Synthesis of D-galactopyranosylphosphonic and (D-galactopyranosylmethyl) phosphonic acids as intermediates of inhibitors of galactosyltransferases

Article abstract of DOI:10.1135/cccc20061659

Employing the Michaelis-Arbuzov reaction of 1-O-acetyl-2,3,4,6-tetra-O- benzyl-D-galactopyranose with triethyl phosphite and trimethylsilyl trifluoromethanesulfonate, α- and β-D-galactopyranosylphosphonic acids were prepared in 33 and 28% yields, respecti

Full text of DOI:10.1135/cccc20061659

Inhibitors of Galactosyltransferases
1659
SYNTHESIS OF D-GALACTOPYRANOSYLPHOSPHONIC AND
(D-GALACTOPYRANOSYLMETHYL)PHOSPHONIC ACIDS AS
INTERMEDIATES OF INHIBITORS OF GALACTOSYLTRANSFERASES
a
a
a1
b1
Helena HEISSIGEROVÁ , Petr KOČALKA , Martina HLAVÁČKOVÁ , Anne IMBERTY ,
Christelle BRETONb2, Valerie CHAZALET and Jitka MORAVCOVÁ
*
b3
a2,
a
Department of Chemistry of Natural Compounds, Institute of Chemical Technology, Prague,
166 28 Prague 6, Czech Republic; e-mail: martina.hlavackova@vscht.cz,
jitka.moravcova@vscht.cz
Centre de Recherches sur les Macromolecules Végétales, CNRS (affiliated with
1
2
b
Université Joseph Fourier), 601 rue de la Chimie, BP 53, 38041 Grenoble cedex 9, France;
1
2
e-mail: anne.imberty@cermav.cnrs.fr, christelle.breton@cermav.cnrs.fr,
3
valerie.chazalet@cermav.cnrs.fr
Received Novem ber 3, 2006
Accepted Novem ber 29, 2006
Em ployin g th e Mich aelis–Arbuzov reaction of 1-O-acetyl-2,3,4,6-tetra-O-ben zyl-D-galacto-
pyran ose with trieth yl ph osph ite an d trim eth ylsilyl trifluorom eth an esulfon ate, α- an d
β-D-galactopyran osylph osph on ic acids were prepared in 33 an d 28% yields, respectively.
(α-D-Galactopyran osylm eth yl)ph osph on ic acid was syn th esized by a five-step route from
2,3,4,6-tetra-O-ben zyl-D-galactopyran ose in 52% overall yield. Wh en tested again st bovin e
α-1,3- an d β-1,4-galactosyltran sferases, all th ree com poun ds sh owed, at best, very poor in h i-
bition s. Both en zym es were in h ibited m ore effectively by β-D-galactopyran osylph osph on ic
acid (IC50 = 17 m m ol l–1 at 13.5 µm ol l–1 of UDP-Gal for β4GalT).
Keyw ord s: Carboh ydrates; UDP-Gal; Mich aelis–Arbuzov reaction ; Ph osph on ates; Galactosyl-
tran sferases; C-Glycosides.
In teraction s of lectin -type receptors with a carboh ydrate epitope of
glycocon jugates form th e basis of cell recogn ition processes in biological
system s. Com plex oligosacch arides are syn th esized in th e en doplasm atic re-
ticulum an d Golgi com plex by th e sequen tial tran sfer of m on osacch aride
residue from an activated don or to an appropriate acceptor1. Th us th e selec-
tive in h ibition of en zym es participatin g in th e biosyn th esis of oligo-
sacch aride ch ain s is of great im portan ce as it m ay lead to th e developm en t
of n ew th erapeutic agen ts. Galactosyltran sferases catalysin g th e tran sfer of
D-galactose from uridin e-5′-diph osph o-α-D-galactose (UDP-Gal) can be clas-
sified m ech an istically as eith er in vertin g or retain in g, depen din g on th e rel-
ative an om eric con figuration of th eir substrate an d product. In vertin g
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672
© 2006 Institute of Organic Chemistry and Biochemistry
doi:10.1135/cccc20061659

1660
Heissigerová, Kočalka, Hlaváčková, Imberty, Breton, Chazalet, Moravcová:
β-1,4-galactosyltran sferase (β4GalT) utilizes N-acetylglucosam in e an d all
N-acetylglucosam in e-term in ated oligosacch arides as th e acceptor wh ile re-
tain in g α-1,3-galactosyltran sferase (α3GalT) creates a n ew glycosidic bon d
to N-acetyllactosam in e. Th e m ech an ism of action is n ow rath er well docu-
m en ted for in vertin g glycosyltran sferases, an d it is assum ed to proceed via a
sin gle n ucleoph ilic SN2 displacem en t2,3. However, th e m ech an ism of retain -
in g glycosyltran sferases rem ain s un clear. Recen tly, it h as been assum ed to
in volve a double n ucleoph ilic displacem en t with th e form ation of a cova-
len t glycosyl-en zym e in term ediate. In th is case, th e participation of two
catalytic residues h as been widely accepted alth ough th e catalysis of on ly
on e base species was also reported4. As th e existen ce of a covalen t in term e-
diate h as n ot been eviden ced so far, a sin gle, fron t-side displacem en t SNi re-
action can also be taken in to accoun t.
In particular, we are in terested in th e developm en t of α3GalT in h ibitors.
Un like m ost oth er m am m als, h um an s an d Old-World prim ates do n ot pos-
sess α3GalT activity, wh ich is relevan t for a h yperacute vascular rejection
observed in pig-to-h um an xen otran splan tation s. A strategy to th is poin t
could be a pretreatm en t of pigs usin g th e α3GalT specific in h ibitor5. Differ-
en t approach es h ave been used to design galactosyltran sferase in h ibitors;
h owever, on ly a lim ited success h as been ach ieved, in particular for α3GalT.
Th e kn own syn th etic in h ibitors of GalT (Fig. 1) in clude an alogues of both
th e tran sition state6–8 an d sugar n ucleotide9–12 alth ough sim ple m eth ylen e
diph osph on ates were also iden tified7 as n ew in h ibitors of β4GalT. Im in o-
cyclitol 1 (Fig. 1) was sh own to be a n on com petitive in h ibitor of α3GalT
with a Ki of 75 (ref.7) or 70 µm ol l–1 (ref.8) again st UDP-Gal but it was also a
very poten t in h ibitor of β-glycosidase8. If th e sim ple m otif of im in oalditol 1
was in troduced in to a m ore com plex structure 2 (Fig. 1), th e in h ibition po-
ten cy of 2 again st α3GalT in creased m arkedly givin g Ki of 4.4 µm ol l–1
wh ilst its activity again st β-glycosidase was suppressed. Th e auth ors sug-
gested th at both 1 an d 2 in teract n ot on ly with th e free en zym e but also
with th e en zym e–substrate com plex8. Th is con clusion is im portan t sin ce
con jugate 2 actually m im ics th e glycosyl don or an d bin din g of such in h ibi-
tors requires gen erally th e presen ce of a n ucleotide m oiety11. Th us, ph os-
ph on o an alogues 3, 4 an d 5 were foun d to be efficien t in h ibitors of β4GalT
with a Ki of 165 (ref.9), 97 (ref.10) an d 62 µm ol l–1 (ref.13), respectively. More
recen tly, in h ibition activity of 2-(galactosyleth yl)ph osph on oph osph ate 6
was evaluated14 towards bovin e m ilk β4GalT displayin g IC50 of 40 µm ol l–1.
Th e fin din g th at sim ple ph osph on ates7 or (2,6-an h ydro-1,3-dideoxy-
D-lyxo-h ept-2-en it-1-yl)ph osph on ate15 are in h ibitors of β4GalT prom pted
us to focus on α- an d β-D-galactopyran osylph osph on ic acids (7 an d 8) an d
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

Inhibitors of Galactosyltransferases
1661
(α-D-galactopyran osylm eth yl)ph osph on ic acid (9) as structure scaffolds
for th e con struction of UDP-Gal an alogues with a m odified diph osph ate
lin kage.
Herein , we would like to report th e syn th esis of th e required startin g
com poun ds 7–9, wh ich were also tested for th eir ability to in h ibit eith er
α3GalT or β4GalT.
FIG. 1
Poten t in h ibitors of α3GalT or β4GalT
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

1662
Heissigerová, Kočalka, Hlaváčková, Imberty, Breton, Chazalet, Moravcová:
RESULTS AND DISCUSSION
Em ployin g a Mich aelis–Arbuzov reaction of 1-O-acetyl-2,3,4,6-tetra-O-
ben zyl-β-D-galactopyran ose with trim eth yl ph osph ite, both galacto-
pyran osylph osph on ic acids 7 an d 8 were already syn th esized in th e form of
th eir per-O-ben zyl dim eth yl esters16. Free α-ph osph on ic acid 7 was ob-
tain ed from th e reaction of 1-O-acetyl-2,3,4,6-tetra-O-ben zyl-α-D-galacto-
pyran ose with tris(trim eth ylsilyl) ph osph ite followed by h ydrolysis an d
catalytic h ydrogen ation 9. In our h an ds, treatm en t of a m ixture of an om ers
of 1-O-acetyl-2,3,4,6-tetra-O-ben zyl-D-galactopyran ose (α/β = 2) with
trieth yl ph osph ite an d trim eth ylsilyl trifluorom eth an esulfon ate (TMSOTf)
accordin g to th e described procedure17 gave correspon din g dieth yl ph os-
ph on ates 10 an d 11 in excellen t 80% yield (Sch em e 1), wh ich was h igh er
th an 58% reported for trim eth yl ph osph ite16. It sh ould be n oted th at th e
SCHEME 1
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

Inhibitors of Galactosyltransferases
1663
yield decreased to 20% if triisopropyl ph osph ite was used in stead of trieth yl
ph osph ite. Accordin g to th e previous fin din g th at th e ratio of products aris-
in g from th e Mich aelis–Arbuzov reaction does n ot depen d substan tially on
th e an om eric con figuration of an in com in g glycosyl acetate15, ph osph on ates
10 an d 11 were obtain ed in th e ratio α/β = 1.5 an d were easily separated by
silica gel ch rom atograph y. Fin ally, th e protectin g ben zyl an d ester groups
were rem oved in on e step usin g an excess of iodotrim eth ylsilan e an d th e
ph osph on ic acids 7 an d 8 were purified by th e precipitation from m eth an ol
by addition of dieth yl eth er18. Th e overall yields of 7 an d 8 were 33 an d
28%, respectively. Th e structure an d stereoch em istry of ph osph on ates 10
an d 11 was con firm ed by NMR spectrom etry an d by com parison of th e data
with th ose publish ed for correspon din g dim eth yl esters16. Accordin g to th e
gen erally accepted rules for ph osph on ates16,17,19, ch em ical sh ift of th e
an om eric axial proton (3.65 ppm for 11) occurred at a h igh er field th an th e
equatorial on e (4.42 ppm for 10). In 13C NMR, th e C-1 sign als of 10 an d 11
were foun d at 69.20 ppm with J(C-1,P) = 157.2 Hz an d 75.16 ppm with
J(C-1,P) = 171.0 Hz, respectively. Th e sign al at a h igh er field with sm aller
couplin g con stan t in dicates th e axial orien tation of th e P-substituen t. Th e
assign m en t of an om eric con figuration s of galactopyran osylph osph on ic
acids 7 an d 8 was based on th e sam e an alysis; α-an om er 7 with axial ph os-
ph on ic group sh owed th e H-1 sign al at 4.24 an d C-1 sign al at 73.15 ppm
with J(C-1,P) = 151.1 Hz, wh ile th e respective NMR sign als of β-an om er 8
were at 3.41 (H-1) an d 76.70 ppm (C-1) with J(C-1,P) = 162.7 Hz. 13C NMR
data of ph osph on ic acid 7 were in accord with th ose previously described9
if experim en tal error was taken in to accoun t, but 1H ch em ical sh ifts were
differen t. Th erefore NOE con tacts for α-an om er 7 were also m easured
(Fig. 2) an d C-1 con figuration was verified un doubtedly.
Two altern ative routes were elaborated for th e syn th esis of n ovel
(α-D-galactopyran osylm eth yl)ph osph on ate 12. Th e m ore straigh tforward
strategy was th e on e step Horn er–Em m on s/Mich ael reaction 20 of sodium
FIG. 2
NOE con tacts in free acids 7 an d 9
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

1664
Heissigerová, Kočalka, Hlaváčková, Imberty, Breton, Chazalet, Moravcová:
tetraeth yl m eth ylen ediph osph on ate with 2,3,4,5-tetra-O-ben zyl-D-galacto-
pyran ose. Th e reaction , h owever, yielded a m ixture of an om ers of (galacto-
pyran osylm eth yl)ph osph on ates 12 an d 13 accom pan ied by a con siderable
am oun t of th e elim in ation product 14 even if th e experim en tal con dition s
were optim ized (Sch em e 2). Th e best result was ach ieved with NaH in
tetrah ydrofuran at room tem perature; th e desired ph osph on ates 12 an d 13
were isolated in 37% yield an d th e con jugated vin ylph osph on ate 14 in 12%
SCHEME 2
yield. If tetraisopropyl m eth ylen ediph osph on ate was used in stead of tetra-
eth yl m eth ylen ediph osph on ate, th e reaction exclusively gave th e elim in a-
tion product un der several con dition s. Th e observation th at 3-ben zyloxy
group can be easily cleaved durin g th is reaction was already m ade wh en
2,3,4,-tetra-O-ben zyl-α-D-glucopyran osyl brom ide21, 2,3,4-tri-O-ben zyl-
L-rh am n ose18,22 or 2,3,4-tri-O-ben zyl-L-fucose18 was in vestigated. Moreover,
both ph osph on ates 12 an d 13 were isolated as a m ixture on ly an d all at-
tem pts to separate th em effectively failed. Th us, we decided to in vestigate
th e m ore gen eral procedure in volved th e Mich aelis–Arbuzov reaction as a
key step in th e reaction sequen ce (Sch em e 3)20. In th e first step, th e Wittig
reaction of 2,3,4,5-tetra-O-ben zyl-D-galactopyran ose with m eth yltriph en yl-
ph osph on ium brom ide afforded21,23 galactoen itol 15 in excellen t yield 81%
if th e reaction was perform ed in THF with butyllith ium at 53 °C for 3 h .
Un der th ese optim ised con dition s, th e form ation 23 of an un desired side
product, dien e 16, arisin g again due to low stability of an in term ediate in
alkalin e solution , was n ot observed. However, th e form ation of dien e 16
was depen den t on th e con dition s used; th e reaction perform ed at 60 °C
gave 15% of 16 accom pan ied by th e yield of th e target product 15 decreas-
in g to 26%. Mercury-m ediated cyclization 23 of galactoen itol 15 provided
stereoselectively th e m ercury derivative 17 isolated after ch rom atograph y
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

Inhibitors of Galactosyltransferases
1665
in a very h igh yield of 89%. Th en , ch lorin e in 17 was substituted with
iodin e in a Fin kelstein -like reaction an d iodide 18 was obtain ed in 92%
yield. Th e reaction of trieth yl ph osph ite with iodide 18 led sm ooth ly to
α-ph osph on ate 12 un der th e con dition s described above in 84% yield. Th e
protectin g groups in ph osph on ate 12 were rem oved in on e step usin g an
excess of iodotrim eth ylsilan e. Free ph osph on ic acid 9 was isolated by pre-
cipitation from m eth an ol by addition of dieth yl eth er21. Th us th e overall
yield of 9 prepared by th is five-step procedure was 52% on a gram scale.
Th e n ew com poun ds 9, 12 an d 15–18 were iden tified in th e usual m an n er
1
(optical rotation , MS, H, 13C an d 31P NMR; see Experim en tal) an d th e C-1
con figuration in 9 was verified by NOE (Fig. 2).
SCHEME 3
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

1666
Heissigerová, Kočalka, Hlaváčková, Imberty, Breton, Chazalet, Moravcová:
Com poun ds 7–9 were tested as in h ibitors of bovin e α-1,3- an d β-1,4-
galactosyltran sferases. α-Galactosylph osph on ic acid 7 sh owed alm ost n o
in h ibition again st α3GalT; it sligh tly in h ibited on ly β4GalT. β-Galactosyl-
ph osph on ic acid 8 an d (α-galactosylm eth yl)ph osph on ic acid 9 were very
poor in h ibitors of both en zym es actin g m ostly in a n on com petitive m ode.
Of th e ph osph on ic acids evaluated, β-galactosylph osph on ic acid 8 dis-
played th e h igh est activity again st both en zym es (for β4GalT: IC50
17 m m ol l–1 at 13.5 µm ol l–1 of UDP-Gal).
=
We elaborated an efficien t syn th esis of ph osph on ic acids 7–9 in reason -
able overall yields. Th ese com poun ds will be con verted in to ph osph on o
an alogues of UDP-Gal as th e n ucleotide m oiety was proved to be an im por-
tan t structural m otif, wh ich can sign ifican tly con tribute to th e bin din g of
in h ibitor in th e active site of en zym e.
EXPERIMENTAL
Optical rotation s were m easured on a Jasco Model DIP-370 polarim eter at 25 °C an d are
given in 10–1 deg cm 2 g–1. NMR spectra were recorded on a Bruker Avan ce DRX-500
(500.1 MHz for 1H, 125,7 MHz for 13C an d 202.2 MHz for 31P). Ch em ical sh ifts are
expressed in parts per m illion down field from Me4Si or ph osph oric acid for 31P spectra.
1H NMR ch em ical sh ifts (δ, ppm ) an d couplin g con stan ts (J, Hz) were obtain ed by first-order
an alysis of th e spectra an d decouplin g experim en ts. 13C NMR spectra were recorded usin g
APT pulse sequen ce. Reaction s were followed on TLC on silica gel (10–40 µm , Merck) or on
cellulose. Colum n ch rom atograph y was carried out on silica gel (100–160 µm , Merck) an d
ch rom atograph ic system s are described in th e text. Mass spectra were recorded on a Q-TOF
Micro spectrom eter (Waters-Microm ass) with electrospray ion isation in n egative m ode. Ele-
m en tal an alysis was perform ed on a CHN-Perkin –Elm er-2400. Solven ts were evaporated at
40 °C/2 kPa an d com poun ds were dried at 13 Pa. Th e usual work-up refers to dilution with
water, extraction with an organ ic solven t (dieth yl eth er or CHCl3), wash in g th e extract with
water to n eutral reaction , dryin g (an h ydrous Na2SO4) an d evaporation of th e solven t.
2,3,4,5-Tetra-O-ben zyl-D-galactopyran ose was prepared from m eth yl β-D-galactopyran oside
(Sigm a–Aldrich , Czech Republic) by alkylation with ben zyl ch loride in th e presen ce of KOH
an d subsequen t acid h ydrolysis (AcOH–H2SO4) of th e glycosidic group24. Fin ally, acetylation
with Ac2O/pyridin e (1:2 v/v) gave 1-O-acetyl-2,3,4,6-tetra-O-ben zyl-D-galactopyran ose (α/β = 2).
En zym e Preparation
Th e gen es of th e bovin e β4GalT (∆114-402) an d α3GalT (∆25-368) were clon ed in dividual-
ly in to pVT-Bac vector. Th e co-tran sfection of Sf9 cells with th e tran sfer vectors an d th e
BaculoGold (BaculoGold, Becton Dickin son Fran ce) lin ear DNA 25 was don e accordin g to th e
m an ufacturer’s in struction s. Recom bin an t protein s are secreted from baculovirus-in fected
cells in th e culture m edium .
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

Inhibitors of Galactosyltransferases
1667
In h ibition Studies
Th e assay used to assess th e in h ibitory poten tial of com poun ds is based on th e tran sfer of
radiolabeled (14C) galactose from th e don or UDP-Gal (Am ersh am , 300 m Ci/m m ol) to th e ac-
ceptor m olecule. Th e un in corporated label was separated from th e radiolabeled product
Gal(14C)-acceptor by adsorption on to AG 1X8 Resin e (form ate). Th e ratio of in corporated to
total added radioactivity is proportion al to th e en zym e activity.
β4GalT Activity
Assays con tain ed 10 m M D-galactose, 1 m M ATP, 10 m M Mn Cl2, 50 m M cacodylate buffer
(pH 7), 3.5 µM UDP-(14C)Gal (106 000 cpm ), UDP-Gal (1–200 µM), GlcNAc acceptor (25 m M),
in h ibitor (0.5–50 m M) an d 5 µl of protein solution in a total assay volum e of 0.05 m l.
In oth er series of assays, th e con cen tration of UDP-Gal was fixed (200 µM) an d th e con cen -
tration s of GlcNAc acceptor were varied (5–40 m M). Assays were perform ed at 37 °C for
10–30 m in . Reaction s were stopped by addition of 1 m l of a m ixedbed resin AG 1X8 slurry
(1:4 w/v in water). Sam ples were vortexed, cen trifuged, an d wash ed th ree tim es with 600 µl
of water. Supern atan ts (4 × 600 µl) were pooled, an d th e radioactivity was m easured by scin -
tillation coun tin g. Parallel con trol reaction s were perform ed in th e absen ce of acceptor. Each
in itial rate was obtain ed at a con sum ption of UDP-Gal lower th an 20%.
α3GalT Activity
Assays con tain ed 1 m M ATP, 10 m M Mn Cl2, 50 m M cacodylate buffer (pH 7), 3.5 µM
UDP-(14C)Gal (106 000 cpm ), UDP-Gal (1–200 µM), N-acetyllactosam in e as acceptor (25 m M),
in h ibitor (0.5–50 m M) an d 5 µl of protein solution in a total assay volum e of 0.05 m l. In
oth er series of assays, th e con cen tration of UDP-Gal was fixed (200 µM) an d th e con cen tra-
tion of lactose acceptor was varied (5–40 m M). Reaction s were th en processed as in dicated
above for th e β4GalT activity.
Dieth yl (2,3,4,6-Tetra-O-ben zyl-α- an d β-D-galactopyran osyl)ph osph on ates (10 an d 11)
Un der argon , TMSOTf (1.1 m l, 6.0 m m ol) was added dropwise durin g 10 m in to a m ixture
of 1-O-acetyl-2,3,4,6-tetra-O-ben zyl-D-galactopyran ose (3.49 g, 6.0 m m ol) an d P(OEt)3 (1.2 m l,
6.6 m m ol) in CH2Cl2 (20 m l) at 0 °C. Th e solution was left to warm to room tem perature
an d stirred un til th e startin g m aterial h ad disappeared (24 h ). Th e reaction was quen ch ed
with water, diluted with AcOEt, th e organ ic layer was extracted with saturated aqueous
NaHCO3, saturated aqueous NaCl an d H2O, dried over an h ydrous Na2SO4 an d con cen trated.
Ch rom atograph y of th e residue on silica gel in th e system AcOEt–CH2Cl2–h exan e (1:1:1) af-
forded com poun d 10 (1.88 g, 2.9 m m ol, 48%), [α]D +51.7 (c 6.4, CHCl3) an d th e β-an om er
11 (1.31 g, 2.0 m m ol, 32%), [α]D +34.3 (c 2.6, CHCl3).
Compound 10: 1H NMR (500 MHz, CDCl3): 7.31–7.24 m , 20 H (H-arom ); 4.75–4.45 m , 8 H
(CH2Ph ); 4.42 dd, 1 H, J(1,2) = 4.5, J(1,P) = 12.9 (H-1); 4.35–4.31 m , 1 H (H-5); 4.20–4.01 m ,
7 H (CH2O, H-2, H-3, H-4); 3.81 dd, 1 H, J(6′,5) = 7.7, J(gem ) = 9.6 (H-6′); 3.58 dd, 1 H,
J(6,5) = 4.2, J(gem ) = 10.7 (H-6); 1.23–1.18 m , 6 H (CH3). 13C NMR: 138.36, 138.25, 138.15,
137.96 (quatern ary C-arom ); 127.47–128.33 (CH-arom ); 75.29 (C-3); 75.26 (C-2); 75.20
(C-5); 73.83 (C-4); 73.83, 73.42, 73.11, 72.97 (CH2Ph ); 69.20 d, J(C,P) = 157.2 (C-1); 67.35
(C-6); 62.93 d, J(C,P) = 6.4 (CH2O); 62.06 d, J(C,P) = 6.4 (CH2O); 16.33 d, J(C,P) = 5.8 (CH3);
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

1668
Heissigerová, Kočalka, Hlaváčková, Imberty, Breton, Chazalet, Moravcová:
16.25 d, J(C,P) = 6.3 (CH3). 31P NMR: 21.71 (J(2,P) = 25.0). For C38H45O8P (660.8) calculated:
69.08% C, 6.86% H; foun d: 69.19% C, 6.98% H.
Compound 11: 1H NMR: 7.34–7.25 m , 20 H (H-arom ); 4.96 d, 1 H, J(gem ) = 11.6 (CH2Ph );
4.93 d, 1 H, J(gem ) = 10.3 (CH2Ph ); 4.86 d, 1 H, J(gem ) = 10.2 (CH2Ph ); 4.74 d, 1 H, J(gem ) =
11.7 (CH2Ph ); 4.70 d, 1 H, J(gem ) = 11.7 (CH2Ph ); 4.58 d, 1 H, J(gem ) = 11.5 (CH2Ph ); 4.45 d,
1 H, J(gem ) = 11.9 (CH2Ph ); 4.41 d, 1 H, J(gem ) = 11.8 (CH2Ph ); 4.23 ddd, 1 H, 4.96 d, 1 H,
J(2,1) = J(2,3) = J(2,1) = J(2,P) = 10.0 (H-2); 4.19–4.07 m , 4 H (CH2O); 3.97 bs, 1 H (H-4);
3.65 dd, 1 H, J(1,2) = J(1,P) = 9.8 (H-1); 3.62–3.52 m , 4 H (H-3, H-6′, H-6, H-5); 1.26–1.19 m ,
6 H (CH3). 13C NMR: 138.69, 138.43, 138.14, 137.76 (quatern ary C-arom ); 128.41–127.42
(CH-arom ); 84.66 d, J(C,P) = 17.6 (C-3 or C-5); 78.78 d, J(C,P) = 16.6 (C-3 or C-5); 75.34
(C-2); 75.19 (CH2Ph ); 75.16 d, J(C,P) = 171.0 (C-1); 74.44 (CH2Ph ); 73.54 (C-4); 73.47, 72.44
(CH2Ph ); 68.64 (C-6); 63.17 d, J(C,P) = 6.4 (OCH2); 62.42 d, J(C,P) = 6.4 (OCH2); 16.20 d,
J(C,P) = 6.4 (CH3); 16.31 d, J(C,P) = 6.4 (CH3). 31P NMR: 20.41. For C38H45O8P (660.8) calcu-
lated: 69.08%C, 6.86% H; foun d: 69.25% C, 7.02% H.
α-D-Galactopyran osylph osph on ic Acid (7)
Un der argon , a solution of 10 (1.81 g, 2.7 m m ol) in CH2Cl2 (10 m l) was cooled to 0 °C an d
Me3SiI (2.5 m l, 18.4 m m ol) was added. After 3 h , th e solven t was rem oved un der dim in -
ish ed pressure; th e brown residue was dissolved in a few m l of MeOH an d precipitated by
addition of dieth yl eth er affordin g h ygroscopic am orph ous 7 (0.45 g, 1.84 m m ol, 69%), [α]D
+58.1 (c 0.4, H2O). 1H NMR (D2O): 3.63 dd, 1 H, J(5,6a) = 4.1, J(6a,6b) = 12.0 (H-6a);
3.75 dd, 1 H, J(5,6b) = 7.9 (H-6b); 3.98 bs, 1 H (H-4); 4.01–4.04 m , 1 H (H-5); 4.06 bs, 1 H
(H-3); 4.11 m , 1 H (H-2); 4.24 dd, J(1,2) = 6.2, J(1,P) = 11.0 (H-1). 13C NMR: 61.57 (C-6);
68.52 (C-2); 69.19 (C-4); 71.02 (C-3); 73.13 d, J(C,P) = 151.1 (C-1); 77.30 d, J(C,P) = 3.0
(C-5); 31P NMR: 19.16. HR-MS: for C6H12O8P (M – H)– calculated: 243.0210, foun d:
243.0350.
β-D-Galactopyran osylph osph on ic Acid (8)
Treatm en t of 11 (0.97 g, 1.5 m m ol) in CH2Cl2 (8 m l) with Me3SiI (1.8 m l, 13.2 m m ol), as
described for 7, afforded h ygroscopic am orph ous 8 (0.31 g, 1.27 m m ol, 85%), [α]D –75.8
(c 0.5, H2O). 1H NMR (D2O): 3.88 d, 1 H, J(4,3) = 3.1 (H-4); 3.81 ddd, 1 H, J(2,3) = J(1,2) =
J(2,P) = 9.8 (H-2); 3.71 dd, 1 H, J(6′,5) = 8.4, J(gem ) = 11.6 (H-6′); 3.63–3.54 m , 3 H (H-3,
H-5, H-6); 3.41 dd, 1 H, J(1,2) = J(1,P) = 9.8 (H-1). 13C NMR: 81.48 d, J(C,P) = 14.5 (C-5);
76.70 d, J(C,P) = 162.7 (C-1); 75.23 d, J(C,P) = 16.2 (C-3); 70.28 (C-4); 68.58 (C-2); 62.70
(C-6). 31P NMR: 17.77. HR-MS: for C6H12O8P (M – H)– calculated 243.0210, foun d: 243.0532.
MS-ESI: 243 (M – H)–, 225 (M – H – H2O)–.
Dieth yl [(2,3,4,6-Tetra-O-ben zyl-α- an d β-D-galactopyran osyl)m eth yl]-
ph osph on ates (12 an d 13)
To a suspen sion of NaH (1 g, 41.7 m m ol) in THF (50 m l), at 0 °C un der argon , CH2[PO(OEt)2]2
(7 m l, 28.2 m m ol) was slowly added. Th e clear solution obtain ed was stirred at 0 °C for
30 m in . Th en a solution of 2,3,4,6-tetra-O-ben zyl-D-galactopyran ose (1.60 g, 2.9 m m ol) in
THF (40 m l) was added. Th e solution was warm to room tem perature an d stirred for 18 h .
Th e m ixture was th en wash ed with saturated aqueous NH4Cl an d th e water layer extracted
with CH2Cl2. Th e com bin ed organ ic layers were dried over an h ydrous Na2SO4 an d con cen -
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

Inhibitors of Galactosyltransferases
1669
trated. Th e residue was ch rom atograph ed on a colum n of silica gel in th e system petroleum
eth er–AcOEt (1:4). Ph osph on ates 12 an d 13 (0.74 g, 37%) were obtain ed as an in separable
m ixture followed by dieth yl [(1E,3E,5R,6R)-3,5,7-tris(ben zyloxy)-6-h ydroxyh epta-1,3-dien -
1-yl]ph osph on ate (14) (0.20 g, 0.4 m m ol, 14%). 1H NMR (500 MHz, CDCl3): 7.30–7.25 m ,
6 H (H-arom ); 6.95 dd, 1 H, J(6,7) = 17.1, J(6,P) = 21.8 (H-6); 6.02 dd, 1 H, J(7,6) = 17.4,
J(7,P) = 18.5 (H-7); 5.45 d, 1 H, J(4,3) = 9.6 (H-4); 4.70 s, 2 H (CH2Ph ); 4.52–4.46 m , 3 H
(CH2Ph ); 4.42 dd, 1 H, J(3,4) = 9.6, J(3,2) = 5.4 (H-3); 4.24 d, 1 H J(gem ) = 11.7 (CH2Ph );
4.14–4.05 m , 4 H (OCH2); 3.70 dd, 1 H, J(2,3) = 10.4, J(2,1) = 5.2 (H-2); 3.47 d, 2 H, J(1,2) =
5.2 (H-1, H-1′); 1.35 t, 6 H, J(CH3CH2) = 7.0 (CH3). 13C NMR: 154.86 d, J(5,P) = 23.9 (C-5);
143.43 d, J(6,P) = 6.9 (C-6); 137.94 (2 × quatern ary C-arom ); 136.47 (quatern ary C-arom );
128.56–127.57 (CH-arom ); 122.78 (C-4); 116.21 d, J(7,P) = 190.7 (C-7); 74.35, 73.50
(CH2Ph ); 73.29 (C-3); 72.97 (C-2); 70.70 (CH2Ph , C-1); 62.00 d, J(H,P) = 5.4 (OCH2); 16.37 d,
J(H,P) = 5.6 (CH3). 31P NMR: 18.87. For C32H39O7P (566.6) calculated: 67.83% C, 6.94% H;
foun d: 68.00% C, 7.02% H.
3,4,5,7-Tetra-O-ben zyl-1,2-dideoxy-D-galacto-h ept-1-en itol (15)
Method A. To a suspen sion of MePh 3PBr (2.34 g, 6.5 m m ol) in THF (15 m l), un der argon ,
butyllith ium (1.6 M in h exan e, 4.9 m l, 7.8 m m ol) was added dropwise. After 30 m in of vig-
orous stirrin g, a solution of 2,3,4,6-tetra-O-ben zyl-D-galactopyran ose (1.23 g, 2.3 m m ol) in
THF (12 m l) was added; th e m ixture was h eated at 60 °C an d stirred for 3 h . Th e organ ic
layer was th en wash ed several tim es with saturated aqueous Na2CO3 an d th e com bin ed
aqueous layers were extracted with dieth yl eth er. Th e com bin ed organ ic layers were dried
over an h ydrous Na2SO4 an d con cen trated. Th e resultin g syrup was subjected to silica gel
ch rom atograph y in system petroleum eth er–AcOEt (4:1) to afford th e m ain product 15
(0.32 g, 0.6 m m ol, 26%), [α]D –6.8 (c 1.1, CHCl3). 1H NMR (500 MHz, CDCl3): 7.34–7.17 m ,
20 H (H-arom ); 5.87 ddd, 1 H, J(2,3) = 7.9 (H-2); 5.87 d, 1 H, J(1′,2) = 17.4 (H-1′); 5.29 d,
1 H, J(1,2) = 10.4 (H-1); 4.75 s, 2 H (CH2Ph ); 4.64 d, 1 H, J(gem ) = 11.8 (CH2Ph ); 4.47 d,
1 H, J(gem ) = 11.9 (CH2Ph ); 4.44–4.32 m , 4 H (CH2Ph ); 4.12 dd, 1 H, J(6,7) = J(6,7′) = 6.0
(H-6); 4.08 dd, 1 H, J(3,4) = 4.1 (H-3); 3.82–3.78 m , 2 H (H-4, H-5); 3.58–3.46 m , 2 H (H-7′,
H-7); 3.09 s, 1 H (OH). 13C NMR: 138.23, 138.17, 138.12, 138.01 (quatern ary C-arom );
135.70 (C-2); 128.26–127.51 (CH-arom ); 119.08 (C-1); 82.13 (C-4 or C-5); 80.75 (C-3); 76.63
(C-4 or C-5); 75.16, 73.11 (2 × CH2Ph ); 73.06 (C-7); 71.17, 70.30 (2 × CH2Ph ); 69.67 (C-6).
For C35H38O5 (538.7) calculated: 78.04% C, 7.11% H; foun d: 77.82.25% C, 6.98% H.
Th en th e elim in ation product (2R,3S,4E)-1,3,5-tris(ben zyloxy)h epta-4,6-dien -2-ol (16) was
isolated (0.16 g, 0.3 m m ol, 15%), [α]D –5.8 (c 0.5, CHCl3). 1H NMR (500 MHz, CDCl3):
7.36–7.26 m , 15 H (H-arom ); 6.24 dd, 1 H (H-6); 5.58 d, 1 H, J(7′,6) = 17.4 (H-7′); 5.28 d,
1 H, J(7,6) = 10.9 (H-7); 5.12 d, 1 H, J(4,3) = 9.7 (H-4); 4.78 d, 1 H, J(gem ) = 11.3 (CH2Ph );
4.74 d, 1 H, J(gem ) = 11.3 (CH2Ph ); 4.54–4.48 m , 3 H (CH2Ph ); 4.42 dd, 1 H, J(3,2) = 6.0
(H-3); 4.27 d, 1 H, J(gem ) = 11.7 (CH2Ph ); 3.73–3.69 m , 1 H (H-2); 3.51–3.45 m , 2 H (H-1′,
H-1); 2.72 s, 1 H (OH). 13C NMR: 156.79 (C-5); 138.34, 138.12, 137.10 (quatern ary C-arom );
131.84 (C-6); 128.50–127.51 (CH-arom ); 116.46 (C-7); 115.10 (C-4); 73.47 (CH2Ph ); 73.43
(C-3); 73.40 (CH2Ph ); 73.28 (C-2); 70.99 (C-1); 70.25 (CH2Ph ). For C28H30O4 (430.6) calcu-
lated: 78.11% C, 7.02% H; foun d: 77.95% C, 7.22% H.
Method B. Usin g MePh 3PBr (6.18 g, 17.3 m m ol) in THF (45 m l), butyllith ium (1.6 M in
h exan e, 13 m l, 20.8 m m ol) an d a solution of 2,3,4,6-tetra-O-ben zyl-D-galactopyran ose (3.32 g,
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

1670
Heissigerová, Kočalka, Hlaváčková, Imberty, Breton, Chazalet, Moravcová:
6.1 m m ol) in THF (15 m l) an d keepin g reaction tem perature in th e procedure described
above at 53 °C, com poun d 15 was prepared in 82% yield (2.69 g, 5.0 m m ol).
2,6-An h ydro-1,3,4,5-tetra-O-ben zyl-7-(ch loroh ydrargyrio)-7-deoxy-D-glycero-
L-galacto-h eptitol (17)
A solution of 15 (2.69 g, 5.0 m m ol) in THF (20 m l) was stirred with m ercuric acetate (1.9 g,
6.0 m m ol) un til th e startin g m aterial disappeared (24 h ). Th en aqueous KCl (c 0.34 g/m l,
8 m l) was added an d th e m ixture was stirred for 2 h . After th e usual work-up, th e resultin g
syrup was purified by ch rom atograph y on silica gel in petroleum eth er–AcOEt (3:1) to afford
17 (3.44 g, 4.5 m m ol, 90%), [α]D +29.3 (c 1.2, CHCl3). 1H NMR (500 MHz, CDCl3):
7.37–7.30 m , 20 H (H-arom ); 4.76 d, 1 H, J(gem ) = 11.8 (CH2Ph ); 4.68–4.48 m , 7 H (CH2Ph );
4.38–4.32 m , 1 H (H-2); 4.19 bs, 1 H (H-6); 4.00–3.92 m , 2 H (H-5, H-7′); 3.80–3.73 m , 1 H
(H-4); 3.67 dd, 1 H, J(7,6) = 3.2, J(gem ) = 10.6 (H-7); 3.60–3.54 m , 1 H (H-3); 2.02 dd, 1 H,
J(1′,2) = 7.5, J(gem ) = 11.9 (H-1′); 1.55 dd, 1 H, J(1′,2) = 3.1 (H-1). 13C NMR: 138.26, 138.20
(2×); 137.06 (quatern ary C-arom ); 128.86–127.54 (CH-arom ); 76.32 (C-3); 75.13 (C-4); 73.83
(C-5); 73.43 (CH2Ph ); 73.20 (C-6); 73.09 (2×); 72.65 (CH2Ph ); 68.20 (C-2); 66.71 (C-7); 28.85
(C-1). For C35H37ClHgO5 (773.7) calculated: 54.33% C, 4.82% H; foun d: 54.25% C, 4.14% H.
2,6-An h ydro-1,3,4,5-tetra-O-ben zyl-7-deoxy-7-iodo-D-glycero-L-galacto-h eptitol (18)
Com poun d 17 (2.61 g, 3.4 m m ol) was dissolved in an iodin e solution (0.86 g, 3.4 m m ol) in
CH2Cl2 (30 m l). Th e m ixture was stirred for 2 h , wash ed with 10% aqueous Na2SO3 an d
th en 5% aqueous KI, an d worked up as usual to afford 18 as oil (2.06 g, 3.1 m m ol, 91%).
1H NMR (500 MHz, CDCl3): 7.36–7.26 m , 20 H (H-arom ); 4.70–4.52 m , 8 H (CH2Ph );
4.13–4.08 m , 1 H (H-6); 4.07–4.04 m , 1 H (H-2); 4.02 dd, 1 H, J(5,4) = 2.8, J(5,6) = 3.8 (H-5);
3.94 dd, 1 H, J(3,4) = 6.4, J(3,2) = 3.6 (H-3); 3.91 dd, 1 H, J(1′,2) = 7.3, J(gem ) = 10.6 (H-1′);
3.73 dd, 1 H, J(1,2) = 4.5 (H-1); 3.71 dd, 1 H (H-4); 3.40 dd, 1 H, J(7′,6) = 6.2, J(gem ) = 10.4
(H-7′); 3.31 dd, 1 H, J(7,6) = 8.7 (H-7). 13C NMR: 138.32, 138.28, 138.23, 137.84 (quatern ary
C-arom ); 128.44–127.50 (CH-arom ); 75.85, 75.77 (C-4, C-3); 73.70 (C-5); 73.48, 73.26
(CH2Ph ); 73.13 (C-2); 72.92 (2 × CH2Ph ); 71.63 (C-6); 66.83 (C-1); 2.72 (C-7). For C35H37IO5
(664.6) calculated: 63.26% C, 5.61% H; foun d: 63.35% C, 5.72% H.
Dieth yl [(2,3,4,6-tetra-O-ben zyl-α-D-galactopyran osyl)m eth yl]ph osph on ate (12)
Un der argon , th e iodo derivative 18 (2.06 g, 3.1 m m ol) was dissolved in P(OEt)3 (4 m l,
23.0 m m ol). Th e m ixture was h eated un der reflux for 3 h . Th e excess of reagen t was
rem oved un der reduced pressure an d th e residue was ch rom atograph ed on silica gel in
petroleum eth er–AcOEt (1:1) to afford 12 (1.75 g, 2.6 m m ol, 84%), [α]D +45.2 (c 0.5, CHCl3).
1H NMR (500 MHz, CDCl3): 7.30–7.25 m , 20 H (H-arom ); 4.69 d, 1 H, J(gem ) = 11.8
(CH2Ph ); 4.67 d, 1 H, J(gem ) = 12.1 (CH2Ph ); 4.61–4.43 m , 7 H (CH2Ph , H-1); 4.08–4.02 m ,
5 H (CH2O, H-5); 4.01 dd, J(2,3) = 2.8, J(2,1) = 3.1 (H-2); 3.91–3.86 m , 1 H (H-4); 3.85–3.80
m , 1 H (H-6′); 3.71 dd, 1 H, J(6,5) = 5.3, J(gem ) = 10.1 (H-6); 3.63 dd, 1 H, J(3,4) = 7.02
(H-3); 2.17–2.14 m , 1 H (CH2P); 2.12–2.10 m , 1 H (CH2P); 1.27 t, 1 H, J(CH3,CH2) = 7.1
(CH3); 1.22 t, 1 H, J(CH3,CH2) = 7.1 (CH3). 13C NMR: 138.43, 138.35, 138.27, 138.07 (qua-
tern ary C-arom ); 128.34–127.45 (CH-arom ); 76.28, 76.20 (C-3, C-4); 73.70 (C-2, C-5); 73.28,
73.21, 73.11, 72.75 (CH2Ph ); 67.20 (C-1); 67.15 (C-6); 61.67 d, J(C,P) = 6.4 (CH2O); 61.58 d,
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

Inhibitors of Galactosyltransferases
1671
J(C,P) = 6.4 (CH2O); 24.79 d, J(C,P) = 142.5 (CH2P); 16.37 d, J(C,P) = 6.3 (CH3). 31P NMR:
29.72. For C39H47O8P (674.8) calculated: 69.48% C, 7.02% H; foun d: 69.68% C, 6.95% H.
(α-D-Galactopyran osylm eth yl)ph osph on ic Acid (9)
A solution of 12 (1.47 g, 2.2 m m ol) in CH2Cl2 (15 m l) un der argon was cooled to 0 °C an d
Me3SiI (1.7 m l, 12.5 m m ol) was added. After 3 h th e solven t was rem oved un der reduced
pressure, th e brown residue was dissolved in a few m l of MeOH an d precipitation by addi-
tion of eth yl m eth yl keton e afforded am orph ous com poun d 9 (0.52 g, 2.0 m m ol, 91%), [α]D
+86.9 (c 0.2, H2O). 1H NMR (500 MHz, D2O): 4.30–4.22 m , 1 H (H-1); 3.87–3.78 m , 2 H
(H-2, H-4); 3.70 dd, J(5,6) = J(5,4) = 5.1 (H-5); 3.60–3.55 m , 2 H (H-6′, H-3); 3.51 dd, 1 H,
J(gem ) = 11.5 (H-6); 2.15 dd, 1 H, J(gem ) = 16.0, J(H,P) = 28.8 (CH2P); 1.95 dd, 1 H J(H,P) =
16.8 (CH2P). 13C NMR: 72.93 (C-5); 72.39 (C-1); 70.48 (C-3); 69.67 (C-4); 68.71 d, J(C,P) =
12.3 (C-2); 61.65 (C-6); 23.85 d, J(C,P) = 138.2 (CH2P). 31P NMR: 28.39. HR-MS: calculated
for C7H14O8P (M – H): 257.0427, foun d: 257.0676.
We thank Dr D. Sýkora for MS and Dr H. Dvořáková for NMR measurement. This work was done
within the frame of the project of the Grant Agency of the Czech Republic No. 203/05/0355 and of the
project of the Czech Ministry of Education, Youth and Sports MSM 6046137305.
REFERENCES
1. Varki A.: Glycobiology 1993, 3, 97.
2. Unligil U. M., Rini J. M.: Curr. Opin. Struct. Biol. 2000, 10, 510.
3. Tarbouriech N., Charnock S. J., Davies G. J.: J. Mol. Biol. 2001, 314, 655.
4. Gastinel L. N., Bignon C., Misra A. K., Hindsgaul O., Shaper J. H., Joziasse D. H.: EMBO J.
2001, 20, 638.
5. Parker W., Saadi S., Lin S. S., Holzknecht Z. E., Bustos M., Platt J. L.: Immunol. Today
1996, 17, 373.
6. Hashimoto H., Endo T., Kajihara Y.: J. Org. Chem. 1997, 62, 1914.
7. Takayama S., Chung S. J., Igarashi Y., Ichikawa Y., Sepp A., Lechler R. I., Wu J., Hayashi Z.,
Siuzdak G., Wong C. H.: Bioorg. Med. Chem. 1999, 7, 401.
8. Kim Y. J., Ichikawa M., Ichikawa Y.: J. Am. Chem. Soc. 1999, 121, 5829.
9. Vaghefi M. M., Bernacki R. J., Halley K., Wilson B. E., Robins R. K.: J. Med. Chem. 1987,
30, 1383.
10. Vaghefi M. M., Bernacki R. J., Hennen W. J., Robins R. K.: J. Med. Chem. 1987, 30, 1391.
11. Endo T., Kajihara Y., Kodama H., Hashimoto H.: Bioorg. Med. Chem. 1996, 4, 1939.
12. Wang R., Steensma D. H., Takaoka Y., Yun J. W., Kajimoto T., Wong C. H.: Bioorg. Med.
Chem. 1997, 5, 661.
13. Schmidt R. R., Frische K.: Bioorg. Med. Chem. Lett. 1993, 3, 1747.
14. Vidal S., Bruyere I., Malleron A., Augé C., Praly J.-P.: Bioorg. Med. Chem. 2006, 14, 7293.
15. Frische K., Schmidt R. R.: Justus Liebigs Ann. Chem. 1994, 297.
16. Czollner L., Baudin G., Bernet B., Vasella A.: Helv. Chim. Acta 1993, 76, 1013.
17. Meuwly R., Vasella A.: Helv. Chim. Acta 1986, 69, 25.
18. Cipolla L., La Ferla B., Panza L., Nicotra F.: J. Carbohydr. Chem. 1998, 17, 1003.
19. Walliman K., Vasella A.: Helv. Chim. Acta 1990, 73, 1359.
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672

1672
Heissigerová, Kočalka, Hlaváčková, Imberty, Breton, Chazalet, Moravcová:
20. Casero F., Cipolla L., Lay L., Nicotra F., Panza L., Russo G.: J. Org. Chem. 1996, 61, 3428.
21. Nicotra N., Ronchetti F., Russo G.: J. Org. Chem. 1982, 47, 4459.
22. Borodkin V. S., Ferguson A. J., Nikolaev A. V.: Tetrahedron Lett. 2001, 42, 5305.
23. Pougny J. R., Nassr M. A., Sinay P.: J. Chem. Soc., Chem. Commun. 1981, 8, 375.
24. Adinolfi M., Barone G., Iadonisi A., Mangoni L.: Tetrahedron Lett. 1998, 39, 2021.
25. Tessier D. C., Thomas D. Y., Khouri H. E., Laliberte F., Vernet T.: Gene 1991, 98, 177.
Collect. Czech. Chem. Commun. 2006, Vol. 71, Nos. 11–12, pp. 1659–1672