вход по аккаунту


Biosynthesis of Vitamin B1 (Thiamin) An Instance of Biochemical Diversity.

код для вставкиСкачать
Biosynthesis of Vitamin B, (Thiamin): An Instance of Biochemical Diversity
Ian D. Spenser” and Robert L. White
The elucidation of the biosynthetic
pathway to thiamin (Vitamin B,) and its
pyrophosphate ester, the important
coenzyme “cocarboxylase”, has challenged researchers for many years and
continues to do so. The problem of the
origin of thiamin can be separated into
three parts: the independent pathways
to the pyrimidine moiety 4-amino-5-hydroxymethyl-2-methylpyrimidine and
to the thiazole moiety 5-(2-hydroxyethyl)-4-methylthiazole, and the route
from these subunits to the vitamin. The
steps in the latter process were fully established some twenty years ago, and it
was shown that the route in aerobic bacteria and yeast differs to some extent
from that in enteric bacteria. The pathways to the subunits, on the other hand,
are still not clarified. Significant differences exist in the routes whereby each of
the two subunits, the pyrimidine moiety
and the thiazole moiety, originate in
bacteria and yeast. One difficulty that
delayed progress was that the incorporation patterns of labeled precursors,
which were observed by different research groups in different microorganisms, could not be reconciled on the basis of a single pathway to each of the two
subunits. It is now accepted that in each
case different pathways exist in enteric
bacteria and yeast, and that the biosynthesis of Vitamin B, represents an instance of biochemical diversity. A second factor that added to the difficulties
is the minute amount of thiamin synthesized in microbiological cultures (about
15 pg per L culture). This limited the investigations until very recently either to
the use of radioactive tracers or to the
use of stable isotopes in conjunction
1. Introduction
Vitamin B, ( 6 ) . the nutritional factor that prevents and reverses the symptoms of beri-beri, was first isolated from rice
bran in 1926 by B. C. P. Jansen and W. F. Donath.”. 2a1 Its structure was elucidated in 1936 by R. R. Williams, and its synthesis
reported soon thereafter.“, Zb*31 The biochemical function of the
compound was recognized in 1937,[41when the corresponding
pyrophosphate (8) was identified as cocarboxylase, the coenzyme of pyruvate decarboxylase (EC 4.1.1. I), pyruvate dehydrogenase (EC 1.2.4. I), transketolase (EC 2.2.1. I), and several other enzymes
[*] Prof. Dr. I . D. Sprnser
Department of Chemistry
McMaster University
Hamilton, Onrario L8S 4M1 (Canada)
Fax’ Int. code + (905) 522-2509
e-mail: spenserco mcmasterxa
Prof. Dr R . L. White
Department of Chemistry, Dalhousie University
Halifax No%a Scotia B3H 4G3 (Canada)
e-mail: RLWhite(o cheml
A n p w C h i w lni
€(I. Efl‘nyl. 1997, 36, 1032- 1046
with mass spectrometric analysis. It is
widely recognized that both methods are
associated with pitfalls in the interpretation of results. High-field 3CNMR, the
most powerful modern method available for the determination of incorporation patterns, has only very recently
been successfully employed in investigations of thiamin biosynthesis. As a result
of the conceptual and experimental
problems, even the primary precursors
of each of the two relatively simple heterocyclic subunits of thiamin are still not
completely established. A search for
committed intermediates, the study of
the enzymes, and identification of the
genes that are involved are the matter of
current research.
Keywords: biosynthesis - enzymes
NMR spectroscopy vitamins
Thiamin (6)15]
is readily available. The annual world production of synthetic Vitamin B, (thiamin) exceeds 4000 metric tons,
and its wholesale price is 30-35 US dollars per kg (1993 figures) .I6] The recommended daily intake is approximately
1.5 mg,[71and to prevent the recurrence of deficiency diseases, in
the Western world the synthetic compound is routinely added to
It is therefore most surprising that knowledge of the biosynthesis of thiamin (that is, of the steps whereby it is generated
from primary precursors within the organisms that produce it)
and of the biochemical mechanism, the enzymology, and the
genetics of these steps is still fragmentary. The current state of
knowledge of the biosynthesis of thiamin and cocarboxylase is
summarized in the present review.
Biosynthesis of thiamin occurs in most microorganisms and
higher plants. A convergent pathway (Scheme 1) generates thiamin monophosphate (7)from the phosphorylated forms of two
independently synthesized subunits, 4-amino-5-hydroxymethyl2-methylpyrimidine (1, the “pyrimidine unit”) and 5-(2-hydroxyethyl)-4-methylthiazole (4, the “thiazole unit”). In enteric
VCH Verlagsgesell.whufimhH, D-69451 Weinlzeim. 1997
0570-0833197/3610-lU33$ 17.50+ 50
I. D. Spenser and R. L. White
R = H
R = H
R = P
R = P
R = H
R = P
Scheme 1. Vitamin B, (thiamin, 6), its subunits (1 and 4), and their phosphate and
pyrophosphate esters.
bacteria, the active coenzyme thiamin pyrophosphate (8, cocarboxylase) is formed from 7 by phosphorylation. However, in
yeast and aerobic bacteria, 7 is hydrolyzed to thiamin (6),and
a pyrophosphate group is added in one step to yield 8. Most of
the enzymes controlling the coupling and phosphorylation processes have been partially purified, and almost all the genes
responsible for the production of these enzymes have been located in Escherichia coli and in yeast. New insights concerning the
enzymology and the genetics that have been gained since the
publication of earlier reviews[*]will be described, and adequate
documentation is provided to summarize the current state of
knowledge of the final stages of thiamin biosynthesis.
Precursor-product relationships for the heterocyclic subunits of thiamin have only been established with certainty in the
past 3 5 years. Other than (5-aminoimidazo1e)ribotide (AIR, 9),
no committed intermediate has yet been recognized, and the
enzymology leading to the two subunits is entirely unknown.
Recent investigations have provided initial genetic information
for these pathways.[’] Whether the biosynthetic pathways to the
heterocyclic subunits produce the free pyrimidine (1) and free
thiazole (4) units or their phosphate esters (2 and 5, respectively)
is not yet established.
Thiamin biosynthesis is a remarkable example of biochemical
diversity: not only is there diversity in the final stages of coenzyme biosynthesis, but two entirely different pathways are established for both the pyrimidine and thiazole units, one in
enteric bacteria and the other in yeast. The information available indicates that aerobic bacteria utilize the yeast pathway for
Ian Spenser studied Chemistry at the University of Birmingham,
and at King’s College, University of London, England, where he
completedhis Ph.D. in Biochemistry in 1952 under the direction
of William Robson and his D. Sc. in 1969. From 1952 until 1957
he was a lecturer in the Department of Biochemistry of the
Medical College of St. Bartholornew’s Hospital, University of
London. In 1953- 1954, as postdoctoral fellows in Leo Marion’s
laboratory at the National Research Council of Canada Laboratories in Ottawa, he and the late Eddie Leete, who became a life
long friend, carried out thejirst radioactive tracer experiments
to test biogenetic hypotheses of the origin of naturalproducts. In
R. L. White
I. D. Spenser
1957 he joined the Department of Chemistry of McMaster University, Hamilton, Canada, as Assistant Professor and now
holds the rank of Professor Emeritus. At McMaster he continued his tracer studies of alkaloid biosynthesis. In the late 1960s
he turned his attention to the biosynthesis of two B vitamins, pyridoxine and thiamin, work that is continuing today. He was
Visiting Professor at the Technical University of Denmark (with A . Kjaer, 1977), at the University of Tokyo (with U . Sankawa,
1983), and at the Universities of Karlsruhe (with J: Retey and H. Musso, 1981) and Bonn (with E. Leistner, 1989), and
“Akademischer Gast” at the ETH. Zurich (with D. Arigoni, 1971, 1989). Among his honors are the Fellowship of the Royal
Society of Canada (1980), a NATO Senior Scientist Award (1982) and the John Labatt Ltd. Award of the Chemical Institute
of Canada (19831.
Robert L . White was born in Halifax, Nova Scotia, on the day that Watson and Crick’s paper on the structure of DNA was
published in Nature. He studied at Dalhousie University and McMaster University, where he received his Ph.D. on thiamin
biosynthesis under the direction of Ian D. Spenser in 1979. As a postdoctoral fellow, he worked on penicillin biosynthesis with
Sir Edward Abraham and Professor Jack Baldwin at the University of Oxford, and on enzyme inhibitors at Syntex, Inc. He is
an Associate Professor of Chemistry at Dalhousie University, engaged in research on the biosynthesis of natural products,
particularly nonprotein amino acids and antibiotics. In 1994, he became a Fellow of the Chemical Institute of Canada.
Angew. Chem. Inr. Ed. Engl. 1997,36, 1032-1046
Biosynthesis of Vitamin B,
thiazole biosynthesis and that plants use the route defined in
enteric bacteria to form the thiazole unit.
The work that led to the recognition of the primary precursors will be briefly summarized, and the biogenetic anatomy of
the pyrimidine and thiazole units of thiamin, in prokaryotes
and in eukaryotes, will be compared and contrasted. The
reader is referred to the original literature for details of the
methods that were used to establish the sites of incorporation
of the labeled substrates employed to trace the origin of the
pyrimidine and thiazole units. Only those contributions that
gave conclusive results will be included. Earlier reviews provide
derailed summaries of the biosynthetic process in eukaryotest'O1
and prokaryotes[lO,"l as it stood in 1984/85. In this review
we intend to stress the advances that have been made during
the past ten years, to compare and contrast the diverse pathways, and to direct attention to the problems that remain
to be solved. In particular, we shall deal with current investigations of the biosynthesis of the pyrimidine and thiazole
1.1. Methodology
Published investigation of the biosynthetic incorporation of
labeled substrates into thiamin have so far been limited to the
use of radioactive tracers or of stable isotopes in conjunction with mass spectrometric analysis. High-field 13C NMR,
the most powerful modern method available for the determination of incorporation patterns, has only very recently been
successfully employed in investigations of thiamin biosynthesis,[12. 131
The radioactive tracer method demands not only the isolation
of thiamin and its separation from other ionic water-soluble
compounds such as amino acids, a procedure whose difficulty
explains the persistent contamination of the desired product by
radioactive impurities, but also the preparation and purification
of derivatives by chemical manipulation on a minute scale. The
lack of rigorous purification of the compounds whose radioactivity was being determined has sometimes led to erroneous
conclusions. This was probably the origin of results that appeared to indicate, wrongly, that methionine was implicated in
the biosynthesis of the thiazole unit of thiamin. In most cases
published reports of radioactive tracer experiments do not include a description of the methods of purification and degradation that were employed.
Moreover, it is important, for reasons of internal control, to
account for the sites of the total radioactivity within thiamin as
a whole and not merely within one or other of the two subunits:
only when this was done did it become evident that glycine was
incorporated into the pyrimidine unit in enteric bacteria but not
in yeast, whereas it entered the thiazole unit in yeast but not in
enterobacteria. This finding served as conclusive evidence for
the existence of different pathways in prokaryotes and eukaryotes.
Although the mass spectrometric method requires gas chromatographic separation of the compounds whose fragmentation pattern is being recorded, none of the reports employing
this analysis include primary data on the efficiency of the gas
chromatographic separation of impurities from the investigated
Angeir. Clwm. In[. Ed. En,?/. 1997, 36. 1032- 1046
compound. Thus, it is often not possible to assess the purity of
the materials whose radioactivity or fragmentation pattern
serves as the basis of biosynthetic inferences.
In our most recent studies"'. 1 3 ] we have succeeded in detecting, by means of high-field nuclear magnetic resonance spectroscopy (NMR), the incorporation of substrates labeled with
the stable isotopes 13C or 2H as the isotopic markers. In particular, the application as substrates of samples that are fully 13Cenriched at contiguous carbon atoms ("bond-labeled'' samples)
has led to significant progress, since by use of such samples it is
possible to observe the transfer of intact multicarbon units from
precursor into biosynthetic product. Substrates bond-labeled
with fully enriched 2H attached to fully enriched 13C may be
employed similarly. Application of '"C-labeled samples provided a method for the detection of the transfer of individual carbon atoms but not of multicarbon units.
Even though the sensitivity of detection of 13C by NMR is
approximately three orders of magnitude lower than that of I4C
by liquid scintillation counting,['"] application of 13C labeled
and particularly 13C bond-labeled substrates is currently the
method of choice in biosynthetic studies that focus on investigations of precursor- product relationships. A significant advantage of the method lies in the fact that 13C NMR not only
detects the site of labeling, but at the same time confirms the
identity of the labeled sample and determines its degree of chemical purity. Thus, an obligatory component of biosynthetic investigations employing radioactive tracers," 51 namely the chemical manipulation of very small samples (which was required to
prepare chemical derivatives of the radioactive products in order to determine radiochemical purity and to execute sequences
of controlled chemical degradation in order to determine the
precise sites of labelling), is minimized with the use of I3C labeled substrates.
Tracer studies with 13C isotopes can only be successful if
signal assignments within the I3C NMR spectrum of the
product are reliable and if the level of 13C enrichment within
the product is adequate for detection by high-field NMR
spectroscopy. Low levels of 13C enrichment at single sites in
a product are much more difficult to detect with certainty
than enrichment from bond-labeled samples, since in the
former instance signal enhancement must be determined by
comparison with the intensity of a natural abundance signal,
whereas in the latter instance enrichment by intact incorporation of a pair of contiguous 13C enriched carbon atoms
is indicated by the presence of new satellite peaks that are
absent in the spectrum of a natural abundance sample. Furthermore, the presence of such satellite peaks resulting from incorporation of 13C- 13C contiguously labeled (that is, bondlabeled) substrates in the signals of the 13C NMR spectrum of
a biosynthetic product provides direct evidence for the transfer
from substrate to biosynthetic product of an intact multicarbon
A serious operational difficulty with the method stems from
the fact that very few specialized 13C-labeled(and particularly
I3C bond-labeled compounds) other than the most common
substrates are commercially available, and even these are quite
expensive. For specialized investigations the required labeled
substrates must be synthesized in house from labeled chemicals
that are available.
I. D. Spenser and R. L. White
2. Biosynthesis of the Pyrimidine Unit
Chemically distinct routes for the biosynthesis of the pyrimidine moiety of thiamin are strongly supported by the results of
isotopic experiments carried out in several different organisms,
particularly yeast (Candida u t i h and Saccharomyces cerevisiae)
and enteric bacteria (E. coli and Salmonella typhimurium). Formate is involved in both the yeast and the enteric bacteria
routes, but enters different positions of the pyrimidine ring in
the two pathways (C4 and C2, respectively). Glycine and ribose
are the precursors of the other carbon atoms in prokaryotes,
and in particular, in E. coli and S . typhimurium (Scheme 2). In
eukaryotes, glycine is not utilized in the formation of the pyrimidine unit, and glucose (presumably via pentulose phosphate)
supplies all other carbon atoms (Scheme 3 ) . There is evidence
Scheme 2. The biosynthesis of the pyrimidine unit (1) of thiamin in facultative
anaerobic bacteria. Precursor-product relationships show the intermediacy of 9, a
purine precursor (N' stems from the amide N atom of glutamine; for the steps on
route from 9 to 1 see Scheme 4; the atoms indicated by wedges, circles, or squares
were labeled in several experiments)
concerning the identity of some of the intermediates between the
early precursors and the final product, but committed intermediates have not been identified, and the mechanism whereby the
building blocks are joined to form the final structure is almost
entirely unknown.
2.1. Formate and Glycine as Precursors
Formate was the first precursor of the pyrimidine unit to be
identified by isotopic tracer experiments.[16]Chemical degradation of labeled thiamin derived from ['4C]formic acid showed
that in E. coli B1l7] and other prokaryotes['8s'91 label from
I'4C]formate entered C2 of the pyrimidine unit and no other site
(Scheme 2). On the other hand, in S . cerevisiae[20*211
and other
eukaryotes["] label entered C4 of the pyrimidine unit and no
other site (Scheme 3). In a single reported instance label entered
both C2 and C4 of the pyrimidine unit: about 75% of label was
located at C4 and 25% at C2.['*]
In enterobacteria three atoms of the pyrimidine ring are supplied by glycine. High incorporations of radioactivity from [l14C]- and [2-14C]glycine were observed in a glycine-requiring
mutant of S. typhimurium LT2,[231and it was subsequently
shown by chemical degradation that radioactivity from the carboxyl (Cl) and methylene (C2) groups of glycine entered C4[18]
and C6,[24Jrespectively, of the pyrimidine unit (Scheme 2).
Analogous results were obtained from mass spectrometric analysis of thiamin isolated from E. coli B incubated separately with
[1-'3C]-, [2-I3C]-,and ['sN]glycine.c2s1Whether glycine is incorporated as an intact unit into C6, C4, and N1 of the pyrimidine
remains to be shown.
In eukaryotes, or more particularly in S. cerevisiae,label from
[2-'4C]glycine[261and from [15N]glycine[271
entered the thiazole
unit (see Scheme 6) but not the pyrimidine unit of thiamin. In
other studies label from [l-'4C]glycine,[281 from [''Nlglycine,'291 or from ['sN,2-'3C]glyciner'21 was not incorporated
into the pyrimidine unit.
2.2. (5-Aminoimidazolyl)ribotide, an Advanced Precursor
in Enteric Bacteria
* NH2
* NH2
Scheme 3. The biosynthesis of the pyrimidine unit (1) in eukaryotes (major pathway) and precursor-product relationships. Numbers on carbon atoms refer to the
carbon atoms of glucose (10).N* stems from the amide N atom of glutamine.
Formate and giycine have been known for many years as
precursors of the purine ring system.1301(5-Aminoimidazoly1)ribonucleotide (AIR, 9) serves as an intermediate
(Scheme 2). C2 of the ring system of 9 originates from formate,
while the C,N unit (N3,C4,C5) originates from the intact NCC
chain of glycine. The remaining nitrogen atoms (NI and NH,)
originate from the amide nitrogen of glutamine (Scheme 2). In
E. coli the three nitrogen atoms of the pyrimidine unit are of
corresponding origin: N l is derived from glycine[2s1and the
other two nitrogen atoms are supplied by g l ~ t a m i n e .That
~ ~ ~9]
satisfied the requirements of an S . typhimurium LT2 mutant
blocked in both the purine and the thiamin pyrimidine pathways
led Newel1 and Tucker[311to suggest that 9 is a key intermediate
also in the biosynthesis of the pyrimidine moiety of thiamin. The
incorporation of a labeled sample of 9, prepared biosynthetically from [l-'4C]glycine, into the pyrimidine moiety without dilution of the specific activity,["I and the incorporation of label
Angew. Chem. Int. Ed. Engi. 1997.36, 1032-1046
Biosynthesis of Vitamin B,
from [3-ISN]-and from ["NH,](5-aminoimidazolyl)riboside into
N1 and NH,, respectively, of the
pyrimidine unit,r321fully support
this hypothesis (Scheme 2).
The CH, group (C2') and the
C, unit (C5,CS') of the pyrimidine moiety are not derived from
the aminoimidazole nucleus of
(5-aminoimidazolyl)riboside or
-ribotide. The ribose unit of (5aminoimidazo1e)ribotide was sugH ring
gested as the precursor of C5,CS'
of the pyrimidine unit when it was
observed that C5' was labeled
ring dosure
ring expansion
from [6-'4C]glucose in E. coif B
(IF0 13168).r331 Incorporation
of an intact C, unit derived from
D-[I,2,3,4.5,6-' 3C,]glucose into
the C5,C5' unit of the pyrimidine
in E. coli B mutant WG2 was recently confirmed by 13C NMR
(5aminoimidazo1e)ribotide serves
Scheme 4. A chemically rational, hypothetical mechanism for the intramolecular conversion, in prokaryotes, of
as the precursor of all the carbon
(5-aminoimidazoly1)ribosideinto the pyrimidine unit of thiamin.
atoms of the pyrimidine unit was
demonstrated in Newell and
Tucker's S. typhimurium strain thi 10/T by David and his collab2.3. Glucose as a Precursor in Yeast
orators. The pyrimidine unit labeled with 13Cwas formed from
In yeast the carbon atoms of the pyrimidine nucleus of thi[U5-aminoimidazolyl)riboside in the presence of glucose,
but no labeling of the pyrimidine unit was observed when [Uamin are derived from formate and glucose. While (5-aminoimidazoly1)ribotide is a precursor of nucleic acid purines in
3C]glucose and unlabeled (5-aminoimidazo1yl)ribosidewere
In a separate experiment, label from (5-aminoimidaS. cerevisiae, and presumably in other e~karyotes,[~'~
it is not an
was located mainly at C2' and at C5,C5',
intermediate of thiamin biosynthesis in this yeast. The mode of
the three carbon atoms not derived from the 5-aminoimidazole
incorporation of carbohydrate precursors further distinguishes
ring. Experiments employing radioactive samples of (5the biosynthetic pathway to the pyrimidine unit in yeast from
as well as of (5-aminoimidathat in enterobacteria. In C . utilis label from [1-I4C]-, [2-I4C]-,
demonstrated that C2' of the pyrimand [6-'4C]glucose (10) entered the pyrimidine unit, but efforts
idine unit was derived from C2 of the ribose moiety of
to locate the sites of activity led to ambiguous results.f361In
(5-aminoimidazolyl)riboside.The intramolecular nature of this
S. cerevisiae ATCC 24903 the distribution of label within the
transformation was deduced by analyzing the simultaneous inpyrimidine nucleus (from [1-'4C]-, [2-'4C]-, [6-I4C]-, and [6''C,6-3H]glucose and from [l-'4C]fructose) was fully accounted
corporation of [' 3C]ribose- and [' 5N]imidazole-labeledsamples
for by complete chemical degradation.[221A complex pattern of
of (5-aminoimida~olyl)riboside.[~~~
Thus, in prokaryotes, (5-aminoimidazole)ribotide (9) is an
distribution was observed, which was interpreted in terms of the
intermediate in the biosynthesis not only of the nucleic acid
simultaneous occurrence of two different independent pathways, one major, the other minor. In the major pathway forpurines but also of the pyrimidine unit of thiamin. To account
for the results of the isotopic experiments described above (summate enters C4 of the pyrimidine, and the five carbon atoms C2',
C2, C6, C5, and C5' of the pyrimidine originate from glucose,
marized in Scheme 2), conversion of (5-aminoimidazoly1)riboside (or (5-aminoimidazolyl)ribotide) to the pyrimidine unit
presumably via C1 to C5, respectively, of a pentulose (11; see
(or the corresponding 5'-phosphate ester) requires the cleavage
Scheme 3). In the minor pathway, not depicted in Scheme 3,
formate enters C2, and the glucose-derived carbon atoms C1,C2
of five bonds of the precursor (imidazole C4-C5; ribose C1of a pentulose supply C5,C4 of the pyrimidine. The origin of
C2, C2-C3, C3-C4; and imidazoleNl -riboseCl) with the exC2', C5', and C6 is still unknown. It should be noted that this
pulsion of two carbon atoms (C1 and C3) of the ribose unit and
the creation of three new bonds (imidazole C2 -ribose C2, yieldduality of routes has so far not been observed in any other strain
ing pyrimidine C2,C2'; imidazole C4-ribose C4, yielding pyrimof S. cerevisiae, or indeed in any other eukaryote.
idine C6- C5 ; and imidazole CS-ribose C4, yielding pyrimidine
A recent 13CNMR study on the incorporation of label from
~-[1,2,3,4,5,6-'~C,]glucoseinto thiamin in S. cerevisiae ATCC
C4-C5). The accumulated evidence suggests that this must take
7752 ( = I F 0 1234) provides evidence of the incorporation of an
place intramolecularly. A hypothetical, chemically rational
intact C, unit derived from glucose into the C, unit C2', C2 and
mechanism for such a process is shown in Scheme 4.
Angen. Chrm Ini Ed. En,ql 1997. 36. 1032-1046
I. D. Spenser and R. L. White
of an intact C, unit into the C, unit C6, C5, C5' of the pyrimidine
unit of thiamin!"]
confirming the mode of entry of glucose by
the major pathway demonstrated earlier by the tracer work with
I4C. No indication for the existence in this strain of yeast of the
minor pathway was detected.
2.4. Advanced Pyrimidine Precursors in Yeast
An early report of the isolation, from S . cerevisiae, of a late
intermediate in the evolution of the pyrimidine
has remained unresolved. The compound was reported to show ultraviolet absorption with a maximum at 259 nm, and to yield a
'H NMR spectrum showing three signals: 6 = 2.0 (s, 3H, possibly CH,CO), 6.2 (d, 2H, possibly vinylic CH2CH=), and 8.5
(t, 3H, ?). Without further information a structure cannot be
proposed on this basis, and the substance will have to be reisolated before useful inferences can be drawn.
Recent results appear to implicate h i ~ t i d i n eand
~ ~ ~pyridox]
iner39.401 as the progenitors of the pyrimidine moiety of thiamin
in S. cerevisiae IF0 1234 (Scheme 5). Radioactivity from L[2-14C]histidine enters an unspecified site (assumed to be C4)
within the pyrimidine unit with little change in specific activity.
,j >HN'
N source
e.g.. aspartate
Scheme 5. Reported incorporation pattern of fragments of pyridoxol and of L-histidine into the pyrimidine unit of thiamin in yeast.
The label from [1,3-'5N,]histidine entered N3 and the NH,
group of the pyrimidine unit. Since N1 of histidine is derived
from the amino group of aspartic acid and since label from
["N]aspartic acid entered N3 of the pyrimidine, it was inferred
that N1 of histidine generates N3 of the pyrimidine and that N3
of histidine, which in turn is derived from the amide nitrogen
atom of glutamine, becomes the NH, group of the product. It
should be noted, however, that both N3 and the NH, group of
the pyrimidine are derivable from the amide nitrogen atom of
g l ~ t a m i n e . [15N
~ ~ ] from ~~-[a-amino-'~N]histidine
did not enter the pyrimidine, nor did the other carbon atoms of histidine
play a part in the biosynthetic process.[411These results were
interpreted as showing that an intact NI,C2,N3, fragment of
histidine, derived by cleavage of the imidazole ring, supplies the
intact NCN unit N3, C4, NH, of the pyrimidine. Both the intact
incorporation of the histidine fragment or an early amination
step in the pyrimidine pathway are consistent with the reported
nonincorporation of a deuterated sample of 4-hydroxy-5-hydroxymethyl-2-methylpyrimidinein S. cerevz~iue.'~~~
Incorporation of the intact NCN fragment of the imidazole
ring of histidine has not been tested by direct experiment, however, and another interpretation of the results is plausible: The
catabolism of histidine via urocanic acid and imidazolonepropionic acid is a general metabolic process that has been extensively
studied in bacteria (but has apparently not yet been investigated
in yeast). This process eventually leads to the transfer to folic
acid of the N1 ,C2 unit of the imidazole nucleus of histidine, that
is, of the aspartate-derived nitrogen atom and the formatederived carbon atom, to yield N-5-formiminotetrahydrofolatef4,] and then N-5-formyltetrahydrofolate, two of the classical donors of formate units.1441In our view, it is likely that C2
and possibly also N1 of the histidine nucleus enters the thiamin
pyrimidine by this indirect route.
The remaining C2NC, fragment of the pyrimidine
(C2',C2,NI,CS,CS,C5') is inferred to be derived directly from
the corresponding C,NC, unit of p y r i d o ~ o l , [ ~on~ ]the basis
both of the results of competition experiments with "NHZCIin the presence of various unlabeled substrates including pyrid o ~ o l ,and
' ~ ~of~incorporation experiments with 13C- and 'Hlabeled samples of p y r i d o ~ o l . ' ~
~ ] biosynthesis of pyridoxol
in yeast is entirely unexplored and for obvious reasons cannot
be assumed to follow the same route that is now emerging for
pyridoxol biosynthesis in E. c0k[451What is known about the
latter route, however, is not compatible with a direct entry of a
pyridoxol fragment into the pyrimidine, since the established
mode of incorporation of several labeled substrates (e.g., pyruvate and glycine) into pyridoxol in E. coli, on the one hand,[451
and into the pyrimidine unit in yeast, on the other (see above),
are not reconcilable.
Judgement on the inferences concerning the incorporation of
an intact pyridoxol fragment into the pyrimidine unit will have
to be reserved until something is known about pyridoxol biosynthesis and catabolism in eukaryotes. Pyridoxine catabolism has
been studied in a number of soil bacteria that are capable of
utilizing the vitamin as sole carbon and nitrogen source.146-4S1
A reaction sequence that occurs in these organisms, of a type
that may be of relevance to the present findings, is the oxidation
of both the C 4 and C5' hydroxymethyl groups of pyridoxol,
followed by loss of the carboxyl group at C4. 3-Hydroxy-2methylpyridine-5-carboxylic acid,[46,471 the compound so generated, then undergoes enzyme-catalyzed oxidative cleavage
of the C2-C3 bond of the pyridine nucleus, to yield 4-Nacetylamino-3-carboxybut-3-enoic acid (CH,CONHCH=C(CO,H)CH,CO,H
acidr461),which is then degraded further.r481An analogous
reaction sequence in which the C5' hydroxymethyl group does
not undergo oxidation but is preserved prior to ring cleavage
would generate an intermediate that might account for the
biogenetic relationship between pyridoxol and the pyrimidine
unit of thiamin which appears to be indicated by the tracer
3. Biosynthesis of the Thiazole Unit
Isotopic feeding experiments performed to identify the primary precursors have firmly established that distinct routes are
used by enteric bacteria and yeast for the biosynthesis of the
Angeir. Chern. h i . Ed. Engl. 1997, 36, 1032-1 046
Biosynthesis of Vitamin B,
thiazole unit. In E. coli and S. typhimurium, tyrosine and a deoxypentulose provide the nitrogen and carbon atoms of the
thiazole moiety, whereas glycine and a pentulose are the precursors in yeast (Scheme 6). Other available evidence indicates that
3sS from [35S]sulfate, but many other sulfur containing substances, including Na,S, were also effective in lowering 35Sin-
3.2. Glycine and Tyrosine as Precursors
- glUCOse
Scheme 6. The biosynthesis of the thiazole unit (4) in facultative anaerobic
prokaryotes ( E (,oliand other enterobacteria, as well as higher plants, left) and in
eukaryotes (aerobic bacteria. yeasts; right). Precursor-product relationships are
shown. For the steps between 14 15 and 4. see Scheme 7 and for those between
11 and 4. Scheme X.
green plants utilize the same pathway as enterobacteria, whereas
the yeast pathway is employed by aerobic bacteria (Bacillus subtilis and Pseudomonasputida). The source of the sulfur
atom is not certain, but is assumed to be cysteine. Chemically
rational hypotheses are available to account for the formation
of the thiazole unit from these substrates (see Schemes 7 and 8),
but intermediates of the pathway have not been identified.
3.1. The Source of the Sulfur Atom
Incorporation efficiencies and competition experiments have
been employed to probe this unsolved problem, but definitive
conclusions will be possible only on the basis of results of future
enzymological and genetic experiments. Cysteine is generally
regarded as the precursor of the sulfur atom in enterobacteria.
Label from [35S]cysteine o r [35S]methionine is incorporated
with equal efficiency in E. coli ATCC 9637,[491but higher incorporations of cysteine in E. coli I F 0 13168[50,511 have been reported. Cysteine more effectively dilutes the incorporation of
label from [35S]sulfate in E. coli mutant 70-17[521and from
[34S]thiocysteine in E. coli K-12 hpb>.-.[531 However, in
S. typhimur.iun7ATCC 23592 methionine dilutes the incorporation of ~ysteine.1'~'
Radioactivity from ~ - [ ~ ~ S ] c y s tand
i n e~ - [ ~ ~ S ] m e t h i o nisi nine
corporated into thiamin by S. cerevisiue. Evidence on the relative efficiency of incorporation from the two substrates is confli~ting.[~*,
s ' ] In S. ceuevisiae NCYC 1062, L-methionine was
more effective than L-cysteine in diluting the incorporation of
Angeit. Cheni. In! G l . E n ~ l1997, 36. 1032-1046
In yeast and aerobic bacteria a C N fragment, derived from
glycine by decarboxylation, serves as the precursor of the C2,N
fragment of the thiazole moiety. The label from ['5N]glycine
was incorporated into the thiazole unit by B. subtilis and
P. put id^,[^^] and in S. cerevisi~e[*~.
5 6 1 and several other euk a r y o t e ~ . [ The
~ ~ I only carbon atom labeled by [a- ''C]glycine in
B. subtilis ATCC 6633 is C2 of the thiazole moiety.[571Incorporation of label from [a-14C]glycine into C2 of the thiazole unit
was demonstrated by chemical degradation also in several
", 5 8 , 591 The carboxyl group of
strains of S . cerevisiae
glycine was not incorporated into thiamin.[27,591 Intact incorporation of the l S N - l 3 C unit of [15N,2-'3C]glycine into the
N3,C2 fragment of the thiazole unit was recently demonstrated.[12]
By contrast, label from [~-'~C]glycine
is not incorporated into
the thiazole unit by S. ryphimurium,[601and [' 'Nlglycine is not
incorporated into the thiazole unit by either E. coh' or Enterobucter a e r o g e n e ~ . ~However,
the nitrogen and the a-carbon of
another amino acid, tyrosine, provide the C2,N fragment of the
thiazole unit. ["N]Tyrosine supplies the nitrogen atom of the
thiazole moiety in E.
and E. ~ e r o g e n e s ,and
~ ~ ~label
from [r-14C]tyrosine(14) is incorporated into C2 of the thiazole
in E. c 0 1 i [ ~and
~ ] S. fyphimurium.[601This label entered in the
presence of unlabeled glycine and methionine, making their participation in the biosynthetic construction of the unit unlikely.
Incorporation of radioactivity from [~-'~C]tyrosine
into thiamin has also been reported in spinach leaf chloroplasts, but the
position of the label was not established.[631In E. coli only C2
of the thiazole unit is labeled by [U-'4C]tyrosine,[6221which indicates that the a carbon atom of tyrosine is the only one that is
incorporated into the thiazole unit. The intact incorporation of
the a-C,N unit of tyrosine is likely, but not as yet demonstrated
by experiment. Deuterium from [3',5'-2H,]tyrosine is incorporated into p-hydroxybenzyl alcohol (16)in E. coli B. and the
correlation observed between p-hydroxybenzyl alcohol and thiamin synthesis suggests that the phenolic alcohol is formed during thiamin b i o s y n t h e ~ i s . ' ~ ~ '
N o incorporation of label from ["N]tyrosine into thiamin
was observed in B. subtilis o r in P. put id^,['^] nor was label from
[a- 14C]tyrosine incorporated into thiamin in S. c.erevisiue.[261
The mutually exclusive incorporation results of glycine and tyrosine support the existence of different routes for thiazole
biosynthesis: one in aerobic bacteria and yeast, and another in
enterobacteria and plants (Scheme 6).
3.3. Carbohydrate Precursors in Enteric Bacteria
The initial evidence for the origin of the C, chain of the
thiazole unit comes from incorporation studies in E. coli B with
samples of glucose, glycerol, and pyruvic acid, labeled with
stable isotopes.[651 The multiple incorporation of deuterium
from [l-ZH]glucose and the incorporation of label from [U1039
I. D. Spenser and R. L. White
'3C,Jglucose indicated that a C, unit (giving rise to C5-C7)
extracts.1711 Pyruvate dehydrogenase (EC purified
combines with a C , unit (giving rise to C4,C4). The thiazole
from B. subtilis IF0 13719 and E. coli was subsequently shown
carbons derived from the C, unit were also labeled by [6,6to catalyze the condensation of pyruvate with D-glyceraldehyde.I7'I
2H,]glucose and [5,6,6-2H,]gIucose,[6sJand the incorporation
of (R)-I(RS)-1-2H]glycerol and of (R)-[(RS)-1-2H,1-'80]The pyruvate dehydrogenase-catalyzed formation of the thiaglycerol provides support for the three-carbon precursor.[661
zole precursor 1-deoxy-D-xylulose implicates thiamin pyrophosphate as a required coenzyme in the thiazole pathway of thiamin
That the C, unit is derived from pyruvate is supported by the
intact incorporation of the CD, group of [3-ZH,Jpyruvate,r651 biosynthesis. Although this is remarkable, it is not unique; both
and the incorporation of label from [U-14C]alanineand [3-14C]pyridoxine[731and biotin1741are also implicated in their own
and [2-14C]pyruvaters'1in E. coli. However, [U-'4C]alanine was
biosyn thesis.
Thus in enteric bacteria, the thiazole unit is formed from
not incorporated in S. typhinzuriurn ATCC 23592.[601
Very recently conclusive evidence for the derivation of the
tyrosine (14), 1-deoxy-D-xylulose (15) (or 1-deoxy-D-xylulose
Sphosphate), and a sulfur donor (e.g., cysteine (17)). Initial
five-carbon unit from two smaller fragments derived from gluSchiff base (18) formation followed by a series of tautomerizacose became available : The incorporation into the thiazole unit
of thiamin in E. coli B, mutant WG2, of two intact multicarbon
tion, decarboxylation, elimination, and addition reactions
units derived from ~-[1,2,3,4,5,6-'~C,]glucose
(a C, unit, entergenerates the thiazole unit and p-hydroxybenzyl alcoholr641
ing C4,C4, and a C , unit, entering CS-C7) was demonstrated
(Scheme 7). With 1 -deoxy-D-xylulose as a precursor, an oxidaby 13C NMR.[l2]
tive step is required. This can be accommodated by oxidative
The C, compound that is derived by combination of the C,
cleavage of tyrosine, as shown in Scheme 7, or by oxidation at
C3 of the deoxyxylulose unit and a subsequent reductive fragand the C, fragment and serves as the direct precursor of the
mentation of tyrosine analogous to the deprotonation step in
five-carbon chain of the thiazole unit has been identified in
the glycine pathway (Scheme 8). An investigation ofp-hydroxyE. coli as 1-deoxy-D-xylulose(15) . [ 6 7 + Intact incorporation of
benzyl derivatives of thiamin and the thiazole moiety indicated
[1-2H, ,5-ZH,]-l-deoxy-~-xylulose
was indicated by the presence of deuterium at C 4 and C7 of
that in E. colimutants cleavage of thep-hydroxylbenzyl unit is
the thiazole unit.[671No incorporation of the corresponding
associated with thiazole formation and not with later biosynD-er-ythro isomer was observed.[68J
thetic steps.r751
Very recently conclusive evidence was obtained in support of
Minor modifications of this reaction sequence (Schemes 9
the intact incorporation of 1-deoxy-D-xylulose into the thiazole
and 10) are required to account for the formation of two E. coli
moiety of thiamin: it was
shown by I3C NMR spec22
/troscopy that the intact
I3C-''C bond of a sample
[2,3-'3C,]-1-deoxy-~x y l ~ l o s eentered
~ ~ ~ ~the C,
unit C4,C5 of the thiazole
moiety of thiamin in
E. coli B mutant WG2.f131
Related investigations have
demonstrated that this pentulose is also an advanced
precursor of pyridoxine, anHO
other B vitarnin.[13~45]
1-Deoxy-D-xylulose has
been isolated from cultures
of Streptomyces hygroscopicus UC-5601
and from
partially purified cell-free
extracts of severaI bacterial
species that had been incu19
bated with pyruvate (12),
D-glyceraldehyde (13), and
thiamin pyropho~phate.[~'I
The formation of l-deoxyD-XyhIOSe was detected by
paper chromatography in
cell-free extracts of several
other bacteria, but signifi4
cantly smaller amounts Scheme 7. A hypothetical mechanism for the biosynthesis of the thiazole unit in facultative anaerobic prokaryotes, from L-tyrosine
formed in yeast and fungal (14) and 1-deoxy-D-xylulose(15); for the steps between 18 and 22, see Scheme 10 and for those between 19 and 21, Scheme 9.
Angew. Chem. hi.Ed. Engl. 1991.36, 1032-1046
Biosynthesis of Vitamin B,
from [l-'4C]fructose and [l-'4C]glycerol, determined by chemical degraOH
dation, indicated that the five-carH02C
bon unit of the thiazole moiety is
formed by way of both the oxidative
and the nonoxidative branch of the
pentose phosphate pathway.[791 A
similar pattern of incorporation was
obtained when [I-'"C]-, [2-'4C]-, and
[6-'4C]glucose were administered to
C. ~ t i l i s . 801
[ ~ ~ ,Furthermore, an
experiment with [6-'4C,6-3H]gluCH$OC@H 12
C O S ~ [ 791
~ ~ in
' S. cerevisiae indicated
intact incorporation of the double la-L
be1 into C7 of the hydroxyethyl side
chain, but incorporation of 14C label
' 20
only into the methyl group ( C 4 ) .
" C02H
These results are not consistent with
Scheme 8. A hypothetical mechanism for the biosynthesis of the thiazole unit in eukaryotes. from glycine and a
the intermediacy of l-deoxy-D-xy~uu-pentulose (11)
lose. They are consistent with the
intermediacy of either D-xylulose or D-ribulose (or their 5-phosphates). Thus, the probable first step on the route to the thiazole
in yeast is the formation of a Schiff base (20) of glycine with
a D-pentulose o r a D-pentulose 5-phosphate (Scheme 8).
Scheme 9. Derivation of the thiazolecdrboxylic acid (21) as a by-product of the
thiazole pathway in lacultative anaerobic prokaryotes.
metabolites, 5-(2-hydroxyethyl)-4-methylthiazole-2-carboxylic acid (21)[761
and 5-( 1,2-dihydroxyethyl)-4-methylthiazole (22),[77]that are structurally related to the thiazole
unit. Label from [~arboxyl-'~C]tyrosine
is incorporated into 21. These compounds are probably by-products rather
than intermediates of thiamin biosynthesis.
+ NH3
3.4. Carbohydrate Precursors in Yeast
The five-carbon chain of the thiazole unit in S . cerevisiae ATCC 24903[78.791 also originates from carbohydrate sources. Labeling data suggest, however, that it is not
formed by combination of a C, with a C, unit via l-deoxyD-xylulose but is generated from an intact pentulose. The
evidence, in S. cerevisiae, is of necessity indirect, since yeast
does not grow on a medium of pentoses, and incorporation
of labeled pentoses can thus not be tested directly. The
indirect evidence is strong, however. Not only was
label from [3-'4C]pyruvate,15 791 [U-'4C]alanine,[s'1 and
[U-'4Ci-iactater791not incorporated, but the labeling pattern from [1-I4c]-, [2-I4C]-, and [6-'4C]glucose, as welt as
Angen.. Chwn. h t . E d Engl. 1997. 36. 1032-1046
Scheme 10. Derivation of dihydroxyethylthiazole (22) as a by-product of the thiazole
pathway in facultative anaerobic prokaryotes.
I. D. Spenser and R. L. White
A sequence of reactions, analogous to that leading to the
product from tyrosine/deoxyxylulose (Scheme 7), generates the
thiazole unit, but an oxidative step is not required, because a
pentulose precursor rather than a deoxypentulose is involved in
the process.
In confirmation of the inferences based on the tracer work
with radioactive isotopes, the incorporation into the thiazole
unit of thiamin in S. cerevisiae ATCC 7752 (= I F 0 1234) of an
intact C, unit derived from D-[I ,2,3,4,5,6-'3C,]glucose, entering
the C, chain, C4,C4-C7, was demonstrated by I3CNMR.l'Z1
4. Coupling of the Two Subunits and
Phosphorylation of Thiamin
Identification of the various phosphorylated derivatives that
are intermediates of the biosynthesis of the vitamin (6)and of its
pyrophosphate ester, cocarboxylase (S), and the isolation and
enrichment of the enzymes catalyzing the various steps of the
process in E. coli and S. cerevisiue, was accomplished by
l o , ''1 These final steps of cocarboxylase biosynthesis are
presented in Scheme 11. Relevant literature on the coupIing of
the subunits and the phosphorylation steps leading to 8 is summarized in Table 1, and only the more recent contributions are
described below.
4.1. Enzymes
Enzyme activities for the phosphorylation and coupling of the
heterocyclic subunits and for the phosphorylation of thiamin
and thiamin monophosphate were detected in crude extracts
prepared from mutants of E. coli, and the hydroxymethylpyrim-
thiamin phosphate
(thiamin phosphate
synthase)(3+5 -7)
[75,81,82][a] [83,84,85l[b]
hydroxymethylpyrimidine kinase (1 2 )
2.7.1 S O
kinase (4 -t 5 )
2.7 4.7
phosphomethylpyrimidine kinase (2 + 3)
thiamin phosphate
kinase (7 + 8)
thiamine kinase (6 -7)
thiamin pyrophosphokinase (6 -, 8)
thiamin phosphate
phosphdtase (7 + 6 )
[l 01,102,103][d]
idine kinase (EC purified 3 3 0 0 - f 0 l d . ~Thiamin
phosphate pyrophosphorylase (EC purified from
recombinant E. ~ o l i , " ~and
~ an enzyme preparation possessing
both hydroxyethylthiazole kinase (EC thiamin
phosphate pyrophosphorylase activity (EC was isolated from S. c e r e v i . s i ~ e .This
[ ~ ~ ~bifunctional protein is an octamer
of 60 kDa subunits and appears to have separate catalytic sites
for the two activities. Thiamin phosphate pyrophosphorylase
and thiamin phosphate phosphatase from the plant Arubidopsis thulium were separated and partially purified by gel filtraTH/4, thi2, nmf2 (yeast)
thiA (P. sativum)
thi3, nmtl (S.
thiC (P. sativumf
Table 1 Literature summary of the biosynthesis of thiamin pyrophosphate (8):
enzymes of the phosphorylation steps and of the coupling steps of the two subunits.
TH16, thi4
thb, thiE
TH16, thi4 (yeast) 5
fhd, th-7 (plants)
C b
Scheme 11. Summary of the final
stages of the biosynthesis of thiamin
(Vitamin B,. 6) and its pyrophosphate ester (cocarboxylase, 8 ) . including the enzymes and genes
( E . coli genes unless otherwise mdicdted) that are implicated in individual steps.
Chein. In!. Ed Engl 1997, 36, 1032-1046
Biosynthesis of Vitamin B,
Table 2. Genes of thiamin pyrophosphate (8) biosynthesis
E. coii
S. cerevisiae
XVI [115] 540[ 1 151
518 [ 1071
XVI[115] 54011 151
( n t f l +) [128]
627 191
NO295 I1051
thiazole pathway
90 PI
thiazole pathway
thiazole pathway
thiazole pathway
pyrimidine kinase
pyrimidine P kinase
thiazole kinase
rl7iG [9]
t / d [ 1121
90 [9]
9.5 [ 1 121
subunit coupling
thiL[88,118] 9[88,118]
rhiK[88,1181 25[88,118]
P kinase
PP kinase
thl3 11021
(nmt/) [ 1061
thi2 [102,110]
(nm12)[1 111
pyrimidine pathway
340 [lo51
326 [l 081
21 1 [9]
regulatory gene
regulatory gene
P H 0 3 [ 101,119,120] I1 [121]
TH/80[ 1231
(R) [I231
( P H 0 6 ) [ 1251
TH13[ 1261
IV [126]
3 19 [1 231
569 [ 1 161
la] Chromosome location: L and R represent the left and right arms of the chromosome [b] Number of amino acid residues in the polypeptide encoded by the gene
t i ~ n . [ ~A' ]glycosylated acid phosphatase (EC has been
purified approximately tenfold from S. c e r e v i s i ~ eand
[ ~ ~from
~ a
transformed strain of S . pornbe."".
These enzymes were
shown to act as thiamin phosphate phosphatases.['''. 1''
deglycosylated enzyme retained acid phosphatase activity and
had a molecular mass of 56 kDa by SDS-gel electrophoresis.
spanning domain. Four genes tnrl, tnr2, tnr3, and thif regulate
ihi3,['291and the promoter of nmtl is regulated by ntff +,a gene
identical with thif that encodes a protein homologous to transcriptional factors containing the characteristic Cys, zinc finger.r'28]For the higher plant Pisurn sativum, a gene in the pyrimidine pathway, thiC, has been mapped to chromosome 3,[1301
and an alt mutant is normalized by supplying the pyrimidine
5. Genetics of the Biosynthesis of Thiamin
In addition to the diversity of pyrimidine pathways described
in Section 2, recent genetic studies have indicated that a route to
(5-aminoimidazo1e)ribotide distinct from the purine pathway
for nucleic acids permits S. iyphimurium mutants blocked in
(5-aminoimidazole)ribotide formation to grow without the addition of exogenous thiamin.['321The alternative route is expressed when the bacterium is cultured anaerobically[' 33aI or
aerobically on certain carbon sources.[133b1 A gene from this
pathway has been cloned and sequenced, and sequence analysis
indicates that the gene product contains an NAD/FAD binding
pocket.['341More recently a second gene has been identified and
assigned to an alternative pathway for the conversion of
(5-aminoimidazoly1)ribotide into the pyrimidine
Genetic studies of the pathways to the two subunits of thiamin are at a preliminary stage. Neither genes nor the corresponding gene products have been unambiguously assigned to
specific steps in the pathways. By contrast, the genes encoding
the coupling enzyme and the kinases involved in the final steps
of thiamin pyrophosphate biosynthesis have been identified in
E. coLiand the eukaryotes S. cerevisiae and S. pombe. The genetic information collected for these organisms is summarized in
Table 2.
5.1. Genetics of the Route to the Pyrimidine Unit
Single genes in the pathway to the pyrimidine unit of thiamin
have been described in several organisms and characterized to
different extents. The E. coli gene, thiC, was sequenced and
shown to complement pyrimidine-requiring mutants.['] In the
eukaryote S. pombe, thi3 mutants required the pyrimidine unit
for growth,['021and the thi3 gene is allelic to the thiarnin-repressible gene nmtl, which has been cloned and sequenced.['061
Homologues of nmtl have been found in Aspergillus parasiticus
and S. ~ e r o ~ i . ~ i u e .The
~ ' ' ~amino
acid sequences of the three
eukaryotic gene products are more than 60% identical, and
each contains a cysteine cluster within a predicted membraneAnRnv. Chen?. fnr. Ed. EngI. 1997. 36, 1032-1046
5.2. Genetics of the Route to the Thiazole Unit
Thiazole genes have been located in a cluster of tightly linked
thiamin genes on both the E. c01i[~~
and S. t ~ p h i r n u r i u m [ ' ~ ~ ]
chromosomes at 90 min. Transcription of the S. typhimurium
cluster is regulated by thiamin, and the E. coli cluster (Table 2)
has been cloned and sequenced. The thiF, thiG, and thiH mutants did not respond to 1-deoxy-o-xylulose; these genes are
therefore required for synthesis of the thiazole unit from this
precursor. The measured molecular masses for the polypeptides
expressed from thiF and thiG were considerably smaller than
I. D. Spenser and R. L. White
those predicted from the DNA sequences.[g1The predicted sequence of the rhiFpolypeptide was homologous with MoeB, the
protein thought to be involved in sulfur insertion in the biosynthesis of molybdopterin. The genetic evidence indicates that at
least four enzyme-catalyzed steps are required for thiazole
biosynthesis in enteric bacteria, but the nature of the individual
steps remains to be determined:
Single genes for the thiazole pathway (Table 2) have been
identified in several eukaryotes. A thiazole gene in P. sativum,
thiA, has been mapped on chromosome 2,[1301and the sequences of the genes thif (Zeumays),[1371thi2 or nmt2
( S . pombe),["ll and TH14 or M O L f ( S . cerevisiue)"Osl are
homologous to that of the Fusarium gene sti35, which encodes
a stress-inducible protein.[1381A putative dinucleotide binding
site is present in each deduced amino acid sequence, suggesting
that an NAD or FAD enzyme is employed in the thiazole pathway." 371 Expression of thi2 is repressed by either thiamin or the
thiazole unit and controlled by four regulatory genes, t n r f , tnr2,
and the allelic nmt2 gene is coordinately
tnr3, and thil,["Og
regulated with nmtl.[llll
5.3. Genetics of the Coupling Process
The genes encoding the coupling enzyme and the five kinases
involved in the final steps of thiamin pyrophosphate biosynthesis are located in four regions of the E. coli chromosome
(Table 2). The thiE gene was initially assigned to the thiazole
but subsequently shown to code for an enzyme that
couples the phosphorylated
It has been sequencedrgl and overexpressed in E. coli BL21 (DE3).[751The
overexpressed polypeptide catalyzes the coupling of the heterocyclic subunits (3 and 5), but a thiazole unit with an attached
p-hydroxylbenzyl group is not a substrate; the cleavage of this
fragment of the tyrosine precursor (Scheme 7) thus occurs prior
to the subunit coupling step. An E. coli transformant carrying
the thiK (thiamin kinase) and thiL (thiamin monophosphate
kinase) genes had elevated levels of thiamin phosphates.[1391
Less genetic information is available for other bacteria, but a
B. subtilis operon containing thiamin genes has been seq ~ e n c e d , " ~and
~ ] genes that complement E. coli thi mutants are
located on a megaplasmid in Rhizobium m e l i l ~ t i . " ~ ' ~
Unlike in E. coli, where individual genes and enzymes have
been characterized for the phosphorylation and subunit coupling steps, the eukaryotic genes THZ6 (S. cerevisiae)[' * '] and thi4
( S .pombe)[102,
encode bifunctional enzymes possessing
thiazole kinase and thiamin phosphate pyrophosphorylase activities. The molecular mass calculated for the THI6 gene
product matches that measured for the purified enzyme,[*'] and
regions of the amino acid sequence are homologous to those
predicted for the E. coli thiE and thiF gene products.["'] The
S. pombe gene thi4 is repressed by thiamin and, to some extent,
by the thiazole unit.[1071
The genes coding for a thiamin-repressible acid phosphatase (EC 3 . 1 . 3 . 2 ) , P H 0 3 ( S . cerevisiae)L'o'*llgland pho4
( S .pombe),"O'*103.1zz1 have been sequenced and show a similarity of only 27%.L1031
However, each has a much higher homology with their corresponding phosphate-repressible acid
For each yeast,
phosphatase genes, PH05'1191and phof
the gene sequences indicate that the phosphate- and thiamin-repressible acid phosphatases have cleavable amino acid signal
sequences and identical numbers of potential glycosylation
Gene disruption experiments demonstrated that the thiamin pyrophosphokinase (EC genes TH180 (S. cerev i s i ~ e ) [ and
' ~ ~ ~trn3 ( S .pombe)["61 are essential for growth.
Reduced levels of thiamin pyrophosphokinase activity were detected in thi80[1431and trn3"
mutants, and the carboxylterminal, amino acid sequence of the trn3 polypeptide is 31 70
identical[1161 with the shorter polypeptide encoded by
Thiamin biosynthesis in yeast is controlled by several regula12'1
tory genes. Two positive regulatory genes TH12(PH06)r124*
and TH13[1261in S. cerevisiae are required to express both the
thiamin-repressible acid phosphatase and the enzymes required
for phosphorylation and coupling of the pyrimidine and thiazole units.['25In S . pombe, three negative regulatory
genes (tnrf, tnr2, and tnr3) and a positive regulatory gene
(thif)['07. l Z 9 ] control pho4 and chi4 together with thi3 (pyrimidine pathway) and thi2 (thiazole pathway). The tnr3 gene
product, thiamin pyrophosphokinase, has both a catalytic and
a regulatory role.[1161The sequence of the positive regulator thif
shows homology with S . cerevisiae transcription factors that
contain a Cys, zinc-finger domain.[1271
In higher plants, the th-1 mutant of A . thaliana is deficient in
thiamin phosphate pyropho~phorylase,[~~]
and the P. sativum
gene, thiB, has been mapped to chromosome 6.[1301Since thiB
mutants responded only to supplements of intact thiamin, the
this gene product is associated with subunit coupling or a later
step of the pathway.
6. Summary and Outlook
Investigations in plants and several species of microorganisms
have uncovered biochemically diverse routes for the biosynthesis of thiamin.
Most of the enzymes and genes associated with the phosphorylation and coupling steps (Scheme 11) have been identified.
Thiamin pyrophosphate is formed directly from thiamin
monophosphate in enteric bacteria and from thiamin in yeast
and aerobic bacteria. Phosphorylation of the thiazole unit and
coupling of the subunits is catalyzed by a single enzyme in yeast,
whereas distinct enzymes for each step are employed in enteric
bacteria. At present very little mechanistic information is available for these enzyme-catalyzed reactions.
In yeast and enteric bacteria, different primary precursors for
each heterocyclic subunit have been identified by isotopic tracer
experiments. Although the exact pyrimidine precursors have not
been established in yeast, the evidence strongly supports two
distinct pathways: one in yeast and the other in enteric bacteria.
Less information is available concerning the identity of the primary precursors of the two subunits in aerobic bacteria and in
higher plants.
The pathway to each subunit is likely to proceed through
several enzyme-catalyzed steps, but no committed intermediates
have yet been identified nor have pathway specific enzymes been
detected. Moreover, Only a few genes have been located. BeAngrn. Chem. In<.Ed. Engl. 1997, 36, 1032-1046
Biosynthesis of Vitamin B,
cause only minute amounts of thiamin are formed in microorganisms, it is likely that further progress in this area will be
obtained only by studying the overexpressed gene products.
In our view, a combination of genetic, enzymological, and
chemical (isotopic and synthetic) approaches is needed to establish the intermediates and the mechanistic details of the biosynthetic pathways to the pyrimidine and thiazole units. It is our
hope that this review of the current state of knowledge will
encourage researchers to employ combinations of these methods to solve the difficult problem of thiamin biosynthesis.
Research grantsfrom the National Institute of General Medical
Sciences, U . S. Public Health Service (GM50778 to I. D. S . ) and
from the Natural Sciences and Engineering Research Council of
Canada ( t o R . L. IK)are gratefully acknowledged.
Received: March 18, 1996
Revised version: October 28, 1966 [A189IE]
German version: Angew. Chem. 1997,109, 1096-11 11
[I] B. C. P. Jansen in The Viraniins Chemisrry. Physiology, Pathology, Methods,
& I / . 5 (Eds.: W. H. Sebrell, Jr.. R. S. Harris), 2nd ed., Academic Press, New
York. 1972,pp 99-102.
[2] a) R. R Williams, R. D. Spies, VifaminB, (Thiamin) andirs Use in Medicine,
MacMillan. New York, 1938,pp. 138-143; b) ibid.. pp. 144-157; c) ibid., pp.
123 129: d ) ihid., pp. 134-135.
[3] Other accounts of the discovery [2c], isolation [2a], structure elucidation. and
synthesis [Zb] of thiamin can be found in many monographs and reviews, for
example a) E. P. Steyn-Parve, C. H. Monfoort, in Comprehensive Biochemirrrr, Val. 11. (Eds.: M . Florkin, E. H. Stotz) Elsevier, Amsterdam 1963,pp.
3 - 22: b) W. Friedrich, Vitamins, Walter de Gruyter, Berlin, 1988. pp. 341 346.
(41 K. Lohmann, P. Schuster, Naturwisseiischaften 1937,25,26-27.
[51 The term thiamin was proposed in 1937 [2d] and recommended by the International Union of Biochemistry in 1965 for6. IUPACInforrnarionBulletin N o
25. February 1966,p. 19 or Eur. J Biochem. 1967,2,1-2. This name supersedes earlier designations (e.g., anti-beriberi vitamin, aneurin).
161 G . Moine. H.-P Hohmann in UllmannS Encyclopedia of’lndusrrialChemisrry,
&J/. 27. 5th ed., (Eds.: B. Elvers, S. Hawkins), VCH, Weinheim, Germany
1996. pp 506-521.
[7] National Research Council (U. S.). Subcommittee on the 10th Edition of the
RDAs Food and Nutrition Board Commission on Life Sciences, Recommendrd Diefar).Allowances, 10th ed., National Academy Press, Washington, 1989,
pp. 125 132
[8] a ) G . M. Brown in Merabolic Pathways, Val. 4 (Ed.: D . M. Greenberg), 3rd
ed.. Academic Press. New York. 1970,pp. 369-381, b) ComprehensiveBiodtenti,stri. R J ~
2f. (Eds.: M. Florkin, E. H . Stotz), Elsevier, Amsterdam, 1971,
pp. 1 - 9. c) 1. G . Leder in Metabolic Pathxays, Vol. 7(Ed : D. M. Greenberg),
3rd ed.. Academic Press, New York, 1975,pp. 57-85.
[9] P. 8. Vdnder Horn, A. D. Backstrom, V. Stewart, T. P. Begley, J Bacterial.
[lo] D.W Young. Nur Prod. Rep. 1986,3, 395-419.
[ l l ] G. M. Brown. J. M. Williamson in Escherichia coliand Saimonella ryphimuriurn CeUidar and Molecular Biolop, Vol. 1 (Ed.: F C. Neidhardt), American
Society for Microbiology, Washington, 1987,pp. 521 -538.
[I21 K. Himmeldirk. I D. Spenser, unpublished results.
[13] K Himmeldirk, I. A. Kennedy, R. E. Hill. B. G. Sayer, I. D. Spenser, Chem.
Crtmmun. 1996. 1187-1188.
[I41 See footnotes 6 and 7 in G. Grue-Ssrensen, I. D. Spenser, Can. J Chem. 1989,
67,998 -1009.
[15] R . E. Hill. I . D. Spenser in Viramin 86 Pyridonal Phosphate. Chemical, Bio~ h m ~ i c and
d . Medical Aspecrs, Purr A (Eds.: D. Dolphin, R. Poulson, 0 .
Avramocic), Wiley. New York, 1986,pp. 417-476.
[16] M. J Pine, R. Guthrie. J Bacreriol. 1959,78,545-549.
1171 H. Kumaoka, G. M. Brown, Arch. Biochem Biophys. 1967,122. 378-384.
[I81 B. Estramareix, M . Lesieur, Biochim. Biopltys. Acra 1969,192,375-377.
(191 K . Ydmada. M. Morisaki, H. Kumaoka, Biochim. Biophys. Aria 1983,756,
41 48
1201 S David. 8. Estramareix, H. Hirshfeld, Biochim. Biophys. Acra 1967, 148,
1 1 21.
[2l] K. Uchida, K. Yamada, H. Kumaoka, Biramin 1982,56,235-240.
1221 G. Grue-Ssrensen, R. L. White, I. D. Spenser, J Am. Chem. Soc. 1986,108,
146- 15X.
[23] P.C Newell, R. G. Tucker, Biochem. J . 1968,106,271-277.
(241 B. Estramareix. Biochim. Biophys. Acra 1970,208,170-171.
Angiw. Chrnt. I n / . Ed. Engl. 1997,36, 1032-1046
[25] R. H. White, F. B. Rudolph, Biochemistry 1979,18. 2632- 2636.
[26] R. L. White, I. D . Spenser, Biochem. J . 1979. 179, 315-325.
1271 P. E. Linnett, J. Walker, Biochem. J. 1968,109,161 - 168
K. Yamada, K. Tazuya, H. Kumaoka, Bitamin 1990, 64. 295-299.
K. Tazuya, M. Tanaka, M. Morisaki, K. Yamada. H . Kumaoka, Biochem. l n t .
1987, 14, 769-777.
For a recent summary of the early literature, see: T. Kozluk, I. D. Spenser, J.
Am. Chem. Sac 1987,109,4698-4702.
P. C. Newell, R. G . Tucker, Biochem. J . 1968,106.279-287.
B Estramareix, S. David, Biochim. Biophys. Acra 1990. 1035. 154-160.
a) K. Yamada, H. Kumaoka, Biochem. Inr. 1982.5,771-776; b) J Nurr. Sci.
Viraminol. 1983,29,389-398.
B. Estramareix, M. Therisod, J Am. Chem. Soc. 1984,106. 3857-3860.
B. Estramareix, S. David, Biochem. Bi0phy.s Res. Commim. 1986,134,1136
K. Tazuya, M . Yamamoto, M. Hayashiji, K. Yamada, H. Kumaoka, Biochem.
Int. 1986,12, 661-668.
W. H. Stratling, FEBSLerr. 1970,9,27-29.
a) K . Tazuya. K . Yamada, H. Kumaoka. Biochim. Biophj,~.Acra 1989,990,
73-79; b) K. Tazuya, C. Azumi, K. Yamada, H. Kumaoka. Biramin 1993,67.
681 -688.
K. Tazuya, K. Yamada. H. Kumaoka, Blochem. Mol 3101.Inr. 1993,30,
a) K. Tazuya, C. Azumi, K. ydmada, H. Kumaoka, Bioc.hem. Mol. Bid. I n / .
1994.33, 769-774; b) ibid. 1995.36. 883-888.
K. Yamada, personal communication.
R. L. Baxter, A B. Hanley, H . W. S. Chan, J Chem. Sac. Perkin Trans. 1
1990,2963 - 2966.
D. M. Greenberg in Metabolic Parhwuys, Vol. 3 (Ed.: D. M.Greenberg), 3rd
ed., Academic Press, New York, 1969,pp. 129-136.
V. Herbert, J. R. Bertino in The Vitamins Chemisrr).,P/ij..siologj,Pathology,
Merhods, Val. 7 (Ed.: P. Gyorgy, W. N . Pearson), 2nd ed., Academic Press,
New York, 1967,pp. 243-276.
I. D. Spenser, R. E. Hill, Nut. Prod. Rep. 1995. 12,555--565.
L. G. Sparrow, P. P. K. Ho, T. K. Sundaram, D Zach, E. J. Nyns. E. E. Snell,
J Bid. Chem. 1969,244,2590-2600
a) T. Yagr, G. M. Kishore, E. E. Snell. J. Biol. Chem 1983,258,9419-9425;
b) M.J. K . Nelson, E. E. Snell, ihid. 1986,261, 15 1 15 I5 120.
M. S. Huynh, E. E. Snell, J Biol. Chem. 1985,260.2379 -2383.
H. Kumaoka, J C’itaminol. 1963,9, 188-190.
K . Tazuya, K. Yamada, K . Nakamura, H. Kumaoka, Biochim. Biophys Acru
K. Yamada, Birumin 1992.66, 1-24.
B. Estramareix, D. Gaudry, M. Therisod, Biochimie 1977,59,857-859.
E. DeMoll, W. Shive, Blochem. Biophys. Re.%Cummun. 1985. 132,217-222.
E. Bellion, D . H. Kirkley. Biochim. Bioph?~.Acra 1977.497, 323-328.
C. H.S. Hitchcock, J. Walker, Biochem. J. 1961,80. 137 -148.
K. Tazuya, M. Morisaki, K. k’dmada, H. Kumaoka. K Saiki. Biochem. Int.
1987,14, 153-160.
R. L. White, unpublished results.
P. E. Linnett, J. Walker. J Chem. Sac. C 1967. 796 -799.
P. E. Linnett, J. Walker, Biochim. Biophys. Acra 1969. 184. 381 -385.
E. Bellion, D . H. Kirkley, J. R. Faust, Biochim. Biaphw A r m 1976. 437,
229 -237.
R. H. White. F. B. Rudolph, Biochim. Biopbys. Acra 1978,542,340-347.
B. Estramareix, M. Therisod, Biuchim. Biup/?j.s.Acra 1972,273,275-282.
J.-H. Julliard, R. Douce. Proc. Narl. Acad. Sci. USA 1991.88,2042-2045.
R H. White, Biochim. Biopltys. Acru 1979,583. 55-61.
R. H. White, Biochemi.srrj, 1978,17,3833-3840.
R. H. White, Experrentin 1980,36,637-638.
a) S. David. B. Estramareix, J-C. Fischer, M. Thcrisod, J. Ant. Chem.
Sac. 1981. 103, 7341-7342; b) J Chem. Soc. Perkin Trans 1 1982, 213121 37.
M. Therisod, J:C. Fischer, B. Estramareix. Biochem. Biophys. Res. Commun.
I. A. Kennedy, T. Hemscheidt, I. F. Britten, I. D. Spenser, Can. J Chem. 1995,
73, 1329-1337.
a) L. Slechta. L. E. Johnson, J Antihior. 1976,29,685.687; b) H.Hoeksema,
L. Baczynskyj, ibid. 1976,29, 688-691.
A. Yokota, K.-I. Sasajima, Agric. Biol. Chem 1984.48. 149-158.
A. Yokota. K.-I. Sasajima. Agric. B i d . Chem. 1986.50,2517-2524.
C.Drewke, M. Klein. D. Clade, A. Arenz. R. Miiller, E Leistner, FEES Lert.
1996.390,179- 182.
I. Sanyal. S.-L. Lee, D. H . Flint, J Am. Chem. Sac. 1994. 116, 26372638
A. D. Backstrom, R. A. S. McMordie, T. P. Begley. J. A m . Chem Sac. 1995,
117, 2351 -2352.
B. Estramareix. M. Therisod, Biochem Biuphys. R c . Cummun. 1980, 95,
1017- 1022.
M. Therisod, D Gaudry, B. Estramareix, Nouv. J Chim 1978.2, 119-121.
R. L. White, I. D. Spenser. J Am. Chem. Soc. 1979. 101,5102- 5104.
[79] R. L. White, 1. D. Spenser, J. Am. Chem. Soc. 1982, 104,4934-4943.
[SO] K. Yamada, M. Yamamoto, M. Hayashiji, K. Tazuya, H Kumaoka, Biochem.
lnr. 1985, 10, 689-694
I811 Y. Kayama, T. Kawasaki, Arch. Biochem. Biophys. 1973, 158, 242-248.
[82] Y. Nose, Y. Tokuda, M. Hirabayashi, A. Iwashima, J. Virumrnol 1964, 10,
105 - 110.
[83] G. W. Camiener, G. M. Brown, J. Bioi. Chrm. 1960,235,2411-2417.
[84] a) I. G. Leder, J. Bid. Chem. 1961, 236, 3066-3071; b) Methods Enzj~mol.
1970, 18A. 207-212.
[85] Y Kawasaki, J. Bucteriol. 1993, 175, 5153-5158.
I861 a) H. Mitsuda, T. Tanaka, F.Kawai, J. Vitaminol. 1970, 16, 263-267; b) H.
Mitsuda, T. Tanaka, Y Takii, F. Kawai, ibid. 1971, 17, 89-95.
[87] Y. Komeda, M. Tanaka, T. Nishimune. Plunr Physiol. 1988, 88. 248-250.
[SS] H Nakayama, Eitumin 1990, 64, 619-632.
I891 a) L M. Lewin. G. M. Brown, J. Btol. Chem. 1961,236,2768-2771; b) G. M.
Brown, Methods Enzymol. 1970, 18A, 203-207.
[901 Suzuoki-ZirB, A. Kobata, J. Biochem. (Tokyo) 1960,47, 262-265.
1911 a) H. Nishino, A. Iwashima, Y. Nose, Biochem. Biophys. Res. Commun. 1971,
45, 363-368; b) H. Nishino, J Biochem. (Tokyo) 1972, 72, 1093-1100.
[92] a) A. Iwashima, H. Nishino, Y. Nose, Blochim. Biophys. Acru 1972, 258,
333-336; b) H. Nakayama, R. Hayashi, J. Bucteriol. 1972, 109, 936-938;
ibid. 1972, 112, 1118-1126.
[93] a) H. Sanemori, Y Egi, T. Kawasaki, J. Bucteriol 1976, 126, 1030-1036;
b) H. Sanemori, T. Kawasaki, J. Biochem. (Tokyo) 1980,88,223-230.
[94] a) E. P. Steyn-Parve, Biochim. Biophys. Actu 1952,8, 310-324; b) N. Shimazono, Y. Mano, R. Tanaka, Y. Kaziro, J. Biochem. (Tokyo) 1959, 46, 959961; C) Y. Kaziro, rbid. 1959, 46, 1523-1539; d) Y. Kaziro, R. Tanaka, Y
Mano, N. Shimazono, hid. 1961, 49,472-476.
[95] A. I. Voskoboyev, Y. M. Ostrovsky, Ann. N Y A c u d . Sci. 1982,378,161-176.
[96] a) H. Mitsuda, Y. Takii. K. Iwami, K. Yasumoto, J. Nutr. Sci. Vituminol. 1975,
21, 19-26; b) ihid 1975, 21, 103-115; c) ibid. 1975, 21, 189-98.
[97] a) F. Leuthardt, H. Nielsen, Helv. Chim. Acta 1952, 35, 1196-1209; b) Y
Mano, J. Brochem. (Tokyo) 1960, 47, 283-289; c) C. J. Gubler, Method.7
Enzymol. 1970,18A, 219-221; d) Y. Egi, S. Koyama, T. Shioda, K. Yamada,
T. Kawasaki, Biochim. Biophys. Actu 1992, 1/60? 171- 178.
[98] T. Kawasaki, Y. Egi, H. Sanemori, J Bucteriol. 1977, 130, 542-544.
[99] K. Nosaka, H. Nishimura, A. iwashima, Biochim. Biophys. Actu 1988, 967.
[loo] K. Nosaka, Biochim. Brophys. Actu 1990, 1037, 147-154.
[loll M. E. Schweingruber, R. Fluri, K. Maundrell, A.-M. Schweingruber, E.
Dumermuth, J. B i d Chem. 1986,261,15877-15882.
[lo21 A. M. Schweingruber, J. Dlugonski, E. Edenharter, M. E. Schweingruber,
Curr. Genet. 1991, 19, 249-254.
[lo31 J. Yang, M. E. Schweingruber, Curr. Genet. 1990, 18, 269-272.
[lo41 a) T. Kawasaki, Y. Nose, J. Biochem. (Tokyo) 1969, 65, 417-425; b) B. J.
Bachmann, K. B. Low, A. L. Taylor, Bacteriol. Rev. 1976,40, 116-167.
[lo51 1.Van Dyck, A. Pascual-Ahuir, B. Purnelle, A. Goffeau, Yeusf 1995, 11,
[lo61 K. Maundrell, J. Brol. Chem. 1990,265, 10857-10864.
[lo71 A. Zurlinden, M. E. Schweingruber, J. Bacteriof. 1994, 176, 6631 -6635.
[lo81 a) U. M. Praekelt, P. A. Meacock, Yeast 1992,8,699-710; b) U. M. Praekelt,
K. L. Byrne, P. A. Meacock, ibid. 1994, 10, 481-490.
[lo91 J. Skala, A. Nawrocki, A. Goffeau, Yeast 1995, 11, 1421-1427.
[110] A. Zurlinden, M. E. Schweingruber, Gene 1992, 117, 141-143.
[11 I] A. G. 0. Manetti, M. Rosetto, K. G. Maundrell, Yeast 1994, 10, 1075- 1082.
[112] T. P. Begley, Nut. Prod. Rep. 1996, 13, 177-185.
I. D. Spenser and R. L. White
[113] N. Imamura, H. Nakayama, Experientiu 1981,37, 1265-1266.
11141 T. Mizote, H. Nakayama, J. Bacteriol. 1989, 171, 3228-3232.
[115] K. Nosaka, H. Nishimura, Y. Kawasaki, T. Tsujihara, A. Iwashima. J. Bioi
Chem. 1994, 269, 30510-30516.
[116] H. Fankhauser, A. Zurlinden, A.-M. Schweingruber, E. Edenharter, M. E.
Schweingruber, J Biol. Chem. 1995, 270, 28457-28462.
[117] a) T. Kawasaki, T. Nakata, Y. Nose, J. BucrerioI. 1968. 95, 1483-1485;
b) A. L. Taylor, Bacteriol. Rev. 1970, 34, 155-175.
[118] N. Imamura, H. Nakayama, J. Bucterinf. 1982, 151, 708-717; corrigendum:
ibid. 1982. 152, 1308.
[119] W. Bajwa, B. Meyhack, H. Rudolph, A,-M. Schweingruber, A. Hinnen, Nucl.
Acids Res. 1984, 12, 7721 -1739.
[120] K. Nosaka, Y. Kaneko, H. Nishimura, A. Iwashima, F E M S Microbid L e f t .
1989,60, 55-59,
11211 A. Toh-e. Y. Ueda, S -1. Kakimoto, Y. Oshima, J Bucterrol. 1973, 113,
[122] K. Maundrell, P. Nurse, F. Schonholzer, M. E. Schweingruber, Gene 1985,39,
[123] K. Nosaka, Y Kaneko, H. Nishimura, A. Iwashima, J. Bid. Chem 1993,268,
11241 H. Nishimura, Y. Kawasaki, Y. Kaneko, K. Nosaka, A. Iwashima, FEBS
Lett. 1992, 297, 155-158.
[125] Y. Kawasaki, K. Nosaka, Y Kaneko, H. Nishimura, A. Iwashima. J Bucteriol. 1990, 172, 6145-6147.
[126] H Nishimura, Y. Kawasaki, Y. Kaneko. K. Nosaka, A. Iwashima, J. Bucteriol. 1992, 174,4701-4706.
[127] H. Fankhauser, M. E. Schweingruber, Gene 1994, 147, 141-144.
[128] C. S. 1. Tang, A. Bueno, P. Russell, J. Bioi. Chem. 1994, 269, 1192111 926.
[129] A.-M. Schweingruber, H. Fankhauser, J. Dlugonski, C. Steinmann-Loss,
M. E. Schweingruber, Genetics 1992, 130,445-449.
[130] S. Kumar, S. B. Sharma, Mu/. Gen. Genet. 1986, 204, 473-476.
[131] W. M. Proebsting, S. P. Maggard, W. W. Guo, J. Plant Physiol. 1990, 136,
231 -235.
[132] D. M. Downs, J. R. Roth, J. Buctrriol. 1991, 173,6597-6604.
[133] a) D. M Downs, J. Bucteriol. 1992, 174, 1515-1521; b) L. Petersen, J. EnosBerlage, D. M. Downs, Genetics 1996, 143, 37-44.
[134] D. M. Downs, L. Petersen, J. Bucterioi. 1994, 176, 4858-4864.
[135] 1.Petersen. D. M. Downs, J. Bucteriol. 1996, 178, 5676-5682.
[136] E. Webb, F. Febres, D. M. Downs, J. Bucteriol. 1996, 178, 2533-2538.
[I371 F. C. Belanger, T. Leustek. B. Chu, A. L. Kriz, Piunt Mol Bid. 1995, 29,
809 - 821.
[138] G. H. Choi, E. T. Marek, C. L. Schardl, M. G. Richey, S. Chang, D. A.
Smith, J. Bucterioi. 1990, 172, 4522-4528.
[139] T. Fujio, M. Hayashi, A. Iida, T. Nishi, T. Hagihara (Kyowa Hakko Kogyo
Company, Ltd.), EP-A 417953, 1991 [Chem. Absrr. 1991,114, 227508tl.
[140] P. Glaser, F. Kunst, M. Arnaud, M.-P. Coudart, W Gonzales, M.-F. Hullo,
M. Ionescu, B. Lubochinsky, L. Marcelino, 1. Moszer, E. Presecan, M. Santana, E. Schneider, J. Schweizer, A. Vertes, G. Rapoport, A. Danchin, Mu/.
Microbioi. 1993, 10, 371 -384.
[141] T. M. Finan, B. Kunkel, G. F. De Vos, E. R. Signer, J. Bacreriol. 1986, 167,
[142] S. Elliott, C.-W. Chang, M. E. Schweingruber, J. Schaller, E. E. Rickli, J.
Carbon, J B i d . Chem. 1986,261,2936-2941.
[143] H. Nishimura, Y. Kawasaki, K. Nosaka, Y. Kaneko, A. Iwashima, J. Bucteri01. 1991, 173, 2716-2719.
Angew. Chem. Int. Ed. Engl. 1997,36,1032-1046
Без категории
Размер файла
2 375 Кб
vitamins, diversity, thiamine, instance, biochemical, biosynthesis
Пожаловаться на содержимое документа