L-selenohomocysteine: One-step synthesis from L-selenomethionine and kinetic analysis as substrate for methionine synthases

Article abstract of DOI:10.1016/S0960-894X(00)00498-4

A single-step convenient synthesis of L-selenohomocysteine (SeHcy) from L-selenomethionine (SeMet) using sodium in liquid ammonia is described. Methionine synthases convert SeHcy to SeMet at rates comparable to their rates of conversion of L-homocysteine (Hcy) to L-methionine (Met). This study suggests that SeHcy generated from SeMet metabolism can be efficiently recycled to SeMet in mammals. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00498-4

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2471±2475
L-Selenohomocysteine: One-Step Synthesis from
L-Selenomethionine and Kinetic Analysis as Substrate for
Methionine Synthases
Zhaohui S. Zhou,a April E. Smithb and Rowena G. Matthewsa,*
aBiophysics Research Division and the Department of Biological Chemistry, The University of Michigan, Ann Arbor,
MI 48109-1055, USA
bDepartment of Medicinal Chemistry, The University of Michigan, Ann Arbor, MI 48109-1055, USA
Received 10 July 2000; accepted 22 August 2000
AbstractÐA single-step convenient synthesis of l-selenohomocysteine (SeHcy) from l-selenomethionine (SeMet) using sodium in
liquid ammonia is described. Methionine synthases convert SeHcy to SeMet at rates comparable to their rates of conversion
of l-homocysteine (Hcy) to l-methionine (Met). This study suggests that SeHcy generated from SeMet metabolism can be
eciently recycled to SeMet in mammals. # 2000 Elsevier Science Ltd. All rights reserved.
Selenium is an essential trace element for humans,
animals, and other living organisms.1 3 Selenium de®-
ciency in certain areas of China previously caused
Keshan disease, which was associated with heart failure
and cardiogenic shock.4 Selenium supplementation has
practically eliminated the incidence of this disease. In
the past decade, changes in selenium metabolism have
been associated with oxidant damage, cancer develop-
ment, male infertility,5 and other disease conditions.1,2
A prevention trial study in the United States involving
1362 patients has shown that selenium supplementation
of 200 mg per day signi®cantly reduces both total mor-
tality and total incidence of cancer of the lung, prostate,
colon, and rectum.6 Because of its bene®cial e€ects,
selenium has been added to many nutritional supple-
ments. On the other hand, excess selenium uptake is
toxic and causes injuries to plants, animal, and
humans.1 3 The toxicology of selenium is not well
understood yet.
intake level at 400 mg per day. It is worth noticing that
these numbers refer to the total amount of selenium.
Selenium exists in many forms in food, supplements,
and human bodies.3 For example, selenomethionine and
Se-adenosylselenohomocysteine account for 85% of the
total selenium in Selenomax from Nutrition 21 (Pur-
chase, New York), the selenium-enriched yeast used in the
cancer prevention trial mentioned earlier.8 On the other
hand, most health supplements contain inorganic forms of
selenium. Detailed studies of the inter-conversions of dif-
ferent selenium compounds will not only provide a bet-
ter understanding of the biological roles and toxicity of
selenium, but also provide us with information to opti-
mize the selenium supplementation to achieve maximal
bene®ts and minimal side e€ects.3
To date, the biological functions of selenium have been
associated with several natural selenoproteins, including
glutathione peroxidase, iodothyronine deiodinase, and
thioredoxin reductase.1,2,9 Selenocysteine has been
found to be the only essential seleno-amino acid in these
proteins. Special proteins are used to synthesize seleno-
cysteinyl-tRNA and deliver it to the ribosome for
incorporation at UGA codons, which normally serve to
terminate translation.10,11 Nonetheless, at elevated con-
centrations, selenocysteine can be randomly incorpo-
rated into proteins,10 replacing cysteine, which may
contribute to selenium toxicity. In comparison, seleno-
methionine can be randomly incorporated into proteins
in place of methionine without perturbation of the
To provide general guidelines for the usage of and fur-
ther research on selenium, the Institute of Medicine of
the National Academy of Sciences recently issued an
updated Dietary Reference Intake for selenium.7 The
report sets a recommended intake level for selenium at
55 mg per day, and for the ®rst time, sets the upper
*Corresponding author. Tel.: +1-734-764-9459; fax: +1-734-764-
3323; e-mail: rmatthew@umich.edu
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00498-4

2472
Z. S. Zhou et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2471±2475
normal structures and functions.12 Nevertheless, excess
selenomethionine is toxic to human cells. The conver-
sion of selenomethionine to selenocysteine via seleno-
homocysteine (as shown in Scheme 1) has been
hypothesized to cause the toxicity in mammals.13
required a partially puri®ed enzyme that is not readily
available.
Since l-selenomethionine is now commercially avail-
able, we have developed a convenient single step synth-
esis of l-selenohomocysteine from l-selenomethionine
using metallic sodium in liquid ammonia as depicted in
Scheme 2.19 The protocol was modi®ed from a literature
procedure for the preparation of l-homocysteine from
l-methionine.20,21 l-Selenohomocysteine was oxidized
in air to its diselenide form for storage, and was easily
regenerated by in situ reduction with dithiothreitol
(DTT) or tris(2-carboxyethyl)phosphine (TCEP) in
aqueous solution.22
Each individual pathway in Scheme 1 has been char-
acterized at the enzyme level, except the conversion
from selenohomocysteine to selenomethionine (boxed in
Scheme 1).13,14 The biochemical pathways of selenium
amino acids and their derivatives have been found to be
very similar to those of their sulfur counterparts. In all
cases, the enzyme for each individual step transforms
the sulfur compounds and their selenium analogues at
comparable rates.13,14 For example, selenohomocysteine
is converted to selenocystathionine by rat liver cysta-
thionine beta-synthase at a rate 69% of that for con-
version of homocysteine to cystathionine; and
selenocystathionine is transformed to selenocysteine by
rat liver cystathionine gamma-lyase at about 3 times the
rate of cystathionine elimination.14 Because seleno-
homocysteine is at a branch point of biochemical
pathways to selenomethionine or selenocysteine,
determination of whether selenohomocysteine can be
converted to selenomethionine by methionine synthases
and the reaction kinetics of that conversion are
obviously crucial for having a complete picture of
selenium metabolism.
The synthesis of selenohomocysteine reported here is
more convenient than the literature procedures. The
modest product yield (62%) was partially due to the
slow oxidation of the selenol to the diselenide by oxygen
in the air. After the synthesis had been completed, we
found that our yields of product were low because it
took a longer time than we expected to complete the
oxidation, particularly in solutions at low pH. Addition
of a catalytic amount of FeCl3 has been reported to
facilitate the oxidation process, which will conceivably
improve the product yield.17 The optical purity of
amino acids normally does not change under the reaction
conditions applied here, as evidenced by the complete
retention of chirality on conversion of l-methionine to
l-homocysteine under similar conditions.21 The result-
ing selenium product was optically active,21 [a]D +31ꢀ
(1.0 N HCl). Only l-homocysteine, but not the d-iso-
mer, is a substrate for cobalamin-dependent methionine
synthase (MetH). A fragment of MetH containing the
substrate binding sites, MetH(2-649), catalyzed the
quantitative conversion of the selenohomocysteine we
synthesized and methyl cobalamin to cob(I)alamin and
selenomethionine, suggesting that the selenohomocys-
teine obtained was in the l-con®guration. In sum-
mary, a convenient single-step synthesis of optically
pure l-selenohomocysteine from l-selenomethionine
was achieved.
In this paper, we report the one-step convenient synth-
esis of selenohomocysteine from selenomethionine using
sodium in liquid ammonia shown in Scheme 2, and the
kinetic characterization of the conversion of selenoho-
mocysteine to selenomethionine by methionine syn-
thases. This work provides key information relevant to
the biochemical pathways of selenium metabolism.
Synthesis of Selenohomocysteine (SeHcy)
Selenohomocysteine has been synthesized chemically
from a-amino-butyrate lactone via an a-amino-g-ben-
zylseleno-butyric acid intermediate,15 17 and enzymati-
cally from homoserine and sodium diselenide (Na2Se2)
using O-acetylhomoserine sulfydrylase.18 The former
procedure involved multiple steps, and the later
Conversion of SeHcy to SeMet by Methionine
Synthases
There are two methionine synthases, the cobalamin-
dependent methionine synthase (MetH, E.C. 2.1.1.13)
and the cobalamin-independent methionine synthase
(MetE, E.C. 2.1.1.14).23 The overall reaction they nor-
mally catalyze is the same, namely, the transfer of a
methyl group from 5-methyltetrahydrofolate to the thiol
group of homocysteine to form tetrahydrofolate and
methionine as depicted in Schemes 3 and 4. The MetH
enzyme has a prosthetic group, cobalamin (B12), which
mediates the methyl transfer.24 In humans and
Scheme 1. Biochemical interconversions of selenium amino acids and
their derivatives in mammals. The reaction catalyzed by methionine
synthases is boxed.
Scheme 2. Synthesis of l-selenohomocysteine and l,l-selenohomocystine from l-selenomethionine.

Z. S. Zhou et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2471±2475
2473
mammals, MetH is the sole form of methionine synthase.
The mammalian and Escherichia coli MetH enzymes are
highly similar, both in their amino acid sequences (which
are 55% identical) and in their catalytic properties.25 Since
human MetH has not been puri®ed to homogeneity, the
E. coli enzyme was used to study the conversion of sele-
nohomocysteine to selenomethionine.
The cobalamin-independent methionine synthase
(MetE) catalyzes methyl transfer from methyltetrahy-
drofolate to homocysteine without formation of a
detectable intermediate.23,27 The enzyme catalyzes the
terminal step of de novo synthesis of methionine in
plants and other organisms that can not utilize cobala-
min.28 Both the MetE and MetH enzymes exist in E.
coli. The activity assay of MetE is based on the conver-
sion of the tetrahydrofolate product and formic acid to
methenyltetrahydrofolate, which can be monitored at
350 nm, as previously reported.29 The diselenide of
selenohomocystine or the disul®de of oxidized DTT
deactivates the MetE enzyme by releasing the catalyti-
cally essential zinc in MetE,30 and excess TCEP inter-
feres with the formation of methenyltetrahydrofolate.
Thus, selenohomocysteine was generated by mixing
excess solid selenohomocystine with an aqueous solu-
tion containing 100 mM TCEP in 500 mM bis[2-
hydroxyethyl]-iminotris[hydroxymethyl]methane (Bis-
Tris) at pH 6.2. The excess diselenide is insoluble at this
A homocysteine-methylcobalamin assay for the E. coli
MetH enzyme has been previously described.26 In this
assay (boxed in Scheme 3), the substrate binding region,
MetH (2-649), was used to catalyze the transfer of the
methyl group from exogenous methylcobalamin to the
selenohomocysteine substrate, forming selenomethio-
nine. Whereas conformational changes appear to limit
the rate of the physiological reaction catalyzed by
MetH, this assay directly assesses the reactivity of sele-
nohomocysteine as a substrate. This assay was per-
formed anaerobically, resulting in the formation of the
cob(I)alamin product. The demethylation of methylco-
balamin was monitored at 538 nm.26 Selenohomocys-
teine was generated by the reduction of the diselenide
with excess TCEP in aqueous solution containing 100
mM potassium phosphate at pH 7.2. An excess of
TCEP did not interfere with the assay.
As previously described, the reaction of MetH (2-649)
with exogenous methylcobalamin is ®rst order in
methylcobalamin.26 The maximum second-order rate
constant for selenohomocysteine in this assay was
s
1, which is
1
determined to be (27Æ1)Â103
M
approximately four times that for homocysteine,
(6.5Æ0.5)Â103 M 1 s 1. The Km for selenohomocysteine
was determined to be 16Æ5 mM as compared with the
Km for homocysteine of 69Æ18 mM (Fig. 1).
Figure 1. Dependence of initial reaction rates on substrate concentra-
tion for MetH-catalyzed methyl transfer from methylcobalamin. Data
for selenohomocysteine and homocysteine are shown with ®lled and
open circles, respectively, and were ®t to the Michealis±Menten equation.
Scheme 3. Reactions catalyzed by cobalamin-dependent methionine
synthase (MetH). The half-reaction reported in this paper is boxed.
Figure 2. HPLC traces of authentic PTC-selenomethionine (bottom
trace) and PTC-modi®ed products of the MetH-catalyzed reaction
(top trace). Amino acids were modi®ed by treatment with phenylthio-
cyanate and separated by reverse phase HPLC.
Scheme 4. Reaction catalyzed by cobalamin-independent methionine
synthase (MetE).

2474
Z. S. Zhou et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2471±2475
pH and was removed by centrifugation. The speci®c
activity for selenohomocysteine was 0.11Æ0.01 mmol
Acknowledgements
1
1
1
min mg in comparison to 0.23Æ0.02 mmol min
1
mg for the homocysteine substrate. The Km values of
homocysteine and selenohomocysteine could only be
estimated to be below 125 mM, due to limitations of the
assay.29
We are indebted to Professor Ronald W. Woodard for
his suggestion to use the reported synthetic procedure.
We also thank Professor James K. Coward, Dr. Vahe
Bandarian, and Jessica Pankuch for helpful suggestions.
The research was partially supported by NIH Research
Grant R37 GM24908 to R.G.M., and by Pharmaco-
logical Sciences Training Grant GM07767 to A.E.S.
HPLC amino acid analysis con®rms the formation of
selenomethionine as the reaction product, as shown in
Figure 2. After derivatization of the amino acids with
phenylisothiocyanate (PTC) to form PTC-modi®ed amino
acids, reverse-phase HPLC analysis was carried out on the
Applied Biosystems separation system at the Protein and
Carbohydrate Structure Facility at the University of
Michigan. PTC-modi®ed selenomethionine (14.82 min)
eluted after PTC-modi®ed methionine (14.37 min).
References and Notes
1. Burk, R. F. Selenium in Biology and Human Health;
Springer-Verlag: New York, 1994.
2. Foster, L. H.; Sumar, S. Crit. Rev. Food Sci. Nutr. 1997, 37,
211.
3. Allan, C.; Lacourciere, G.; Stadtman, T. Annu. Rev. Nutr.
1999, 19, 1.
4. Keshan Disease Research Group Chin. Med. J. 1979, 92,
471.
Discussion and Conclusion
5. Ursini, F.; Heim, S.; Kiess, M.; Maiorino, M.; Roveri, A.;
Wissing, J.; Flohe, L. Science 1999, 285, 1393.
This study, for the ®rst time, establishes that selenoho-
mocysteine can readily be converted to selenomethio-
nine by both puri®ed cobalamin-dependent and
cobalamin-independent methionine synthases at rates
comparable to those for the conversion of homocysteine
to methionine. A previous study with rat liver enzymes
has shown that selenohomocysteine can be transformed
to selenocysteine with similar eciency to its sulfur
counterpart.14 The extent of conversion of selenohomo-
cysteine to selenocysteine will be governed by the
competition between methionine synthase and b-
cystathionine synthase for selenohomocysteine.31,32 b-
cystathionine synthase is activated by S-adenosyl-
methionine,33 so partitioning of selenohomocysteine will
be critically dependent on the levels of S-adenosyl methio-
nine and possibly also Se-adenosylselenomethionine in
human cells.
6. Clark, L. C.; Combs, G. F., Jr.; Turnbull, B. W.; Slate, E.
H.; Chalker, D. K.; Chow, J.; Davis, L. S.; Glover, R. A.;
Graham, G. F.; Gross, E. G.; Krongrad, A.; Lesher, J. L., Jr.;
Park, H. K.; Sanders, B. B., Jr.; Smith, C. L.; Taylor, J. R. J.
Am. Med. Assoc. 1996, 276, 1957.
7. Dietary Reference Intakes for Vitamin C, Vitamin E, Sele-
nium, and Beta-Carotene, and other Carotenoids; National
Academy Press: Washington, DC, 2000.
8. Kotrebai, M.; Birringer, M.; Tyson, J. F.; Block, E.; Uden,
P. C. Anal. Commun. 1999, 36, 249.
9. Stadtman, T. C. J. Biol. Chem. 1991, 266, 16257.
10. Muller, S.; Senn, H.; Gsell, B.; Vetter, W.; Baron, C.;
Bock, A. Biochemistry 1994, 33, 3404.
11. Hat®eld, D. L.; Choi, I. S.; Ohama, T.; Jung, J.-E.; Dia-
mond, A. M. In Selenium in Biology and Human Health; Burk,
R. F., Ed.; Springer-Verlag: New York, 1994; p 221.
12. Smith, J. L.; Thompson, A. Structure 1998, 6, 815.
13. Kajander, E. O.; Harvima, R. J.; Eloranta, T. O.; Marti-
kainen, H.; Kantola, M.; Karenlampi, S. O.; Akerman, K.
Biol. Trace Elem. Res. 1991, 28, 57.
14. Esaki, N.; Nakamura, T.; Tanaka, H.; Suzuki, T.; Mor-
ino, Y.; Soda, K. Biochemistry 1981, 20, 4492.
15. Klosterman, H. J.; Painter, E. P. J. Am. Chem. Soc. 1947,
69, 2009.
Because selenium has similar, but nevertheless di€erent,
chemical and physical properties as compared to sulfur,
selenohomocysteine can be used as an excellent probe to
study the interactions between homocysteine and
methionine synthases. Indeed, we have successfully
applied X-ray absorption spectroscopic analysis to
complexes of selenohomocysteine and methionine syn-
thases (K. Peariso, Z. S. Zhou, A. E. Smith, R. G.
Matthews and J. E. Penner-Hahn, submitted for pub-
lication). For example, because the methionine syn-
thases do not contain selenium, the presence of a
selenium in the coordination sphere of the active site
zinc in the selenohomocysteine±enzyme complexes
unambiguously proves the direct ligation of substrate to
the zinc ion.
16. Painter, E. P. J. Am. Chem. Soc. 1947, 69, 232.
17. Skupin, J. Roczniki Chem. 1962, 36, 631.
18. Chocat, P.; Esaki, N.; Tanaka, K.; Soda, K. Agric. Biol.
Chem. 1985, 49, 1143.
19. Experimental details for synthesis: To a solution of l-(+)-
selenomethionine (0.98 g, 5 mmol; purchased from Acros
Organics) in 40 mL liquid ammonia at 80 ꢀC under stirring,
small pieces of metallic sodium (ꢁ0.26 g, 11 mmol) were care-
fully added in 45 min until the solution remained blue for 15
min. The solution was stirred for another 50 min at 80 ꢀC,
solid ammonium chloride (1.27 g, 24 mmol) was then added to
neutralize sodium amide. The reaction mixture was opened to
air and slowly warmed up to room temperature overnight. The
resulting yellow solid was mixed with 100 mL water. The
mixture was adjusted to neutral pH and stirred vigorously
open to air overnight. The pale-yellow precipitate collected by
®ltration was a mixture of selenomethionine and selenohomo-
cystine as judged by NMR analysis. Selenohomocysteine is
much more soluble than selenomethionine in aqueous solution
In conclusion, a convenient synthesis of selenohomo-
cysteine has been developed, and methionine synthases
have been shown to convert selenohomocysteine to
selenomethionine at rates comparable to those for their
sulfur analogues. This study will provide a better
understanding of selenium metabolism, and the role of
selenium in disease development and prevention.

Z. S. Zhou et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2471±2475
22. Gunther, W. H. H. J. Org. Chem. 1967, 32, 3931.
2475
at pH 4±6, and selenohomocystine can be readily reduced
by dithiothreitol (DTT) or tris(2-carboxyethyl)phosphine
(TCEP). Thus, the diselenide was reduced to the selenol and
separated from selenomethionine by ®ltration. The crude pro-
duct was suspended in 35 mL of 250 mM aqueous citric acid
bu€er, pH 5.8, and dithiothreitol (0.5 g, 3 mmol) was then
added. After stirring under nitrogen for 30 min at room tem-
perature, the suspension was ®ltered, and the ®ltrate was stir-
red overnight in open air, allowing selenohomocysteine to be
oxidized to selenohomocystine. The precipitate was collected
by ®ltration, washed with water, and dried in vacuo to give a
pale yellow solid as desired compound (0.39 g, 62% yield
based on selenomethionine consumed). 1H NMR (300 MHz, 1
N DCl in D2O) d 3.63 (2H, t, J=6.41 Hz), 2.43 (4H, m), 1.72±
1.92 (4H, m); 13C NMR (300 MHz, 1 N DCl in D2O with 2%
(v/v) CD3OD) d 171.3, 52.5, 31.2, 22.8. DEPT 135: CH-CH3, d
52.5; CH2, d 31.2, 22.8. HRMS (MALDI, TFA/H2O) calcd for
C8H16N2O4Se2+H 362.9527, found 362.9208. Mp (uncor-
rected): 235 ꢀC (decomp, literature 214 ꢀC,18 260 ꢀC16,17). [a]2D1
+31ꢀ (1.0 N HCl).
23. Gonzalez, J. C.; Banerjee, R. V.; Huang, S.; Sumner, J. S.;
Matthews, R. G. Biochemistry 1992, 31, 6045.
24. Drennan, C. L.; Huang, S.; Drummond, J. T.; Matthews,
R. G.; Ludwig, M. L. Science 1994, 266, 1669.
25. Matthews, R. G. In Chemistry and Biochemistry of B12;
Banerjee, R., Ed.; John Wiley & Sons: New York, 1999; pp
681±706.
26. Goulding, C. W.; Postigo, D.; Matthews, R. G. Biochem-
istry 1997, 36, 8082.
27. Whit®eld, C. D.; Steers, E. J., Jr.; Weisbach, H. J. Biol.
Chem. 1970, 245, 390.
28. Eichel, J.; Gonzalez, J. C.; Hotze, M.; Matthews, R. G.;
Schroder, J. Eur. J. Biochem. 1995, 230, 1053.
29. Drummond, J. T.; Jarrett, J.; Gonzalez, J. C.; Huang, S.;
Matthews, R. G. Anal. Biochem. 1995, 228, 323.
30. Zhou, Z. S.; Peariso, K.; Penner-Hahn, J. E.; Matthews,
R. G. Biochemistry 1999, 38, 15915.
31. Finkelstein, J. D.; Martin, J. J. J. Biol. Chem. 1984, 259,
9508.
20. Ramalingam, K.; Woodard, R. W. Tetrahedron Lett.
1985, 26, 1145.
21. Dolphin, D.; Endo, K. Anal. Biochem. 1970, 36, 338.
32. Selhub, J. Annu. Rev. Nutr. 1999, 19, 217.
33. Finkelstein, J. D.; Kyle, W. E.; Martin, J. L.; Pick, A. M.
Biochem. Biophys. Res. Commun. 1975, 66, 81.