Metabolic compartmentation of vertebrate glutamine synthetase: putative mitochondrial targeting signal in avian liver glutamine synthetase.

Metabolic Compartmentation of Vertebrate Glutamine Synthetase: Putative Mitochondrial Targeting Signal in Avian Liver Glutamine Synthetase’ James W. Campbell and Darwin D. Smith, Jr-.= Department of Biochemistry and Cell Biology and Laboratory of Biochemical and Genetic Engineering, Rice University

Introduction

Downloaded from https://academic.oup.com/mbe/article-abstract/9/5/787/1009228 by guest on 15 January 2019

The evolution of uricoteley as a mechanism for hepatic ammonia detoxication in vertebrates required targeting of glutamine synthetase (GS) to liver mitochondria in the sauropsid line of descent leading to the squamate reptiles and archosaurs. Previous studies have shown that in birds and crocodilians, sole survivors of the archosaurian line, hepatic GS is translated without a transient, N-terminal targeting signal common to other mitochondrial matrix proteins. To identify a putative internal targeting sequence in the avian enzyme, the amino acid sequence of chicken liver GS was derived by a combination of sequencing of cloned cDNA, direct sequencing of mRNA, and sequencing of polymerase chain reaction (PCR) products amplified from reverse-transcribed mRNA. Analysis of the first 20 or so N-terminal amino acids of the derived sequence for the chicken enzyme shows that they are devoid of acidic amino acids, contain several hydroxy amino acids, and can be predicted to form a positively charged, amphipathic helix, all of which are characteristic properties of mitochondrial targeting signals. A comparison of the Nterminus of chicken GS with the N-termini of cytosolic mammalian GSs indicates that at least three amino acid replacements may have been responsible for converting the N-terminus of the cytosolic mammalian enzyme into a mitochondrial targeting signal. Two of these, HisI and Lyslg, result in additional positive charges, as well as in changes in hydrophilicity. Both could have resulted from third-base-codon substitutions. A third replacement, Ala ,*, may contribute to the helicity of the Nterminus of the chicken enzyme. The N-terminus of the cytosolic chicken brain GS (positions l-36) was found to be identical to that of the liver enzyme. The complete sequence of chicken retinal GS is also identical to that of the liver enzyme. GS is coded by a single gene in birds, so these sequence data suggest that, unlike the situation in other tissue-specific compartmental isozymes, differential targeting of avian GS to the mitochondrial or cytosolic compartments is not dependent on the sequence of the primary translation product of its mRNA but may involve some other tissue-specific factor(s) .

Glutamine synthetase [ GS; L-glutamate ammonia ligase ( ADP-forming ); E.C. 6.3.1.21 is the primary ammonia-detoxifying system in liver of uricotelic vertebrates 1. Key words: glutamine synthetase, evolution of uricoteley, signal sequence evolution, species-specific compartmentation, tissue-specific compartmentation. 2. Present address: Abbott Laboratories, Abbott Park, Illinois 60064.

Address for correspondence and reprints: James W. Campbell, Department of Biochemistry and Cell Biology, Rice University, P.O. Box 1892, Houston, Texas 7725 1. Mol. Biol. Evol. 9(5):787-805. 1992. 0 1992by The U&ersity of Chicago. All rights resend 0737-4038/92/0905-0003$02.00

787

788 Campbell and Smith

Material and Methods Material


and is therefore analogous in function to the combined actions of carbamyl phosphate synthetase-I (CPS-I) and ornithine transcarbamylase (OTC) in liver of ureotelic vertebrates (Campbell 199 1). GS may also function as a means of ammonia detoxication in mammalian liver, but this is more as a “fail-safe” mechanism (Htiussinger and Gerok 1986), because its expression in mammalian liver is restricted to only a few cells surrounding the terminal venules (Gebhardt and Mecke 1983; Gebhardt et al. 1988), which do not express CPS-I (de Groot et al. 1987; Moorman et al. 1988). In keeping with its primary physiological function in ammonia detoxication, GS is expressed in all avian hepatocytes. CPS-I has been “silenced” in the avian line, so it is also not expressed in these cells (Smith and Campbell 1988 ). At the subcellular level, GS is localized within the cytosolic compartment in mammalian liver (Wu 1963) whereas it is localized in the mitochondrial matrix in avian liver (Vorhaben and Campbell 1972, 1977). Since the latter is the main site of ammonia release during amino acid gluconeogenesis (Campbell 199 1)) translocation of GS into the mitochondrial matrix during tetrapod evolution allowed it to function efficiently as the primary hepatic ammonia-detoxifying system (Campbell et al. 1987). Most mitochondiral matrix proteins are encoded in the nucleus and are translated in the cytosol as larger precursor peptides containing an N-terminal targeting signal that is cleaved during import (Attardi and Schatz 1988; Grivell 1988; Pfanner and Neupert 1990). However, this is not true for either chicken (Smith and Campbell 1983) or alligator (Smith and Campbell 1987) liver GS, suggesting that these enzymes may contain internal targeting signals. While internal targeting signals are common for inner-membrane proteins (Grivell 1988; Zara et al. 199 1)) they are rare for matrix proteins. Internal signals nevertheless appear to be utilized by at least three other matrix enzymes; these are adenylate kinase (Watanabe and Kubo 1982), mediumchain acylCoA dehydrogenase (Kelly et al. 1987 ), and 3-ketoacyl-CoA thiolase (Mori et al. 1985 ) . The targeting information for the latter has been shown to reside at the N-terminus of the mature protein (Amaya et al. 1988; Arakawa et al. 1990). The present study was therefore directed toward identifying the putative mitochondrial targeting signal in avian liver GS, in order to understand what evolutionary events may have led to the differential compartment of the enzyme in mammalian and avian liver.

Restriction enzymes, avian myeloblastosis virus ( AMV) reverse-transcriptas and sequencing reagents, the transcription system, and other reagents were obtained from Bethesda Research Laboratories, Pharmacia, United State Biochemical, Stratagene, or Promega. The translation system was as described elsewhere (Smith and Campbell 1983 ) . [32P] -labeled nucleotides and [ 35S]-labeled L-methionine were from ICN Radiochemicals. Bluescribe vector was from Stratagene, and M13mp9 was from Pharmacia. Chickens (white leghorn) were from Rich Glo, Stafford, Tex. Oligonucleotides were synthesized with a Biosearch 8600 DNA synthesizer in the Rice University Laboratory of Biochemical and Genetic Engineering. 7’uq polymerase

cDNA Cloning Total RNA was prepared from chicken brain and liver tissue that had been snapfrozen in liquid nitrogen by using guanidine thiocyanate extraction and cesium chloride gradient centrifugation (Chirgwin et al. 1979). This was followed by oligo-dT chro-

Avian GS Mitochondrial

Targeting Signal

789

Nucleic Acid Sequencing


matography for poly (A)+ mRNA isolation (Aviv and Leder 1972). The isolated poly (A)+ mRNA was further fractionated on sucrose gradients (Williamson et al. 197 1). Fractions enriched in GS mRNA were identified by dot-blotting (Thomas 1980) using the bovine retinal GS clone pGS1 (Smith and Campbell 1988). The mRNA in these fractions was precipitated with ethanol. A cDNA library was synthesized from 2 ug of the enriched mRNA by using AMV reverse transcriptase and RNAase H (Gubler and Hoffman 1983; Haymerle et al. 1986). BamHI blunt-end adaptors were ligated to the resulting cDNA. This was then inserted into the BumHI site of Bluescribe, which contains T7 and T3 RNA promoter sites. Escherichia coli JM109 was transformed and plated (Hanahan 1983). Transformed colonies were screened using the pGS 1 probe [ 32P]-labeled by random priming (Feinberg and Vogelstein 1983 ).

Double-stranded sequencing (Chen and Seeburg 1985 ) of isolated clones utilized primers to the T7 and T3 regions of Bluescribe and either a Sequenase or a Klenow fragment ( DNA polymerase I). Restriction-endonuclease fragments were subcloned into Bluescribe, for double-stranded sequencing, or into M 13-mp9, for single-stranded dideoxy sequencing. Wedge gradient gels were used for electrophoresis. Primers derived from sequence data were used to extend sequence information for some clones or to confirm the sequence by using the complementary strand. Direct mRNA sequencing was with primers based on the main cDNA clone (pCGS11; see below), and this sequence information was used to walk the sequence toward the 5’end (Geliebter et al. 1986). Twenty micrograms of poly (A)+ mRNA from brain or liver was annealed with 5 ng 32P 5’end-labeled 20-mer primers at 42°C for 2 h in 10 mM Tris chloride pH 8.3,225 mM potassium chloride. This was followed by sequencing using AMV reverse transcriptase for primer extension analysis (Williams and Mason 1985 ) . Sequence data from direct mRNA sequencing were used to generate 20-mer primers to permit polymerase chain reaction (PCR) amplification of the 5’ end of GS mRNA by using Taq DNA polymerase (Saiki et al. 1988 ). Portions of this reaction mixture were used for asymmetric single-stranded DNA production by the PCR for sequencing (Gibbs et al. 1989). Northern and Southern Blotting

Glyoxylated mRNA was separated on 1% agarose gels and then was transferred, either by capillarity to nitrocellulose (Thomas 1980) or by electroblotting to Genescreen (Stellway and Dahlberg 1980). Hybridization was with either random-primed intact pCGS 11 plasmid or the purified insert. Genomic DNA ( lo-20 pg) (Blin and Stafford 1976; Bowtell 1987) was digested with various restriction enzymes, was electrophoresed in 1% agarose gels, and was transferred to nitrocellulose (Southern 1975). pCGSl1 was randomly primed with [ 32P]dATP ( l-5 X lo8 cpm/pg) and used as the probe. Autoradiography was with an intensifying screen. Construction

of Full-Length GS Coding Plasmid

pCGS 11 was cut with BumHI, and this was followed by purification of the largest fragment, which contained most of the coding region and a portion of the 3’noncoding region (fig. 1) . cDNA to the N-terminal GS mRNA coding region was synthesized as described above, by using both a primer to the 5 ’noncoding region and one overlapping the 5’-most BumHI site (fig. 1), and was amplified by the PCR. The PCR product resulted from 30 cycles of the following sequence: 2 min at 42°C (annealing), 2 min

790

Campbell and Smith

CODING

:i?

START

3:

$ SE ma,

3:

L

z

E

-0

X

c--KK

X

.

*

X -0

.

0 *

-a0

m

*

v

co m

c

0

0 *


.

f-r--

z

c

0

u)o (uu)

iii

=

0

FIG. 1.-Sequencing strategy and partial restriction map of full-length chicken liver GS construct (i.e., pCGSl1N). Sequencing was with T7, T3, or M 13 primers (0) or with synthetic oligonucleotide primers determined from prior sequencing (X) as described in the text.

polymerization using Tuq polymerase at 72°C and 1 min denaturation at 94°C. A final polymerization at 72°C was for 15 min. The PCR product was cut with BumHI and was purified by agarose gel electrophoresis. The BamHI blunt-ended PCR product was ligated onto the large BumHI fragment of pCGS 11. The construct (designated “pCGS 11N”) was purified by phenol extraction and ethanol precipitation. Unligated ends were polished with T4 DNA polymerase prior to ligation into BumHI-cut Bluescribe. The resulting plasmid was inserted into E. coli JM 109. Clones were characterized by restriction digestion and sequencing, to verify the construct. Transcription

of pCGS 11N Construct and Translation of mRNA

The plasmid containing the full-length construct was linearized with EcoRI, which cuts 3’of the insert in the Bluescribe polylinker region and 5’in the noncoding region. The linearized plasmid was then transcribed with T3 RNA polymerase with or without the capping analogue m7GpppG (Contreras et al. 1982 ). The reaction conditions were 40 mM Tris hydrochloride pH 7.9; 6 mM magnesium chloride; 2 mM each of ATP, GTP, CTP, and UTP; 40 units of T3 polymerase, and various amounts of plasmid. When added, m7GpppG was 5 mM. Incubation was at 37°C for 1 h. The resulting mRNA was purified using phenol-chloroform extraction and ethanol precipitation. pCGS11N mRNA was translated in vitro by using a rabbit reticulocyte system (Smith and Campbell 1983 ) . mRNA levels were titrated from 10 ng to 4 l.tg, per each 10 pl of reaction mixture. After translation, nascent peptides were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis in 10% gels, before and after immunoprecipitation with rabbit anti-chicken liver GS y-globulin (Smith et al. 1983). Gels were fluorographed using an enhancer and Kodak X-Omat film. Software

Computer-assisted sequence manipulations were with the MacVector (IBI) program. Helical wheel projections were with the program Wheel, supplied by Mr. Martin Jones and Dr. Jere P. Segrest of the Artherosclerosis Research Unit, University of Alabama at Birmingham (see Segrest et al. 1990).

Avian GS Mitochondrial Targeting Signal

79 1

Results and Discussion Glutamine Synthetase cDNA Clone Isolation

Northern and Southern Blotting Northern blot analysis of chicken liver, brain, breast muscle, pancreas, spleen, and kidney (fig. 2; kidney is not shown) demonstrated high levels of GS expression in liver and brain, with little or no detectable expression in the other tissues. As shown previously, the major hybridization signal for most vertebrate GS mRNAs corresponds to -3.2 kb (Sanders and Wilson 1984; Smith and Campbell 1988; Pu and Young 1989). A signal corresponding to approximately one-half this size may also be obtained (Sanders and Wilson 1984)) possibly because of the use of alternative polyadenylation sites (Hayward et al. 1986). Patejunas and Young ( 1987) first proposed that avian GS is coded by a single gene. This has now been isolated and characterized (Pu and Young 1989). We also found that Southern blot analysis of chicken genomic DNA after restriction with EcoRI, BarnHI, HindIII, and PstI gave a pattern indicative of a single gene (data not shown). A similar pattern has been found with elasmobranch genomic DNA (Campbell and Anderson 199 1), suggesting that all vertebrate GSs are encoded in a single gene. Nucleotide Sequence of Chicken Liver Mitochondrial Synthetase cDNA

Glutamine

The nucleotide sequence for chicken liver GS cDNA derived from the various approaches is shown in figure 3. The nucleotide shown at position 225 is a T in pCGS11 but is a C in the mRNA derived by PCR amplification. This does not effect the coding sequence and may possibly reflect individual differences. The nucleotide sequence found for chicken liver GS cDNA differs in only minor ways from that for chicken retinal cDNA reported by Pu and Young ( 1989 ) . The nucleotides at positions 4 and 19 (G and T, respectively) are equivocal in our sequence data (C’s in Pu and Young’s sequence). The sequence at position 1443-1444 in the 3’noncoding region is unequivocal in our data, whereas Pu and Young report two additional T’s between the two positions. Again, these differences do not affect the coding sequence. Expression of pCGS11B in E. coli As shown in figure 4, the expression of the full-length GS cDNA construct resulted in a protein the same size as that of the authentic chicken liver GS subunit ( -42


The library created from chicken liver poly (A)+ RNA enriched for GS mRNA was screened with pGS 1, a bovine retinal GS cDNA containing the carboxyl one-third of the coding region (Smith and Campbell 1988), as described in the Material and Methods section. From a screen of 60,000 clones, two were isolated that continued to hybridize with pGS1 after tertiary screening. One clone of 1,200 bp, designated “pCGS3,” contained 400 bp of sequence similar to mammalian GS sequences, but the remaining 5’sequence was unrelated to any known GS sequences. This clone was judged to be either a cloning artifact or a processing intermediate and was not further characterized. A second clone, designated “pCGS11,” contained 1,622 bp. On sequencing, this clone was found to contain -85% of the protein coding sequence, with the remainder made up of 3’flanking noncoding sequence (see below). It was therefore used to determine tissue expression of chicken GS mRNA, for Southern blotting of genomic DNA, and to obtain additional sequence data.


PCR GAGECCCTGCCCGCAGCC~AGCCCAGGACAGCCCTCGCCAGCTCCGCAGCCATGGCCACC MAT --RNA,-1 TCGGCGAGCTCCCACCTGAGCAAAGCCATCAAGCCATGTGCTGCCGCAGGGT SASSHLSKAIKHMYMKLPQG

t

60 120

GAGAAGGTCCAAGCCATGTACATCTGGATCGACGGGACTGGGGAGCACCTCCGCTGC~ EKVQAMYIWIDGTGEH TmRhA+“Pt?R-flACCCGCACTCTGGACCACGAACCCAAGAGCCTGGAiiATCTCCC GAGTGGAACTTTGAT TRTLDHEPKSLEDLPEWNFD

180 240 300

ATGTTCCGGGACCCTTTTCGCAAGGATCCCAACAAATTAG MFRDPFRKDPNKLVLCEVFK

360

TACAACCGCCAGTCTGCAGACACAAATCTTCTTCGGCACACCTGTAGGCGGATTATGGATATG YNRQSADTNLRHT CRRIMDM

420

GTGTCCAACCAGCACCCCTGGTTTGGGATGGAGCAGGAGTACACCCTTCTGGG~CAGAT VSNQHPWFGMEQEYT L L G

480 T

D

GGTCATCCGTTTGGCTGGCCTTCCAATTGCTTGCTTCCCTGGACCCC~GGTCCGTACTACTGC GHPFGWPSNCFPGPQGPYYC

540

GGTGTAGGAGCTGACAAAGCCTATGGCAGAGACATTGTGGAGGCCCACTACCGAGCGTGC GVGADKAYGRDIVEAHYRAC

600

CTGTATGCTGGTGTGAAAATTGGAGGAACCAACGCAGAAGCCCAGTGGGAG LYAGVKIGGTNAEVMPAQWE

660

TTCCAGGTGGGACCGTGCGAGGATTGAGATGGGGGATCACCTCTGGATAGCACGTTTC FQVGPCEG IEMGDHLWIARF

720

ATCCTCCACCGGGTGTGCGAGACTTTGGTGTCATTGTGTCCTTCGATCCC~CCCATC ILHRVCEDFGVIVSFDPKPI

780

CCTGGGAACTGGAACGGTGCTGGCTGTCACACCAACTTCA PGNWNGAGCHTNFSTKNMRE

840

GATGGAGGTCTCllAGCACATCGAGGAGGCCATCGAGAAGCTGAGCAAGCGCCACCAGTAC DGGLKHIEEAIEKLSKRHQY

900

CACATCCGTGCCTACGACCCCAAAGGAGGAGGGCTGGAC~CGCCCGGCGCCTGACGGGCTTC HIRAYDPKGGLDNARRLTGF

960

CACGAGACGTCCAGCATCCACGAGTTCTCCGCCGGCGTGGCC~CCGCGGCGCCAGCATC HETSSIHEFSAGVANRGASI

1020

CGCATCCCACGCAACGTGGGCCATGAGAAGAAAGGCTACTTCGAGGACCGCGGGCCTTCA RIPRNVGHEKKGYFEDRGPS

1080

GCCAACTGCGATCCCTACGCCGTGACGGAGGCCCTGGTCCGTACGTGTCTCCTC~CG~ AN C D P YAVTEALVRTCL L

1140 N

E

ACCGGGGACGAGCCTTTTGAGTACAAGAACTAAGTGGACT~GTGGACTCGTGCCCACAGACACCGCCT TGDEPFEYKNX

1200

TCCCCCTCCCCCCACCCCCCCCGTGCTCCCCGTACCCCTACTTCCCTTCTAGTTGTAA

1260

TCCTGAGGGTACAAGATAACACCTTCGTGTCTCAGTAACTCTTGTTGTTTTGAGGTGGGG

1320

GAGGAGGGCAGGTTTAGTTTTATTAATGTCTGTCTGTTTGTCATTGACTCTTCA~GGCGAGA

1380

GGAGGAGGGGGGTGGGGGGGGGGTAGATGATTTTTAACCATGTTTTGTTTTGGTTTTTT

1440

TT~ETCTAATTCATCTGTACATCTGGTGGGATCCAGTGGCTTCAGGGACAGGCTACACAC

1500

AAAGCCAGCAGACGAGAGGAGTGGAAAATTATGGCCGACTTCCCC~TTTGATTGTT

1560

AGGCCCAGCGCTGGGGGGCTCTCTCCCAGCGGCTGATGGTTCTC~GGCATAGCATCCCA

1620

AGAAGGGCAGGCAGGGACTGGGATGGAACCCAGCTCTGTCCCGTGGTGTGATGGGG~GT

1680

CTTCAGCCTATTGGCACTGTCACATTTCTAGGCTCCTTCTTGGCATCGAGCGTGTGGTGT

1740

AGGAGCACCTAGGAGAGCTCCTCGTCCTGTCATTGGAAGTA

1800

ACACAGAGTGAAGTAGTATTTTTCTATTTAAATGTAAA


GGdTCCAGCACCTTCCAAGCCGAAGGCTCCAACAGCGACATGTACCTGCGACCTGCTGCC GSSTFQAEGSNSDMYLRPAA

FIG. 3.-Nucleotide sequence of chicken liver GS cDNA construct and its derived amino acid sequence. Regions analyzed by the PCR and by direct mRNA sequencing are shown by arrows. Asterisks (* ) indicate bases that differ from those reported by Pu and Young ( 1989) (see text). The start of the pCGSl1 clone is indicated by the downward-pointing arrowhead.



Chicken Human Rat Hamster Mouse


795

MATSASSHLNKGI AMYIWVDGTGE

PKCVEELPEWNFDGSSl@QSEGSNSDMY PKCVEELPEWNFDGSSTFQSEGSNSDMY

150 VFKYNR&&$#NLRH IMDMVSNQHPWFGMEQEYTLETDGHPFGWP KRIMDMVSNQHPWFGMEQEYTLMGTDGHPFGWP VFKYNRKPAETNLRH VFKYNRKPAETNLRHSCKRIMDMV$&HPWFGMEQEYTLMGTDGHPFGWP VFKYNRKPAETNLRHSCKRIMDMVSNQHPWFGMEQEYTLMGTDGHPFGWP VFKYNRKPAETNLRHRCKRIMDMVSNQHPWFGMEQEYTLMGTDGHPFGWP


SWFPGPQGPYYCGVGADKAYGRDIVEAHYRACLYAGVK SNGFPGPQGPYYCGVGADRAYGRDIVEAHYRACLYAGVK SNGFPGPQGPYYCGVGADKAYGRDIVEAHYRACLYAG~ITGTNAEVMPA GFPGPQGPYYCGVGADKAYGRDIVEAHYRACLYAGVKITGTNAEVMPA GFPGPQGPYYCGVGADKAYGRDIVEAHYRACLYAGVKITGTNAEVMPA


QWEF-PCEGI GDHLWjI]ARFILHRVCEDFGVIwDPKPIPGNWNGA GDHLWVARFILHRVCEDFGVIATFDPKPIPGNWNGA QWEFQIGPCEGI QWEFQIGPCEGIRMGDHLWVARFILHRLCEDFGVIATFDPKPIPGNWNGA QWEFQIGPCEGIRMGDHLWVARFILHRVCEDFGVIATFDPKPIPGNWNGA QWEFQIGPCEGI/&GDHLW@@FILHRVCEDFGVIATFDPKPIPGN~


GCHTNFSTK#+RuG GCHTNFSTKAMREENG GCHTNFSTKAMREENG GCHTNFSTKAMREENG GCHTNFSTKAMREENGL

200 TNAEVMPA

250

300

Chicken Human Rat Hamster Mouse Chicken Human Rat Hamster Mouse

IEEAIEKLSKRHQYHIRAYDPKGGLDNARRL IEEAIEKLSKRHQYHIRAYDPKGGLDNARRL



FIG. 5.-Amino acid sequences of vertebrate GSs. Sources are chicken (Gal/us gallus), Pu and Young ( 1989) and present study; human, Gibbs et al. ( 1987); rat (Rattus norvegicus), van de Zande et al. ( 1990); hamster (Cricetulus griseus), Hayward et al. ( 1986); and mouse (Mus musculus), Bhandari et al. ( 1991). For the rat sequence, Mill et al. ( 199 1) report a K at position 310 and an I at position 355.

sequence similarity. In fact, mitochondrial malate dehydrogenase shows more sequence similarity with the Escherichia coli enzyme than with its cytosolic counterpart (Joh et al. 1987 ) . In contrast, the compartmental isozymes of GS in higher vertebrates are highly conserved. The sequence similarity between chicken and human GS is 90%; that between chicken and hamster GS is 88%; and that between chicken and mouse

796

Campbell and Smith

Identification of Putative Targeting Signal in Chicken GS


and rat is 87%. Of the differences between the chicken and human enzymes, only 8% of the replacements are “radical” in that they result in marked differences in local functionality. As indicated above, all vertebrate GSs appear to be coded for a single gene despite being compartmental isozymes. However, in Drosophila melanogaster, the mitochondrial and cytosolic isozymes of GS are coded by separate genes (Caizzi et al. 1990). It has been suggested that these two insect GS genes diverged about the same time as did the invertebrates and the vertebrates (Pesole et al. 1991). Drosophila cytosolic GS sequence is more similar to that of the rat (62%; Pesole et al. 199 1) and chicken (5 1%; present data) than is the sequence of the mitochondrial enzyme (46% and 43%, respectively), and Pesole et al. ( 1991) therefore suggest that the cytosolic Drosophila GS gene shared a common ancestor with the vertebrate gene. Drosophila mitochondrial cDNA encodes a larger peptide than does the cytosolic cDNA (Caizzi et al. 1990)) presumably because of the presence of a mitochondrial targeting presequence. The mitochondrial isozyme of GS is also translated as a larger-MT precursor in certain chondrichthyian fish tissues (Smith et al. 1987; Campbell and Anderson 199 1). Thus, were an invertebrate gene similar to that coding for the Drosophila cytosolic isozyme to have been ancestral to the vertebrate gene, coding information for a mitochondrial targeting signal would have had to have been acquired early in the vertebrate lineage, on the basis of the presence of such a signal in extant chondrichthyians. The N-terminus of Drosophila mitochondrial GS translation product contains an additional 35 amino acids compared with Drosophila cytosolic GS (Caizzi et al. 1990). However, on the basis of the common motifs described by Gavel and von Heijne ( 1990), there are no potential cleavage sites in this extension. Secondarystructure predictions indicate that no more than the first 20 amino acids are in a helical configuration, and, while this region contains three basic amino acids (Arg4, Lys, , , and LyQ, it also contains an acidic amino acid ( G~u,~) and is not enriched in hydroxy amino acids, especially Ser (von Heijne et al. 1989; see below). Thus, if this region contains a targeting signal, it is an unorthodox one.

As can be seen in figure 5, the 20 N-terminal amino acids in vertebrate GS are the most variable: among the five species, there are a total of 12 replacements in this region, which is more than for any other run of 20 amino acids elsewhere in the molecule. Because the internal mitochondrial targeting signals of several other “leaderless” mitochondrial enzymes have been localized to the N-terminus, the hypothesis is that the main information for targeting of chicken liver GS to the mitochondrial matrix resides in its first 20 amino acids. Specific amino acid replacements in this region of the cytosolic enzymes during divergence of the mammalian and avian lines may therefore have served to convert it from a nontargeting sequence to a targeting sequence. There is no consensus sequence for targeting to mitochondria, but transient, Nterminal targeting presequences do share certain common properties. These include high basic and hydroxy amino acid contents, the general absence of acidic amino acids, and, most important, the ability to form amphipathic helices (Lemire et al. 1989; von Heijne et al. 1989). Although a considerable amount is known about targeting proteins to the mitochondrial matrix via such N-terminal presequences, far less is known about targeting to this compartment via internal signals, such as occurs with chicken liver GS (Grivell 1988; Pfanner and Neupert 1990). For the three or so other enzymes in this category, experimental evidence for a functional internal sequence is


191


available for only one, 3-ketoacyl-coA thiolase. The mature thiolase has been shown to inhibit import of a protein containing a presequence (preOTC; Mori et al. 1985), indicating competition for a common receptor (Steger et al. 1990). The N-terminal 14-16 amino acids of the thiolase have also been shown to target a passenger protein (i.e., OTC) to mitochondria ( Arakawa et al. 1990), thereby localizing the signal to this region. At least two inner membrane proteins-the subunit of cytochrome c oxidase (Duhl et al. 1990) and the uncoupling protein (Liu et al. 1988)-have also been shown to have internal targeting signals at the N-terminus. Mutations at the N-terminus of proteins whose targeting presequences have been genetically deleted may restore import (Vassarotti et al. 1987)) and redundant targeting information may also be present in this region (Bedwell et al. 1987). These observations therefore support the N-terminal localization of the targeting sequence for chicken liver GS. However, targeting information could also be present in other regions of the molecule. This has, for example, been shown for the inner membrane ADP-ATP translocator (Smagula and Douglas 1988) and apocytochrome c (Nye and Scarpulla 1990). With respect to chicken GS, there are some radical substitutions at the C-terminus, including a Glyfor-Arg replacement at position 34 1 and a Glu-for-Gln at position 370 (fig. 5 ) . One of the most characteristic properties of mitochondrial targeting N-terminal presequences is their ability to form amphipathic helices (Lemire et al. 1989; von Heijne et al. 1989). As shown in figure 6, secondary-structure predictions based both on the algorithm of Chou and Fasman and on that of Robson and Garnier show a core helical structure at the N-terminus of chicken liver GS. Only the Robson-Gamier algorithm predicts helical structure in this region of the cytosolic human GS, the GS most similar to the chicken enzyme. Helical wheel projections of the amino acids

50

100

l50

200

WI

300

350

FIG. 6.-Secondary-structure predictions for mammalian (human; A) and chicken (B) GSs. The ChouFasman (CF; Chou and Fasman 1978) and Robson-Gamier ( RG; Gamier et al. 1978) methods were used as executed by the MacVector 3.5 program, which also gives a combination of the two (CfRg). Arrows indicate N-terminal helical regions in chicken GS that were not predicted by the Chou-Fasman method for mammalian GSs.


Role of Specific Amino Acid Replacements in Chicken GS


involved have been routinely used to predict amphipathic helices (Schiffer and Edmundson 1967; Segrest et al. 1990), and, as shown in figure 7B, such a projection of the first 20 amino acids of chicken liver GS demonstrates amphipathic structure. A less “complete” amphipathic structure is shown with mammalian cytosolic GSs, as illustrated in figure 7A with the human enzyme. The hydrophobic moment/residue for the first 20 amino acids of all vertebrate GSs, with the exception of Chinese hamster GS, is -0.3 (range 0.297-0.308), which is near the mean for several types of amphipathic helices (Parker and Song 1990; Segrest et al. 1990). The value for Chinese hamster GS is 0.09. The N-terminal sequence of Drosophila mitochondrial GS has a hydrophobic moment/residue of 0.11, which again indicates that this would be an unorthodox targeting signal. Helical wheel projections of the 20 or so N-terminal amino acids of medium-chain acyl-coA dehydrogenase (Kelly et al. 1987) and of mitochondrial, but not cytosolic, adenylate kinase [J. W. Campbell, unpublished data from Schulz et al. ( 1986)] show that they also can be predicted to form amphipathic helices. However, unlike the putative N-terminal targeting signal of chicken GS, the N-termini of these two proteins do not contain an excess of hydroxy amino acids. None of the proteins targeted to the matrix via internal signals contain acidic amino acids in their N-termini.

The replacement of HisI and Lyslg at the N-terminus of the chicken GS serves to increase and expand the hydrophilicity on one side of the predicted helix (fig. 7B). Of the amino acids occurring at position 19 in the cytosolic mammalian enzymes, Ser, the most hydrophilic, still has only one-half the hydrophilicity value of Lys (Eisenberg et al. 1984). The substitution of HisIs does not increase the hydrophilicity, but it does provide for an additional positive charge that may serve to increase the interaction between the N-terminus and the negatively charged cardiolipins and other phospholipids in the inner membrane (Endo et al. 1989; Tamm and Bartoldus 1990)) which results in the eventual translocation due to Aw (Martin et al. 199 1) . A key role for His in mitochondrial targeting signals has been demonstrated with aspartate aminotransferase in which the replacement of His5 by Gly abolishes its import (Nishi et al, 1989). The replacement of HisI and Lyslg in chicken GS could have occurred via third-base substitutions in genes coding one or the other of the mammalian cytosolic GSs. In the case of HisiS a Gln-to-His replacement could have occurred in any ancestral vertebrate GS (fig. 5), whereas in the case of Lyslg an Asn-to-Lys replacement in an ancestral protein similar to that now found in the rat would suffice. His and Lys are the only basic amino acids found in the putative targeting signal of chicken GS; this is unusual, in that Arg is almost universally the predominant basic amino acid found in other mitochondrial targeting sequences (von Heijne et al. 1989). At the N-terminus of chicken liver GS there are possibly two other replacements that may have been critical in converting it from a nontargeting to a targeting function. One is Alar2 . Although there is some controversy with respect to the a-helix-forming capacity of Ala (Kemp et al. 199 1)) several experimental approaches suggest that it has a high propensity for this (Chakrabartty et al. 199 1; Daggett et al. 199 1; Padmanabhan and Baldwin 1991). The presence of Alar2 in the chicken enzyme may therefore serve to convert the first 20 or so amino acids to the a-helical configuration necessary for targeting (fig. 6). The other possibly important replacement is Serlo. All cytosolic mammalian GSs have an Asn at this position. Whether this replacement is critical for a targeting function in the chicken enzyme is not known, but, in the signal


flG. 7.-Helical wheel projections of the first 20 N-terminal amino acids of human (A) and chicken (B) GSs. The hydrophobic moments/residue the A and B helices, respectively.

are 0.299 & 0.29 1 for


peptide of the OmpA secretory protein, the replacement of Iles by an Asn renders it incapable of penetrating the hydrophobic core of membrane lipid bilayers (Goldstein et al. 199 1; Hoyt and Gierasch 199 1). Tissue-specific Targeting of Chicken GS

Conclusions


Chicken brain GS occurs in the cytosolic compartment (Smith and Campbell 1983 ), so we compared the sequences of the liver and brain enzymes, especially their N-termini. Direct sequencing of brain GS mRNA by using AMV reverse transcriptase as described in Material and Methods indicated that the first 36 N-terminal amino acids of the brain enzyme are identical to those of the liver despite its cytosolic localization. There is indirect evidence for a cytosolic localization of GS in the chicken neural retina (Soh and Sakar 1978 ) , and the amino acid sequence derived from retinal GS cDNA as well as from the avian GS gene itself (Pu and Young 1989) indicates that the complete sequence of cytosolic GS in neural tissue of the chicken is identical to that of mitochondrial GS in liver. This therefore raises the question of the mechanism responsible for the tissue-specific compartmentation of GS in birds. Heretofore, the information for differential compartmentation of isozymes has been known to be contained in the primary sequences of their mRNA translation products. In the case of mitochondrial versus cytosolic isozymes-such as aspartate aminotransferase (Christen et al. 1985 ) , phosphoenolpyruvate carboxykinase (Hod et al. 1982)) and malate dehydrogenase (Joh et al. 1987), which are coded by separate genes-differential compartmentation requires the differential expression of either the gene that codes for the targeting signal or that which does not, and this expression may be tissue specific (Kelly et al. 1989 ). The mitochondrial form of these three enzymes is targeted by a transient, N-terminal presequence. However, differential gene expression is also used for enzymes such as adenylate kinase, the mitochondrial isozyme of which has an internal targeting signal (see Suminami et al. 1988). Even in cases where the two compartmental isozymes are coded by a single gene, there are mechanisms for generating translation products with or without the mitochondrial targeting signal (Surguchov 1987). This may be via the use of different transcriptional initiation sites to generate separate mRNAs (Tropschug et al. 1988) or, possibly, via the use of alternative translation start sites from the same mRNA (Suzuki et al. 1989). No other consensus initiation sites exist 5 ’of the designated start site for chicken GS (Pu and Young 1989; fig. 3 ) , which would appear to eliminate an mRNA coding for a mitochondrial targeting signal too small to be detected by the methods used. The differential compartmentation of chicken GS would thus appear to be due to a tissue-specific factor(s) rather than to the presence or absence of a targeting signal in the mRNA translation product.

1. A comparison of the derived amino acid sequence of chicken liver mitochondrial GS with the sequences of four mammalian liver cytosolic GSs has revealed 87%90% conservation of structure despite the differential compartmentation of the enzymes and the long time since the divergence of birds and mammals. 2. The N-terminus of higher-vertebrate GS is one of its more variable regions, and a comparison and analysis of the first 20 N-terminal amino acids of the chicken enzyme indicates that these may serve as an internal mitochondrial targeting signal. This region contains several hydroxy and basic amino acids, does not contain an acidic amino acid, and can be predicted to form a strong amphipathic a-helix, all of which are properties shared by mitochondrial targeting sequences.

Avian GS Mitochondrial Targeting Signal 80 I

Acknowledgments


3. A comparison of the N-termini of mammalian liver GSs with the N-terminus of avian liver GS indicates that there may have been three main amino acid replacements responsible for converting the latter to a mitochondrial targeting signal. These would have been Alar*, HisIs, and Lyslg, all of which contribute to the amphipathic a-helix configuration of the N-terminus of the avian liver enzyme. 4. As shown previously by others and confirmed here, avian GS is coded by a single gene. An analysis of the first 36 amino acids of chicken brain GS indicated that they are identical to those of liver GS. The complete published sequence for the chicken retinal enzyme is also identical to that of chicken liver GS. The brain enzyme is known to be cytosolic, and there is evidence that this is also true for the retinal enzyme. Information for the differential compartmentation of GS in the chicken thus appears not to reside in the primary sequence of the mRNA translation products, as it does in all other compartmental isozymes. Other tissue-specific factors may therefore be responsible for the localization of avian GS to different cellular compartments.

This work was supported by NSF grant DCB x7-18042. J.W.C. would like to thank Dr. Susan Chacko for help with the Wheel program and Dr. Nadine Ritter for help with fig. 4. LITERATURE CITED

AIVIAYA,Y., H. ARAKAWA, M. TAKIGUCHI, Y. EBINA, S. YOKOTA, and M. MORI. 1988. A noncleavable signal for mitochondrial import of 3-oxoacyl-CoA thiolase. J. Biol. Chem. 263: 14463-14470. ARAKAWA, H., Y. AMAYA, and M. MORI. 1990. The NH*-terminal 14-16 amino acids of mitochondrial and bacterial thiolases can direct mature omithine carbamoyltransferase into mitochondria. J. Biochem. 107: 160- 164. ATTARDI, G., and G. SCHATZ. 1988. Biogenesis of mitochondria. Annu. Rev. Cell Biol. 4:289333. AVIV, H., and P. LEDER. 1972. Purification of biologically active globin messenger RNA by chromatography on oligothymidylic-cellulose. Proc. Natl. Acad. Sci. USA 69: 1408- 14 12. BEDWELL,D. M., D. J. KLIONSKY, and S. D. EMR. 1987. The yeast F,-ATPase p subunit precursor contains functionally redundant mitochondrial protein import information. Mol. Cell. Biol. 7:4038-4047. BHANDARI,B., W. J. ROESLER,K. D. DELISIO, D. J. KLEMM, N. S. Ross, and R. E. MILLER. 1991. A functional promoter flanks an intronless glutamine synthetase gene. J. Biol. Chem. 266:7784-7792. BLIN, N., and D. W. STAFFORD. 1976. A general method for isolation of high molecular weight DNA from eukaryotes. Nucleic Acids Res. 3:2303-2308. BOWTELL,D. D. L. 1987. Rapid isolation of eukaryotic DNA. Anal. Biochem. 162:463-465. CAMPBELL,J. W. 1991. Excretory nitrogen metabolism. Pp. 277-324 in C. L. PROSSER,ed. Environmental and metabolic animal physiology. Comparative animal physiology, 4th ed. Wiley-Liss, New York. CAMPBELL,J. W., and P. M. ANDERSON, 199 1. Evolution of mitochondrial enzyme systems in fish: the mitochondrial synthesis of glutamine and citrulline. Pp. 43-75 in P. W. H~CHACHKA and T. P. MOMMSEN,eds. Biochemistry and molecular biology of fishes. Vol. 1: Phylogenetic and biochemical perspectives. Elsevier, Amsterdam. CAMPBELL,J. W., J. E. VORHABEN,and D. D. SMITH, JR. 1987. Uricoteley: its nature and origin during the evolution of tetrapod vertebrates. J. Exp. Zool. 243:349-363. CAIZZI, R., M. P. BOZZETTI, C. CAGGESE,and F. RITOSSA. 1990. Homologous nuclear genes



encode cytoplasmic and mitochondrial glutamine synthetase in Drosophila melanogaster. J. Mol. Biol. 212: 17-26. CHAKRABARTTY,A., J. A. SCHELLMAN,and R. L. BALDWIN. 1991. Large differences in the helix propensities of alanine and glycine. Nature 351:586-588. CHEN, E. Y., and P. H. SEEBURG. 1985. Supercoil sequencing: a fast and simple method for sequencing plasmid DNA. DNA 4: 165- 170. CHIRGWIN, J. M., A. E. PRZYBYLA,R. J. MACDONALD,and W. J. RUTTER. 1979. Isolation of biologically active ribonucleic acids from sources enriched in ribonucleases. Biochemistry l&5294-5299. CHOU, P. Y., and G. D. FASMAN. 1978. Prediction of the secondary structures of proteins from their amino acid sequence. Adv. Enzymol. Relat. Areas Mol. Biol. 47:45-148. CHRISTEN, P., U. GRAF-HAUSNER, F. BOSSA,and S. DOONAN. 1985. Comparison of covalent structures of the isozymes of aspartate aminotransferase. Pp. 173- 185 in P. CHRISTENand D. E. METZLER, eds. Transaminases. Elsevier, Amsterdam. CONTRERAS,R., H. CHEROUTRE,W. DEGRAVE,and W. FIERS. 1982. Simple, efficient in vitro synthesis of capped RNA useful for direct expression of cloned eukaryotic genes. Nucleic Acids Res. 10:6353-6362. DAGGETT, V., P. A. KOLLMAN,and I. D. KUNTZ. 199 1. A molecular dynamics stimulation of polyalanine: an analysis of equilibrium motions and helix-coil transitions. Biopolymers 31: 1115-1134. DEGROOT,C. J., G. H. J. TENVOORDE,R. E. VAN AANDEL,A. TE KORTSCHOT,J. W. GAASBEEK JANZEN, R. H. WILSON, A. F. M. MOORMAN, R. CHARLES,and W. H. LAMARS. 1987. Reciprocal regulation of glutamine synthetase and carbamoylphosphate synthetase levels in rat liver. Biochim. Biophys. Acta 908:23 l-240. DUHL, D. M., T. POWELL,and R. 0. POYTON. 1990. Mitochondrial import of cytochrome c oxidase subunit VIIa in Saccharomyces cerevisiae: identification of sequences required for mitochondrial localization. J. Biol. Chem. 265:7273-7277. EISENBERG,D., E. SCHWARTZ,M. KOMAROMY,and R. WALL. 1984. Analysis of membrane and surface protein sequences with the hydrophobic moment plot. J. Mol. Biol. 179:125142. ENDO, T., M. EILERS,and G. SCHATZ. 1989. Binding of a tightly folded artificial mitochondrial precursor protein to the mitochondrial outer membrane involves a lipid-mediated conformational change. J. Biol. Chem. 264:295 l-2956. FEINBERG, A. P., and B. VOGELSTEIN. 1983. A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6- 13. GARNIER,J., D. J. OSGUTHROPEand B. ROBSON. 1978. Analysis ofthe accuracy and implication of simple methods for predicting the secondary structure of globular proteins. J. Mol. Biol. 120:97-120. GAVEL, Y., and G. VON HEIJNE. 1990. Cleavage-site motifs in mitochondrial targeting peptides. Protein Eng. 4:33-37. GEBHARDT, R., A. EBERT, and G. BAUER. 1988. Heterogeneous expression of glutamine synthetase in rat liver parenchyma revealed by in situ hybridization and Northern blot analysis of RNA from periportal and perivenous hepatocytes. FEBS L&t. 241:89-93. GEBHARDT,R., and D. MECKE. 1983. Heterogenous distribution of glutamine synthetase among rat liver parenchymal cells in situ and in primary culture. EMBO J. 2:567-570. GELIEBTER,J., R. A. ZEFF, R. W. MELVOLD,and S. G. NATHENSON. 1986. Mitotic recombination in germ cells generated two major histocompatibility complex mutant genes shown to be identical by RNA sequence analysis: Kbmg and Kbm6. Proc. Natl. Acad. Sci. USA 83: 3371-3375. GIBBS, C. S., K. E. CAMPBELL,and R. H. WILSON. 1987. Sequence of a human glutamine synthetase cDNA. Nucleic Acids Res. 15:6293. GIBBS,R. A., P.-N. NGUYEN, L. J. MCBRIDE, S. M. KOEPF, and C. T. CASKEY. 1989. Identi-

Avian GS Mitochondrial Targeting Signal 803


fication of mutations leading to the Lesch-Nyhan syndrome by automated direct DNA sequencing of in vitro amplified DNA. Proc. Natl. Acad. Sci. USA 86: 19 19- 1923. GOLDSTEIN, J., S. LEHNHARDT, and M. INOUYE. 1991. In vitro effect of asparagine in the hydrophobic region of the signal sequence. J. Biol. Chem. 266: 144 13- 144 17. GRIVELL, L. A. 1988. Protein import into mitochondria. Int. Rev. Cytol. 111: 107-141. GUBLER, U., and B. J. HOFFMAN. 1983. A simple and very efficient method for generating cDNA libraries. Gene 25:263-269. HANAHAN,D. 1983. Studies on transformation of Escherichiu coli with plasmids. J. Mol. Biol. 166:557-580. H.XUSSINGER,D., and W. GEROK. 1986. Metabolism of amino acids and ammonia. Pp. 253291 in G. THURMAN, F. C. KAUFFMAN, and K. JUNGERMANNeds. Regulation of hepatic metabolism. Plenum, New York. HAYMERLE,H., J. HERZ, G. M. BESSAN,R. FRANK, and K. K. STANLEY. 1986. Efficient construction of cDNA libraries in plasmid expression vectors using an adaptor strategy. Nucleic Acids Res. 14:86 15-8624. HAYWARD,B. E., A. HUSSAIN, R. H. WILSON, A. LYONS, V. WOODCOCK,B. MCINTOSH, and T. J. R. HARRIS. 1986. The coding and nucleotide sequence of cDNA for an amplified glutamine synthetase gene from the Chinese hamster. Nucleic Acids Res. 14:999-1008. HOD, Y., M. F. UTTER, and R. W. HANSON. 1982. The mitochondrial and cytosolic forms of avian phophoenolpyruvate carboxykinase (GTP) are encoded by different mRNAs. J. Biol. Chem. 257:13787-13794. HOYT, D. W., and L. M. GIERASCH . 199 1. Hydrophobic content and lipid interactions of wildtype and mutant OmpA signal peptides correlate with their in vivo function. Biochemistry 30:10155-10163. JOH, T., H. TAKESHIMA,T. TSUZUKI, C. SETOYAMA,K. SHIMADA,S. TANASE,S. KURAMITSU, H. KAGAMIYAMA,and Y. MORINO. 1987. Cloning and sequence analysis of cDNA encoding mammalian cytosolic malate dehydrogenase. J. Biol. Chem. 262: 15 127- 15 13 1. KELLY,D. P., J. I. GORDON, R. ALPERS,and A. W. STRAUSS.1989. The tissue-specific expression and developmental regulation of two nuclear genes encoding rat mitochondrial proteins. J. Biol. Chem. 264:18921-18925. KELLY, D. P., J.-J. KIM, J. J. BILLADELLO,B. E. HAINLINE, T. W. CHU, and A. W. STRAUSS. 1987. Nucleotide sequence of medium-chain acyl-CoA dehydrogenase mRNA and its expression in enzyme-deficient human tissue. Proc. Natl. Acad. Sci. USA 84:4068-4072. KEMP, D. S., J. G. BOYD, and C. C. MUENDEL. 1991. The helical s constant for alanine in water derived from template-nucleated helices. Nature 352:45 l-454. LIU, X., A. W. BELL, K. B. FREEMAN,and G. C. SHORE. 1988. Topogenesis of mitochondrial inner membrane uncoupling protein: rerouting transmembrane segments to the soluble matrix compartment. J. Cell Biol. 107:503-509. LEMIRE,B. D., C. FANKHAUSER,A. BAKER,and G. SCHATZ. 1989. The mitochondrial targeting function of randomly generated peptide sequences correlated with predicted helical amphiphilicity. J. Biol. Chem. 264:20206-202 15. MARTIN, J., K. MAHLEK, and N. PFANNER. 1991. Role of an energized inner membrane in mitochondrial protein import: AY drives the movement of presequences. J. Biol. Chem. 266:18051-18057. MILL, J. F., K. M. MEAROW, H. J. PUROHIT, H. HALEEM-SMITH,R. KING, and E. FREESE. 1991. Cloning and functional characterization of the rat glutamine synthetase gene. Mol. Brain Res. 9: 197-207. MOORMAN,A. F. M., P. A. J. DE BOER, W. J. C. GEERTS, L. VAN ZANDE, W. H. LAMAR&and R. CHARLES. 1988. Complementary distribution of carbamoylphosphate synthetase in rat liver acinus is regulated at a pretranslational level. J. Histochem. Cytochem. 36:75 l-755. MORI, M., H. MATSUE, S. MIURA, M. TATIBANA, and T. HASHIMOTO. 1985. Transport of proteins into mitochondrial matrix: evidence suggesting a common pathway for 3-ketoacyCoA thiolase and enzymes having presequences. Eur. J. Biochem. 149: 18 l- 186.

804 Campbell and Smith NISHI,T., F. NAGASHIMA, S.TANASE, Y. FUKUMOTO, T. JOH, K. SHIMADA,Y. MATSUKADO,


Y. USHIO, and Y. MORINO. 1989. Import and processing of precursor to mitochondrial aspartate aminotransferase: structure-function relationships of the presequences. J. Biol. Chem. 264:6044-605 1. NYE, S. H., and R. C. SCARPULLA . 1990. Mitochondrial targeting of yeast apoiso- 1-cytochrome c is mediated through functionally independent structural domains. Mol. Cell. Biol. 10: 5763-5771. PADMANABHAN,S., and R. L. BALDWIN. 1991. Short-chain non-polar amino acids are good helix-formers in water. J. Mol. Biol. 219: 135-l 37. PARKER,W., and P.-S. SONG. 1990. Location of helical regions in tetrapyrrole-containing proteins by a helical hydrophobic moment analysis. J. Biol. Chem. 265: 17568- 17575. PATFXJNAS,G., and A. P. YOUNG. 1987. Tissue-specific regulation of avian glutamine synthetase expression during development and in response to ghtcocorticoid hormones. Mol. Cell. Biol. 7:1070-1077. F%SOLE,G., M. P. BOZZETTI, C. LANAVE,G. PREPARATA,and C. SACCONE. 1991. Glutamine synthetase gene evolution: a good molecular clock. Proc. Natl. Acad. Sci. USA 88:522-526. PFANNER, N., and W. NEUPERT. 1990. The mitochondrial protein import apparatus. Annu. Rev. Biochem. 59:331-353. Pu, H., and A. P. YOUNG. 1989. The structure of the chicken glutamine synthetase-encoding gene. Gene 81:169-175. SAIKI, R. K., D. H. GELFAND, S. STOFFEL, S. J. SCHARF, R. HIGUCHI, G. T. HORN, K. B. MULLIS, and H. A. ERLICH . 1988. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487-49 1. SANDERS,P. R., and R. H. WILSON. 1984. Amplification and cloning of the Chinese hamster glutamine synthetase gene. EMBO J. 3:65-7 1. SCHIFFER, M., and A. B. EDMUNDSON. 1967. Use of helical wheels to represent the structures of proteins and to identify segments with helical potential. Biophys. J. 7: 12 l- 135. SCHULZ,G. E., E. SCHILTZ,A. G. TOMASSELLI,R. FRANK, M. BRUNE, A. WITTINGHOFER,and R. H. SCHIRMER. 1986. Structural relationships in the adenylate kinase family. Eur. J. Biochem. 161: 127- 132. SEGREST,J. P., H. DE LOOF, J. G. DOHLMAN, C. G. BROUILLETTE,and G. M. ANANTHARA MAIAH . 1990. Amphipathic helix motif: classes and properties. Proteins 8: 103-l 17. SMAGULA,C. S., and M. G. DOUGLAS. 1988. ADP-ATP carrier of Saccharomyces cerevisiae contains a mitochondrial import signal between amino acids 72 and 111. J. Cell. Biochem. 36:323-327. SMITH, D. D., and J. W. CAMPBELL.1983. Subcellular localization of chicken brain glutamine synthetase and comparison with chicken liver mitochondrial glutamine synthetase. J. Biol. Chem. 258:12265-12268. ---. 1987. Glutamine synthetase in liver of the American alligator, Alligator mississippiensis. Comp. Biochem. Physiol. 86B3755-762. -. 1988. Distribution of glutamine synthetase and carbamoyl-phosphate synthetase in vertebrate liver. Proc. Natl. Acad. Sci. USA 85: 160- 164. SMITH, D. D., JR., N. M. RITTER, and J. W. CAMPBELL.1987. Glutamine synthetase isozymes in elasmobranch brain and liver tissues. J. Biol. Chem. 262: 198-202. SMITH, D. D., JR., J. E. VORHABEN,and J. W. CAMPBELL.1983. Preparation and cross-reactivity of anti-avian glutamine synthetase antibody. J. Exp. Zool. 226:29-35. SOH, B. M., and P. K. SARKAR. 1978. Control of glutamine synthetase messenger RNA by hydrocortisone in embryonic chick retina. Dev. Biol. 64~3 16-328. SOUTHERN, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503-5 17. STEGER, H. F., T. SOLLNER,M. KIEBLER,K. A. DIETMEIER,R. PFALLER,K. S. TRULZSCH,M. TROPSCHUG, W. NEUPERT, and N. PFANNER. 1990. Import of ADP/ATP carrier into mitochondria: two receptors act in parallel. J. Cell Biol. 111:2353-2363.

Avian GS Mitochondrial Targeting Signal 805

RICHARD K. KOEHN, reviewing Received

November

Accepted

March

editor

15, 199 1; revision

3 1, 1992


STELLWAY,E. J., and A. E. DAHLBERG. 1980. Electrophoretic transfer of DNA, RNA and protein onto dizobenzyloxymethyl (DBM)-paper. Nucleic Acids Res. 8:299-3 17. SUMINAMI,Y., F. KISHI, T. TORIGOE, and A. NAKAZAWA. 1988. Structure and complete nucleotide sequence of the gene encoding cytosolic adenylate kinase. J. Biochem. (Tokyo) 103: 61 l-617. SURGUCHOV,A. P. 1987. Common genes for mitochondrial and cytoplasmic proteins. Trends Biochem. Sci. 12:335-338. SUZUKI, T., M. SATO, T. YOSHIDA, and S. TUBOI. 1989. Rat liver mitochondrial and cytosolic fumarases with identical amino acid sequences are encoded from a single gene. J. Biol. Chem. 264:2581-2586. TAMM, L. K., and I. BARTOLDUS. 1990. Secondary structure of mitochondrial signal peptide in lipid bilayer membranes. FEBS Lett. 272:29-33. THOMAS, P. S. 1980. Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc. Natl. Acad. Sci. USA 77:5201-5205. TROPSCHUG,M., D. W. NICHOLSON,F.-U. HARTL, H. K~HLER, N. PFANNER, E. WACHTER, and W. NEUPERT. 1988. Cyclosporin A-binding protein (cyclophilin) of Neurospora crassa: one gene codes for both the cytosolic and mitochondrial forms. J. Biol. Chem. 263: 1443314440. VAN DE ZANDE, L., W. T. LABRUYERE,A. C. ARNBERG, R. W. WILSON, A. J. W. VAN DEN BOGAERT,A. T. DAS, D. A. J. VAN OORSCHOT,C. FRIJTERS, R. CHARLES,A. F. M. MOORMAN, and W. H. LAMARS. 1990. Isolation and characterization of rat glutamine synthetaseencoding gene. Gene 87~225-232. VASSAROTTI,A., R. STROUD, and M. DOUGLAS. 1987. Independent mutations at the amino terminus of a protein act as surrogate signals for mitochondrial import. EMBO J. 6:705711. VON HEIJNE, G., J. STEPPUHN,and R. G. HERRMANN. 1989. Domain structure of mitochondrial and chloroplast targeting peptides. Eur. J. Biochem. 180:535-545. VORHABEN,J. E., and J. W. CAMPBELL. 1972. Glutamine synthetase: a mitochondrial enzyme in uricotelic species. J. Biol. Chem. 247:2763-2767. -. 1977. Submitochondrial localization and function of enzymes of glutamine metabolism in avian liver. J. Cell Biol. 73:300-3 10. VORHABEN,J. E., D. D. SMITH and J. W. CAMPBELL. 1982. Characterization of glutamine synthetase from avian liver mitochondria. Int. J. Biochem. 14:747-756. WATANABE,K., and S. KUBO. 1982. Mitochondrial adenylate kinase from chicken liver. Eur. J. Biochem. 123:587-592. WILLIAMS,J. G., and P. J. MASON. 1985. Hybridization in the analysis of RNA. Pp. 139-160 in B. D. HAMES and S. J. HIGGINS, eds. Nucleic acid hybridization, a practical approach. IRL, Oxford. WILLIAMSON,R., M. MORRISON, G. LANYON, R. EASON, and J. PAUL. 1971. Properties of mouse globin messenger ribonucleic acid and its preparation in milligram quantities. Biochemistry lo:30 14-302 1. WV, C. 1963. Glutamine synthetase. II. The intracellular localization in rat liver. Biochim. Biophys. Acta 77:482-493. ZARA, V., J. RAssow, E. WACHTER, M. TROPSCHUG, F. PALMIERI, W. NEUPERT, and N. PFANNER. 1991. Biogenesis of the mitochondrial phosphate carrier. Eur. J. Biochem. 198: 405-410.

received

March

3 1, 1992

Regulation of synthesis of glutamine synthetase by adenylylated glutamine synthetase.

Regulation of Glutamine Synthetase Activity.

Discovery of glutamine synthetase cascade.

Immunochemical characterization of glutamine synthetase from Neurospora crassa glutamine auxotrophs.

Photorespiratory nitrogen cycling: evidence for a mitochondrial glutamine synthetase [proceedings].

Ontogeny of glutamine synthetase in rat brain.

Glutamine synthetase interpretation in hepatocellular adenoma.

Reconstruction of glutamine synthetase using computer averaging.

Crystals of glutamine synthetase from Escherichia coli.

Glutamine synthetase: glial localization in brain.

Glutamine synthetase of Chlamydomonas: Rapid reversible deactivation.

Intrahippocampal kainic acid reduces glutamine synthetase.

Streptomyces hygroscopicus has two glutamine synthetase genes.

Effect of glutamine on the degradation of glutamine synthetase in hepatoma tissue-culture cells.

Secondary NAD+ deficiency in the inherited defect of glutamine synthetase.

Genetic control of glutamine synthetase in Klebiella aerogenes.

Glutamine regulates glutamine synthetase expression in skeletal muscle cells in culture.

Intracellular localization of glutamine synthetase in Chlorogloeopsis fritschii.

Neuron-glial interactions involved in the regulation of glutamine synthetase.

Blockade of Glutamine Synthetase Enhances Inflammatory Response in Microglial Cells.

The control of glutamine synthetase level in Lemna minor L.

Reduction of glutamine synthetase mRNA in hypertrophied skeletal muscle.

Glutamine synthetase in the chloroplasts of Vicia faba.

Neural control of glutamine synthetase activity in rat skeletal muscles.