Ornithine decarboxylase and S-adenosylmethionine decarboxylase in

Shared by: peisty474

Tags

ornithine decarboxylase, s-adenosylmethionine decarboxylase, odc activity, polyamine metabolism, spermidine synthase, adenosylmethionine decarboxylase, arginine decarboxylase, polyamine synthesis, j. biol. chem, odc gene, plasmodium falciparum, cell cycle, biosynthetic enzymes, cell proliferation, spermine synthase

Stats

views:: 9
posted:: 12/18/2009
language:: English
pages:: 4

Public Domain

Document Sample

314 Biochemical Society Transactions (2007) Volume 35, part 2

Ornithine decarboxylase and
S-adenosylmethionine decarboxylase
in trypanosomatids
L. Persson1
Department of Experimental Medical Science, Lund University, BMC F:13, S-221 84 Lund, Sweden

Abstract
The production of polyamines has been shown to be an effective target for a drug against the West African
form of sleeping sickness caused by Trypanosoma brucei gambiense. T. brucei belongs to the group of
protozoan parasites classed as trypanosomatids. Parasitic species of this group are the causative agents
of various tropical diseases besides African sleeping sickness, e.g. Chagas’ disease (Trypanosoma cruzi),
cutaneous (Lesihmania spp.) and visceral (Leishmania donovani) leishmaniasis. The metabolism of poly-
amines in the parasites is a potential target for the development of new drugs for treatment of these
diseases. The key steps in polyamine synthesis are catalysed by ODC (ornithine decarboxylase) and AdoMetDC
(S-adenosylmethionine decarboxylase). In the present paper, some of the available information on ODC and
AdoMetDC in trypanosomatids will be described and discussed.

Introduction e.g. Chagas’ disease (Trypanosoma cruzi), and cutaneous
Infectious diseases caused by protozoan parasites represent a (Lesihmania spp.) and visceral (Leishmania donovani) leish-
major public health problem around the world. It is estimated maniasis. The trypanosomatids contain a unique redox sys-
that almost 50% of the world’s population is at risk and ap- tem, in which trypanothione replaces the general glutathione
prox. 500 million people suffer from illnesses related to these system [6,7]. Trypanothione [bis(glutathionyl)spermidine] is
diseases each year (http://www.who.int/tdr/index.html). The synthesized from spermidine, which is the major polyamine
drugs available against these diseases are still of relatively in these parasites [6,7]. Thus interference with their poly-
low potency and/or high toxicity, and, despite much effort, amine production may make them more vulnerable to the
very few highly effective drugs have been developed in recent oxidative defence encountered in the human host. However,
years. Hence there is an urgent need for the development of DFMO is not very effective against trypanosomatids other
new drugs in this area. than T. brucei, indicating important differences. Some of these
The polyamine biosynthetic pathway has been demon- differences may be in the target of the drug, namely ODC.
strated to be a successful target for treatment of the West The key enzymes in the polyamine biosynthetic pathway,
African form of sleeping sickness, caused by the protozoan ODC and AdoMetDC, have been cloned and characterized
parasite Trypanosoma brucei gambiense [1,2]. The drug used from a variety of trypanosomatids, including those causing
is eflornithine, also known as DFMO (difluoromethylornith- tropical diseases [8–14]. Some of the available information is
ine), which is a highly specific irreversible inhibitor of the presented and discussed in the present paper.
enzyme catalysing the first step in polyamine biosynthesis,
namely ODC (ornithine decarboxylase). The drug was ini- ODC
tially developed as a cancer drug targeting mammalian ODC, ODC catalyses the first step in the polyamine biosynthetic
but was later also demonstrated to be effective against para- pathway, which is conversion of ornithine into putrescine. In
sites causing human sleeping sickness [3]. In addition, drugs mammalian cells, ODC is highly regulated and its turnover
targeting AdoMetDC (adenosylmethionine decarboxylase), is extremely rapid with half-lives of usually less than 1 h
which catalyses another key enzyme in the polyamine bio- [3,15]. In contrast with the mammalian enzyme, ODC
synthetic pathway, have been shown to be effective against has been demonstrated to be a stable enzyme in the try-
these parasites [4,5]. T. brucei belongs to the group of panosomatids T. brucei [8,16] and L. donovani [9]. The
closely related protozoan parasites classed as trypanosoma- slow turnover of T. brucei ODC is suggested to contribute
tids. Parasitic species of this group are the causative organisms to the effectiveness of the ODC inhibitor DFMO against
of various tropical diseases besides African sleeping sickness, the West African form of sleeping sickness [8]. DFMO is
an enzyme-activated irreversible inhibitor of ODC. If the
Key words: decarboxylase, drug target, ornithine decarboxylase (ODC), polyamine, protozoan enzyme has a slow turnover, the enzyme will be inactive
parasite, S-adenosylmethionine.
for a long time. However, if ODC has a fast turnover,
Abbreviations used: AdoMetDC, S-adenosylmethionine decarboxylase; ODC, ornithine
decarboxylase. the inactivated enzyme molecule will be rapidly replaced
1
email Lo.Persson@med.lu.se by newly synthesized enzymatically active ODC molecules.

C 2007 Biochemical Society
Health Implications of Dietary Amines 315

Like other short-lived proteins mammalian ODC is degraded part of mammalian ODC (which contains a PEST sequence),
by the 26 S proteasome [15,17]. However, ODC degradation it has two regions which contain sequences which fulfil the
is not triggered by ubiquitination which otherwise is the requirements of a PEST region. One is located in the N-termi-
usual mechanism for targeting proteins for degradation by nal extension and the other is located close to the C-terminal
the 26 S proteasome [18]. Instead, the degradation of ODC end [10]. The PEST sequence in the N-terminal extension is
by the 26 S proteasome appears to be induced by the binding a very strong PEST sequence according to the algorithm by
of an inducible protein named antizyme to the enzyme Rogers et al. [25]. Both these PEST regions are missing in
[15,19]. The polyamines induce the degradation of ODC by the mammalian ODC. Although there is a strong homology
affecting the synthesis of antizyme. The synthesis of antizyme between C. fasciculata ODC and L. donovani ODC, the latter
is dependent on a unique mechanism involving ribosomal does not have a region corresponding to the first PEST region
frame shifting [20]. Besides targeting ODC for degradation, of C. fasciculata ODC. Nevertheless, L. donovani ODC has
the binding of antizyme inactivates the enzyme. Antizyme a PEST sequence close to the C-terminal end, corresponding
binds to the monomer of ODC with a very high affinity, to the second PEST region of C. fasciculata ODC, although
which results in a shift from enzymatically active dimers to this is a much stronger PEST sequence in C. fasciculata
enzymatically inactive monomers. ODC than in L. donovani ODC. Thus L. donovani ODC,
Molecular analysis of mammalian ODC has revealed that which is a metabolically stable enzyme, lacks the N-terminal
the C-terminal part of the enzyme is crucial for the rapid PEST region and has a much weaker second PEST region
degradation of the enzyme [21,22]. Truncations or mutations compared with C. fasciculata ODC, which is a metabolically
in this part of the ODC protein convert the enzyme into a unstable enzyme [9,10]. However, it should be stressed that
metabolically stable protein. It is believed that the binding of the importance of these PEST regions for the rapid turnover
antizyme to ODC induces some kind of structural changes of C. fasciculata ODC remains to be confirmed.
exposing the C-terminal end of the protein to the 26 S Interestingly, C. fasciculata ODC is also rapidly degraded
proteasome [23]. ODCs from T. brucei and L. donovani, in mammalian cells as well as in a rabbit reticulocyte lysate
both of which are stable proteins, lack this C-terminal do- system, whereas L. donovani ODC is not [10,28]. The rabbit
main [8,9]. Furthermore, combining T. brucei ODC with reticulocyte lysate contains all the components necessary for
the C-terminus of mammalian ODC confers a short half- degradation of mammalian ODC by the 26 S proteasome,
life to the fusion protein when expressed in mammalian and changes in the ODC protein sequence which affect the
cells [16,24]. Interestingly, the C-terminal part of mammalian half-life in vivo also affect the degradation rate in the re-
ODC contains a so-called PEST region, which are regions ticulocyte lysate [22]. Thus C. fasciculata ODC contains
rich in proline (P), glutamic acid (E), aspartic acid (D), serine degradation signals that are also functionally active in mam-
(S) and threonine (T). These regions have been shown to be malian cells. Inhibitors of the 26 S proteasome markedly
more abundant in proteins with short half-lives, indicating a reduce the degradation rate of C. fasciculata ODC in the
role in the rapid turnover of the protein [25]. mammalian systems, indicating the involvement of the 26
Both T. brucei and L. donovani are digenetic parasites S proteasome in the degradation of C. fasciculata ODC in
which inhabit two highly specific hosts during their life mammalian cells [28]. However, whether C. fasciculata ODC
cycle. Dr Israel Algranati and colleagues discovered that some is degraded by a similar proteolytic system in the parasite is
monogenetic trypanosomatids, such as Crithida fasciculata not known. Nevertheless, in contrast with the degradation
and Phytomonas T274, contain ODCs with short half-lives of mammalian ODC, the degradation of C. fasciculata ODC
[26,27]. C. fasciculata is a monogenetic trypanosomatid that does not seem to be induced by a protein corresponding to
grows in the digestive tract of infected flies and Phytomonas the mammalian antizyme [28].
T274 is a plant trypanosomatid. Algranati and colleagues As mentioned above, DFMO is highly effective against the
demonstrated that inhibiting protein synthesis in C. fascicu- parasite T. brucei gambiense causing the West African form
lata and Phytomonas T274 with cycloheximide resulted in a of sleeping sickness. This has, at least partly, been suggested
rapid decrease in ODC activity with a half-life of approx. to be due to fact that ODC in T. brucei is a stable enzyme [8]
30 min [26,27]. and that polyamines are substrates for the synthesis of tryp-
The cloning of the gene encoding C. fasciculata ODC re- anothione [6,7], which is essential for the protection of the
vealed a high degree of sequence homology with ODC from parasite against reactive oxygen species produced by the host.
L. donovani [10], with an identity of approx. 70% between However, DFMO is not very effective against the trypano-
the two sequences. Both ODCs from C. fasciculata and somatid parasite causing visceral leishmaniasis, L. donovani,
L. donovani contain long N-terminal extensions of more in spite of the fact that this parasite also has a stable ODC
than 200 amino acids compared with ODCs from T. brucei and uses polyamines for trypanothione production. DFMO
and mouse [10]. The sequencing of C. fasciculata ODC also was originally developed as a cancer agent for use against
revealed that this protein lacks the region corresponding mammalian ODC. It is conceivable that it may be less
to the C-terminal domain of mammalian ODC, indicating effective as an inhibitor of ODCs from these protozoan
that C. fasciculata ODC contains other signals targeting the parasites, which are highly separated phylogenetically from
enzyme for rapid degradation [10]. Although C. fasciculata mammals. However, a recent report suggests that the Odc
ODC does not have a region corresponding to the C-terminal gene in T. brucei has been replaced by an Odc gene from a

C 2007 Biochemical Society
316 Biochemical Society Transactions (2007) Volume 35, part 2

vertebrate source [29]. Eukaryotic phylogenies from full- similar to antizyme has yet been discovered. Instead, it is
length genes for ODC revealed that the Odc genes from vari- conceivable that the degradation of mammalian AdoMetDC
ous African trypanosomes (T. brucei, Trypansoma congolense by the 26 S proteasome is ubiquitin-dependent [39]. Never-
and Trypanosoma vivax) clustered with Odc genes from theless, the synthesis of AdoMetDC is highly regulated by
vertebrates rather than with Odc genes from other trypano- the cellular polyamine levels [38]. Spermidine and spermine
somatids, indicating a horizontal gene transfer from a verte- inhibit the synthesis of AdoMetDC at the levels of transcrip-
brate species [29]. The conclusion of a loss of the Odc gene tion as well as translation. Mammalian AdoMetDC is synthe-
in the Trypanosoma lineage followed by acquisition of a new sized as a pro-enzyme, which has to be cleaved in order to
copy from a vertebrate host in the African branch of the genus produce the two different subunits of the enzyme (which is
was supported by analysis of the regions around the Odc a heterotetramer containing two subunits of each) [40]. The
genes in Leishmania major, T. brucei and T. cruzi [29]. cleavage of the proenzyme generates a pyruvate covalently
DFMO has been demonstrated to be without effect on bound to the larger subunit [40]. Mammalian AdoMetDC is
Chagas’ disease, which is the American form of trypano- one of the very few enzymes that use pyruvate, instead of
somiasis. The causative agent of Chagas’ disease is T. cruzi, pyridoxal 5 -phosphate, as a cofactor [41]. The conversion
which is closely related to T. brucei. All attempts to measure of the AdoMetDC proenzyme into the subunits is stimulated
any ODC activity in this parasite have been unsuccessful in vitro by putrescine [42]. Putrescine also exerts a direct
and it was suggested that the parasite contains an alternative stimulatory effect on the activity of the enzyme [42].
pathway for the production of putrescine from arginine, via AdoMetDC in trypanosomatids has been demonstrated to
agmatine [30]. However, gene analysis as well as metabolic be a potential target for drugs against various diseases caused
studies using radiolabelled ornithine and arginine together by protozoan parasites of this group [4,5]. However, the
with the relevant enzyme assays indicate that T. cruzi is molecular understanding of AdoMetDC and its regulation
unable to synthesize significant amounts of putrescine, and in trypanosomatids is much more meagre than that of the
is dependent on scavenging this diamine from an external mammalian enzyme. AdoMetDC appears to be a stable
source (like the host) [31–33]. The parasite has been shown enzyme in T. brucei [43] as well as in L. donovani [13]. No
to contain highly effective high-affinity transporters for the reports exist describing the half-life of T. cruzi AdoMetDC.
uptake of putrescine [34] and spermidine [35]. Thus, during Analysis of AdoMetDC in C. fasciculata showed that the
evolution, T. cruzi may have developed into a putrescine enzyme has a very fast turnover in the parasite, with a half-
auxotrophic parasite by the loss of its Odc gene. T. cruzi life as short as 3 min, which appears to be the shortest half-life
replicates inside a cell in the host and it is conceivable that it ever recorded for a eukaryotic AdoMetDC [44]. As for the
may scavenge putrescine or other polyamines from the host mammalian enzyme, putrescine has a direct stimulatory effect
cell. T. brucei, on the other hand, replicates extracellularly on AdoMetDCs from T. brucei [45], T. cruzi [11,46] and C.
in the bloodstream of the host and is thus more dependent fasciculata [44]. No information on whether this is true also
on endogenous polyamines (the polyamine concentration in for AdoMetDC from Leishmania spp. is available, but it is
blood is relatively low). The introduction of the Odc gene highly likely.
from C. fasciculata into the genome of T. cruzi makes the AdoMetDC have been cloned and sequenced from a vari-
R
parasite overcome the exogenous putrescine requirement for ety of trypanosomatids, e.g. T. brucei (GenBank accession
growth [36]. Interestingly, the C. fasciculata is a stable enzyme number U20092), T. cruzi [11], L. donovani [13], Leishmania
when expressed in T. cruzi [36]. The recent identification and infantum [14] and C. fasciculata (L. Persson, unpublished
characterization of a polyamine transporter from L. major work). The enzymes exhibited high homology between the
may facilitate the identification of the specific putrescine various trypanosomatid species (with identity values ranging
transporter responsible for the uptake of putrescine in T. cruzi between 62 and 85%), but less with the mammalian enzyme
[37]. The molecular characterization of this transporter may (30–33% identity). Interestingly, there is a close homology
form the basis for future design of specific drugs against (85% identity) between C. fasciculata AdoMetDC and L.
Chagas’ disease. donovani AdoMetDC, in spite of the large difference in
turnover rates in the parasites (L. Persson, unpublished
work). The trypanosomatid AdoMetDCs are very similar
AdoMetDC in size, with molecular masses of approx. 42–44 kDa, which
AdoMetDC catalyses the production of decarboxyl- is somewhat bigger than that of mammalian AdoMetDC
ated AdoMet, which provides the aminopropyl groups in the (38 kDa). All of the trypanosomatid AdoMetDCs contain
synthesis of spermidine and spermine from putrescine and a highly conserved region around an ESS (Glu-Ser-Ser)
spermidine respectively. Also, mammalian AdoMetDC is a sequence. This is similar to the site in mammalian AdoMetDC
highly regulated enzyme with a fast turnover [38]. Like ODC, where the cleavage of the proenzyme into the two subunits
mammalian AdoMetDC appears to be degraded by the occurs. The cleavage occurs between the glutamic acid and
26 S proteasome [39]. However, so far not much is known the serine residues, which also transforms the serine into the
about the mechanisms involved. Unlike the degradation of covalently bound pyruvate. In contrast with the processing
ODC, the degradation of AdoMetDC is not affected by of mammalian AdoMetDC, putrescine does not appear to
polyamines. Furthermore, no AdoMetDC-binding protein stimulate the conversion of the trypanosomatid AdoMetDC

C 2007 Biochemical Society
Health Implications of Dietary Amines 317

pro-enzyme into the two subunits as studied in a rabbit 20 Matsufuji, S., Matsufuji, T., Miyazaki, Y., Murakami, Y., Atkins, J.F.,
reticulocyte lysate using AdoMetDC from T. cruzi [46] and Gesteland, R.F. and Hayashi, S. (1995) Cell 80, 51–60
21 Ghoda, L., van Daalen Wetters, T., Macrae, M., Ascherman, D. and
C. fasciculata (L. Persson, unpublished work). Cofﬁno, P. (1989) Science 243, 1493–1495
22 Rosenberg-Hasson, Y., Bercovich, Z. and Kahana, C. (1991) Eur. J.
Biochem. 196, 647–651
Conclusions 23 Li, X. and Cofﬁno, P. (1993) Mol. Cell. Biol. 13, 2377–2383
The polyamine biosynthetic pathway in various trypanoso- 24 Ghoda, L., Sidney, D., Macrae, M. and Cofﬁno, P. (1992) Mol. Cell. Biol.
12, 2178–2185
matids is a potential target for drugs against various diseases 25 Rogers, S., Wells, R. and Rechsteiner, M. (1986) Science 234, 364–368
caused by these protozoan parasites. The target has been ´
26 Ceriani, C., Gonzalez, N.S. and Algranati, I.D. (1992) FEBS Lett. 301,
validated using different enzyme inhibitors [1,47], as well as 261–264
27 Carrillo, C., Cejas, S., Cortes, M., Ceriani, C., Huber, A., Gonzalez, N.S. and
null mutants of some of the parasites, i.e. T. brucei [48] and Algranati, I.D. (2000) Biochem. Biophys. Res. Commun. 279, 663–668
L. donovani [13,49,50]. The understanding of species-specific 28 Nasizadeh, S., Jeppsson, A. and Persson, L. (2003) Cell. Physiol. Biochem.
differences in the enzymes involved in parasitic polyamine 13, 321–328
29 Steglich, C. and Schaeffer, S.W. (2006) Infect. Genet. Evol. 6, 205–219
production may form the basis for future design of drugs 30 Kierszenbaum, F., Wirth, J.J., McCann, P.P. and Sjoerdsma, A. (1987)
targeting these parasites. Proc. Natl. Acad. Sci. U.S.A 84, 4278–4282
31 Hunter, K.J., Le Quesne, S.A. and Fairlamb, A.H. (1994) Eur. J. Biochem.
226, 1019–1027
32 Ariyanayagam, M.R. and Fairlamb, A.H. (1997) Mol. Biochem. Parasitol.
References 84, 111–121
1 Bacchi, C.J. and Yarlett, N. (2002) Mini Rev. Med. Chem. 2, 553–563
33 Carrillo, C., Cejas, S., Huber, A., Gonzalez, N.S. and Algranati, I.D. (2003)
2 Burri, C. and Brun, R. (2003) Parasitol. Res. 90, (Suppl. 1), S49–S52
J. Eukaryotic Microbiol. 50, 312–316
3 Wallace, H.M., Fraser, A.V. and Hughes, A. (2003) Biochem. J. 376, 1–14
34 Le Quesne, S.A. and Fairlamb, A.H. (1996) Biochem. J. 316, 481–486
4 ¨
Bacchi, C.J., Brun, R., Croft, S.L., Alicea, K. and Buhler, Y. (1996)
35 Carrillo, C., Canepa, G.E., Algranati, I.D. and Pereira, C.A. (2006)
Antimicrob. Agents Chemother. 40, 1448–1453
Biochem. Biophys. Res. Commun. 344, 936–940
5 ¨
Brun, R., Buhler, Y., Sandmeier, U., Kaminsky, R., Bacchi, C.J., Rattendi, D.,
36 Carrillo, C., Cejas, S., Gonzalez, N.S. and Algranati, I.D. (1999) FEBS Lett.
Lane, S., Croft, S.L., Snowdon, D., Yardley, V. et al. (1996) Antimicrob.
454, 192–196
Agents Chemother. 40, 1442–1447
37 Hasne, M.P. and Ullman, B. (2005) J. Biol. Chem. 280, 15188–15194
6 Fairlamb, A.H. and Cerami, A. (1992) Annu. Rev. Microbiol. 46, 695–729
38 Pegg, A.E., Xiong, H., Feith, D.J. and Shantz, L.M. (1998)
7 Muller, S., Liebau, E., Walter, R.D. and Krauth-Siegel, R.L. (2003)
Biochem. Soc. Trans. 26, 580–586
Trends Parasitol. 19, 320–328
39 Yerlikaya, A. and Stanley, B.A. (2004) J. Biol. Chem. 279, 12469–12478
8 Phillips, M.A., Cofﬁno, P. and Wang, C.C. (1987) J. Biol. Chem. 262,
40 Stanley, B.A., Pegg, A.E. and Holm, I. (1989) J. Biol. Chem. 264,
8721–8727
21073–21079
9 Hanson, S., Adelman, J. and Ullman, B. (1992) J. Biol. Chem. 267,
41 Pegg, A.E. (1977) FEBS Lett. 84, 33–36
2350–2359
42 Stanley, B.A. and Pegg, A.E. (1991) J. Biol. Chem. 266, 18502–18506
10 ¨ ¨
Svensson, F., Ceriani, C., Lovkvist Wallstrom, E., Kockum, I., Algranati, I.D.,
¨ ¨ ¨
43 Selzer, P.M., Hesse, F., Hamm-Kunzelmann, B., Muhlstadt, K., Echner, H.
Heby, O. and Persson, L. (1997) Proc. Natl. Acad. Sci. U.S.A. 94, 397–402
and Duszenko, M. (1996) Eur. J. Cell Biol. 69, 173–179
11 Persson, K., Åslund, L., Grahn, B., Hanke, J. and Heby, O. (1998)
44 Nasizadeh, S. and Persson, L. (2003) FEBS Lett. 553, 131–134
Biochem. J. 333, 527–537
45 Tekwani, B.L., Bacchi, C.J. and Pegg, A.E. (1992) Mol. Cell. Biochem. 117,
12 Kinch, L.N., Scott, J.R., Ullman, B. and Phillips, M.A. (1999)
53–61
Mol. Biochem. Parasitol. 101, 1–11
46 Clyne, T., Kinch, L.N. and Phillips, M.A. (2002) Biochemistry 41,
13 Roberts, S.C., Scott, J., Gasteier, J.E., Jiang, Y., Brooks, B., Jardim, A., Carter,
13207–13216
N.S., Heby, O. and Ullman, B. (2002) J. Biol. Chem. 277, 5902–5909
47 Muller, S., Coombs, G.H. and Walter, R.D. (2001) Trends Parasitol. 17,
14 Taladriz, S., Ramiro, M.J., Hanke, T. and Larraga, V. (2002) Parasitol. Res.
242–249
88, 421–426
48 Li, F., Hua, S.B., Wang, C.C. and Gottesdiener, K.M. (1996) Mol.
15 Cofﬁno, P. (2001) Nat. Rev. Mol. Cell Biol. 2, 188–194
Biochem. Parasitol. 78, 227–236
16 Ghoda, L., Phillips, M.A., Bass, K.E., Wang, C.C. and Cofﬁno, P. (1990)
49 Jiang, Y., Roberts, S.C., Jardim, A., Carter, N.S., Shih, S., Ariyanayagam,
J. Biol. Chem. 265, 11823–11826
M., Fairlamb, A.H. and Ullman, B. (1999) J. Biol. Chem. 274,
17 Murakami, Y., Matsufuji, S., Kameji, T., Hayashi, S., Igarashi, K., Tamura,
3781–3788
T., Tanaka, K. and Ichihara, A. (1992) Nature 360, 597–599
50 Roberts, S.C., Jiang, Y., Jardim, A., Carter, N.S., Heby, O. and Ullman, B.
18 Rosenberg Hasson, Y., Bercovich, Z., Ciechanover, A. and Kahana, C.
(2001) Mol. Biochem. Parasitol. 115, 217–226
(1989) Eur. J. Biochem. 185, 469–474
19 Hayashi, S., Murakami, Y. and Matsufuji, S. (1996) Trends Biochem. Sci.
21, 27–30 Received 18 October 2006

C 2007 Biochemical Society