ARCHIVES

OF BIOCHEMISTRY

AND

BIOPHYSICS

Vol. 297, No. 2, September, pp. 285-290, 1992

Primary Structure of Locust Flight Muscle Fatty Acid Binding Protein’ Heather M. Price,* Robert 0. Ryan,* and Norbert H. Haunerlandtl’ *Lipid

and Lipoprotein

252; and TDepartment

Research Group and Department of Biochemistry, University of Alberta, Edmonton, Alberta, Canada T6G Sciences, Simon Fraser University, Burnaby, British Columbia, Canada V5A lS6

of Biological

Received March 30, 1992

The amino acid sequence of the fatty acid binding protein (FABP) from flight muscle of the locust, Schistocerca gregaria, has been determined. The sequence of the Nterminal 39 amino acid residues, determined by automated Edman degradation, was used to prepare a degenerate oligonucleotide that corresponded to amino acid residues 16-23. cDNA coding for FABP was constructed from flight muscle mRNA and amplified by the polymerase chain reaction using the degenerate oligonucleotide and an oligo dT-Not1 primer adapter as primers. The amplification product was cloned and sequenced. Additionally, a cDNA library of flight muscle mRNA was prepared and screened with a 414-bp probe prepared from the clone. The primary structure of locust FABP was compared with the proteins in the Swiss protein databank and found to have significant homology with mammalian FABPs over the entire 133-residue sequence. The best match was versus human heart FABP (41% identity), attesting to the highly conserved nature of this protein. The results suggest that locust muscle FABP is a member of the lipid binding protein superfamily and may provide valuable insight into the evolution of this abundant protein class. 0 1992 Academic Press, 11~2.

Fatty acid binding proteins (FABPs)~ are low molecular weight intracellular proteins believed to be involved in Sequence data from this article have been deposited with the EMBL/ GenBank Data Libraries under Accession No. M95918. i This work was supported in part by grants from the U.S. National Heart Lung and Blood Institute (HL 34786 to R.O.R.), the Natural Science and Engineering Research Council of Canada (OGP0041988 to N.H.H.), and the British Columbia Health Research Foundation (46/ 1990 to N.H.H.). R.O.R. is a Medical Scholar of the Alberta Heritage Foundation for Medical Research. * To whom correspondence should be addressed. 3 Abbreviations used: ALBP, adipocyte lipid binding protein; CRBP, cellular retinol binding protein; FABP, fatty acid binding protein; IFABP, intestinal fatty acid binding protein; L-FABP, liver fatty acid binding protein; M-FABP, muscle fatty acid binding protein; MyP2, myelin P2 protein; PCR, polymerase chain reaction. 0003-9861/92 $5.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form

fatty acid transport and metabolism (1). Since their discovery by Ockner et al. (Z), these proteins have been found in several tissues of numerous vertebrate species. It now appears that there are at least three different classes of FABP: intestinal (I-FABP), hepatic (L-FABP), and cardiac (H-FABP). Sequence comparison between FABPs from different mammalian tissues has revealed that, together with cellular retinol binding protein (CRBP), adipocyte lipid binding protein (ALBP), cellular retinoic acid binding proteins, mammary-derived growth inhibitor, myelin P2 protein (MyPB), and gastrotropin, FABPs comprise a superfamily of lipid binding proteins (3). With the exception of gastrotropin, all members of this superfamily have been shown to bind at least one molecule of long chain fatty acid. While the primary structure appears to be highly conserved in this protein superfamily, it is evident that a greater degree of sequence identity exists between FABPs from the same tissue of different vertebrate species than between FABPs from different tissues within one species. Based on this observation, several authors have proposed that ALBP, H-FABP, and MyPP have diverged from a common ancestor relatively recently, as have CRBP, LFABP, and I-FABP. These two lines are believed to have originated from a progenitor at some point before the vertebrate-invertebrate divergence (3,4). These conclusions are solely based on sequences of mammalian lipid binding proteins, however, while little is known about the more distant evolution of these proteins. During our studies of lipid utilization in the flight muscle of the desert locust, Schistocerca gregaria, we discovered a cytosolic protein that shares biochemical characteristics with mammalian FABPs (5). Locust muscle fatty acid binding protein (M-FABP) is a major protein in adult flight muscle, comprising more than 10% of total cytosolic proteins. Like mammalian FABPs, it is an acidic (PI 5.2), low molecular weight protein (1M, 15,000) which binds a single long chain fatty acid. In an effort to elucidate the relationship between locust M-FABP and mammalian 285

Inc. reserved.

286

PRICE, RYAN, AND HAUNERLAND

members of the lipid binding protein superfamily we have determined the amino acid sequence of locust M-FABP. MATERIALS

AND

METHODS

Znsects and tissue preparation. S. gregaria were raised conditions at 32°C under continuous light. Adults, 4-7 last molt, were used as the source of flight muscle tissue. was dissected, rinsed with ice-cold phosphate-buffered sodium phosphate, 0.154 M NaCl, pH ‘7.4), immediately liquid Na, and stored at -8O’C until further use.

under crowded days after the Flight muscle saline (0.01 M frozen under

Materials. Restriction endonucleases and DNA-modifying enzymes were obtained from Bethesda Research Laboratories (Gaithersburg, MD) and U.S. Biochemical Corp. (Cleveland, OH), which also sold the Sequenase kit and DNA binding protein. [y-32P]ATP, [cy-35S]dATP, [a32P]dATP, and [a-32P]dCTP were obtained from Amersham Corp. (Arlington Heights, IL). PolyATtract mRNA Isolation System, Riboclone cDNA Synthesis Kit, and Taq DNA polymerase were obtained from Promega (Madison, WI). Fastract mRNA isolation kit, TA Cloning kit, PCR 1000 vector, and the pUC/MlS forward primer were obtained from Invitrogen (San Diego, CA), while the reverse sequencing primer came from Boehringer Mannheim (Indianapolis, IN). Zap cDNA synthesis kit and Gigapack Gold packaging extracts were from Stratagene (La Jolla, CA). Centricon 100 microconcentrator columns were from Amicon (Beverly, MA) and Sephacryl S-400 spin columns from Pharmacia (Dorval, P.Q.). All other chemicals used were of the highest grade available. FABP isolation and N-terminal amino acid sequence determination. Locust FABP was isolated from mature adult locusts essentially as described earlier (5), by gel permeation chromatography and isoelectric focusing. Purified FABP was dialyzed against deionized water and lyophilized prior to N-terminal sequence determination. Automated Edman degradation was performed on an Applied Biosystems (Foster City, CA) gas phase sequencer. Probe preparation. Using the N-terminal amino acid information, the possible codons for the mRNA structure were deduced. A mixture of eight distinct 23mers, corresponding to amino acid residues 16 through 23 of the N-terminal sequence, was synthesized in which the thymidine base analog 5-bromodeoxyuridine (B) was substituted in those positions where cytidine or thymidine was indicated: (5’-AAB TTB GAA(G) GAA(G) TAB ATG AAA(G) GC). RNA isolation. Total RNA was isolated from freshly excised or previously frozen flight muscle as described by Chomczynski and Sacchi (6) and polyadenylated RNA was selected with a magnetic separation system. cDNA preparation and polymerase chain reaction. Double-stranded cDNA was prepared from 5 pg of polyadenylated RNA obtained from flight muscle of adult locusts with a commercial kit in which reverse transcription of the mRNA was primed with an oligo dT-Not1 primer adaptor (5’-AATTCGCGGCCGC(T),,). The RNA strand was then replaced with DNA using RNAse H and T4 DNA polymerase. Following phenol extraction and ethanol precipitation the cDNA was subjected to size selection using a Sephacryl S 400 spin column and amplified by the polymerase chain reaction (PCR). Purified double-stranded DNA was chosen for PCR instead of RNA because this leads to fewer nonspecific amplification products. All reactions were performed in the TR96-1 thermal reactor unit (Tyler Research Instruments, Edmonton, AB). Reaction components included 20 ng cDNA, 10 pmol each primer (degenerate 23mer and oligo dT(dATP, Not1 primer adaptor), 0.1 mM deoxynucleoside triphosphates dCTP, dGTP, dTTP), 50 mM KCl, 10 mM Tris*HCl (pH 9.0 at 25’C), 1.5 mM MgCl*, 0.01% gelatin (w/v), 0.1% Triton X-100 and 2 units Taq DNA polymerase. Thermal cycling parameters used were: initial denaturation at 94°C for 2 min prior to addition of enzyme, followed by a low, nonstringent annealing temperature of 45OC for 20 s, ramping slowly (0.25”C/s) to an extension temperature of 75°C (1 min) and

denatured at 94°C (15 s). This cycle was repeated twice, after which the reactions were subjected to an additional 30 cycles at a more stringent annealing temperature (94”C, 20 s; 55”C, 15 s; 75”C, 1 min). Reaction products were analyzed by gel electrophoresis through 1.5% agarose containing ethidium bromide and compared to DNA size markers. Cloning of PCR product. The PCR reactions containing the expected size amplification product (predicted based upon the molecular weight of the protein) were combined and concentrated using the Centricon 100 microconcentrator for removal of low molecular weight products. Concentrated PCR product (50-100 ng) was ligated directly into the PCR 1000 vector, a pUC/M13 derivative containing single 5’ dT overhangs. This feature of the vector takes advantage of the inherent terminal transferase activity of the Taq DNA polymerase which results in addition of single 3’ dA residues on to PCR amplification products (7). The vector also contains the P-Gal gene for blue/white color selection. Competent cells were transformed and grown on plates containing 50 pg/ml kanamycin and 25 pg/ml X-Gal. White transformants selected for further analysis were transferred onto new plates and nitrocellulose replicas were made. The colonies on replica filters of these plates were lysed and bacterial debris was removed by several washings. Membranes were hybridized with a 32P end-labeled probe complementary to the degenerate 23mer used in PCR. Three of the 25 colonies screened hybridized strongly to the probe and were selected for plasmid DNA purification. DNA purification and restriction analysis. Small scale plasmid DNA isolation was performed using the alkaline lysis method (8). Large scale plasmid DNA isolation was performed by alkaline lysis followed by discontinuous cesium chloride density gradient ultracentrifugation in the presence of ethidium bromide. Purified plasmids (1 ,ug) were digested with Not1 and KpnI restriction enzymes to verify that the size of the insert was -500 bp, as expected from the molecular weight of FABP. One of the three positively hybridizing plasmids analyzed by restriction digestion, which had the expected size insert, was sequenced. mRNA used for library construction cDNA library and screening. was isolated from total RNA on oligo dT cellulose using the Fastract mRNA isolation kit. cDNA was synthesized with the ZAP cDNA synthesis kit and ligated into the Uni-Zap XR vector. The vector was packaged using Gigapack Gold packaging extracts. Plaques were screened with an FABP-specific probe generated by PCR. Inclusion of digoxigenin II dUTP in the polymerase chain reaction served to label the probe (9). Purified positive plaques were excised in uiuo into the Bluescript phagemid and DNA was isolated for double-stranded dideoxy sequencing. DNA sequencing. Double-stranded DNA for sequencing was prepared by digestion of plasmid minipreps with ribonuclease A followed by phenol extraction and ethanol precipitation. DNA (4 pg) was denatured with 1 M NaOH and precipitated with ethanol prior to use in the sequencing reaction. Sequencing was carried out using the dideoxy chain termination method of Sanger et al. (10) using Sequenase and 35S-labeled dATP. Primers used were the pUC/M13 forward and reverse sequencing primers. From the sequence derived using the above two primers, two additional specific 19mers were synthesized, one on each strand, for use as sequencing primers. When bands on the sequencing gel were compressed and ambiguous due to the presence of G + C-rich regions, 0.5 fig single-stranded DNA binding protein was included in the reaction mixture. Sequence comparison. Sequence comparison between locust MFABP and other published sequences was determined by searching the Swiss-protein (release 20) and GenPept data bank (release 64.3) with the FASTA search alogorithm of Pearson and Lipman (11). Sequences with optimized initN scores larger than 100 were individually aligned with M-FABP and the ratio of identical to different residues was determined. The match between human H-FABP and M-FABP was graphically displayed using the Macintosh version of the PLFasta program.

RESULTS

AND

DISCUSSION

Isolated M-FABP was subjected to automated Edman degradation. In contrast to most FABPs, locust M-FABP

LOCUST

FATTY

ACID

BINDING

is not blocked at its N-terminus. The amino-terminal 39 residues were determined as H,N-Val-Lys-Glu-Phe-AlaGly-Ile-Lys-Tyr-Lys-Leu-Asp-Ser-Gln-Thr-AsnPhe-Glu-Glu-Tyr-Met-Lys-Ala-Ile-Gly-Val-GlyAla-Ile-Glu-Arg-Lys-Ala-Gly-Leu-Ala-Leu-Ser-Pro-. This sequence was compared with sequences in the Swiss protein databank and found to share significant sequence identity with mammalian FABPs. A degenerate oligonucleotide was synthesized from the N-terminal sequence (residues 16-23) and used in PCR amplification of flight muscle cDNA together with the oligo dT primer adapter originally used in the preparation of the first strand cDNA. To allow for hybridization of the degenerate primer, the PCR conditions included a low, nonstringent, annealing temperature of 45°C in the initial two cycles of amplification that was found to be necessary for product formation. The inclusion of a slow ramp time between the annealing and extension temperatures was also important during this phase of the amplification (12). The stringency of annealing was then increased to 55°C over the ensuing 30 cycles. The resulting 537-nucleotide product was directly cloned into the PCR 1000 vector, enabling elimination of several steps required for conventional cloning techniques. Bacterial colonies harboring the amplification product were screened with the 32P-labeled degenerate probe and positives isolated and sequenced. The sequencing strategy employed is shown in Fig. 1. Initially, nucleotide sequences located within the vector were used as primers. The sequence obtained in this way was used to prepare specific primers for sequencing the internal region of the clone. This sequence, shown in Fig. 2, contains a single open reading frame with a 144-nucleotide 3’-untranslated region. To correct for the possibility of nucleotide incorporation errors in the PCR sequence we prepared a cDNA library of flight muscle and screened this library with a PCRgenerated probe. Two positive clones were sequenced. While neither of these clones was full length (lacking 66

1

133 amino acid residues I 537 nucleotides

1

B C

4 b . . . . . .._...............................................~........................ Q,o. .‘“,I

FIG. 1. Sequencing strategy. The sequence coding for M-FABP is represented by the box, in which the shaded area indicates the N-terminal portion of the protein for which the amino acid sequence has been determined. Arrows show the extent and direction of each sequence determination. A. Sequenced regions of the PCR-generated clone, using oligonucleotide sequences from the vector as primers. B. Sequenced regions of the PCR-generated clone, using the probes shown in Fig. 2. C. Sequenced regions from the cDNA clones.

PROTEIN GTN

AAN

GAN

287

SEQUENCE TTY

GCN

GGN ATN

AAN

TAY

AAN

YTN

GAY

NNN

CAR

ACN

GGC

G’X

ATC

GAG

val-lv~-alu-~he-ala-~-ile-lvs-tvr-lvs-leu-~-ser-aln-thr* MT

TTT

GAG

GAA

TAT

***

l **

l **

***

l **

l **

AT’S

AAG

GCC

ATC

GGT

GTG

asn-~he-a1u-alu-tvr-met-lys-ala-ile-gly--ala-ile-~luCGG AAG GCA GGT '2% GCG CTG TCG CCG GTG ATC GAG CTG GAG An: ara-lvs-ala-aly-leu-ala-leu-ser-Dro-val-ile-glu-leu-glu-ile'2% GAC GGT GA‘! AAG ITC AAG CTC ACC TCC AAG ACT GCC ATC AAG leu-asp-gly-asp-lys-phe-lys-leu-thr-ser-lys-thr-ala-ile-lysAAC ACC GAG TK AK TX AAG CTG GGC GAG GAG TTC GAC GAG GAG asn-thr-glu-phe-thr-phe-lys-leu-gly-glu-glu-phe-asp-glu-gluACC CTG GAC GGC CGC AAG GTC AAG TCC ACC ATC thr-leu-asp-gly-arg-lys-val-lys-ser-thr-ile-thr-gln-asp-gly-

ACT

CAG GAC GSA

CCC MC AAG CTT GTC CAC GAG CAG AAG GCX GAC CAC CCC ACC pro-asn-lys-leu-val-his-glu-gln-lys-gly-asp-his-pro-thr-ile-

ATC

AX An: CGC GAG TX TCC AAG GM CAG TGC GTT ATC ACG ATT AAA ile-ile-arg-glu-phe-ser-lys-glu-gln-cys-val-ile-thr-ile-lysGGC GAC CTG GTG GCA ACG AGA ATA TAC leu-gly-asp-leu-val-ala-thr-arg-ile-tyr-lys-ala-gln-stop

CT-0

GGCACCCGTCCCACCGACTCT'PE~

AAG

GCC

CAG

TM ACG

t*t**tttt*t GCA'PTAAATGTAATTTTCAGTTTACTGT

******** GTGACCCGAAGTTTTTTAAAAGTAATATGTGC~XTGTATCTMTT AAATCATTTlTTACAATPT

FIG. 2. Nucleotide and amino acid sequence. Amino acid residues that have been determined through protein sequence analysis are underlined. The oligonucleotide serving as sequencing primers have been labeled with asterisks. The nucleotide sequence for residues 1 to 20 has been deduced from the amino acid sequence. N, unknown nucleoside; R, purine nucleoside; Y, pyrimidine nucleoside.

nucleotides from the 5’-end), their nucleotide sequence was identical to that of the PCR product in the region corresponding to amino acids 23-133. The results confirmed that the sequence obtained by cloning of the PCR product was correct. The entire amino acid sequence of M-FABP could be determined by combining the amino acid sequence data obtained from direct protein sequencing with that deduced from nucleotide sequencing of the cloned PCR amplification product. The amino acid and nucleotide sequences overlap over a region corresponding to 23 amino acid residues. In this region the nucleotide sequence predicts an amino acid sequence identical to that determined by automated Edman degradation, confirming that the clone obtained is that for M-FABP. Combined, the sequences predict a polypeptide of 133 amino acids with a molecular weight of 14,939 Da, a value identical to that previously reported (5). Within the 3’-untranslated region is a consensus sequence for polyadenylation 20 base pairs upstream from the polyadenylation site. This sequence (ATTAAA), which differs from the common polyadenylation signal AATAAA (13) by substitution of a single nucleotide, has been previously identified for chicken lysozyme mRNA (14) and mouse pancreatic a-amylase (15). The sequence of M-FABP was compared with sequences in the protein data bank. Significant sequence

288

PRICE,

RYAN,

AND

identity (26-41% over the entire sequence) was observed with all members of the lipid binding protein superfamily. The highest degree of sequence homology was found with FABP from human heart, which is identical to the FABP from human skeletal muscle (16) (Fig. 3). In addition, another 41% of the amino acids are conservatively substituted. The high degree of identity persists throughout the entire sequence, as is evident from the matrix plot shown in Fig. 4. Similar sequence homology was also found for H-FABP from other mammalian sources, as well as for MyP2 and ALBP. I-FABP, L-FABP, and CRBP, however, show less sequence identity, indicating a more distant evolutionary relation. It has been proposed that the lipid binding protein superfamily itself can be divided in two families that diverged from a common ancestor gene (3, 4). One family contains L-FABP, I-FABP, and CRBP, while the other is composed of H-FABP, ALBP, and MyP2. By comparing the sequences of FABP from different mammalian sources, it was estimated that these two lines diverged between 900 and 690 million years ago, at some point before the vertebrate-invertebrate divergence ca. 600 million years ago (3). These estimates are relatively crude, since many assumptions with regard to evolutionary clock speed had to be made. Our results represent evidence that the vertebrate-invertebrate divergence indeed occurred after the divergence of the two FABP families. This is particularly evident when the numbers of substitutions per site between each pair of proteins are compared. Table I demonstrates that MFABP is more similar to human H-FABP, ALBP, and MyP2 than to L-FABP, I-FABP, or CRBP. Furthermore, the difference between human members of the first family (H-FABP, ALBP, and MyP2) and human members of the second family (I-FABP, L-FABP, CRBP) is always greater than the difference between M-FABP and human members of the first family. Recently, three other invertebrate FABPs have been identified. Moser et al. (17) found a polypeptide in the

PI-FABP E-PASP nyp2

20 30 40 50 10 VKEFAGIKYKLDSCTNFEBYHKAIGVGAIE RKAGLALSPVIELBILFKLTSKTAIK .DA.L.-TW..VDSK..DD...SL . ..FAT.QVASMTK.'FTII.K-N..ILT.KTSSTF. SNK.L.-'IW..V.SB..D.....L...LAT..L.NLAK.RVIISK-K..IITIRTBSPP. 20 30 40 10 50

M-FASP E-FABP MYP2

SO 90 100 70 110 NTDFTFKLGKBFDEKTLDGSXVKSTITQDGPNKLVS,!QK-GDHFTIII~FS~~VITI ..EIS....V....T.A.D.....IV.L..-G . . ..L..WDGQE.TKV..LIDGKLIL.L ..BIS....Q..E.T.A.N..T... V.L-ARGS.NQV..WNGNE.T.K.KLVDGKH.VEEC SO 60 70 90 100 110

M-FABP H-FASP

*=

60

120 130 KLGDLVATRIYKAQ TA.TA.S..T.EKEA .NK.V.C . . . . EKV 120 130

FIG. 3. Aligned sequence of M-FABP. The amino acid sequence of M-FABP was individually aligned with human H-FABP and bovine myelin P2 protein as described under Materials and Methods. In the sequences of H-FABP and MyP2, a dot indicates that the amino acid is identical to the corresponding amino acid in the M-FABP sequence. The hyphen represents a gap (deletion) as introduced in the alignment process. The numbers above the sequences relate to the amino acids in M-FABP, while the lower numbers belong to both MyP2 and H-FABP.

HAUNERLAND

locust M-FABP FIG. 4. Matrix plot. A matrix plot showing the aligned sequence of locust M-FABP and human H-FABP was drawn with the PLFasta program as described by Pearson (23). The solid line represents optimized local similarity scores larger than 200.

flat worm, Schistosoma mansoni, that is homologous to the FABP gene superfamily. Its sequence shows a similar degree of homology to mammalian H-FABP, MyP2, ALBP, and locust M-FABP, with less to L-FABP, IFABP, and CRBP. This indicates that the flat worm protein also evolved from an early precursor of the H-FABP/ MyP2/ALBP family. Since the phylum Platohelminthes is believed to have branched out of the arthropod line shortly after the vertebrate-invertebrate divergence, it is not surprising that locust M-FABP and the S. mansoni protein possess similar sequence homology to each other and the other members of this FABP family. Two more invertebrate FABPs have been isolated from the midgut of the tobacco hornworm, Munduca sexta, named MFBl and MFB2 (18). Although isolated from an insect species, their sequence homology to all other FABPs is relatively low, including locust M-FABP, a finding that suggests that these proteins may have branched off at a similar time as the major FABP families diverged, well before the vertebrate-invertebrate diversion. Taken together with the biochemical data previously reported (5) our results establish a close relationship between the locust protein and mammalian FABPs. Although their precise physiological function has not been confirmed, FABPs appear to be involved in fatty acid transport and the protection of cells against a high concentration of potentially damaging free fatty acids. The fact that invertebrate species possess closely related FABPs suggests that these proteins may play important roles in the regulation of lipid metabolism throughout the entire animal kingdom. Recently Buelt et al. (19) reported that ALBP, which contains a tyrosine kinase substrate phosphorylation consensus sequence (Glu-Asn-PheAsp-Asp-Tyrlg), is phosphorylated by the soluble kinase domain of the human insulin receptor exclusively at Tyri’.

LOCUST

FATTY

ACID

BINDING TABLE

PROTEIN

289

SEQUENCE

I

Number of Substitutions per Site, Comparing Locust M-FABP with Members of the FABP Superfamily M-FABP M-FABP ALBP MyPX H-FABP CRBP L-FABP I-FABP SMFABP MFBl MFB2

0.61 0.62

0.59 0.73 0.68 0.75 0.64 0.71 0.67

MyP2

H-FABP

0.32 0.37 0.66 0.68

0.38 0.63 0.71

0.67 0.68

0.72

-

0.78 0.64 0.74

0.77 0.58 0.80

0.73 0.58 0.72

0.76 0.72 0.76

0.75 0.71 0.75

0.73

-

0.79

0.73

-

0.70

0.75

0.66

0.75

0.76

0.80

0.73

0.44

CRBP

L-FABP

I-FABP

SMFABP

MFBl

ALBP

-

Since this phosphorylation was affected by fatty acid binding it is possible that fatty acid binding to ALBP is regulated by a tyrosine kinase. Interestingly, M-FABP has a similar sequence (Thr-Asn-Phe-Glu-GluTyr”) and therefore could also be subject to phosphorylation. The three-dimensional structure of all FABPs appears to be similar, as evident from high-resolution Xray crystallographic structure analysis which has been carried out for rat I-FABP (20) as well as bovine MyP2 (21). The overall structure is a barrel of two orthogonal P-sheets that is effectively closed off by a helix-loophelix motif. In MyP2 the first helix is amphiphilic with F16, Y19, M20, and L23 (and G33 from the second helix) pointing into the barrel while the hydrophilic residues D17, E18, and K21 point outward. When the sequence of M-FABP is aligned to the sequence of bovine MyP2, identical or equivalent residues appear in these positions: F17, Y20, M21, and 124, as well as E18, E19, and K22, which would belong to the first helix, and G34, which would belong to the second. The other end of the barrel is closed by a ring of hydrophobic side chains of F4,142, 151, F64, and L91, amino acids that are equivalent to F4, L43, L53, F66, and L94 in locust M-FABP. Most charged amino acid side groups in MyP2 form salt bridges, with the exception of R106 and R126, that are totally buried in the barrel and form, together with Y 128, the binding site for the carboxy group of the fatty acid. These residues are present in M-FABP as R108, R128, and Y130. The remaining part of the binding pocket of MyP2 is composed of V115, 1104, Y19, D76, F16, and M20, residues that are also present in the aligned structure of M-FABP (1117, 1106, Y20, D78, F17, and M21). It appears, therefore, that despite great evolutionary distance the structural elements essential for fatty acid binding by MyP2 have been conserved in M-FABP. Among the most active muscles known, locust flight muscle exclusively utilizes fatty acids during extended flight. The sequence homology between locust muscle

-

FABP and human heart FABP is particularly interesting in light of the suggested functional similarities between these tissues (22); both are nonfibrillar muscles that share high oxidative capacity, large numbers of mitochondria, and the ability to sustain prolonged muscular activity. The biochemical and structural similarities and the evolutionary relationship between locust M-FABP and mammalian H-FABP make insect muscle an attractive model for studying functional aspects of FABP during muscle activity. ACKNOWLEDGMENTS We thank Joan M. Chisholm for technical assistance, Dr. Nollaig Parfrey for helpful discussions, and Sunil Mangal for contributing to the development of this research.

REFERENCES 1. Veerkamp, J. H., Peeters, R. A., and Maatman, Biochim. Biophys. A& 1081, l-24. 2. Ockner,

R. K., Manning, J. A., Poppenhausen, W. K. L. (1979) Science 177,56-58.

R. G. H. J. (1991) R. B., and Ho,

3. Matarese,

V., Stone, R. L., Waggoner, D. W., and Bernlohr, (1989) Prog. Lipid Res. 28, 245-272.

D. A.

4. Chan, L., Wei, C-F., Li, W-H., Yang, C-Y., Ratner, P., Pownall,

Gotto Jr., A. M., and Smith, C. (1985) J. Biol. Chem. 260,

H., 2629-

2632. 5. Haunerland, N. H., and Chisholm, Acta 1047,233-238. 6. Chomczynski,

J. M. (1990) Biochim. Biophys.

P., and Sacchi, N. (1987) Anal. Biochem. 162,

156-

159. 7. Clark, J. M. (1988) Nucleic Acids Res. 16, 9677-9686. 8. Birnboim,

H. C., and Doly, J. (1979) Nucleic Acids Res. 7, 1513-

1523. 9. Lion, T., and Haas, 0. (1990) Anal. Biochem. l&3,335-337. 10. Sanger, F. S., Nicklen, S., and Coulson, A. R. (1977) Proc. NC&. Acad. Sci. USA 74, 5463-5467. 11. Pearson, W. R., and Lipman, D. J. (1988) Proc. Natl. Acad. Sci. USA 85,2444-2448. 12. Compton, T. (1990) in PCR Protocols: A Guide to Methods and Applications (M. A. Innis, D. H. Galfand, J. J. Sninsky, T. J. White, Eds.), pp. 39-45, Academic Press, San Diego.

290

PRICE,

RYAN,

AND

13. Fitzgerald, M., and Shenk, T. (1980) Cell 24, 251-260. 14. Jung, A., Sippel, A. E., Grez, M., and Schutz, G. (1980) Proc. N&l. Acad. Sci. USA 7’7,5759-5763. 15. Hagenbuchle, O., Bovey, R., and Young, R. A. (1976) Cell 21,179187. 16. Peeters, R. A., Veerkamp, J. H., van Kessel, A. G., Kanda, T., and Ono. T. (1991) Biochem. J. 276. 203-207. 17. Moser, D., Tendler, M., Griffith, G., and Klinkert, M.-Q. (1991) J. Biol. Chem. 266,8447-8454. 18. Smith, A. F., Tsuchida, K., Hanneman, E., Suzuki, Teri C., and Wells, M. A. (1992) J. Biol. Chem. 267, 380-384.

HAUNERLAND 19. Buelt, M. K., Shekels, L. L., Jarvis, B. W., and Bernlohr, (1991) J. Biol. C&m. 266, 12,266-12,271.

D. A.

20. Sacchettini, J. C., Gordon, J. I., and Banaszak, L. J. (1989) J. Mol. Biol. 208.327-339. 21. Jones, T. A., Bergfors, J. 7, 1597-1604.

T., Sedzik, J., and Unge, T. (1988) EMBO

22. Crabtree, B., and Newsholme, E. A. (1975) in Insect Muscle (P. N. R. Usherwood, Ed.), pp. 405-491, Academic Press, London. 23. Pearson, W. R. (1990) in Methods in Enzymology (Doolittle, Ed.), Vol. 183, pp. 63-98, Academic Press, San Diego.

R. F.,