Human Liver Nicotinamide N-Methyltransferase

THE J o m m OF BIOLOGICAL CHEMISTRY 0 1994 hy The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 269, No. 20, Issue of May 20, pp. 14635-14840, 1994 Printed in U.S.A.

Human Liver Nicotinamide N-Methyltransferase cDNA CLONING, EXPRESSION, AND BIOCHEMICAL CHARACTERIZATION* (Received for publication, February 1, 1994, and in revised form, March 8, 1994)

Saime Aksoy, Carol L. Szumlanski, and RichardM. WeinshilboumS From the Department of Pharmacology, Mayo Medical School /MayoCliniclMayo Foundation, Rochester, Minnesota55905

Nicotinamide N-methyltransferase (NNMT) catalyzes the N-methylation of nicotinamide and otherpyridines. Human liver NNMT activity has a bimodal frequency distribution, an observation which raises the possibility that this enzyme activity might be regulated by a genetic polymorphism, a polymorphism that could have functional implications for individual differences in drug andxenobiotic toxicity.As a first step toward testing that hypothesis, we set out to clone and express a Human liver NNMT was cDNA for human liver partially purified, photoaffinity-labeled, subjected to limited proteolysis, and partialamino acid sequence information was obtained. The polymerasechain reaction was then used to amplify a 550-nucleotidesequence with human liver cDNA as template and primers designed on the basis of the NNMT amino acid sequence. The5’- and 3’-ends of a human liver NNMT cDNA were obtained by use of the rapid amplification of cDNA ends. The combined use of these approaches resulted in the isolation of a human liver NNMT cDNA that was 969 nucleotides in length, with a 792-nucleotide open reading frame that encoded a 264-amino acid protein with a calculated molecular mass of 29.6 kDa. The humanliver NNMT cDNA was transcribed in vitro and translated with a reticulocyte lysate system to yield a protein with a molecular mass of approximately 29 kDa that comigrated during SDS-polyacrylamide gel electrophoresis with photoaffinity-labeledhuman liver NNMT. The NNMT cDNA was also subcloned into the eukaryotic expression vector p91023(B). COS-1 cells transfected with this construct expressed a high level of NNMT enzymatic activity,and the biochemical properties of this activity were similar to those of human liver NNMT. Human liver NNMT and values for transfected COS-1 cellNNMT had apparent K,,, the two cosubstrates for the reaction, nicotinamide and S-adenosyl-L-methionine,of 0.43 and 0.38 m~ and of 1.8 and 2.2 p ~respectively. , IC, values for the inhibition of NNMT by N’-methylnicotinamide were60 and 30 p for human liver and COS-1 cell-expressed NNMT, respectively. Cloning of a cDNA for human liver NNMT will help make it possible to test the hypothesis that inheritance may play a role in the regulation of individual differences in human liver NNMT activity.

NNMT.

N-methyltransferase (NNMT,’ EC 2.1.1.1) catalyzes the N-methylation of nicotinamide and other pyridines form to pyridiniumions (2). S-Adenosyl-L-methionine (Ado-Met) is the methyl donor for this reaction (3). Wilhelm His first described the methylconjugation of an exogenous compound, pyridine,in 1884 (4), and the enzyme which catalyzes that reaction, NNMT, was characterized by Cantoni in 1951 (5). Many other drugmetabolizing methyltransferase enzymes have also been identified, and variation in the activities of several of those enzymes in humans are controlled by genetic polymorphisms (6-9). “Pharmacogenetic” variation in these enzyme activities can contribute to individual differences in the metabolism, therapeutic effect, and toxicity of drugs that undergo methylation (1).Human liver NNMT activity displays a &fold variation among individuals and has a bimodal frequency distribution, with approximately 25% of samples included in a subgroup with high levels of enzyme activity (10).Those observations raise the possibility that human liver NNMT activity, like the activity of several other methyltransferaseenzymes, might be regulated by a genetic polymorphism. Molecular genetic techniques could be used to study possible the role of inheritance in the regulation of individual differences in NNMT activity in humans. As a step toward making it possible to perform such experiments, we setoutto clone a cDNA for humanliver NNMT. Thestrategy involved purification of humanliver NNMT, obtaining partial aminoacid sequence,and directPCRbased cloning with the useof the rapid amplification of cDNA ends (RACE) (11). Transient expression of the protein encoded by this cDNA demonstrated thatit catalyzed the methylationof nicotinamide and had biochemical propertiessimilar to, or identical with, thoseof human liverNNMT. Cloning of a cDNA for human liverNNMT will make itpossible to studymolecular mechanisms responsible for individual differences of NNMT activity in humans.

MATERIALS AND METHODS Tissue Acquisitionand Preparation-Hepatic tissue obtained during autopsy under guidelines approved by the MayoClinic Institutional Review Board was stored at -80 “C. Prior to enzyme purification, the tissue was homogenized in 5 mM potassium phosphate buffer, pH 7.5, with a Polytron homogenizer (Brinkman Instruments, Inc., Westbury, N Y ) , and a 100,000 x g “high-speed supernatant” was prepared as described previously (10). NNMT and Protein Assays-NNMT activity was measured with the radiochemical enzymaticassay of Rini et al. (IO). This assay is based on Methylation is an important pathway in the biotransformathe conversion of nicotinamide to radioactively labeled N’-methylnicotion of many drugs and xenobiotic compounds (1).Nicotinamide tinamide with [methyl-’4ClAdo-Met (58 pCi/pmol, DuPont NEN) as the methyl donor. One unit of NNMT activity representedthe formation of * This work was supported in part by National Institutes of Health 1 nmol of P-methylnicotinamideihof incubation at 37 “C. Protein conGrants GM 28157 andGM 35720. The costsof publication of this article centrations were measuredby the method of Bradford (12) with bovine were defrayed in part by the payment of page charges. This article must serum albumin as standard. thereforebeherebymarked“aduertisement” in accordancewith18 U.S.C. Section 1734 solely to indicate this fact. The abbreviations used are: NNMT, nicotinamide N-methyltransThe nucleotide sequence(s) reported in this paper hasbeen submitted ferase; Ado-Met, S-adenosyl-L-methionine;RACE, rapid amplification to the G‘enBank“1EMBL Data Bank with accession nunbeds) U08021. of cDNA ends; PCR,polymerasechain reaction; ORF,open reading $To whom all correspondence and reprint requests shouldbe ad- frame; UTR, untranslated region;MOPAC,mixedoligonucleotide dressed: Tel.: 507-284-2246;Fax: 507-284-9111. primed amplification of cDNA.

14835

Human Liver NNMT cDNA

14836

TABLEI Primers andprobes used in the cloning of human liver NNMT cDNA Restriction enzyme recognition sites incorporated into primer sequences are indicatedby lines below the appropriate sequences. The abbreviations usedinclude: MOPAC, mixed oligonucleotide primed amplificationof cDNA,RACE, rapid amplificationof cDNA ends; andORF, open reading frame. Designation

MOPAC primers MOPACl MOPAC2 MOPAC3 Screening primers NNMTl NNMT2 5'-RACE primers 5'-RACE1 5'-RACE2 5'-Anchor adapter

Sequence

5'-ATGGA(A/G)(A/T)(G/C)(C/T)GG(A/G/C)TT(C/T)AC-3' 5'-TT(A/G)TT(A/G/C)AG(A/G)GC(A/G)CG(C/T)C(G/T)(A/G)TA(A/G)TA-3' 5'-CC(C/T)CT(G/C/T)CC(C/T)CC(C/T)GC(C/T)GA(C/T)TG(C/T)GT-3' 5'-GGGCAGGCGGCATCCAGACA-3' 5"AGTGTGCTGAGCACGCAGTCAGC-3'

5'-CCTTCTCTGGACCCTTGACTCTG-3' 5'-GTAGTCAGTGACGACGATCTCCTTAAAGGATTC-3' 5'-CACGAATTCACTATCGATTCTGGAACCTTCAGAGG-3' EcoRI site

5"Anchor primer 3'-RACE primers 3'-RACE1 3'-RACE2 3'-Anchor-d(T),,

5"CTGGTTCGGCCCACCTCTGAAGGTTCCAGAATCGATAG-3' 5"GACATCGGCTCTGCCCCCACTATC-3' 5"GAGCCAGAGGCCTTTGACTGGTCCC-3'

5"AACTGGAAGAATTC EcoRI site

ORF amplification primer 5'-ORF

GCGGCCGCAGGAA( T)18-3' Not1 site

5"TGTCGAATTCGTGCTCCAGTGGTACAGAAGTGAG-3' EcoRI site

Oligonucleotide primers for the PCR were designed on the basis of amino acid sequence information obtained after limited proteolysis of human liver NNMT. The amino acid sequences MESGFT and PLPPADCV wereusedtodesignsenseprimers that weredesignated MOPACl (551 bp) MSACB + MOPACland MOPAC3 (Table I). Theaminoacidsequence Y(Y/ 1 c C)RAALN was used to designan antisense primer, MOPAC2 (Table I). MOPACP The degree of degeneracy of the primers was decreased by use of a MoPAC' (513 bp) + human codon usage table (19). When the PCR reaction was performed 2 tt with MOPACl and MOPAC2 as primers, and human liver cDNA as NNMTZ NNMT1 template, the amplification product was approximately 550 nucleotides in length (Fig.1).This PCR product was usedas template for a nested 3 3 PCR performed with primers MOPAC2 and MOPAC3 (Table I, Fig. 1). S'ANCHOR The new amplification product was 92 nucleotidesin length. Two nonPRIMER + 141.1 I._- hn\ degenerate oligonucleotides, designated NNMTl andNNMT2 (Table I), 4 were then designed on the basis of the sequence of the 92 nucleotide % $O R f 5'RACEZ ADAPTER PCR amplification product. A 513-nucleotide amplification product was obtained whenthe PCR SORF + (914 bp) 5 5 was performed with MOPACl and NNMTl as primers and human liver f B'ANCHOR-d(T)la cDNA as template (Fig. 1). Sequencing of this product made it possible 3'-RACE1, 3'-RACE2, 5'-RACE1, and 5'-RACE2 FIG.1. Cloning strategy forhuman liver NNMT cDNA. The full- todesignprimers length NNMT cDNAis depictedat the topof the diagram. The box in the (Table I). To obtain the 3'-end of the human liver NNMT cDNA, a diagram at the top represents the ORF, whereas solid lines represent 3'-RACE protocol was used (11).First-strand cDNA synthesis was perthe 5'- and 3-UTRs. The lines designated 1 through 5 represent sequen- formed with1pg of human liver poly(A)' RNA (Clontech, PaloAlto, CA) tial PCR reactions used in thecloning of a human liver NNMT cDNA. and a 3'-Anchor-d(T),, primer (Table I) that hadbeen provided as part See text for details. of a first-strand cDNA synthesis kit (Pharmacia LKB Biotechnology Inc.). An anchored PCR reaction was then performed with this firststrand cDNA as template and 3'-RACE1 and3'-Anchor-d(T),, as primNNMT Purification and Photoafinity Labeling-NNMT was partially purified from human liver high-speed supernatant with a modi- ers. The amplification product obtained was approximately 670nuclein template turn, for a nested fication of the method described by Van Loon and Weinshilboum (131. otides in length. This product was used, as The purification procedureinvolved DEAE-agarose ion exchange chro- PCR reaction performed with 3'-RACE2 and 3'-Anchor-d(T),, as primmatography, followed by Sephadex G-100 SF gel filtration chromatog- ers (Table I, Fig. 1).The product of this reaction was sequenced by raphy. Photoafinity labelingof the partially purified enzyme was per- automated DNA sequencing. 5'-RACE was then used t o obtain the formed as described by Van Loon et al. (14)with [meth~l-~HIAdo-Met 5'-end of the cDNA (11). The 5'-AmpliFINDER RACE Kit (Clontech) (75.3 pmol, 5.49 pCi) and approximately 10 pg of protein of the partially was used to synthesize first-strand cDNA from human liver poly(A)+ RNA with 5'-RACE1 as a primer, followed by ligation of the 5"Anchor purified human liver NNMT preparation. adapter (TableI, Fig. 1). PCR performed with thiscDNA as a template Limited Proteolysis and Amino Acid Sequencing-NNMT that had been partially purified and identified by photoafinity labeling was sub- and with 5'-RACE2 and 5"Anchor primer as primers (Table I, Fig. 1) yielded a n amplification product 413 nucleotides in length. This product jected to limited proteolysis. Enzymatic proteolysis was performed by the method of Cleveland et al. (15)with subtilisin and papain. Cleavage was also sequenced with the automated DNA sequencer. In Vitro Dunslation and COS-1 Expression-PCR amplification of was also performed with cyanogen bromide (16). Amino acid sequencing of peptide fragments electroblotted onto polyvinylidene difluoride mem- the NNMT cDNA open reading frame (ORF) and 3'-untranslated region and 3'-Anchor-d(T),, as primers branes was performed in the Mayo Research Resource Protein Coreas (UTR) was performed with 5'-ORF (Table I, Fig. 1). This PCR product was subcloned into the EcoRI-Not1 described elsewhere (17). sites of pBluescript SK(+) (Stratagene, La Jolla, CA) and theEcoRI site Molecular Cloning of cDNA-The procedure referred to as "mixed oligonucleotide primed amplificationof cDNA" (MOPAC)was originally of the eukaryotic expressionvector p91023(B) (20,21). The NNMT-pBluescript construct was usedfor in vitro transcription developed to amplifyspecific DNA sequences by using degenerate DNA Coupled Reticulocyte Lysate primers designed on the basis of a known amino acid sequence (18). and translation performed with the TnTTM

Human Liver NNMT cDNA -117

TGAACTCTGGATGCTGTTAGCCTGAGACTCAGGAAGACAACTTCTGCAGGGTCACTCCCT

-57

GGCTTCTGGAGGAAAGAGAAGGAGGGCAGTGCTCCAGTGCTCCAGTGGTACAGAAGTGAGACATAATG M

4 64

FIG.2. Human liver NNMT cDNA nucleotide and deduced amino acid sequences. Nucleotides are numbered in the 5’ to 3’ direction with the “A”in the translation initiation codon designatedas +l. Underlined areas in the deduced amino acid sequencerepresent sequences obtained by partial sequencing of human liver NNMT. The polyadenylation signal sequence ATTAAA is also underlined.

14837

GAATCAGGCTTCACCTCCAAGGACACCTATCTAAGCCATTTTAACCCTCGGGATTACCTA E S G F T S K D T Y L S H F N P R D Y L GAAAAATATTACAAGTTTGGTTCTAGGCACTCTGCACTCTGCAG~GCCAGATTCTTAAGCACCTT E K Y Y K F G S R H S A E S Q I L K H L

124

CTGAAAAATCTTTTCAAGATATTCTGCCTAGACGGTGTG~GGGAGACCTGCTGATTGAC L K N L F K I F C L D G V K G D L L I D

184

ATCGGCTCTGGCCCCACTATCTATCAGCTCCTCTCTGCTTGTGAATCCTTTAAGGAGATC I G S G P T I Y Q L L S A C E S F K E I

244

GTCGTCACTGACTACTCAGACCAGAACCTGCAGGAGCTGCAGGA~TGGAGAAGTGGCTGAAGAAAGAG V V T D Y S D Q N L Q E L E K W L K K E

304

CCAGAGGCACTCTGCCTTTGACTGGTCCCCAGTGGTGACCTATGTGTCAGA

P

E

A

F

D

W

S

P

V

V

T

Y

V

C

D

L

E

G

N

R

364

GTCAAGGGTCCAGAGAAGGAGGAGAAGTTGAGACAGGCACTCTGCGGTCAAGCAGGTGCTGAAGTGT V K G P E K E E K L R Q A V K Q V L K C

424

GATGTGACTCAGAGCCAGCCACTGGGGGCCGTCCGTCCCCTTACCCCCGGCTGACTGCGTGCTC D V T Q S Q P L G A V P L P P A D C V L

484

AGCACACTGTGTCTGGATGCCGCCTGCCCAGACCTCCCCACCTACTGCAGGGCGCTCAGG S T L C L D A A C P D L P T Y C R A L R

544

AACCTCGGCAGCCTACTGAAGCCAGGGGGCTTCCTGGTGATCAT~ATGCGCTCAAGAGC N L G S L L K P G G F L V I M D A L K S

604

AGCTACTACATGATTGGTGAGCAGAAGTTCTCCAGCCTCCCCCTGGGCCGGGAGGCAGTA S Y Y M I G E Q K F S S L P L Q R E A V

664

GAGGCTGCTGTGAAAGAGGCTGGCTACACAATCG~TGGTTTGAGGTGATCTCGCAAAGT E A A V K E A Q Y T I E W F E V I S Q S

724

TATTCTTCCACCATGGCCAACAACG~GGACTTTTCTCCCTGGT~CGAGGAAGCTGAGC Y S S T Y A N N E G L F S L V A R K L S

784

AGACCCCTGTGATGCCTGTGACCTCAATTAAAGCAATTCCTTTGACCTGTCW

R 844

P

L

*

AAAAAAAAA

System (Promega, Madison, WI)(22). Products of the reaction were tein was 29 kDa as estimated by SDS-PAGE. analyzed by SDS-polyacrylamide gel electrophoresis (PAGE)and autoCleavage of NNMT was performed with papain, subtilisin, radiography.TheNNMT-p91023(B) construct wasusedto transfect and cyanogen bromide.Amino acid sequencewas then obtained COS-1 cells using the DEAE-dextran method (23, 24) as described by by automated Edman degradationperformed with 24- and 29Honchel et al. (17). kDa cyanogen bromide, 8-kDa papain, and7- and 15-kDa subNorthern Blot Analysis-A human multiple tissue Northern blot (Clontech) was probed with the NNMT cDNA ORF plus 3’-UTR that tilisin peptide fragments.Amino acid sequences obtainedfrom had been radioactively labeled by random priming with CCI-~~PI~CTP these peptides were ESGFTSKDXY for both the 24- and 29(25). Hybridizationwas performed overnight at 42 “C in 5 x SSPE, 50% kDa cyanogen bromidefragments, AWLPPADCVLSTLCLDAA freshly deionized formamide, 10 x Denhardt’s solution, 2% SDS, and for the papain fragment, LDAACPDL(L/P)TY(Y/C)RAA100 pg/ml sonicated salmon sperm DNA. The blot was then washed at (?)L(?)NN(?)G for the 7-kDa subtilisinfragmentand (D/ room temperature in 0.2 x SSC and 0.1% SDS. Southern Blot Analysis-Aliquots of 5 pg of human lymphocyte G)TYLSHFN for the 15-kDa subtilisin fragment. Residues 9 genomic DNA (Promega)were digested with excess amounts of BgZII, and 12 in the 7-kDa subtilisin fragment and residue 1 in the KpnI, S a d , and XbaI, electrophoresed ina 0.8% agarose gel, and trans- 15-kDa subtilisin fragmentcould have been either of the amino ferred to a nylon membrane. The hybridization probe was the same as acids indicated; residues15, 16, and 18 in the 7-kDa subtilisin that used for the Northern blot analysis. Hybridization was performed overnight at 65 “C in 1%SDS, 1 M NaCl, 10% dextran sulfate, 50 mM fragment were ambiguous (?); and residue 9 in the cyanogen bromide fragment wasunresolved ( X ) .Assuming no repeats in Tris, pH 8.0, and 100 pg/ml sonicated salmon sperm DNA. The membrane was washed once at room temperature in 3 x SSC and twice at the amino acid sequence, the papain and the7-kDa subtilisin 65 “C in 3 x SSC plus 1%SDS. fragments contained the overlappingsequence LDAA (underData Analysis-TheUniversity of WisconsinGeneticsComputer lined in the preceding sequences). Combining information from Group (GCG) software package (26) was used to analyze sequenceinthese two fragments yielded a sequence that was 34 amino formation.IC,, values were calculated with the GraphPAD Inplot curvefitting program (GraphPADInPlot Software, San Diego, CAI. Apparent acids in total length. Cloning of Human Liver NNMT cDNA-Cloning of a cDNA K , values were calculated by the method of Wilkinson (27) with a computer program writtenby Cleland (28). for human liverNNMT was performed with a direct PCR-based strategy as described in detail under “Materials and Methods” RESULTS (Table I, Fig. 1). Thefull-length NNMT cDNA consisted of 969 Protein Purification, Photoafinity Labeling, Limited Prote- nucleotides with a 792-nucleotide ORFthat encoded 264 amino olysis, andPartial Amino Acid Sequencing-Human liver acids (Fig.2). There were no additional in- or out-of-frame NNMT was partially purified by sequential ion exchange and initiation or termination codons in close proximity to either the gel filtration chromatography. Afterthese two chromatographic start or stopcodons shown inFig. 2. The sequence environment steps, the enzyme was purified approximately 150-fold when of the ATG initiation codon closely approximated the optimal compared with a 100,000 x g hepatic supernatant preparation. sequence for translation initiation in higher eukaryotes (29). The final step in the purification utilized SDS-PAGE, with The 3’-UTR ended with a poly(A) tract, and the polyadenylaidentification of NNMT by photoaffhity labeling performed tion signal ATTAAA (30) was located 19 nucleotides upstream with [rnethyl-3HlAdo-Met, the methyl donor for the enzymatic from the poly(A) tract (Fig. 2). reaction. The molecular mass of the photoaffinity labeled proI n Vitro Danslation-The human liver NNMT cDNA was


14838 Human Human Liver NNMT :DNLi

'

HUMAN NNMT

Liver NNMT 3H-Ado-Met

kDa 66.2

SUBSTRATE KINETICS AND INHIBITION CURVE

-

A

k O.Wl2

42.7

31 .O

I

-3

-2

-1

I

l/[NlCOTINAMIDE], mM-'

21.5

B 14.4 L a 3 hr 40 hr FIG.3. Human liver NNMT cDNA translation in a reticulocyte lysate system.SDS-PAGE of [35Slmethionine-labeled protein obtained by translation of human liver NNMT cDNA is shown on the left. The cDNA encoded a protein with an apparentmolecular mass of approximately 29 kDa. SDS-PAGE of purified human liverNNMT photoaffinity labeled with [rnethyL3H1Ado-Metis shown on the right for comparison.

transcribed and translated in vitro as described in the "Material and Methods." The translation product had an apparent molecular mass of approximately 29 kDa as estimated by SDSPAGE, and this proteincomigrated with photoaffinity labeled human liver NNMT (Fig. 3). The molecular mass of human liver NNMT calculated on the basisof the aminoacid sequence deduced from the sequence of the cDNA was 29.6 kDa. Expression of NNMT cDNA-COS-1 cells were transfected with the human liver NNMT cDNAin bothsense and antisense orientations in the eukaryoticexpression vector p91023(B) as well as with p91023(B) that lacked insert (20, 21). NNMT activity was then measured inhigh-speed supernatant prepared from these cells under optimal conditions for the assay of human liver NNMT activity (10). COS-1 cell preparations transfected with the human liver cDNA clone in the sense orientation were capableof catalyzing themethylation of nicotinamide (8700 units/mg of protein), but there wasno detectable NNMT activity in cells transfected withvector that contained insert in the antisense orientation, cells transfected withvector alone, or cells exposed only to the transfectionbuffer. Biochemical Properties of Expressed NNMT-Substrate kinetic andenzyme inhibition experiments were performed with both human liver NNMT and preparations from COS-1 cells transfected with human liver NNMT cDNA. When a total of 10 different concentrations of nicotinamide that rangedfrom 0.02 to 10 mM were tested, apparent K, values of human liver and COS-1 cell expressed NNMT for nicotinamide were 0.43 and 0.38 mM, respectively. After exposure to six different Ado-Met , K, concentrations that ranged from 0.313 to 10 p ~ apparent values of human liver and COS-1 cell expressed NNMT for , These apparentK, Ado-Met were 1.8 and 2.2 p ~respectively. values were very similar to those reported previously for human liver NNMT (10). Double inverse plots of a portion of the data for both substrates are depicted graphically in Fig. 4, A and B. Human liver and COS-1 cell expressed NNMT were also used to estimateIC,, values for inhibition by N'-methylnicotinamide, a potent inhibitor of NNMT (10). IC,, values were calculated on the basis of data obtained with 10 different concentrations of N'-methylnicotinamide that ranged from 0.008 to 2 mM. Those IC,, values were 60 and 30 1.1~for human liver and COS-1expressed NNMT, respectively (Fig. 4C), values

I

-1

4.5

1

0

2

l/IAdoMet], pM1

I

-5.0

-4.5

4.0

-3.5

-3.0

-2.5

LOO ~"M€rHYLNICOllNAMIDEJ, M

1 0 0

Human Uver NNMT COS1 Expressed NNMT

FIG.4. Human liver NNMT cDNA expression in COS-1 cells. Double inverse plots of the relationship between nicotinamide( A )and Ado-Met ( B ) concentrationsand NNMT activity for COS-1 cell expressed NNMT a s well as partially purified human liver enzyme are shown. N'-Methylnicotinamide inhibition of the two enzyme preparations is shown in C.

that were also similar to that reportedpreviously for the inhibition of human liver NNMT by N1-methylnicotinamide (10). Northern and Southern Blot Analyses-Northern blot analysis of mRNA from eight different human tissues was performed with a probe that consisted of the NNMT cDNA ORF plus 3'-UTR. A single mRNA species approximately 1kilobase in length, very similar to the size of the human liver NNMT cDNA, was detected in liver (Fig. 5). When the autoradiogram was exposed for a longer time, a similar sized transcript was also present in heart, placenta, lung, skeletal muscle, and kidney, but no transcript was seen in brain or pancreas (Fig. 5). Southern blot analysis was performed with human lymphocyte genomic DNA that had been digested with four different restriction enzymes (Fig. 6). Restrictionsites for these enzymes were not present in the sequence of the human liver NNMT cDNA. The probe was the same as that used to perform Northern blot analysis. The resultsobtained with the Southernblot were compatible with the conclusion that only a single or very few genes for NNMT are present in the human genome. The patterns shown in these Southernblots might also be useful in


14839 hNNMT

mTEMT hPNMT rPNMT

‘I 2.4

MNMT bPNMT

FIG.7. Relationships amongthe sequences of human ( h ) NNMT, mouse ( m ) thioether S-methyltransferase, and phenylethanolamine N-methyltransferase from four different species (human ( h ) ,rat ( r ) ,mouse ( m ) ,and bovine ( b ) )are shown. Amino acid sequences were clustered by use of the PILEUP program (39)from the GCG package (Universityof Wisconsin).

1.35

48 hr .,.”

exposure 1

and bovine phenylethanolamine N-methyltransferase, respectively (31-35). The dendrogram in Fig. 7 shows the relationships among the sequences of this group of methyltransferase enzymes.

9 h r exposure DISCUSSION FIG.5. Northern blot of human NNMT. A multiple tissue Northern Methylation is an important pathway in thebiotransformablot (2 pg of poly(A)+RNMane, Clontech) was probed with the human tion of many drugs andxenobiotic compounds (1).NNMT cataliver NNMT cDNA ORF and 3’-UTR as a probe. Comparable loadingof RNA in eachlane was verified by reprobingthe blot with a cyclophyllin lyzes the N-methylation of nicotinamide and otherpyridines to probe (data not shown). form pyridinium ions (2). Several N-methylpyridinium compounds are toxic. Examples includethe herbicide paraquat (36) and the neurotoxin 1-methyl-4-phenylpyridiniumion, a metabolite of l-methyl-4-phenyl-l,2,3,6-tetrahydropyridine (37, kb 38).Therefore, the possibility exists thatselected pyridine substrates for NNMT might function as “protoxicants.” The obser21.2 vation of large individual variations in human hepatic NNMT activity could have implications for individual differences in xenobiotic and drug toxicity. Furthermore, the existence of a subgroup of subjects with high levels of hepatic NNMT activity raises thepossibility of a genetic polymorphism for this enzyme activity (10). Pharmacogenetic studies of other cytosolic methyltransferases in humans have demonstrated that the activities of several of those enzymes are controlled by genetic poly5.0morphisms (1,6-9). I t would be important to clone a cDNA for the enzyme to 4.3 pursue molecular genetic mechanisms involved in the regulation of NNMT activity in humans.Therefore, we set outto clone 3.5a cDNA for NNMT from human liver as a first step in studies of molecular mechanisms responsible for individual variation in NNMT activity. The cloning strategy involved purification of human liver NNMT, obtaining partial amino acid sequence of the enzyme after limited proteolysis, followed by direct PCR2.0based cloning performed with RACE protocols (11).Translation of the NNMT cDNA clone in a reticulocyte lysate system FIG.6. Southern blot of human N NMT.Five pg of human lympho- yielded a protein with a molecular mass of approximately 29 cyte genomic DNA was exhaustively digested with BglII, Sad, KpnI, and Xbd; separated by electrophoresis on a 0.8% agarose gel; trans- kDa. When the human liver cDNA clone wasexpressedin ferred to a Micron Separations, Inc. (Westboro, M A ) nylon membrane, COS-1 cells, the expressed enzyme had biochemical properties and probed with the human liver NNMT cDNA ORF plus 3’-UTR. similar to, or identical with, those of human liver NNMT. Northern blot analysis showed that mRNA for NNMT was prefuture attempts to detect restriction fragment length polymor- sent predominantly in liver, but detectable transcripts were phisms for NNMT in humans. also present in kidney, lung, skeletal muscle, placenta, and Amino Acid Sequence Homology-The nucleotide sequence heart. Theamino acid sequence encoded by human liver NNMT within the ORF and the deduced amino acid sequence of the cDNA showed homology with the sequence of mouse thioether protein encoded by the human liver NNMT cDNA clone were S-methyltransferase andwith those of phenylethanolamine Ncompared with sequences in the GenBank Genetics Sequence methyltransferase from several species. Theserelationships Data Bank, the EMBL Nucleotide Sequence Data Base, and the are of interest because, unlike many other “families” of funcSwiss-Prot Protein Sequence Data Base. The sequence of the tionally related enzymes,with the exception of orthologues protein encoded by the human liver NNMT cDNA was 52, 37, across species lines, little homology has been reported among 39, 38 and 39% identical with the amino acid sequences of mammalian cytoplasmic methyltransferase enzymes (17). The mouse thioether S-methyltransferase and human, rat, mouse, cloning and expression of a cDNA for NNMT from human liver

-

-


14840

represents a potentially important step toward determining the molecular basis for the regulation of individual differences in levels of this enzyme activity in humans and, ultimately, a determination of the potential functional consequences of those differences. Acknowledgments-We thank Luanne Wussowfor assistance with the preparation of this manuscript, Dr. Ibrahim A. Aksoy and Diane Otterness for their advice and assistance, and Dr. Eric Wiebenfor reading the manuscript. We also thank Dr. Daniel McCormick and Benjamin Madden, Mayo Research Resource Protein Core, for performing amino acid sequencing of the peptide fragments. REFERENCES 1. 2. 3. 4. 5. 6.

7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

Weinshilboum, R. (1989) Pharmucol. Ther 43,77-90 DSouza, J., Caldwell, J., and Smith, R. L. (1980)Xenobiotica 10, 151-157 Cantoni, G. L. (1951)J. Biol. Chem. 189, 745-754 His, W. (1884) Arch. Exp. Pathol. Pharmakol. 22, 253-260 Cantoni, G. L. (1951)J . Biol. Chem. 189, 203-216 Weinshilboum, R. M., and Raymond, F.A. (1977) Am. J. Hum. Genet. 29, 125-135 Weinshilboum, R. M., and Sladek,S . L. (1980)Am.J. Hum. Genet. 32,651-662 Price, R. A,, Keith, R. A,, Spielman, R. S., and Weinshilboum, R. M. (1989) Genet. Epidemiol. 6, 6 5 1 4 6 2 Price, R. A,, Scott,M. C., and Weinshilboum, R. M. (1993)Genet. Epidemiol. 10, 123-131 Rini, J., Szumlanski,C., Guerciolini, R., and Weinshilboum, R. M. (1989) Clin. Chim. Acta 186,359-374 Frohman, M. A,, Dush, M. K., and Martin, G. R. (1988)Proc. Natl. Acad. Sci. U. S. A. 86,8998-9002 Bradford, M. M. (1976) Anal. Biochem. 72,248-254 Van Loon, J. A,, and Weinshilboum, R.M. (1990) Drug Metab. Dispos. 18, 632438 Van Loon, J.A., Szumlanski, C. L., and Weinsbilboum, R. M. (1992) Biochem. Pharmucol. 44, 775-785 Cleveland, D. W., Fischer, S. G., Kirschner, M. W., and Laemmli, U. K. (1977) J . Biol. Chem. 262, 1102-1106 Jahnen, W., Ward, L. D., Reid, G. E., Moritz, R. L., and Simpson,R. J. (1990)

Biochem. Biophys. Res. Commun. 166, 139-145 17. Honchel, R.,Aksoy, I., Szumlanski, C., Wood, T.C., Otterness, D. M., Wieben, E. D., and Weinshilboum, R. M. (1993)Mol. Pharmucol. 43,878-887 18. Lee, C. C., Wu, X., Gibbs, R. A,, Cook, R. G., Muzny, D. M., and Caskey, C. T. (1988)Science 239, 1288-1291 19. Wada, K., Aota, S., Tmchiya, R., Ishibashi, F., Gojobori, T., and Ikemura, T. (1990) Nucleic AcidsRes. 18, (suppl.) 2367-2411 20. Wong, G. G., Witek, J. S., Temple, P. A,, Wilkens, K. M., Leary, A. C., Luxemberg, D. P., Jones, S. S., Brown, E. L., Kay, R. M., Orr, E. C., Shoemaker, C., Golde, D. W., Kaufman, R. J., Heurick, R. M., Wang, E. A,, and Clark, S. C. (1985)Science 228,810-815 21. Kaufman, R. J. (1985)Proc. Natl. Acad. Sei. U.S. A. 82, 689-693 22. Pelham, H. R. B., and Jackson, R. J. (1976)Eur J . Biochem. 67, 247-256 23. McCutchan, J. H., and Pagano, J. S. (1968)J. Natl. Cancer Inst. 41, 351-357 24. Kaufman, R.J., Davies, M. V., Pathak, V. K., and Hershey, J . W. B. (1989)Mol. Cell. Bid. 9, 9 4 6 9 5 8 25. Feinberg, A. P., and Vogelstein, B. (1983) Anal. Biochem.1 3 2 , 6 1 3 26. Devereux, J., Haeberli, P., and Smithies, 0.(1984) Nucleic Acids Res. 12, 387-395 27. Wilkinson, G. N. (1961) Biochem.J. 80,324-332 28. Cleland, W.W. (1963) Nature 198,463-465 29. Kozak, M. (1991)J. Biol. Chem. 266, 19867-19870 30. Wahle, E., and Keller, W. (1992)Annu. Rev. Biochem. 61,419-440 31. Warner, D. R. (1992) Cloning and Active Site Labeling of Thioether S-Methyltransferase. Ph.D. dissertation, Department of Biochemistry, University of Louisville 32. Kaneda, N., Ichinose, H., Kobayashi, K., Oka, K., Kishi, F., Nakazawa, A,, Kurosawa, Y., Fujita, K., and Nagatsu,T.(1988)J. Biol. Chem. 263, 76727677 33. Weisberg, E.P., Baruchin, A,, Stachowiak, M. K., Stricker, E. M., Zigmond, M. J., and Kaplan, B. B. (1989)Mol. Brain Res. 6, 159-166 34. Morita, S., Kobayashi, K., Hidaka, H., and Nagatsu, T. (1992)Mol. Brain Res. 13,313-319 35. Baetge, E. E., Suh, Y. H., and Joh, T. H. (1986)Proc. Natl. Acad. Sci. U.S. A. 83,5454-5458 36. Haley, T.J . (1979) Clin.Toxicol. 14, 1-46 37. Langston, J. W. (1985) Dends Neurosci. 8, 79-83 38. Singer, T. P., Castagnoi, N., Jr., Ramsay, R. R., and Trevor, A. J. (1987) J. Neurochem. 49, 1-8 39. Feng, D.-F., and Doolittle, R. E (1987)J. Mol. Euol. 25, 351-360