Received for publication October 23, 1991Accepted March 28, 1992
Identification and Expression of a cDNA Clone EncodingAspartate Aminotransferase in Carrot'
Frank J. Turano2, Jane M. Weisemann, and Benjamin F. Matthews*United States Department of Agriculture, Agricultural Research Service, Plant Molecular Biology Laboratory,
Beltsville, Maryland 20705
ABSTRACT
A full-length cDNA clone encoding aspartate aminotransferase(AAT) has been identified from a carrot root cDNA library. Degen-erate oligo primers were synthesized from the known amino acidsequence of AAT form I from carrot (Daucus carota L. cv Danvers).These primers were utilized in a polymerase chain reaction toamplify a portion of a carrot AAT gene from first strand cDNAsynthesized from poly(A)' RNA isolated from 5-d-old cell suspen-sion cultures. The resulting 750-bp fragment was cloned, mapped,and sequenced. The cloned fragment, mpAAT1, was used as aprobe to identify a full-length cDNA clone in a library constructedfrom poly(A)' RNA isolated from carrot roots. A 1.52-kb full-lengthclone, AAT7, was isolated and sequenced. AAT7 has 54% nucleo-tide identity with both the mouse cytoplasmic and mitochondrialAAT genes. The deduced amino acid sequence has 52 and 53%identity with the deduced amino acid sequences of mouse cyto-plasmic and mitochondrial AAT genes, respectively. Further analy-sis of the sequence data suggests that AAT7 encodes a cytoplasmicform of carrot AAT; the evidence includes the (a) absence of atransit or signal sequence, (b) lack of "m-residues," or invariantmitochondrial residues, in the carrot AAT sequence, and (c) highdegree of sequence similarity with the amino acid sequence pre-viously obtained for form I of carrot, a cytoplasmic isoenzyme.High- and low-stringency hybridizations to Southern blots of carrotnuclear DNA with AAT7 show that AAT is part of a small multigenefamily. Northern blot analysis of AAT7 suggests that AAT is ex-pressed throughout cell culture up to 7 d and is highly expressedin roots but not in leaves.
AAT3 (EC 2.6.1.1) catalyzes the reversible interconversionbetween the amino acids aspartate and glutamate and theircorresponding keto acids, oxaloacetate and a-ketoglutarate.The enzymic transfer of amino groups plays an importantrole in basic metabolism in both plant and animal systems.AAT is important in the transport of reducing equivalentsacross membranes, in transport of fixed carbon between cells
1 This research was supported in part by the U.S. Department ofAgriculture Competitive Grants Office (No. 87-CRCR-1-2284). Thenucleotide sequence(s) reported in this paper has been submitted tothe GenBank/EMBL databases with accession number M92660.
2 Present address: U.S. Department of Agriculture, AgriculturalResearch Service, Climate Stress Laboratory, Beltsville, MD 20705.
3Abbreviations: AAT, aspartate aminotransferase; PLP, pyridoxal5'-phosphate; PCR, polymerase chain reaction; poly(A)+, polyade-nylation; TBE, lx = 89 mm Tris-HCI (pH 8.0), 89 mm boric acid, and2 mm EDTA; SSC, standard sodium citrate.
(in some C4 plants), and in nitrogen distribution (for reviewssee refs. 10, 12, 15). Because of its central role in variousmetabolic processes, AAT has been extensively studied inboth animal and plant systems. The biochemical and physicalcharacteristics of AAT appear to be well conserved in bothkingdoms, i.e. the kinetic properties, cofactor dependency,and mode of action of the enzymes examined to date aresimilar. Additionally, in both systems, the native enzyme hasa relative molecular mass of 90,000 to 120,000 D and isusually composed of two identical subunits of 40,000 to45,000 D.The multifunctionality of the enzyme is correlated with
multiple isoenzymes that have been identified in differentcellular locations in both animals and plants. A review of theanimal literature shows that there are two isoenzymic formsof AAT, each in different cellular locations. One isoenzymehas been localized in the cytoplasm and the other in mito-chondria. The two isoenzymes are encoded by distinct genesthat have 52% nucleotide identity (20). Likewise, in differentplant systems, AAT has been shown to have numerousisoenzymic forms that are located in different cellular loca-tions, namely, the cytoplasm, mitochondria, chloroplasts, andperoxisomes (10, 12). There are genetic and immunologicaldata that strongly suggest that the isoenzymes are encodedby distinct genes. Genetic analyses of varieties of maize (22)and wheat (4) have shown that some of these isoenzymicforms are under independent genetic control. Further evi-dence for the presence of independent AAT genes in plantsinclude the characterization of immunologically distinct an-tibodies to different forms of AAT isolated from alfalfanodules (7). Likewise, in the C4 plant Panicum maximum,immunologically distinct antibodies to different forms ofAAThave been identified (19).A goal of this laboratory is to determine the molecular
mechanism(s) that regulates and orchestrates the expressionof AAT isoenzymes. Our goal is to gain an understanding ofthe regulation of the enzymic reaction at the level of protein,mRNA, and DNA. To that end, we have conducted anextensive study on the predominant form of AAT from carrot(Daucus carota L.) suspension cultures. This study has in-cluded the purification and the physical and biochemicalcharacterization of the isoenzyme (24) and the subcellularlocalization of the isoenzyme in cell suspension cultures (25).In this paper, we report the identification and characterizationof a full-length cDNA clone encoding a cytoplasmic isoen-zyme of AAT from carrot. Additionally, we present data for
expression of the gene in cell suspension cultures and carrotorgans.
MATERIALS AND METHODS
Materials
Guanidinium isothiocyanate, agarose, distilled phenol,oligo(dT)-cellulose, T4 ligase, 1-kb DNA markers, and restric-tion enzymes were purchased from Bethesda ResearchLaboratories4. [a-32P]dCTP (3000 Ci/mmol) was purchasedfrom New England Nuclear. DNA primers were synthesizedcommercially by Synthecell. Vector DNA, pIBI31, and M13mpl8 and mpl9 were purchased from IBI. Modified T7 DNApolymerase (Sequenase 2.0) was purchased from U.S. Bio-chemical. Gigapack II Plus packaging extracts were purchasedfrom Stratagene. Nitrocellulose filters were purchased fromSchleicher and Schuell, and nylon membranes were pur-chased from Gelman. The following kits were purchased:cDNA Synthesis System Plus from Amersham; gene ampli-fication kit from Perkin-Elmer Cetus; Riboclone EcoRI Adap-tor Ligation System and Xgtl 1 arms from Promega, and arandomly primed DNA-labeling kit from Boehringer Mann-heim. All other chemicals were of analytical grade.
Cell Cultures
Cell suspension cultures of carrot (Daucus carota L. cvDanvers) were transferred and maintained as previously de-scribed (24).
RNA Isolation and Poly(A)+ RNA Purification
Total RNA and poly(A)+ RNA were harvested from 5-d-old carrot suspension cultures or from 30-d-old plants usingthe guanidinium isothiocyanate procedure and affinity chro-matography using oligo(dT)-cellulose as described by Man-iatis et al. (16). The purity and integrity of the poly(A)+ RNAwere determined by gel electrophoresis using formaldehyde-formamide gels as described by Maniatis et al. (16). Poly(A)+RNA samples were stored in 70% ethanol at -800C.
Gene Amplification of First-Strand cDNA and Sequencingof the PCR Product
First-strand cDNA was synthesized from 5 ig of poly(A)+RNA using a cDNA synthesis kit. One one-hundredth of thefirst-strand cDNA reaction served as the template in a PCRusing a gene amplification kit. The sequences for the primers5PRFI (5'-CCC GCC AT[CTA] CA[AG] GA[AG] AA-3') and18CATFI (5'-CCA [AG]TC [AT]CC NGG NGT NCC-3') werederived from the amino acid sequences obtained from carrotAAT form I (25). The gene amplification reaction was con-ducted at 940C for 30 s, 420C for 45 s, and 720C for 90 s,for 25 cycles. The resulting PCR fragment was separated ona 1.0% lx TBE agarose gel, and the band was excised from
'Mention of trademark, proprietary product, or vendor does notconstitute a guarantee or warranty of the product by the U.S.Department of Agriculture and does not imply its approval to theexclusion of other products or vendors that may be suitable.
the gel and extracted with phenol. EcoRI adapters were ligatedto the fragment, and the fragment was cloned into M13mpl8. The DNA insert from the clone, mpAAT1, wasmapped and sequenced by the dideoxy chain terminationmethod (21) using modified T7 DNA polymerase (Sequenase2.0). The sequence was analyzed with the University ofWisconsin GCG sequence analysis package running on aVAX 8250 system.
cDNA Library Construction and Isolation of the CarrotAAT cDNA Clone
Total RNA was extracted from 2-month-old (60 d) carrotroots, and poly(A)+ RNA was isolated from total RNA usingoligo(dT)-cellulose as described above. Three micrograms ofpoly(A)+ RNA were utilized to synthesize cDNA. EcoRI adap-tors were added, and the cDNA was ligated with Xgtl 1 armsand packaged into X phage heads.A 750-bp DNA fragment was gene amplified from mp-
AATI and was purified from an agarose gel by phenol ex-traction as described above. The fragment was labeled with[a-32P]dCTP using the random-priming method. Approxi-mately 240,000 plaques were screened using the radiolabeledPCR probe. Lifts of nitrocellulose filters were performed asdescribed by Maniatis et al. (16). The nitrocellulose filterswere prehybidrized in 50% formamide, 5x Denhardt's solu-tion, 5x SSC, 0.1% SDS, 50% dextran sulfate, and 100 ,tg/ml of denatured salmon sperm DNA at 420C for 1 to 4 h.Labeled probe was added to the prehybridization solution(final concentration of 1 X 106 cpm/mL), and the filters werehybridized at 420C for 18 h. The final wash was at 450Cwith 0.1X SSC, 0.1% SDS. From this screen, four stronglypositive plaques were found and analyzed further.The DNA sequence was determined for the largest clone,
AAT7. X phage DNA was digested with restriction endonu-cleases EcoRI, BamHI, and PstI, and fragments were sub-cloned into plasmid vector pIBI31 or M13 vectors mpl8 andmpl9 and sequenced as described above.
Restriction Endonuclease Digestion and Electrophoresis
Restriction endonuclease digestions were performed ac-cording to the manufacturer's recommendations. DNA frag-ments were separated in 1.0% agarose gels in lx TBE.
Carrot DNA Isolation
Carrot nuclear DNA was isolated from 5-d-old carrot cellsuspension cultures. Thirty grams of cells were frozen inliquid nitrogen and ground to a powder with a mortar andpestle. The powder was resuspended in extraction buffer (50mi Tris-HCl [pH 8.0], 1% hexadecylatrimethylammoniumbromide, 50 mm EDTA, 1 mm 1,10 o-phenanthroline, 0.7 MNaCl, and 0.1% f3-mercaptoethanol), and debris was re-moved by centrifugation at 8000g for 5 min. The supematantwas transferred to a sterile tube and extracted twice withbuffered phenol and once with chloroform:isoamyl alcohol(24:1). After the addition of salt and ethanol, the large DNAwas removed by spooling on a glass rod. The spooled DNAwas transferred to another tube and washed several timeswith 70% ethanol.
Carrot nuclear DNA was digested with various restrictionendonucleases, and the fragments were separated by gelelectrophoresis. DNA in the gels was depurinated in 0.25 NHCl for 10 min and transferred to Gelman nylon membranesby capillary transfer overnight with 0.4 N NaOH. Blots wererinsed in 2 X SSC and dried at room temperature. RNA wasseparated by gel electrophoresis in formaldehyde-formamidegels as described by Maniatis et al. (16) and transferred toGelman nylon membranes by capillary transfer in 20X SSC.
High-stringency hybridization of a genomic Southern blotwas performed with the entire 1523-bp fragment from AAT7as a probe. Prehybridization, hybridization, and washes wereperformed at high stringency as described above with thefollowing exceptions. The final washes were in 0.1X SSC and0.1% SDS at 65OC. Low-stringency hybridizations were per-formed as follows: filters were prehybidrized in 15% form-amide, 5X Denhardt's solution, 5X SSC, 0.1% SDS, and 100jig/mL of denatured salmon sperm DNA at 420C for 1 to 4h. Labeled probe was added to the prehybridization solution,and the filters were hybridized at 420C for 18 h. The filterswere washed at 420C with lx SSC, 0.5% SDS for 24 to72 h.Northern blots of poly(A)+ RNA were hybridized with the
full-length AAT cDNA fragment or full-length cDNA encod-ing f3-tubulin from carrot (Daniel J. Prochaska, Earlham Col-lege, Richmond, IN) under stringent conditions as describedabove. The j3-tubulin probe was used as a control to indicateequal loading and show the integrity of the RNA. Results ofthe northern analysis were quantified on a Betagen counter(Waltham, MA). The blots were stripped of probe by immer-sion in 0.1x SSC, 0.1% SDS for 30 min at 900C.
RESULTS AND DISCUSSION
We report the identification of a full-length cDNA clonecoding for the cytoplasmic isoenzymic form ofAAT in carrots.We have used oligonucleotide primers derived from theamino acid sequences of form I of AAT from carrot (25) toamplify a 750-bp fragment of cDNA encoding a portion ofAAT (Fig. 1). The 750-bp fragment was cloned into M13mpl8 (mpAAT1) for restriction endonuclease mapping andsequence analysis. A partial restriction map of mpAAT1revealed one site each for EcoRI, HindIll, PstI, and SstI. Theclone contained a 750-bp insert that had 53 and 55% nucleo-tide identity to the mouse cytoplasmic and mitochondrialAAT genes, respectively (data not shown). The 750-bp frag-ment was used as a probe to identify a full-length clone.A 1523-bp full-length cDNA clone (AAT7) encoding for a
cytoplasmic AAT isoenzyme was identified from a cDNAlibrary prepared from poly(A)+ RNA isolated from carrotroots. The restriction map for AAT7 was identical with themap for mpAAT1 over the 750-bp gene amplified region;outside that region, AAT7 has additional sites for BamHI,PstI, and SstI. However, the nucleotide sequences betweenthe two clones differed slightly (99.6% identity). There arethree nucleotide differences at positions 298, 476, and 726(Fig. 2). Two of the changes are transversions and one is atransition. AAT7 has a 1523-bp insert that contains a 1215-
A.
B.
93 T2 AI I
345
w. . (, ! .:;
,~y . ', s .._
C.
-750
Figure 1. Alignment of oligonucleotide primer sequences (under-lined) derived from amino acid sequences of the cytoplasmic formof carrot AAT (25) and the gene amplification product. The oligo-nucleotide primers 5PRFI (A) and 18CATFI (B) were derived fromamino acids 93 to 98 and 345 to 350 of the carrot cytoplasmicisoenzyme, respectively. Amino acid positions are indicated in boldtype before and after the amino acid sequences. Abbreviations forvariable or wobble positions are as follows: R = A or G; H = A, C,or T; and N = A, C, G, or T. Gene amplification product (C) of first-strand cDNA isolated from 5-d-old carrot suspension cultures using5PRFI and 18CATFI as primers (see "Materials and Methods"). Theresulting 750-bp fragment (mpAAT1) was mapped, subcloned, andsequenced.
bp open reading frame. A putative initiator methionine occursat the 5' end of the cDNA clone. A translation productbeginning at this point would encode a polypeptide of 406residues with a calculated molecular mass of 44.2 kD. This isconsistent with the approximate molecular mass of 43 kDpreviously determined by SDS-PAGE for subunits of form Icarrot AAT (24, 25).The nucleotide and deduced amino acid sequences for the
carrot AAT cDNA clone were compared to the mouse cyto-plasmic and mitochondrial and bacterial AAT sequences toverify the identity of the clone and to show divergence andvariability of these genes. Because identity among homolo-gous animal AAT genes is very high (approximately 85%),only nucleotide and amino acid sequence comparisons withthe mouse cytoplasmic and mitochondrial AAT sequences(20) will be discussed. The use of the mouse AAT sequences,versus other animal AATs, for sequence analysis does notalter the interpretation of the results. The carrot clone has54% nucleotide identity with both the mouse cytoplasmicand mitochondrial AAT cDNA clones (data not shown). The
10 20 30 40 50 60 .70 80 90ATG TCT TCC GTC TTC GCT AAC GTC GTT CGC GCT CCT GAG GAT CCT ATC TTA GGG GTC ACT GTT GCT TAT CAC AAA GAT CAA AGT CCT AATmet ser ser val phe ala asn val val arg ala pro glu asp pro lie leu gly val thr val ala try his lys asp gln ser pro asn
100 110 120 130 140 150 160 170 180AAG TTG AAT TTG GGT GTT GGT GCT TAT CGA ACT GAG GAA GGA AAA CCT CTT GTT CTT AAT GTT GTG AAG AAA GCA GAG CAG ATG CTT GTAlys leu asn leu gly val gly ala tyr arg thr glu glu gly lys pro leu val leu asn val val lys lys ala glu gln met leu val
190 200 210 220 230 240 250 260 270AAT GAT CAG TCT CGA GTA AAA GAG TAT CTT CCT ATT GTT GGA CTG GCA GAC TTT AAC AAA TTG AGT GCC AAG CTG ATT TTC GGC GCT GATasn asp gln ser arg val lys glu tyr leu pro ile val gly leu ala asp phe asn lys leu ser ala lys leu ile phe gly ala asp
280 290 300 310 320 330 340 350 360T
AGT CCT GCC ATT CAA GAG AAT AGA GTC GCA ACT GTC CAG TGC TTA TCT GGT ACT GGC TCA CTG AGG GTT GGG GGT GAA TTC CTG GCC AGAser pro ala ila gin giu amn Av al ala thr xal zin cysile.= gi thr gly mar ia= i xi gln gny giu nDI Ilu Ala ann
Sax
370 380 390 400 410 420 430 440 450CAT TAT CAT GAA CAT ACA GTG TAT ATC CCC CAG CCA ACT TGG GGA AAC CAT CCA AAA ATT TTC ACT TTA GCA GGC TTA TCG GTG AAG ACA
WAgin rxWAMI hisII& g= 2 = = 1 iAn b M iXI ith I& ri=a a gin iesez Ii it h
460 470 480 490 500 510 520 530 540C
TAC CGC TAT TAT AAT CCA GAA ACG CGT GGA TTG GAT TTT GAA GGC ATG CTA GAG GAT CTT GGT TCT GCT CCA TTG GGA GCA ATA GTG CTTinx : ix~ .Ia i m in h a= nin ia ar n -aixi gin=el ian g An ia gin. San Ala proleu gin ala ha xMl ia
550 560 570 580 590 600 610 620 630CTT CAT GCA TGT GCT CAC AAT CCA ACC GGT GTT GAT CCA ACC ATT GAG CAG TGG GAG CAG ATC AGA CAG CTG ATG AGA TCG AAA AGC TTGian hin Ala MCa ala b a nan ibr agi an na r Li g g n g g IIa g i= lit A= iu ilesan ile
vAI ala in Lx xii i ginlan1YSL Ia l inn na= it a ane n nan Ia hin g ala irn hia MIi Ala
910 920 930 940 950 960 970 980 990GCG ATT CTC AAA GAT GGT GAC CTC TAT AAT GAA TGG ACA CTG GAG CTG AAA GCA ATG GCT GAC CGA ATC ATC AGC ATG CGA CAG GAG CTCala hL A le i annA in anniaSIlet nnA inInnthr Ileg1n iAU lxi ala ML Al ian ia mit iia mel inn gin g ia
1000 1010 1020 1030 1040 1050 1060 1070 1080TTT AAT GCT TTA CAA GCA AM GGA ACA CCA GGT GAC TGG AGT CAC ATT GTC AAG CAA ATT GGA ATG TTT ACA TTC ACT GGG TTA AAC TCT
phn An ala ian1 giala hlxi g a gin iA= tn ser his ile val lys gln ile gly met phe thr phe thr gly leu asn ser
1090 1100 1110 1120 1130 1140 1150 1160 1170GAG CAA GTC ACC TTC ATG ACC AAT GAG TAT CAT ATA TAT TTG ACA TCT GAT GGG CGG ATC AGC ATG GCA GGT CTT AGT TCC CGA ACA GTT
glu gln val thr phe met thr asn glu tyr his ile tyr leu thr ser asp gly arg ile ser met ala gly leu ser ser arg thr val
1180 1190 1200 1210 1220 1230 1240 1250 1260CCT CAT CTT GCA GAT GCT ATT CAT GCT GCA GTT ACT GGC AAG GCC TAG AT ACTGTTGTGT TATATATACA TGTTTCATGA GATTTTTGGGpro his leu ala asp ala ile his ala ala val thr gly lys ala ***
Figure 2. Nucleotide sequence of AAT cDNA clones. The nucleotide sequences of AAT7 and mpAAT1 are shown in the mRNA sense strand.Nucleotides are numbered from the putative start codon. The encoded amino acids are denoted below. Regions of the nucleotide sequencefrom the PCR fragment mpAAT1 are underlined. Nucleotides different in mpAAT1 are denoted above, and amino acids are denoted below.The 5' in-frame termination codon is indicated by bold type.
nucleotide identity between the carrot AAT gene and theEscherichia coli aspC gene (9) is 50%, whereas the nucleotideidentity with the gene encoding AAT in the thermoacido-philic archaebacterium Sulfolobus solfataricus (2) is 38% (datanot shown).The deduced amino acid sequence of the carrot gene was
compared to the deduced amino acid sequences of the mousecytoplasmic and mitochondrial isoenzymes (Fig. 3), and theamino acid identities were 52 and 53%, respectively. Analysisof the sequences by the method described by Needleman andWunsch (18), which determines the evolutionary relationshipbetween amino acid sequences, shows 68 and 67% similaritybetween the cytoplasmic and mitochondrial mouse genes,respectively. Amino acid identity and similarity between thededuced carrot AAT amino acid sequence and the deducedamino acid sequences for E. coli are 45 and 65%, and for S.solfataricus they are 17 and 41%, respectively. These datashow that plant AAT is more similar to animal and eubacterialAATs than the archaebacterial AAT. This is not surprisingbecause archaebacterial proteins usually do not have high
Figure 3. Comparison of the deduced aminoacid sequence encoded by carrot AAT7(caAAT) with the deduced amino acid se-quences of the mouse cytoplasmic (mcAAT)and mitochondrial (mmAAT) cDNA clones (4).Dashes represent spaces inserted in the se-quences to maximize homology. Amino acidsare numbered beginning with the N-terminalalanine of mcAAT. Identical amino acids be-tween the plant and animal sequences are in-dicated by an asterisk (*). Residues involved inthe formation of an active site are indicated bybold type. Invariant amino acid residues in AATand tyrosine and histidinol-phosphate amino-transferases (17) are denoted with a small opencircle above. Regions of the deduced carrotamino acid sequence that are identical withamino acid sequence obtained from the cyto-plasmic carrot isoenzyme (25) are underlined.Invariant m-residues from chicken and pig mi-tochondrial AAT sequences (1) are doubleunderlined.
mcAAT
similarity with corresponding eukaryotic or eubacterial pro-teins (3, 5). Furthermore, the plant sequence is more similarto the animal AATs than the eubacterial AAT. However, thesurprising result is the lack of high amino acid identitybetween the carrot AAT and either the cytoplasmic or mito-chondrial isoenzymes. In animal systems, there is a highdegree of amino acid identity among homologous isoen-zymes, i.e. identity among cytoplasmic AATs is approxi-mately 83% and among mitochondrial AATs is approximately86%, whereas identity between different isoenzymes is muchlower, i.e. identity between cytoplasmic and mitochondrialisoenzymes is approximately 45%.
In animals, there are clear distinctions between cytoplasmicand mitochondrial amino acid AAT sequences. Christen etal. (1) analyzed cytoplasmic and mitochondrial amino acidsequences from pig and mouse and identified nine regionsand 36 amino acid residues, 'm-segments' and 'm-residues,characteristic of mitochondrial polypeptides. The designationof the m-segments and m-residues is based on comparisonsof both mitochondrial and cytoplasmic sequences within the
two species, mouse or pig. However, these segments andresidues are present in all animal mitochondrial polypeptidessequenced to date. The identification of m-segments is basedon two criteria: (a) the mitochondrial sequences must have ahigher identity in a designated region than the average iden-tity between the entire mitochondrial sequences and (b) thecytoplasmic sequences in the same designated region musthave a lower identity than the homology between the entirecytoplasmic sequence. The carrot AAT only has one con-served m-segment (120-149). However, much of the similar-ity is in the absence of amino acid residues as opposed tohigh sequence identity among the two sequences. Of the 36m-residues found in mouse mitochondrial AAT, only 12(33%) are identical with the carrot AAT sequence; even withconserved substitutions, the amino acid similarity is only44%. These data suggest that the carrot AAT cDNA does notcode for the mitochondrial isoenzyme.
In eukaryotic systems, polypeptides that are encoded inthe nucleus and transported to organelles contain signal ortransit sequences. Based on the putative initiation site, AAT7does not contain a signal or transit polypeptide. The absenceof a signal or transit polypeptide is supported further by thepresence of an in-frame termination codon before the desig-nated initiation site (Fig. 2). This suggests that AAT7 is a full-length clone that encodes a cytoplasmic isoenzyme.The assignment of AAT7 as the gene encoding a cyto-
plasmic isoenzyme was verified by comparison of the de-duced amino acid sequence to the amino acid sequence datapreviously obtained from the cytoplasmic form of AAT incarrot cell suspension cultures (25). The identity between theamino acid sequences was nearly perfect, 97% (31 of 32residues). The only difference was a substitution of an alanineresidue in the cDNA sequence in place of a seine residue atposition 101. However, the PCR-amplified fragment has thesame seine residue as the amino acid data at this position.The difference in the two amino acid residues is due to atransversion at nucleotide 298 (Fig. 2). Whether the transver-sion occurred during cloning of AAT7 or during the PCRreaction is unclear. However, it is clear from these data thatAAT7 encodes for the cytoplasmic isoenzyme of AAT incarrot.A comparison of known animal and E. coli AAT sequences
shows that 107 (26%) amino acid residues are invariant orconserved (1). Alignment of the polypeptides show thatregions of the AAT polypeptides are highly conserved atamino acids 193 to 200, 220 to 231, and 258 to 268 (1). Thesehighly conserved regions also exist in the carrot AAT poly-peptide. These conserved regions contain several of the 13amino acids proposed to be involved in binding at the activesite (8, 14, 20). In carrot, these amino acid residues areconserved in identity and position at residues Tyr70, Trp 40,His143 fiS189, His193, Asp222, Ala224, Tyr225 5er5, , Lys258,Arg266, Arg292, and Arg386. This includes the motif Ser-X-X-Lys, a sequence common to most PLP-binding sites (23).Other amino acid residues that are conserved in animal AATsand are associated with the active site include Val37, Gly38,Ala39, Gly108, Thr109, Asn142, Asn194, Sere96, and Phe360 (13,14); these residues are also conserved in the carrot polypep-tide. The only amino acid that is associated with the activesite in animal AATs and is not conserved in carrot is Asn297,
A B
B E H
Su -m
B E H
- 1e -Us
- 6.0
-4.0
Figure 4. Southern blot analysis of carrot genomic DNA. Carrotgenomic DNA was digested with BamH1 (B), EcoRI (E), and HindIll(H), separated by gel electrophoresis, transferred to nylon mem-branes, and hybridized under high (A) and low (B) stringency (see"Materials and Methods") with a 1520-bp insert from AAT7. Thesizes of several 1-kb markers are indicated (kb).
which is replaced with serine. Serine is not a conservedsubstitution for asparagine. How this substitution affects theformation of the active site is not known. Most of the invar-iant residues bind to PLP at the active site, specifically Lys258,which is covalently bound to PLP via the E-amino group toform an aldimine double bond (11). The presence of theseresidues in carrot AAT shows that the PLP-binding site isconserved in plants and suggests that carrot AAT requiresPLP for enzymic activity, as do all AAT enzymes studied todate. Previously, we reported the complete inhibition of formI (carrot cytoplasmic isoenzyme) in the presence of 2 mmaminooxyacetate and suggested that the enzyme may requirePLP (24).
Further analysis of the carrot amino acid sequence showsthat carrot AAT contains 12 residues that are conserved intyrosine, histidinol-phosphate, and AATs. Most of these res-idues have a known functional or structural role (17). Ingeneral, these data support the view that the conserved orinvariant residues play significant structural and/or func-tional roles that are common to these enzymes.
Genetic and immunological studies of plant AAT isoen-zymes suggest that they are encoded by distinct genes (4, 7,19, 22), thus requiring the presence of a small multigenefamily. Our Southern analysis of carrot genomic DNA sup-ports that hypothesis (Fig. 4). High- and low-stringencyhybridization experiments were conducted with AAT7 (1523-bp full-length cDNA) with identical Southern blots of ge-nomic carrot DNA digested with three restriction endonucle-ases (BamHI, EcoRI, and HindIII). Under high-stringencyhybridization conditions, simple hybridization patterns wereevident (Fig. 4A): three bands were observed with BamHI(12.0, 4.2 and 3.8 kb), two bands with EcoRI (6.0 and 1.3 kb),and three bands with HindIII (12.2, 9.0 and 6.6 kb). Underless stringent conditions (Fig.4B), an additional band was
observed with BamHI (>20 kb), three additional bands wereobserved with EcoRI (16.5, 12.5, and 4.1 kb), and threeadditional bands were observed with HindIII (18.0, 1.0, and0.8 kb). These data indicate that AAT7 is part of a smallmultigene family that includes several distinct genes.
Hybridization of northern blots of poly(A)+ RNA isolatedfrom carrot cell suspension culture 1, 3, 5, and 7 d afterinoculation shows very little difference in AAT messageexpression throughout a 7-d cycle (Fig. 5A), although thereis a slight increase in transcript levels on day 3 and a slightdecrease on day 7. When the relative amount of mRNA fromdays 3 and 7 are normalized against the tubulin control andcompared to days 1 or 5, the changes are less than 2-fold.The estimated size of the AAT mRNA is approximately 1.55kb, which is in close agreement with the size of the cDNAclone. These data also indicate that the cDNA clone is fulllength. Hybridization of northern blots of poly(A)+ RNAisolated from 30-d-old roots and leaves shows that AAT isexpressed differently in both organs (Fig. 5B); quantificationof the data shows that transcript levels are at least 15- to 20-fold higher in the roots than in leaves. AAT isoenzymes havebeen shown to be differentially expressed in various plantorgans (6, 7). Our northern analysis shows that AAT isdifferentially expressed in roots, where it is highly expressed,and in leaves, where there is very little or no expression.These data correlate well with our isoenzyme data (unpub-lished results) in that in leaves there is little or no detectableAAT activity. If 90 to 95% of the detectable AAT activity isthe cytoplasmic isoenzyme, as is the case in cell culture (24),then repression of the cytoplasmic AAT gene in leaves couldaccount for the low amount of detectable AAT activity wehave observed.
In summary, based on three criteria ([a] the lack of a leadersequence in the cDNA clone, [b] the lack of conservation ofinvariant m-residues in the carrot AAT sequence, and [c] thenear-prefect sequence identity of the deduced amino acidsequence from the cDNA clone with the amino acid sequenceobtained from the cytoplasmic form of carrot AAT), wesuggest that the cDNA AAT7 encodes a cytoplasmic AATisoenzyme. In carrot, AAT is part of a small multigene family,
A
AAT |
tubulin A*-"e W.
BR
1.55
w 1-75
Figure 5. Northern blot analysis of poly(A)4 RNA. RNA isolatedfrom (A) 1-, 3-, 5-, and 7-d-old carrot suspension cultures and (B)30-d-old carrot leaves (L) and roots (R) was separated by gel elec-trophoresis, transferred to nylon membranes, and hybridized underhigh stringency (see "Materials and Methods") with a 1 520-bp insertfrom AAT7 or #l-tubulin. The sizes of hybridizing bands are indicatedin kb.
and the cytoplasmic gene is differentially expressed in rootsand leaves and is not differentially expressed in cell suspen-sion cultures up to 7 d.
ACKNOWLEDGMENTS
The authors would like to thank Mark L. Tucker, Ann C. Smigocki,Eugene Vigil, and Kenneth G. Wilson for critical review of thismanuscript and Karen Lewin and Simos Marmaras for their skillfultechnical assistance. We also thank Dr. Daniel J. Prochaska (EarlhamCollege, Richmond, IN) for the carrot fl-tubulin clone.
LITERATURE CITED
1. Christen P, Graf-Hausner U, Bossa F, Doonan P (1985) Com-parison of covalent structures of the isoenzymes of aspartateaminotransferase. In P Christen, DE Metzler, eds, Transami-nases. John Wiley and Sons, New York, pp 173-184
2. Cubellis MV, Rozzo C, Nitti G, Arnone MI, Marino G, SanniaG (1989) Cloning and sequencing of the gene coding foraspartate aminotransferase from the thermoacidophilic ar-chaebacterium Sulfolobus solfataricus. Eur J Biochem 186:375-381
3. Denda K, Konishi J, Oshima T, Takayasu D, YoshidaM (1988)The membrane-assocciated ATPase from Sulfolobus acidocal-darius is distantly related to FI-ATPase as assessed from theprimary structure of its alpha-subunit. J Biol Chem 263:6012-6015
4. Diaz F, Jouve N (1986) Structure of the isozymes of the AAT-2and AAT-3 systems of aspartate aminotransferase in wheat,rye and triticle. Euphytica 35: 129-135
5. Fabry S, Lang J, Niermann T, Vingron M, Hensel R (1989)Nucleotide sequence of the glyceraldehyde-3-phosphate de-hydrogenase gene from the mesophilic methanogenic archae-bacteria Methanobacterium bryantii and Methanobacterium for-micicum. Eur J Biochem 179: 405-413
6. Farnham MW, Griffith SM, Miller SS, Vance CP (1990) As-partate aminotransferase in alfalfa root nodules. III. Genotypicand tissue expression of aspartate aminotransferase in alfalfaand other species. Plant Physiol 94: 1634-1640
7. Farnham MW, Miller SS, Griffith SM, Vance CP (1990) As-partate aminotransferase in alfalfa root nodules. II. Immuno-logical distinction between two forms of the enzyme. PlantPhysiol 93: 603-610
8. Ford GC, Eichele G, Jansonius JN (1980) Three-dimensionalstructure of a pyridoxal-phosphate-dependent enzyme, mito-chondrial aspartate aminotransferase. Proc Natl Acad Sci USA77: 2559-2563
9. Fotheringham IG, Dacey SA, Taylor PP, Smith TJ, HunterMG, Finlay ME, Primrose SB, Parker DM, Edwards RM(1986) The cloning and sequence analysis of the aspC and tyrBgenes from Escherichia coli K12. Comparison of the primarystructures of the aspartate aminotransferase and aromatic ami-notransferase of E. coli with those of aspartate aminotrans-ferase isoenzymes. Biochem J 234: 593-604
10. Givan CV (1980) Aminotransferases in higher plants. In BJMiflin, ed, The Biochemistry of Plants: A ComprehensiveTreatise. Academic Press, New York, pp 329-357
11. Hughes RC, Jenkins WT, Fischer EH (1962) The site of bindingof pyridoxal-5'-phosphate to heart glutamic-aspartic trans-aminase. Proc Natl Acad Sci USA 48: 1615-1618
12. Ireland RJ, Joy KW (1985) Plant transaminases. In P Christen,DE Metzler, eds, Transaminases. John Wiley and Sons, NewYork, pp 376-383
13. Jansonius, JN, Eichele G, Ford GC, Picot D, Thaller C, VincentMG (1985) Spatial structures of mitochondrial aspartate ami-notransferase. In P Christen, DE Metzler, eds, Transaminases.John Wiley and Sons, New York, pp 109-137
14. Jansonius JN, Vincent MG (1987) Structural basis for catalysisby aspartate aminotransferase. In FA Jurnak, A McPherson,eds, Biological Macromolecules and Assemblies: Active Sites
of Enzymes, Vol 3. John Wiley and Sons, New York, pp
187-28515. Jenkins W, Fonda ML (1985) Kinetics, equilibria, and affinity
for coenzymes and substrates. In P Christen, DE Metzler, eds,Transaminases. John Wiley and Sons, New York, pp 217-234
16. Maniatis T, Fritsch EF, Sambrook J (1982) Molecular Cloning:A Laboratory Manual. Cold Spring Harbor Laboratory, ColdSpring Harbor, New York
17. Mehta PK, Hale TI, Christen P (1989) Evolutionary relation-ships among aminotransferases: tyrosine aminotransferaseand histidinol-phosphate aminotransferase and aspartate ami-notransferase are homologous proteins. Eur J Biochem 186:249-253
18. Needleman SB, Wunsch CD (1970) A general method applica-ble to the search for similarities in the amino acid sequence oftwo proteins. J Mol Biol 48: 443-453
19. Numazawa T, Yamada S, Hase T, Sugiyama T (1989) Aspartateaminotransferase from Panicum maximum Jacq. var. trichog-lume Eyles, a C4 plant: purification, molecular properties, andpreparation of antibody. Arch Biochem Biophys 270: 313-319
20. Obaru K, Nomiyama H, Shimada K, Nagashima F, Morino Y(1985) Cloning and sequence analysis of mRNA for mouse
