps of the Cs pathway, but its nonessentiality in Gsa expression remains valid. Therefore, expression of GSA aminotransferase in root nodules probably does not require a regulatory factor that compensates for darkness, because Gsa is not, strictly speaking, a light-induced gene in soybean.. The source of leghemoglobin heme in nodules remainij an open question. It is possible that leghemoglobin heme is a
833
Glutamate 1-Semialdehyde Aminotransferase in Soybean Root Nodules 1 TTTGACGAAGAGTGAGAGAGTCTTATCTGTCGTCTCTGATCTCTGATCGCATCTTCATTC 60
* r r v r e s y l s s l i s d r i f i p
1 .....MAVSAITGARLTLGMSLSSSTRSRTVAM.AVSIDPKTDNKLTLTK 44 II : I : =1 IIII Mill I hi 1 MAGAAAAVASGISIRPVAAPKISRAPRSRSWRAAVSIDEKA. . .YTVQK 47
61 CGAAAATGGCTGTTTCGGCTATCACTGGAGCGAGGCTAACTCTAGGGATGTCTCTTTCCT 120 k M A V S A I T G A R L T L G M S I S S 121 CTTCCACGCGATCACGAACCGTCGCAATGGCCGTATCTATCGACCCCAAGACCGATAACA 180 S T R S R T V A M A V S I D P K T D N K
45 SEEAFAAAKELMPGGVNSPVRAFKSVGGQPIVIDSVKGSRMWDIDGNEYI 94
III I
1 1 1 1 I I M M 1 1 1 1 1 1 1 M M 111 h i 111 I h l I h i II11
48 SEEIFNAAKELMPGGVNSPVRAFKSVGGQPIVFDSVKGSHMWDVDGNEYI 97 95 DYVGSWGPAIIGHADDQVLAALGETMKKGTSFGAPCLLENTLAELVIDAV
144
98 DYVGSWGPAIIGHADDKVNAALIETLKKGTSFGAPCALENVLAQMVISAV
147
181 AACTCACTCTTACCAAGTCCGAGGAAGCTTTCGCTGCGGCCAAGGAGCTGATGCCTGGAG 240 L T L T K S E E A F A A A K E L H P G G 241 GCGTGAACTCCCCAGTTCGTGCCTTCAAATCCGTGGGTGGTCAACCAATTGTGATTGATT V
N
S
P
V
R
A
F
K
S
V
G
G
O
P
I
V
I
D
300
S
K
G
S
R
N
U
D
I
D
G
N
E
Y
I
D
Y
V
G
G
P
A
I
I
G
H
A
D
D
O
V
L
A
A
L
195 GSGVATLGLPDSPGVPKAATFETLTAPYNDTEAIEKLFEANKGEIAAVFL
S
G
M M 1 1 M i l l I ! 111 h i I M I M I M I : h
E
T
421 CCATGAAGAAAGGAACCAGCTTTGGTGCACCCTGTCTGCTGGAAAACACTTTGGCAGAGC M
K
K
G
T
S
F
G
A
P
C
L
L
E
N
T
L
A
E
480
541 CTTGCATGGGTGCGCTCCGTCTGGCCCGTGCTTATACCGGAAGAGAGAAGATCATCAAGT M
G
A
I
R
L
A
R
A
Y
T
G
R
E
K
1
I
K
600
G
C
Y
H
G
H
A
D
P
F
L
V
K
A
G
S
G
V
L
G
L
P
O
S
P
G
V
P
K
A
A
T
F
E
T
720
L
P
Y
N
D
T
E
A
I
E
K
L
F
E
A
N
K
G
E
780
A
V
F
L
E
P
V
V
G
N
A
G
F
I
V
P
K
P
840
D
841 ATTTTCATAGTTTCTTGCGCAAGATCACCAAGGAGAACAATACCCTTCTTGTGTTTGATG 900 F
H
S
F
L
R
K
I
T
K
E
N
N
T
L
L
V
F
D
E
901 AAGTCATGACTGGATTTCGTTTGTCATATGGAGGTGCTCAAGAGTATTTTGGCATAACTC 960 V
M
T
G
F
R
L
S
Y
G
G
A
Q
E
Y
F
G
I
T
345 AMTAGIETLQRIKEPGTYEYLDKITGELVEGIIEAGKRAGHAICGGHIRG
II h I I 1 1 1 1 1 Mhl 11II
Ih:
394
I : II Mlllll
348 AMTAGIHTLKRLMEPGTYEYLDKVTGELVRGILDVGAKTGHEMCGGHIRG 397
MUM MM II I l l l l l l l l h l
MlhlllMIIMIIIIIM
398 MFGFFFAGGPVHNFDDAKKSDTAKFGRFHRGMLGEGVYLAPSQFEAGFTS 447 445 LAHTSDDIKKTIAAAEKVFREI 467
MM Ml Ih I l l l h l I
448 LAHTTQDIEKTVEAAEKVLRWI 470
1
781 TTGCCGCAGTTTTCCTCGAACCTGTTGTTGGAAACGCTGGTTTCATTGTTCCTAAGCCTG A
I II II hll II II II I Mil II II I hll II I MM II III I II MM I
298 FGITPDVTTLGKIIGGGLPVGAYGGRKDIMEMVAPAGPMYQAGTLSGNPL 347
T
721 CAGCCCCCTACAATGACACCGAGGCCATTGAGAAACTCTTCGAGGCCAACAAAGGAGAAA A
295 FGITPDITTLGKI1GGGLPVGAYGGRRDIMEKVAPAGPMYQAGTLSGNPL 344
395 MFGFFFTEGPVYNFADAKKSDTAKFARFFWGMLAEGVYLAPSQFEAGFTS 444
A
661 CCACCTTAGGACTTCCTGATTCTCCCGGTGTCCCCAAAGCTGCCACTTTTGAAACCCTTA T
245 EPWGNAGFIVPKPDFHSFLRKITKENNTLLVFDEVMTGFRLSYGGAQEY 294 I l l l l l l l l l III II H I I I I II I I I I I I I I • I I I I I I I 248 EPWGNAGFIPPQPAFLNALREVTKQDGALLVFDEVMTGFRLAYGGAQEY 297
Mill
F
601 TTGAGGGCTGTTACCATGGCCATGCTGATCCTTTTCTTGTTAAGGCAGGTAGTGGAGTTG 660 E
244
MM IMIIMIM
198 GSGVATLGLPDSPGVPKGATVGTLTAPYNDADAVKKLFEDNKGEIAAVFL 247
L
481 TGGTTATCGATGCCGTCCCCAGCATTGAAATGGTTCGGTTTGTCAATTCAGGCACTGAAG 540 V I D A V P S I E H V R F V N S G T E A C
hi 1111 hll 11 MM 11 HIM
360
361 CCTGGGGTCCTGCAATCATTGGTCACGCTGATGATCAGGTGCTTGCAGCTCTGGGTGAAA 420 U
11 M i l 1 1 1 1 1 1 1 1 1 M l ! I
148 PSIEHVRFVNSGTEACMGALRLVRAFTGREKILKFEGCYHGHADSFLVKA 197
301 CAGTCAAAGGGTCTCGTATGTGGGATATCGATGGCAATGAGTACATTGACTACGTTGGTT V
145 PSIEMVRFVNSGTEACMGALRLARAYTGREKIIKFEGCYHGHADPFLVKA 194
P
Figure 4. Comparison of amino acid sequences of the GSA aminotransferases from soybean (top) and barley (bottom). Solid lines represent identity and dotted lines represent conservative amino acid substitutions. The soybean protein shares 79% identity with that of barley.
961 CAGATATAACAACTCTAGGAAAGATCATTGGTGGAGGTCTGCCGGTAGGCGCTTATGGAG 1020 D I T T L G K I 1 G G G L P V G A Y G G 1021 GGAGGAGGGATATTATGGAGAAGGTGGCACCAGCTGGCCCAATGTATCAGGCTGGGACCT 1080 R
R
D
I
M
E
K
V
A
P
A
G
P
M
Y
O
A
G
T
L
1081 TGAGTGGGAACCCTTTGGCCATGACTGCAGGCATAGAGACCCTGCAGCGTATTAAGGAGC 1140 S
G
N
P
L
A
M
T
A
G
I
E
T
L
O
R
I
K
E
P
1141 CAGGAACTTACGAGTACTTGGACAAAATCACTGGTGAGCTTGTTGAGGGCATCATCGAAG G
T
Y
E
Y
L
D
K
I
T
G
E
L
V
E
G
I
I
E
1200
A
that soybean has the synthetic capacity to make ALA in that organ. In addition, soybean expresses the activity of ALA dehydratase (Nadler and Avissar, 1977), the enzyme that directly utilizes ALA for tetrapyrrole synthesis. Therefore, it is plausible that the plant host can synthesize hemoglobin heme in nodules. Furthermore, B. japonicum may be able to
1201 CTGGGAAGCGGGCAGGCCATGCAATATGTGGTGGGCATATAAGGGGGATGTTTGGGTTTT 1260 G
K
R
A
G
H
A
I
C
G
G
H
I
R
G
M
F
G
F
F
1261 TCTTCACAGAAGGACCAGTGTATAATTTTGCAGATGCCAAAAAGAGTGATACGGCCAAGT F
T
E
G
P
V
Y
N
F
A
D
A
K
K
S
D
T
A
I
C
Etiolated
23 days
1320
F
1321 TTGCTAGGTTCTTTTGGGGAATGCTGGCGGAAGGTGTCTATTTGGCACCTTCCCAGTTTG 1380 A
R
F
F
U
G
M
L
A
E
G
V
Y
L
A
P
S
Q
F
E
1381 AGGCTGGCTTCACCAGCTTGGCACATACTTCTGATGACATAAAAAAGACGATAGCCGCTG A
G
F
T
S
L
A
H
T
S
D
D
I
K
K
T
I
A
A
K
V
F R
E
mRNA
A
1441 CTGAGAAGGTTTTCAGGGAGATCTGATGGTTAAATTTTGTTTTGTTGCAAATTTAATTTT
E
Gsa
1440 1500
1 *
1501 CGGAGGGTGAATTTTTAGGTCAATTTAGATTATTGTTATGGCAGTTGCTTTCGTTATGAT
1560
c»
I7I/7A/4
•»'«• «P
1561 CTGTATCATTTTCGCATCCTGTATCTATCCAGTGTATTATGTTGAGCTGTGAGTTACTTG 1620 1621 AATTTGAAGCTTGTAAGCATTCGAATTCATTGTTTAACTCCTCATTCTAGTTTCACATGT 1680 1681 TATGTTTTTAATGTAGGCATGCATTTTGCTAATGCATTACATATCAGTATACATCTTTGT 1740 1741 GCACTACACACTTTCGGSGTGGGAATAGATAGAGAATTGTCTTTGTGCTAGATAAATTGA 1800
GSA-AT Activity
0.2
1801 ACTAATTTTTGAGCTGTGATCCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
1860
1861 AAA
1863
Figure 3. Nucleotide sequence of soybean nodule cDNA insert of pKN4 and the putative translation product (in uppercase letters). Underlined amino acids represent the putative leader sequence. Underlined nucleotides are possible polyadenylation signals.
bacterial product, but a Rhizobium meliloti hemA mutant elicits alfalfa nodules that are arrested in development (Dickstein et al., 1991). Hence, the hemoglobin-defective phenotype is expected regardless of the source of the heme moiety. The expression of GSA aminotransferase (Table I and Fig. 5) and glutamate-dependent ALA formation activity in soybean nodules (Sangwan and O'Brian, 1991, 1992) demonstrates
0
1.3
2.5
2.6
nd
nd
Figure 5. GSA aminotransferase mRNA and enzyme activities in nodules (N), leaves (L), or uninfected roots (R) of 23-d-old, lightgrown plants or from etiolated plants. The asterisk (*) indicates RNA from tissues of the dark-grown etiolated plants that were exposed to light for the final 24 h before harvesting. The same harvest of plants was used for the enzyme activities and northern blots. Csa and ubiquitin (Ub) mRNA were analyzed by northern blots using the respective cDNAs as radiolabeled probes. Approximately 5 Mg of poly(A)+ was loaded onto each lane, and the filter was probed with Csa cDNA, stripped, and then probed with Ufa cDNA. Ub is a control for a constitutively expressed gene. The figure is a composite of the two autoradiograms, and bands in the same row can be directly compared with each other. GSA aminotransferase activity is expressed as nmol of ALA formed mg~' protein in 20 min and is the average of three trials.
834
Sangwan and O’Brian
take u p a n d metabolize soybean nodule ALA (Sangwan and O’Brian, 1991). Thus plant involvement is likely even if the heme moiety is provided directly by t h e bacterial symbiont. ACKNOWLEDCMENTS We thank Dr. C.G. Kannangara for GSA, Dr. D. Sol1 for E. coli strain GE1377,Dr. M.L. Kahn for the soybean nodule cDNA library, and Dr. D.P.S. Verma for soybean ubiquitin cDNA. Received January 8, 1993;accepted March 31, 1993. Copyright Clearance Center: 0032-0889/93/102/0829/06. The GenBank accession number for the sequence described in this paper is L12453. LITERATURE CITED Appleby CA (1984) Leghemoglobin and Rhizobium respiration. Annu Rev Plant Physiol35 443-478 Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman JG, Smith JA, Struhl K (1987)Current Protocols in Molecular Biology. Wiley Interscience, New York Avissar Y, Beale SI (1989)Identification of the enzymatic basis for 6-aminolevulinicacid auxotrophy in a hemA mutant of Escherichia coli. J Bacterioll71: 2919-2924 Beale SI (1990)Biosynthesis of the tetrapyrrole pigment precursor, 6-aminolevulinic acid, from glutamate. Plant Physiol 9 3 1273-1279 Beale SI, Castelfranco PA (1974a)The biosynthesis of 6-aminolevulinic acid in higher plants. I. Accumulation of 6-aminolevulinic acid in greening plant tissues. Plant Physiol53 291-296 Beale SI, Castelfranco PA (1974b)The biosynthesis of d-aminolevulinic acid in in higher plants. 11. Formation of “C-6-aminolevulinic acid from labeled precursors in greening plant tissues. Plant Physiol53 297-303 Beale SI, Weinstein JD (1991)Biochemistry and regulation of photosynthetic pignent formation in plants and algae. In PM Jordan, ed, Biosynthesis of Tetrapyrroles. Elsevier Science Publishers, New York, pp 155-235 Caetano-Anolles G, Gresshoff PM (1991)Plant genetic control of nodulation. Annu Rev Microbiol45 345~382 Devereaux J, Haeberli P, Smithies O (1984)A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 12: 387-395 Dickstein R, Scheirer DC, Fowle WH, Ausubel FM (1991)Nodules elicited by Rhizobium meliloti heme mutants are arrested at an early stage in development. Mo1 Gen Genet 230 423-432 Elliott T, Avissar YJ,Rhie G-E, Beale SI (1990) Cloning and sequence of the Salmonella typhimurium hemL gene and identification of the missing enzyme in hemL mutants as glutamate-lsemialdehyde aminotransferase. J Bacterioll72 7071-7084 Fluhr R, Harel E, Klein S, Meller E (1975)Control of b-aminolevulinic acid and chlorophyll accumulation in greening maize leaves upon light-dark transitions. Plant Physiol56 497-501 Fortin MG, Purohit SK, Verma DPS (1988)The primary structure of soybean (Glycine max) is identical to other plant ubiquitins. Nucleic Acids Res 1 6 11377 Frustaci Jh4, OBrian MR (1992)Characterization of a Bradyrhizobium japonicum ferrochelatase mutant and isolation of the hemH gene. J Bacteriol174 4223-4229
Plant Physiol. Vol. 102, 1993
Frustaci JM, Sangwan I, OBrian MR (1991)Aerobic growth and respiration of a 6-aminolevulinicacid synthase (hemA) mutant of Bradyrhizobium japonicum. J Bacterioll73: 1145-1 150 Gavel Y, von Heijne G (1990)A conserved cleavage-site motif in chloroplast transit peptides. FEBS Lett 261: 455-458 Grimm B (1990)Primary structure of a key enzyme in plant tetrapyrrole synthesis: glutamate 1-semialdehyde aminotransferase. Proc Natl Acad Sci USA 87: 4169-4173 Grimm B, Bull A,, Breu V (1991)Structural genes of glutamate 1semialdehyde aminotransferase for porphyrin synthesis in i i cyanobacterium and Escherichia coli. Mo1 Gen Genet 225: 1-10 Grimm 8, Bull A, Welinder KG, Gough SP, Kannangara CG (I 989) Purification and partia1 amino acid sequence of the glutamate 1semialdehyde aminotransferase of barley and Synechococcus.Carlsberg Res Commun 5 4 67-79 Guerinot ML, Chelm BK (1986) Bacterial 6-aminolevulinic acid synthase activitV is not essential for leahemoalobin formation in the soybean/BrÚdyrhizobium japonicum &mbi&is. Proc Natl Acad Sci USA 83: 1837-1841 Hoober TK, Kahn A, Ash DE, Goueh S. Kannaneara CG (1988) Biosyrhhesis of A-aminolevkinateyn beening bzrley leaves. IX. Structure of tht: substrate, mode of gabaculine inhibition, and the catalvtic mechanism of elutamate 1-semialdehvde aminoíransferaie. Carlsberg Res CoGmun 53: 11-25 Ilag; LL, Tahn D, Eggertsson G, Soll D 11991) The Escherichia coli &mL gene encodggglutamate 1-semialdehyde aminotransferase. J Bacterioll73 3408-3413 Jahn D, Verkamp E, Soll D (1992) Glutamyl-transfer RNA a precursor of heme and chlorophyll biosynthesis. Trends Biochem Sci 17: 215-218 Klein S, Katz E, Neeman E (1977)Induction of 6-aminolevulinic acid formation in etiolated maize leaves controlled by two light systems. Plant Physiol60 335-338 Li JM, Brathwaite O, Cosloy SD, Russell CS (1989)5-Aminolevulinic acid synthesis in Escherichia coli. J Bacterioll71: 2547-21552 Long SR (1989)Rhizobium-legume nodulation: life together in the underground. Cell56: 203-214 Masoner M, Kasemir H (1975)Control of chlorophyll synthesis by phytochrome. I. The effect of phytochrome on the formatiorl of 5aminolevulinate in mustard seedlings. Planta 126 11 1-1 17 Nadler KD, Avissar YJ (1977) Heme synthesis in soybean root nodules. I. On the role of bacteroid 6-aminolevulinicacid syrlthase and 6-aminolevulinicacid dehydratase in the synthesis of the heme of leghemoglobin. Plant Physiol60 433-436 OBrian MR, Maier RJ (1989)Molecular aspects of the energetics of nitrogen fixation in Rhizobium-legumesymbioses. Biochim Biophys Acta 974: 229-246 Pugh CE, Harwood JL, John RA (1992)Mechanism of glutamate semialdehyde aminotransferase. Roles of diamino- and dioxointermediates in the synthesis of aminolevulinate. J Biol Chein 267: 1584-1588 Sangwan I, O’Brian MR (1991)Evidence for an inter-organismic heme biosynthetic pathway in symbiotic soybean root nodules. Science 251: 1220-1222 Sangwan I, OBrian MR (1992)Characterization of 6-aminolevulinic acid formation in soybean root nodules. Plant Physiol 9 8 1074-1079 Schneegurt MA, Beale SI (1986)Biosynthesis of protoheme and heme a from glutamate in maize. Plant Physiol 81: 965-971 Udvardi MK, Kahn ML (1991)Isolation and analysis of a cDNA clone that encodes an alfalfa (Medicago sativa) aspartate aminotransferase. Mo1 Gen Genet 231: 97-105 Verma DPS (1992)Signals in root nodule organogenesis and endocytosis of Rhizobium. Plant Cell4 373-382 \
I
