Mammalian l-amino acid decarboxylases producing 1,4-diamines: analogies among differences

PROTEINSEQUENCEMOTIF Mammalian L-aminoacid decarboxylasesproducing 1,4-diamines: analogies amongdifferences Mammalian ornithine decarboxylase (ODC, EC 4.1.1.17), histidine decarboxylase (HDC, EC 4.1.1.22) and aromatic-L-amino-acid decarboxylase (DDC, EC 4.1.1.28) are homodimeric pyridoxal-phosphate (PLP)-dependent enzymes that produce the physiological 1,4-diamines: putrescine, histamine and serotonin (and tryptamine), respectivelyL Mammalian HDCs and DDCs exhibit extensive homology and have some common immunological and catalytic properties; however, there is no obvious

TIBS 1 9 - A U G U S T 1 9 9 4

homology between these two enzymes and mammalian ODCsz-4. Mammalian ODC has o n e of the shortest half-lifes described for mammalian proteinsS.6; this is explained by the presence of clusters rich in proline, aspartate, glutamate, serine and threonine (PEST re~ions) and, more specifically, the PEST region located in the carboxyl terminus of the protein between amino acid residues 423 and 449 (in mouse ODC) 6,7. Mammalian ODC turnover is regulated by polyamines; different domains confer constitutive degradation and polyamine responsiveness to this protein 7. Histamine, serotonin and other structurally related 1,4-diamines can behave like polyamine analogs: they regulate mammalian ODC, intracellular polyamine levels, and cell growth in vivo and in vitro 8. HDC, like ODC, is a very

N termini

(a)

C termini 47

HDC

70

505

R. norvegicus

P16453 --

74

63

HDC H. sapiens

P19113~[~LPES~EDPDS@SZFOD'~'~

DDC B. taurus

P27718--1RPLIPTT&PQEPETFEATIEDIEK[

39

DDC H, sapiens

,480

62

48O

57 480 62

'480

6Z

P2071I'---I~LZP~POEPDTF~-DZZ~VF.Xl N termini

562

IN.ETLLDPVDDCFSEE~DATX 1-56z

lETSEPD~DESCLNM~]

DDC C. porcellus P22781---'I~LIPSS~PEEPETYEDIIODIEK I 39

555

541

184

DDC R. norvegicus P14173~[I~,LIPTTM'OEPET¥£DIII~J 39

66 I 529

I KDLTI~LSLESVNEC~DDPVOVILI-Ip,LPGDSLETTMDPFDDCF$EEASDTTE[-- 656 170

P80041~] RPLIPATAPQEPDTFEDILQDVEKI 39

525

6Z

39

DDC S. scrofa

503

II~,QIPSS~EEPDSW~SIFGDIEOZTr~GVVHI 40

523

I~L SLEsv SEC~DDP&Q&P,I

M. muscu~us P23738~[I~J~LI'ASAeEEPDSC~1)SZFeDZER[ 43

HDC

unstable protein 1, and there is evidence for a coordinate regulation of ODC and HDC in several mammalian cel! ,ines and tissues 9,~°.After consideratior of all these data, some common structu al features might be expected between .hese enzymes. By using a computer prJgram developed from the algo;'ithm described by Rogers et al. 6, the deduced primary translation product, of the known ,i,mmalian mRNAs encoding HDC and DDC were analysed for the existence of PEST regions. Interestingly, all deduced products contained at least one sequence (Fig. la) that satisfies the requirements of the PEST regions found in extremely short-half-life eukaryotic proteins: hydrophilic sequences (between positively charged residues) enriched in P, S/T or D/E that tend to reside near the amino or carboxyl termini of enzymes 6.

480

C termini

(b) m

ODC B. taurus P27117(168-98) ODC M. muscu/us P00860 (168-98) ODC M. pahari P27119(168-98) ODC R. norvegicus P09057 (168-98) ODC C. griseus P14019(165-95) ODC H, sapiens Pl1926 (168-98) HDC M. muscu!us P23738 (246-79) HDC R. norvegicus P16453 (242-75) HDC H. sapiens DDC B. taurus DDC S. scrofa DDC R. norvegicus

P19113(239-72) P27718(237-70) P80041(237-70) P14173 (237-70)

DDC C. porce//us P22781 (237-70) DDC H. sapiens P20711 (237-70)

LEnA LEI~A

DV DV DV iDV

DiV L El~lA GVCAFD~IL GVCAFDKL GVCAFDCL SCCSFDNL SCCSFDNL SCCSFDNL

NCCSFDSL TCCSFDNL

IDlY AS AR AR HK iH E

NQ NQ NK

Rgure 1

(a) PESTregions of mammalian HDCs and DDCs, Mammalian species and accession numbers of every deduced sequence in the SWISS-PROTdata bank are indicated. The figure reflects the alignment of the PESTsequences by the PILEUPprogram11.PESTregions are shown in boxes; numbers on boxes refer to the first and last residues of the deduced sequences. (b) Multiple alignment of a fragment of mammalian ODC (residues 168-198 in mouse ODC)with fragments of mammalian HDCs and DDCs. Black boxes indicate residues that are identical between ODCs and some other mammalian decarboxylases. Shaded residues in HDC and DDC fragments present a degree of similarity equal to or higher than 0.8 (Gribstov and Burguess' scores)15with respect to the aligned ODC residues; residues with similarity scores from 0,2--0.8 with respect to those of ODC are shown in white boxes. The first and last residue numbers of every fragment are given in brackets.

318

© 1994,ElsevierScience Ltd 0968-0004/94/$07.00

PROTEINSEQUENCE MOTIF

TIBS 19 - AUGUST 1994 The amino-terminal PEST regions of HDCs and DDCs overlap (PILEUPn). PEST regions were also detected in the carboxyl termini of the HDC-deduced sequences, in the fragment with no counterpart in DDCs; this fragment is apparently removed during post-translational maturation of the HDC subunits2,3. The essential residues for ODC activity (Lys169 and His197) occur in conserved fragments of mouse ODCs4. The program RDF2 (Ref. 12) was used to evaluate the interspecific similarity between these conserved and essential residues of mammalian ODCs and the corresponding fragments of mammalian HDCs (z scores =8-8.5; 200 local shuffles; window = 10 residues) and DDCs (z scores = 3-3.5, same conditions) (Fig. It)). A histidine residue followed by a branched-chain residue at the end of these fragments occurs, not only in the mammalian decarboxylases depicted in Fig. lb, but also in other homologous PLP-dependent prokaryotic and eukaryotic amino acid decarboxylases present in the SWISSPROT data bank (release 27.0), including glutamate decarboxylases (GDC, EC 4.1.1.15). His197 is the proposed proton donor during PLP-mediated decarboxylation in mouse ODC; to our knowledge the relevance of this histidine residue has not been studied in the other amino acid decarboxylases. In addition, Glu186 and Iie189 of mouse ODC were aligned with identical residues of the other mammalian decarboxylases shown in Fig. lb. The consensus sequence [A/V]TLxT[T/S] could be deduced from the amino termini of these fragments (residues 172-177 of mouse ODC). According to the secondary structure prediction of the fragments (PHD method13), these conserved residues are close to loop-enriched zones in all the mammalian ODCs, HDCs and DDCs. Every ODC described so far from vertebrates has three hydroxylated residues (Ser or Thr) aligning to positions 173, 176 and 177 of the mouse enzyme; this is also the case for ODC of Neurospora crassa (P27121, SWISS-PROT) which presents the overall phenomenology of man~:.aalian ODC regulation ~4.HoweveL ~-~.se hydroxylated residues are not present in the homologous fragments of ODCs from lower eukaryotes [Trypenosoma brucei, Leishmania donovani and Saccharomyces cerevisiae (P07805, P27116 and P08432 in SWISS-PROT, respectively)(FASTA~)], in which a loss of ODC protein does not occur in response to polyamines.

Acknowledgements We are grateful to B. Rost, R. Schneider and C. Sander (EMBL, Heidelberg,

Germany) for secondary structure predictions, and to L. Blanco (CBM, Madrid, Spain), !. N06ez de Castro and E R. Canton (F. Ciencias, University of M~laga, Spain) for their helpful discussion of the manuscript. This work was supported by Grants PTR92-0027 and SAF92-0582 (CICYT), Grant PB91-0910 (DGICYT) and Acci6n Integrada UK-189B. To the memory of Jes0s S. Olavarria

References I Hayashi, H. et al. (1990) Annu. Rev. Biochem. 59, 87-110 2 Joseph, D. R. et al. (1990) Proc. Natl Acad. Sci. USA 87, 733-737 3 Tanaka, T. et al. (1989) Proc. Natl Acad. Sci. USA 86, 8142-8146 4 Poulin, R. et al. (1992) J. Biol. Chem. 267, 150-158 5 Heby, O. and Persson, L. (1990) Trends Biochem. Sci. 15, 153-158 6 Rogers, S., Wells, R. and Reichteiner, M. (1986) Science 234, 364-368 7 Ghoda, L., Sidney, D., Macrae, M. and Coffino, P. (1992) Mol. Cell. Biol. 12, 2178-2185 8 Urdiales, J. L e t al. (1992) FEBS Lett. 305, 260-264 9 Bartholeyns, J. and Bouclier, M. (1984) Cancer Res. 44, 639-645 10 Endo, Y. (1989) Biochem. PharmacoL 38, 1287-1292 11 Genetics Computer Group (April 1991) Program manual for the GCG package, version 7, 575 Science Drive, Madison, Wisconsin, USA, 53711 12 Pearson, W. R. (1990) Methods Enzymol. 183, 63-98 13 Rost, B., Schneider,R. and Sander,C. (1993) Trends Biochem. Sci. 18, 120-123 14 Davis, R. H., Morris, D. ~. and Coffino, P. (1992) Mir;'obieL Rev. 56, 280-290 15 Grlbskov, M. and Bur~uess, R. R. (1986) Nucleic Acids Res. 14, 6745-6763 ~51~okITANE'O

ENRIQUE VIGUERA Centro de Investigaciones BiolOgicas CSIC, VelSzquez144, 28006 Madrid, Spain.

"-"L_ _.~ "~,~ /

"'"

V.S REC|TAI~(,,~

| NEITHER OF K~ CAN THIlVK OF .~N~-rt,l l , ~

)

ia

OSWALDO TRELLES Departamento de Arquitectura de Computadores de la Universidad de M~laga, 29071 M~laga, Spain.

JOS(: LUIS URDIALES, JOSI~ MANUEL MATINSAND FRANCISCA SANCHEZ-

JIMENEZ Laboratorio de Bioquimica y Biologia Molecular de la Facultad de Ciencias de la Universidad de M~laga, 29071 M~laga, Spain.

ivl, lri;

.

u ~t~.

II V.d, I r r l r l l l ~

IPhNUk~

|qltl~r..

"Pl^~ TWICE ..... Pete Jeffs is a biologist working in Paris, France.

319