Selective bromochlorination of a homoallylic alcohol for the total synthesis of (-)-Anverene

Article abstract of DOI:10.3762/bjoc.12.129

The scope of a recently reported method for the catalytic enantioselective bromochlorination of allylic alcohols is expanded to include a specific homoallylic alcohol. Critical factors for optimization of this reaction are highlighted. The utility of the

Full text of DOI:10.3762/bjoc.12.129

Selective bromochlorination of a homoallylic alcohol for the
total synthesis of (−)-anverene
Frederick J. Seidl and Noah Z. Burns*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2016, 12, 1361–1365.
Department of Chemistry, Stanford University, Stanford, California
94305, USA
doi:10.3762/bjoc.12.129
Received: 29 April 2016
Accepted: 23 June 2016
Published: 01 July 2016
Email:
Noah Z. Burns* - nburns@stanford.edu
* Corresponding author
This article is part of the Thematic Series "Strategies in asymmetric
catalysis".
Keywords:
enantioselective catalysis; halogenation; natural products; total
synthesis
Guest Editor: T. P. Yoon
© 2016 Seidl and Burns; licensee Beilstein-Institut.
License and terms: see end of document.
Abstract
The scope of a recently reported method for the catalytic enantioselective bromochlorination of allylic alcohols is expanded to
include a specific homoallylic alcohol. Critical factors for optimization of this reaction are highlighted. The utility of the product
bromochloride is demonstrated by the first total synthesis of an antibacterial polyhalogenated monoterpene, (−)-anverene.
Introduction
The directed enantioselective functionalization of olefins is an Several years ago our laboratory developed the first catalytic
extremely powerful tool in synthesis. A preeminent example is enantioselective dibromination of cinnamyl alcohols [6]. Unfor-
the Sharpless asymmetric epoxidation (SAE), which has been tunately, this method suffered from the requirement for elec-
featured in countless syntheses of enantioenriched small mole- tronically biased substrates, and the catalyst system was not
cules [1,2]. While the generality and scope of such catalytic en- applicable to other dihalogenation reactions. In 2015, we
antioselective methods may be large, limitations often exist. In disclosed a much-improved catalytic enantioselective bromo-
the case of the SAE, allylic alcohols are viable substrates, but chlorination protocol that is able to override inherent substrate
homoallylic alcohols undergo epoxidation at slower rates and regioselectivity [7]. This development vastly expanded the sub-
with diminished yields and enantioselectivities [3]. This particu- strate scope to include a larger subset of alkyl-substituted allylic
lar limitation has inspired the development of several remote alcohols (Scheme 1, top); the relative and absolute stereochemi-
epoxidation methods for homoallylic and bishomoallylic alco- cal configurations of all resulting bromochlorides have been
hols [4,5], now allowing chemists to directly access a greater assigned and are shown for the (R,S) Schiff base ligand [7-9].
variety of enantioenriched epoxides.
Furthermore, the reaction conditions could now be modulated to
1361

Beilstein J. Org. Chem. 2016, 12, 1361–1365.
Scheme 1: Selective bromochlorination and possible disconnections for anverene (1).
affect both catalytic enantioselective dichlorination and di- hyde 3, which should readily engage in standard olefination
bromination of allylic alcohols [9]. The utility of this method reactions. As 3 could arise from a bromochlorinated homoal-
has been demonstrated by highly selective total syntheses of lylic alcohol, we viewed this as an impetus for developing
several important polyhalogenated natural product targets in- conditions for the bromochlorination of 4-methyl-3-pentenol
cluding (+)-halomon [8] and (−)-danicalipin A [9].
(homoprenol, 5).
While exploring the scope of this bromochlorination reaction, Optimization
we discovered that slight modifications allow for the bromo- Under our published conditions for allylic alcohols, homo-
chlorination of a homoallylic alcohol with synthetically useful prenol (5) underwent bromochlorination to deliver the target
levels of regio- and enantioselectivity. Disclosed herein, this isomer 6 in 46% yield and 84% ee (Table 1, entry 1). Constitu-
discovery has enabled the first total synthesis of (−)-anverene tional isomer 7 was also observed (4:1 ratio of 6:7), as well as
(1) (Scheme 1, bottom), a secondary metabolite from the algae dibromide 8 and other racemic brominated byproducts (see Sup-
Plocamium cartilagineum with selective antibiotic activity porting Information File 1). The absolute configuration of 6 was
against vancomycin-resistant Enterococcus faecium (VREF) assigned through its subsequent conversion into (−)-anverene
[10]. (−)-Anverene (1) is a member of a large and synthetically (1, see below), the absolute configuration of which was deter-
challenging family of polyhalogenated acyclic monoterpenes, mined by the isolation chemists on natural material by X-ray
for which few stereoselective synthetic approaches exist [11]. crystallography [10]. Interestingly, using the same enantiomer
of ligand, the bromochlorides derived from prenol and homo-
Results and Discussion
prenol (5) have the same absolute configuration. This is in stark
In considering a retrosynthesis of anverene (1) we initially iden- contrast to the SAE, for which the facial selectivity of epoxida-
tified prenol bromochlorotriflate 2 (Scheme 1, bottom) as a use- tion for homoallylic alcohols is the opposite of that seen for
ful synthon, as it is readily accessible from prenol using our allylic alcohols [3]. Additional information on the scope of the
original bromochlorination protocol [7]. In a forward sense, this bromochlorination for other homoallylic alcohols is presented
intermediate could be coupled with a vinyl organometallic to in Supporting Information File 1. In the subsequent optimiza-
introduce the remainder of the carbon skeleton of 1. While we tion, three critical factors came to the fore: concentration, the
have achieved moderate success in the copper-catalyzed cross inclusion of additional Ti(OiPr)4, and stirring rate.
coupling of an analogous dichlorotriflate [9], in our hands 2
failed to engage in productive carbon–carbon bond formation. When unoptimized bromochlorination reactions of homoprenol
An alternative retrosynthesis traced 1 back to bromochloroalde- were quenched at low and high conversion, different distribu-
1362

Beilstein J. Org. Chem. 2016, 12, 1361–1365.
Table 1: Optimization of a selective homoallylic bromochlorination reaction.a
entry
additive
concentration
yield 6/ee 6 (%)
yield 7/ee 7 (%)
cr (6:7)
yield 8, others (%)
1
2
3
4
–
0.1 M
0.025 M
0.1 M
46/84
45/82
50/89
57/89
13/23
10/19
10/18
6/38
4:1
5:1
6, 34
3, 27
7, 28
6, 16
–
Ti(OiPr)4b
Ti(OiPr)4b
5:1
0.025 M
10:1
aReactions were conducted on 0.1 mmol scale with 1.2 equiv NBS, 1.1 equiv ClTi(OiPr)3; b20 mol %.
tions of side products were obtained. Unlike Table 1, entry 1, the slower rate of selective bromochlorination of homoallylic
which was taken to completion, bromochlorinations quenched alcohols, leading to closer competition with nonselective back-
at low conversion exhibited high enantioselectivity for 6, but ground halogenation than the same reaction with allylic alco-
also constitutional isomer ratios (cr) as low as 2:1 for 6:7 (not hols. Consistent results were obtained when bromochlorination
shown). These observations are consistent with a highly selec- reactions were run in round-bottomed flasks, using rod-shaped
tive catalytic reaction competing with several nonselective magnetic stir bars roughly the length of the flask radius, and
background processes. By both lowering the concentration and with stirring at no less than 1500 rpm (see Supporting Informa-
adding 20 mol % of Ti(OiPr)4 (Table 1, entries 2–4), much of tion File 1 for images). Under optimized conditions, bromo-
the background reactivity was minimized; adding more than chloroalcohol 6 was produced in 57% yield and 89% enan-
20 mol % of Ti(OiPr)4 did not result in further improvement. tiomeric excess (ee) with small amounts of undesired bromo-
While the structure of the active bromochlorination catalyst is chloride 7 (6%) and other brominated byproducts (8 + others,
not yet known, we speculate that the procedural changes in 22% combined) (Table 1, entry 4). The trace dibromide 8 was
entries 2–4 do not alter its structure, but rather change the distri- quantitatively and chemoselectively decomposed, via a reduc-
bution of titanium species responsible for off-cycle reactivity. tive de-dihalogenation pathway to the corresponding olefin, by
Titanium alkoxides are known to undergo facile ligand heating the crude material to reflux in acetone with sodium
exchange at −20 °C, and can adopt both monomeric and iodide.
oligomeric structures, several of which can be catalytically
active [12]. When Cl2Ti(OiPr)2 is used in place of ClTi(OiPr)3, Total synthesis of anverene
bromochloride 7 forms very quickly with no enantioselectivity With scalable access to 6, a total synthesis of (−)-anverene (1)
and with inherent substrate regioselectivity. The additional was explored (Scheme 2). Homoprenol (5) was subjected to the
Ti(OiPr)4 may limit the trace formation of these highly reactive two-step bromochlorination/de-dibromination protocol on
dichlorotitanium species early in the reaction, thereby 10 mmol scale, furnishing bromochloride 6 in 64% overall yield
preventing background bromochlorination. Lowering the con- and 89% ee as an 8:1 ratio of constitutionally isomeric
centration of the reaction may favour lower-order titanium bromochlorides. Early incorporation of the sensitive vicinal
aggregates over oligomeric species [12].
bromochloride moiety in 6 necessitated a strategy that avoided
exposure to strong bases or reducing agents, a limitation that
It has been our observation that high stirring rates are critical significantly guided the choice of reagents and conditions for
for obtaining optimal results in these heterogeneous bromo- the subsequent steps. Dess–Martin periodinane oxidation of
chlorination reactions of both allylic and homoallylic alcohols. alcohol 6 followed by Horner–Wadsworth–Emmons olefina-
The origin of this stirring dependence is not definitively known, tion with triethyl 2-phosphonopropionate (9) furnished the
but we suspect that it facilitates the dissolution of N-bromosuc- targeted unsaturated ester (not shown). In initial investigations
cinimide (NBS) into hexanes and/or its incorporation into the into the formation of this unsaturated ester, Masamune–Roush
active catalyst. The bromochlorination of homoprenol (5) conditions [13] were attempted but led to sluggish reactivity
proved to be especially sensitive to stirring. This may be due to and concomitant base-mediated elimination of the bromo-
1363

Beilstein J. Org. Chem. 2016, 12, 1361–1365.
Scheme 2: Selective total synthesis of (−)-anverene. Reagents and conditions: a) NBS (1.2 equiv), ClTi(OiPr)3 (1.1 equiv), 20 mol % Ti(OiPr)4,
10 mol % (S,R)-4, 0.025 M hexanes, −15 °C, 17 h; b) NaI (2 equiv), acetone, 65 °C, 3 h; c) Dess–Martin periodinane (1.2 equiv), NaHCO3 (5 equiv),
CH2Cl2, 0 °C to rt; d) triethyl 2-phosphonopropionate 9 (1.5 equiv), n-BuLi (1.2 equiv), 1:1 THF/acetonitrile, 0 °C, 5 min; e) DIBAL−H (3 equiv),
CH2Cl2, −78 °C, 20 min; f) NBS (1.2 equiv), ClTi(OiPr)3 (1.1 equiv), 20 mol % Ti(OiPr)4, 10 mol % (S,R)-4, 0.025 M hexanes, −15 °C, 42 h; g)
Dess–Martin periodinane (1.2 equiv), NaHCO3 (5 equiv), CH2Cl2, 0 °C to rt; h) CBr4 (2 equiv), PPh3 (4 equiv), CH2Cl2, 0 °C, 10 min; i) Et3N (5 equiv),
dimethyl phosphonate (4 equiv), DMF, rt, 5 min.
chloride. This decomposition pathway was minimized by the Tetrahalide 11 was then oxidized to the corresponding alde-
stoichiometric generation of the lithium carbanion of 9 with hyde. Installation of the vinyl bromide was found to be difficult
n-BuLi prior to addition of the corresponding aldehyde, and by using traditional methods. Takai olefination [14] with CHBr3
maintaining very short (<10 minutes) reaction times at 0 °C. and CrBr2 resulted in significant de-dihalogenation, likely due
This protocol led to complete conversion exclusively to the to the presence of reducing agents such as Cr(II) or residual
E-unsaturated ester without any observed elimination.
LiAlH4 from the preparation of CrBr2. Fortuitously, an effi-
cient and high-yielding alternative to these routes was found.
The unsaturated ester was then reduced with DiBAl–H to allylic Following Ramirez dibromomethylenation [15], 1,1-dibro-
alcohol 10, which was isolated as an unchanged 8:1 mixture of moolefin 12 was isolated and subsequently reduced with
regioisomeric bromochlorides, strongly suggesting that no dimethyl phosphonate [16,17]. While the mechanism of this
stereochemical or regiochemical isomerization had taken place Hirao reduction is not fully understood, the reaction was exquis-
over the previous three steps from 6. When 10 was subjected to itely mild and proceeded with high chemoselectivity for the
ligandless bromochlorination with ClTi(OiPr)3 and NBS in geminal dihalide, delivering (−)-anverene (1) in high yield as a
hexanes, a surprisingly high dr of 7:1 favouring tetrahalide 11 7:1 mixture of E/Z isomers. Following recrystallization, over
was observed, indicating significant inherent substrate dia- 100 mg of 1 was isolated as a single diastereomer and constitu-
stereocontrol for the anti-1,3-dihalide over the syn alternative. tional isomer, culminating a 9-step route with an overall yield
Alternatively, when 10 was subjected to the bromochlorination of 21.8%.
protocol with 10 mol % (S,R)-4, the product tetrahalide 11 was
isolated in >20:1 dr favouring the same anti-1,3-dihalide. Inter- Conclusion
estingly, we observed considerable enrichment in the regioiso- In conclusion, a new protocol for the catalytic enantioselective
meric bromochloride ratio for the left hand side, which had bromochlorination reaction has been developed which has
roughly doubled from 8:1 to 18:1, and the recovered starting enabled the highly efficient and selective synthesis of 1 via
material was found to have a regioisomeric bromochloride ratio homoprenol bromochloride 6. The factors addressed in the opti-
that was lowered to 1.6:1. Taking into account that the minor mization of the bromochlorination for homoprenol, namely con-
bromochloride constitutional isomer (7, Table 1) is formed in centration, the inclusion of additional Ti(OiPr)4, and vigorous
20–40% ee, these results are consistent with the catalyst system stirring, are expected to be generally important to other chal-
being capable of selective resolution between two enantiomers lenging substrates. Work is ongoing in our laboratory to better
(or pseudoenantiomers) of a substrate allylic alcohol.
understand and predict the matched/mismatched effect seen for
1364

Beilstein J. Org. Chem. 2016, 12, 1361–1365.
15.Desai, N. B.; McKelvie, N.; Ramirez, F. J. Am. Chem. Soc. 1962, 84,
1745–1747. doi:10.1021/ja00868a057
chiral substrates, in particular by studying the kinetic resolution
of racemic allylic alcohols. Detailed mechanistic and computa-
tional studies are underway to better understand the active cata-
lyst structure and the origins of selectivity.
16.Abbas, S.; Hayes, C. J.; Worden, S. Tetrahedron Lett. 2000, 41,
3215–3219. doi:10.1016/S0040-4039(00)00353-1
17.Hirao, T.; Masunaga, T.; Ohshiro, Y.; Agawa, T. J. Org. Chem. 1981,
46, 3745–3747. doi:10.1021/jo00331a039
Supporting Information
Supporting Information File 1
Experimental procedures, full characterization of new
compounds, and spectral data.
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-12-129-S1.pdf]
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
Acknowledgements
This work was supported by Stanford University and the
National Institutes of Health (R01 GM114061). We are grateful
to Dr. S. Lynch (Stanford University) for assistance with NMR
spectroscopy.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
References
1. Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102,
doi:10.3762/bjoc.12.129
5974–5976. doi:10.1021/ja00538a077
2. Gao, Y.; Klunder, J. M.; Hanson, R. M.; Masamune, H.; Ko, S. Y.;
Sharpless, K. B. J. Am. Chem. Soc. 1987, 109, 5765–5780.
doi:10.1021/ja00253a032
3. Rossiter, B. E.; Sharpless, K. B. J. Org. Chem. 1984, 49, 3707–3711.
doi:10.1021/jo00194a007
4. Li, Z.; Yamamoto, H. Acc. Chem. Res. 2013, 46, 506–518.
doi:10.1021/ar300216r
5. Xia, Q.-H.; Ge, H.-Q.; Ye, C.-P.; Liu, Z.-M.; Su, K.-X. Chem. Rev. 2005,
105, 1603–1662. doi:10.1021/cr0406458
6. Hu, D. X.; Shibuya, G. M.; Burns, N. Z. J. Am. Chem. Soc. 2013, 135,
12960–12963. doi:10.1021/ja4083182
7. Hu, D. X.; Seidl, F. J.; Bucher, C.; Burns, N. Z. J. Am. Chem. Soc.
2015, 137, 3795–3798. doi:10.1021/jacs.5b01384
8. Bucher, C.; Deans, R. M.; Burns, N. Z. J. Am. Chem. Soc. 2015, 137,
12784–12787. doi:10.1021/jacs.5b08398
9. Landry, M. L.; Hu, D. X.; McKenna, G. M.; Burns, N. Z.
J. Am. Chem. Soc. 2016, 138, 5150–5158. doi:10.1021/jacs.6b01643
10.Ankisetty, S.; Nandiraju, S.; Win, H.; Park, Y. C.; Amsler, C. D.;
McClintock, J. B.; Baker, J. A.; Diyabalanage, T. K.; Pasaribu, A.;
Singh, M. P.; Maiese, W. M.; Walsh, R. D.; Zaworotko, M. J.;
Baker, B. J. J. Nat. Prod. 2004, 67, 1295–1302.
doi:10.1021/np049965c
11.Vogel, C. V.; Pietraszkiewicz, H.; Sabry, O. M.; Gerwick, W. H.;
Valeriote, F. A.; Vanderwal, C. D. Angew. Chem., Int. Ed. 2014, 53,
12205–12209. doi:10.1002/anie.201407726
12.Finn, M. G.; Sharpless, K. B. J. Am. Chem. Soc. 1991, 113, 113–126.
doi:10.1021/ja00001a019
13.Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25,
2183–2186. doi:10.1016/S0040-4039(01)80205-7
14.Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108,
7408–7410. doi:10.1021/ja00283a046
1365