Synthesis of a tetrasubstituted arylphosphonate via the anionic phospho-Fries rearrangement

Article abstract of DOI:10.1016/j.tetlet.2005.04.104

The anionic phospho-Fries rearrangement of phosphoric acid (3,5-di-isopropoxy)phenyl ester diethyl ester (11) gave rise to (2-hydroxy-4,6-di-isopropoxy-phenyl)phosphonic acid diethyl ester (12) in excellent yield. The phenol functionality of 12 was converted to the corresponding triflate which was coupled with vinyltributylstannane, under Stille conditions, to give a styrene. This molecule is intended to serve as the aromatic fragment in the synthesis of a phosphorus-based transition-state analogue for the hydrolysis of the S-(-)-zearalenone lactone.

Full text of DOI:10.1016/j.tetlet.2005.04.104

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 4311–4313
Synthesis of a tetrasubstituted arylphosphonate via the
anionic phospho-Fries rearrangement
Krishanthi P. Jayasundera, Amy J. Watson and Carol M. Taylor*
Institute of Fundamental Sciences, Massey University, Private Bag 11-222, Palmerston North, New Zealand
Received 2 April 2005; revised 15 April 2005; accepted 25 April 2005
Abstract—The anionic phospho-Fries rearrangement of phosphoric acid (3,5-di-isopropoxy)phenyl ester diethyl ester (11) gave rise
to (2-hydroxy-4,6-di-isopropoxy-phenyl)phosphonic acid diethyl ester (12) in excellent yield. The phenol functionality of 12 was con-
verted to the corresponding triflate which was coupled with vinyltributylstannane, under Stille conditions, to give a styrene. This
molecule is intended to serve as the aromatic fragment in the synthesis of a phosphorus-based transition-state analogue for the
hydrolysis of the S-(À)-zearalenone lactone.
Ó 2005 Elsevier Ltd. All rights reserved.
S-(À)-Zearalenone (1, Fig. 1) is a macrolactone of the
resorcylic acid family;1 it is an estrogenic mycotoxin that
has a negative impact on the fertility of livestock. In
efforts directed toward the generation of antibodies that
might catalyze the degradation of S-(À)-zearalenone, we
sought to prepare a transition-state analogue for the
hydrolysis of the lactone, viz. 2. Synthesis of 2 could
the type I (Scheme 1) was our immediate target. It was
envisioned that a precursor II would undergo regioselec-
tive deprotonation ortho to an electron-withdrawing
group X, which would subsequently rearrange to give
the tetrasubstituted aromatic ring. While many simple
examples of this anionic phospho-Fries rearrangement3
have been reported previously, such a highly substituted
example was unprecedented.
draw heavily upon existing syntheses of zearalenone
2
(1), vis-a-vis construction of the aliphatic part of the
`
molecule. However, the aryl phosphonate presented
two significant challenges: formation of the Ar–P bond
and formation of the macrolactone. Herein, we describe
our efforts toward the assembly of the aromatic subunit
in compound 2.
Scheme 1.
In a forward sense, resorcylic acid (3) was converted to
its ethyl ester (Scheme 2). Protection as the mono-p-
methoxybenzyl ether was difficult to achieve in good
yield. However, phosphorylation of the remaining phe-
nol under Atherton–Todd conditions4 gave 4. Treat-
ment of 4 with LDA gave a single product, in good
yield, which we hoped was 5a.
Figure 1.
Wittig chemistry was selected intially for the macro-
cyclization step in our synthesis of 2, thus an aldehyde of
One-dimensional NMR spectra of the isolated material
was consistent with both 5a and 5b. Confirmation of
the formation of the ortho-hydroxyaryl phosphonate
Keywords: Zearalenone; Phosphonate; Fries rearrangement.
*
Corresponding author. Tel.: +64 6 356 9099x3571; fax: +64 6 350
5682; e-mail: C.M.Taylor@massey.ac.nz
functionality was afforded by a doublet in the
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.04.104

4312
K. P. Jayasundera et al. / Tetrahedron Letters 46 (2005) 4311–4313
Scheme 2.
13C NMR at d 100.3 ppm (2JCP = 177 Hz). A sharp sig-
nal was observed for the phenolic proton (d 11.32 ppm)
which is strongly hydrogen bonded to the adjacent phos-
phoryl oxygen. It was the two-dimensional HMBC spec-
trum which verified that the isolated material was in fact
regioisomer 5b, the undesired compound. The 13C signal
of the ester carbonyl at d 165 ppm showed a correlation
to both H2 and H6 (Fig. 2), indicating that there was no
substituent ortho to the COOEt group. Conversely, C5,
bearing the OPMB substituent (d 159 ppm) showed long
range coupling to only H6. Thus, it was concluded that
the product was 5b and that the OPMB group is supe-
rior to the COOEt substituent at directing metallation
to the ortho-position.
phosphate 7. Unfortunately, rearrangement gave com-
pound 8, again the undesired regioisomer.
It seemed we were unlikely to overcome the strong
directing effect of the two meta-related ether substitu-
ents. Rearrangement of a symmetrical substrate IV7
(Scheme 4) could only give one product. The resulting
phenol might be elaborated to a styrene III which would
be useful in a RCM approach8 to target 2.
Scheme 4.
The realization of this approach is illustrated in Scheme
5. Formation of di-isopropyl ether 10 from phloroglu-
cinol (9) was difficult,9 although the mono-protected
derivative could be isolated and resubjected to the reac-
tion conditions. Phosphorylation gave 11 which rear-
ranged cleanly to give 12. We experienced inordinate
difficulty in converting phenol 12 to triflate 13 (vide
supra). Ultimately, anion formation in the presence of
a crown ether, followed by reaction with diphenyltrifla-
mide afforded triflate 13. This was converted unevent-
fully to styrene 14 under Stille conditions.
Figure 2. HMBC correlations in compound 5b.
We were optimistic that replacing the ethyl ester with a
diethyl amide, according to Meyers and Avila,5 would
reverse the regioselectivity. Tribenzylation of resorcylic
acid (3), followed by hydrolysis of the ester gave an acid
which was converted to diethylamide 6 in good overall
yield (Scheme 3). It was possible to cleave one of the
benzyl ethers in 66 and convert the resulting phenol to
The difficulties encountered in the formation of triflate
13 seem to be related to the acidity and/or hydrogen
bonding capacity of its phenol. Some relevant com-
pounds and their estimated pKa values10 are given in
Figure 3. Furstner et al. reported the conversion of 15
¨
to the corresponding triflate using triflic anhydride and
pyridine in dichloromethane.8 These conditions could
be used to convert compound 16 to its triflate ester,
but in poor yield. We found that N-phenyltrifluorome-
thanesulfonamide was more effective than triflic anhy-
dride but the yield was still only 65%. These, and
many other reaction conditions, failed to deliver triflate
13. The relative acidity of 17,7a 187b (and 12) implies that
their conjugate bases are weaker and inter alia maybe
poorer nucleophiles.
Scheme 3.

K. P. Jayasundera et al. / Tetrahedron Letters 46 (2005) 4311–4313
4313
5 equiv. Me2CHBr
OiPr
OH
OiPr
HP(=O)(OEt)2
CCl4, MeCN
K2CO3, KI
LDA
THF
DMF, 50 ˚C, 3 d
O
-78 ˚C
(92%)
(30%)
iPrO
OH
HO
OH
P(OEt)2
iPrO
O
(93%)
9
10
11
OiPr O
P(OEt)2
OiPr O
OiPr O
1. NaH,
nBu3SnCH=CH2
Pd(PPh3)2Cl2
LiCl, DMF
12-crown-4
P(OEt)2
P(OEt)2
THF
2. TfNPh2
(89%)
iPrO
OTf
iPrO
OH
iPrO
(67%)
13
12
14
Scheme 5.
References and notes
1. (a) Stob, M.; Baldwin, R. S.; Tuite, J.; Andrews, F. N.;
Gillette, K. G. Nature (Lond.) 1962, 196, 1318; (b) Urry,
W. H.; Wehrmeister, H. L.; Hodge, E. B.; Hidy, P. H.
Tetrahedron Lett. 1966, 3109–3114; (c) Shipchandler, M.
T. Heterocycles 1975, 3, 471–520; (d) Mirocha, C. J.;
Christensen, C. M.; Nelson, G. H. In Microbial Toxins;
Kadis, S., Ed.; Academic: New York, 1971; Vol VII, pp
107–138, Chapter 4; (e) Betina, V. In Mycotoxins. Chem-
ical, Biological and Environmental Aspects; Elsevier:
Amsterdam, 1989, Chapter 12, pp 271–284.
2. Zearalenone was first synthesized at Merck Sharp and
Dohme in Rahway: Taub, D.; Girotra, N. N.; Hoffsom-
mer, R. D.; Kuo, C. H.; Slates, H. L.; Weber, S.; Wendler,
N. L. Tetrahedron 1968, 24, 2443–2461. An extensive list
of previous synthetic work is beyond the scope of this
letter.
Figure 3.
In summary, we have uncovered some unexpected regio-
chemical outcomes in the anionic phospho-Fries rear-
rangement. For the purposes of our ultimate synthetic
goal, the problem was circumvented by rearrangement
of a symmetrical substrate 11. Elaboration of the result-
ing phenol gave styrene 14 which is being utilized in the
synthesis of compound 2, a transition-state analogue for
the hydrolysis of the lactone in S-(À)-zearalenone. This
will be reported in due course.
3. Taylor, C. M.; Watson, A. J. Curr. Org. Chem. 2004, 8,
623–636, and references cited therein.
4. Atherton, F. R.; Todd, A. R. J. Chem. Soc. 1947, 674–676.
5. Meyers, A. I.; Avila, W. B. Tetrahedron Lett. 1980, 21,
3335–3338.
6. Maleski, R. J. Synth. Commun. 1995, 25, 2327–2335.
7. For the rearrangement of similarly substituted phosphates
see: (a) Melvin, L. S. Tetrahedron Lett. 1981, 22, 3375–
3376; (b) Dhawan, B.; Redmore, D. Phosphorus, Sulfur
Silicon 1991, 61, 183–187.
Acknowledgments
We thank Drs. Chris Miles, Ian Garthwaite, Mark
Dines and Neale Towers of AgResearch (Ruakura,
NZ) for helpful discussions. K.P.J. and C.M.T. thank
the Massey University Research Fund (MURF) for a
postdoctoral fellowship.
8. Use of ring-closing metathesis (RCM) as the macrocycli-
zation step in the synthesis of zearalenone: Furstner, A.;
¨
Thiel, O. R.; Kindler, N.; Bartkowska, B. J. Org. Chem.
2000, 65, 7990–7995.
9. Compound 10 has also been prepared via the monotosy-
late: Chiba, K.; Arakawa, T.; Tada, M. J. Chem. Soc.,
Perkin Trans. 1 1998, 2939–2942.
Supplementary data
10. These values were obtained from SciFinder Scholar, which
states that they were calculated using Advanced Chemistry
Development (ACD/Labs) Software Solaris V4.67, error:
0.2 units.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.
2005.04.104.