6-Deoxyhexoses from l-Rhamnose in the Search for Inducers of the Rhamnose Operon: Synergy of Chemistry and Biotechnology

Article abstract of DOI:10.1002/chem.201602482

In the search for alternative non-metabolizable inducers in the l-rhamnose promoter system, the synthesis of fifteen 6-deoxyhexoses from l-rhamnose demonstrates the value of synergy between biotechnology and chemistry. The readily available 2,3-acetonide of rhamnonolactone allows inversion of configuration at C4 and/or C5 of rhamnose to give 6-deoxy-d-allose, 6-deoxy-d-gulose and 6-deoxy-l-talose. Highly crystalline 3,5-benzylidene rhamnonolactone gives easy access to l-quinovose (6-deoxy-l-glucose), l-olivose and rhamnose analogue with C2 azido, amino and acetamido substituents. Electrophilic fluorination of rhamnal gives a mixture of 2-deoxy-2-fluoro-l-rhamnose and 2-deoxy-2-fluoro-l-quinovose. Biotechnology provides access to 6-deoxy-l-altrose and 1-deoxy-l-fructose.

Full text of DOI:10.1002/chem.201602482

DOI: 10.1002/chem.201602482
Full Paper
&
Chemical Biology
6-Deoxyhexoses from l-Rhamnose in the Search for Inducers of
the Rhamnose Operon: Synergy of Chemistry and Biotechnology
Zilei Liu,[a, d] Akihide Yoshihara,*[b] Ciarꢀn Kelly,[c] John T. Heap,[c] Mikkel H. S. Marqvorsen,[a]
Sarah F. Jenkinson,[a] Mark R. Wormald,[d] Josꢁ M. Otero,[e] Amalia Estꢁvez,[e] Atsushi Kato,[f]
George W. J. Fleet,*[a] Ramꢂn J. Estꢁvez,*[e] and Ken Izumori[b]
Abstract: In the search for alternative non-metabolizable in-
ducers in the l-rhamnose promoter system, the synthesis of
fifteen 6-deoxyhexoses from l-rhamnose demonstrates the
value of synergy between biotechnology and chemistry. The
readily available 2,3-acetonide of rhamnonolactone allows
inversion of configuration at C4 and/or C5 of rhamnose to
give 6-deoxy-d-allose, 6-deoxy-d-gulose and 6-deoxy-l-
talose. Highly crystalline 3,5-benzylidene rhamnonolactone
gives easy access to l-quinovose (6-deoxy-l-glucose), l-oli-
vose and rhamnose analogue with C2 azido, amino and
acetamido substituents. Electrophilic fluorination of rhamnal
gives a mixture of 2-deoxy-2-fluoro-l-rhamnose and 2-
deoxy-2-fluoro-l-quinovose. Biotechnology provides access
to 6-deoxy-l-altrose and 1-deoxy-l-fructose.
lized by E. coli[2] and the metabolism of l-rhamnose is precisely
controlled. l-Rhamnose works as inducer of the operon encod-
ing the enzymes required for its metabolism, rhaBAD.[3] The l-
rhamnose promoter system, PrhaBAD, is used extensively for the
heterologous expression of many genes of interest in E. coli
and other bacteria.[4] Its many advantages over alternative pro-
moter systems include: non-leaky basal expression, titratable
concentration-dependent expression and the ability to achieve
higher levels of expression.[5] However, due to l-rhamnose me-
tabolism and its autorepression of the PrhaBAD promoter,[6] in-
duction is transient, which limits its application. This system
could be improved by the development of a non-metaboliza-
ble analogue of l-rhamnose, an approach which has been de-
veloped for only a few other inducible expression systems.[7]
No 6-deoxyhexose analogues have been tested for this pur-
pose previously.
Introduction
In biotechnology, control over gene expression is often useful,
such as in the production of proteins of interest. This is typical-
ly achieved by assembling a DNA construct in which the gene
of interest is placed under the control of a promoter (a regula-
tory DNA sequence) which responds to a known external stim-
ulus. Numerous inducible gene expression systems have been
reported, exploiting inducible promoters which respond to
natural or synthetic stimuli such as sugars, antibiotics and
metals.[1] l-Rhamnose (6-deoxy-l-mannose) is readily metabo-
[a] Dr. Z. Liu, M. H. S. Marqvorsen, Dr. S. F. Jenkinson, Prof. Dr. G. W. J. Fleet
Chemistry Research Laboratory, Department of Chemistry
University of Oxford, Oxford, OX1 3TA (UK)
E-mail: george.fleet@chem.ox.ac.uk
[b] Prof. Dr. A. Yoshihara, Prof. Dr. K. Izumori
International Institute of Rare Sugar Research and Education
Kagawa University, Miki, Kagawa 761-0795 (Japan)
E-mail: yoshihara@ag.kagawa-u.ac.jp
The chemotherapeutic potential of interfering with l-rham-
nose metabolic pathways in bacteria and in plants has long
been recognized, and is safe because l-rhamnose has no role
in mammalian metabolism.[8] Synthetic inducers/antagonists of
the rhamnose operon might also be useful as antimicrobials,
targeting pathogens requiring rhamnose-incorporation into
their cell walls for survival. Control or suppression of expres-
sion of the genes encoding rhamnose metabolic enzymes may
allow control of biosynthesis of TDP-rhamnose and thus of cell
wall construction of a number of pathogens.[9] For example,
a rhamnose-inducible promoter was reported to be essential
to Burkholderia cenocepacia.[10] Other l-rhamnose metabolic en-
zymes, such as l-rhamnosidases and l-rhamnose isomerase,
are also potential targets. Iminosugars affecting rhamnose
processing enzymes may have potential as antibiotics.[11] The
identification of synthetic inducers or antagonists of the l-
rhamnose metabolic pathway will be valuable tools for bio-
technology.
[c] Dr. C. Kelly, Dr. J. T. Heap
Centre for Synthetic Biology and Innovation, Department of Life Sciences
Imperial College London, SW7 2AZ (UK)
[d] Dr. Z. Liu, Dr. M. R. Wormald
Glycobiology Institute, Department of Biochemistry
University of Oxford, Oxford, OX1 3QU (UK)
[e] Dr. J. M. Otero, Dr. A. Estꢀvez, Prof. Dr. R. J. Estꢀvez
Departamento de Quꢁmica Orgꢂnica and Centro Singular de
Investigaciꢃn en Quꢁmica Biolꢃxica e Materiais Moleculares
Universidade de Santiago de Compostela
15782 Santiago de Compostela (Spain)
E-mail: ramon.estevez@usc.es
[f] Prof. Dr. A. Kato
Department of Hospital Pharmacy, University of Toyama
Toyama 930-0194 (Japan)
Supporting information and the ORCID identification number for the
author of this article can be found under
http://dx.doi.org/10.1002/chem.201602482.
Chem. Eur. J. 2016, 22, 1 – 10
1
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
A recent study, which evaluated the effect of some 35 ana-
logues of rhamnose on the operon, required access to 6-deox-
yhexoses.[12] This study describes the conversion of l-rhamnose
1, the cheapest 6-deoxyhexose,[13] into 15 isomers and other
analogues of rhamnose as free sugars by four different strat-
egies (Scheme 1). Nucleophilic substitution of either pyranose
or furanose derivatives of rhamnose[14] is easy neither to set up
the precursor nor to perform SN2 displacements of sulfonates.
However efficient SN2 reactions in rhamnonolactones can allow
isomerization at C5, C4 and C2. The 2,3-acetonide of rhamno-
nolactone 2 is readily available from l-rhamnose in multigram
scale by a well-established procedure.[15] Lactone 2 allows
access to the three diastereomeric lactols 3–5 as intermediates
in the synthesis of 6-deoxy-d-gulose 6, 6-deoxy-d-allose 7 and
6-deoxy-l-talose 8 (strategy A). Unlike many lactone aceto-
nides, 2 is not crystalline and is usually contaminated with the
corresponding 3,5-acetonide. The highly crystalline benzyli-
dene lactone 9, with only C2 OH free, was prepared from l-
rhamnose on a 50 g scale by a reported procedure (strategy
B).[16] High yield nucleophilic substitution of the epimeric
manno-10 and gluco-11 C2 triflates allowed the syntheses of
the epimer at C2 l-quinovose 12, the dideoxy hexose l-olivose
13 and both epimers of analogues with nitrogen introduced at
C2 14; 2-nitrogen substituted analogues are also useful build-
ing blocks for the synthesis of bioactive glycoconjugates.[17] It
was not possible to introduce fluorine at C2 by an SN2 reac-
tion; strategy C uses diacetyl rhamnal 15 for the non-stereose-
lective introduction of electrophilic fluorine to allow access to
the 2-fluoro analogues of rhamnose 16 and quinovose 17. It is
not easy to access C3 of rhamnonolactone (or rhamnose)
chemically;[18] however, a highly efficient biotechnological syn-
thesis of the C3 epimer 18 allows the preparation of multigram
quantities (strategy D). Rhamnose may also be converted to 19
where the anomeric hydroxyl group has been transposed from
C1 to C5. Thus a combination of biotechnology and chemistry
has allowed the preparation of some 15 analogues of rham-
nose as free sugars for evaluation as inducers of the rhamnose
operon.
Results and Discussion
Synthesis
1. Stereoisomers at C4 and C5 of rhamnose from 2,3-O-isopro-
pylidene rhamnonolactone (2)
The acetonide of rhamnonolactone 2 allowed easy access to
three diastereomeric lactones at C4 and C5 22, 24 and 26
(Scheme 2). There are very few examples of SN2 reactions[19] at
C5 of 2. The difficulty probably arises from the extra steric hin-
drance because of the cis-relationship of the isopropylidene
protecting group and the C4-side chain; SN2 displacements at
C5 of the C4-epimeric talono-lactone 24, where there is a corre-
sponding trans arrangement, are efficient.[20] All attempts to
use oxygen nucleophiles (acetate or trifluoroacetate in DMF,
butanone or pentanone) on either the mesylate 23 or the cor-
responding triflate gave no indication of the formation of the
gulo-epimer 22. Accordingly, inversion of configuration at C5
was achieved by Dess–Martin oxidation of 2 in dichlorome-
thane to give the crystalline ketone 21 (77%). Reduction of 21
with sodium borohydride at room temperature showed little
diastereoselectivity to give crystalline 22[21] (54%) and the d-
gulo-lactone 2 in a 1.5:1 ratio. No improvement in the ratio
was found at lower temperatures or use of cyanoborohydride
and other conditions led to competitive reduction of the lac-
tone; a similar lack diastereoselectivity has been found in anal-
ogous systems.[22] The two lactones 2 and 22 were readily sep-
arable by chromatography. Reduction of 22 with diisobutylalu-
minum hydride (DIBALH) in dichloromethane afforded the lac-
tols 3 (75%). In the reduction of 2,3-protected lactones, no at-
tempts were made to analyze the ratio of 2,3-O-isopropylidene
pyranoses and furanoses formed. Acid ion exchange catalyzed
hydrolysis of the lactols at room temperature gave 6-deoxy-d-
gulose 6,[23] a syrup, (93%) predominantly as the b-pyranose
form.
Esterification of the free alcohol in 2 with mesyl chloride in
pyridine gave the mesylate 23 (83%); treatment of 23 with
aqueous potassium hydroxide in 1,4-dioxane caused inversion
Scheme 1. Strategy for synthesis of rhamnose mimics.
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
2
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 2. i) Dess–Martin periodinane, CH2Cl2, ii) NaBH4, MeOH, RT, iii) DIBALH, CH2Cl2, À788C, iv) DOWEXꢄ 50WX8-100 (H+ form), H2O, RT, v) MeSO2Cl, DMAP,
pyridine, 08C, vi) KOH, H2O/1,4-dioxane, RT, vii) (CF3SO2)2O, pyridine, CH2Cl2, viii) CF3COONa, pentanone.
2. 2-Substituted rhamnose analogues from 3,5-O-benzylidene
rhamnonolactone (9)
of configuration at both C4 and C5 by Payne rearrange-
ments[24] to form the d-allono-lactone 24 (79%). Nucleophilic
displacement at C5 of the allono-lactone 24 where the side
chain is trans to the isopropylidene protecting group was effi-
cient. Thus reaction of 24 with triflic anhydride in dichlorome-
thane in the presence of pyridine gave the corresponding tri-
flate 25 which, with sodium trifluoroacetate in pentanone at
room temperature followed by work-up with methanol,[25]
gave the talono-lactone 26 in 70% yield. DMF is not an appro-
priate solvent since the solvent competes with trifluoroacetate
as a nucleophile; the mesylate of 24 did not give SN displace-
ment by trifluoroacetate.
The highly crystalline benzylidene lactone 9 was directly ob-
tained from rhamnose on a 50 g scale as previously de-
scribed.[28] Esterification of the C2-OH group in 9 with triflic an-
hydride in THF afforded the triflate 10[16] which is an ideal in-
termediate for SN2 reactions at C2 of rhamnose. Although the
crude triflate 10 may be used directly in subsequent reactions,
10 is a stable triflate which can be purified by column chroma-
tography. For the preparation of l-quinovose (6-deoxy-l-glu-
cose) 12, reaction of 10 with cesium trifluoroacetate in buta-
none[29] gave the inverted alcohol 28[16] (90%; Scheme 3). At-
tempts to reduce 28 directly gave complex mixtures but prior
protection of 28 with TBDMS chloride in DMF in the presence
of imidazole afforded the TBDMS ether 29 (71%); DIBALH re-
duction of the fully protected lactone 29 formed the protected
lactols 30 (88%). Removal of the silyl and benzylidene protect-
ing groups in 30 by ion exchange gave l-quinovose 12[30]
An identical two-step procedure for the reduction and hy-
drolysis of the protected allono-24 and talono-26 lactones
gave 6-deoxy-d-allose 7[26] and 6-deoxy-l-talose 8,[27] in overall
yields of 83 and 72%, via the lactols 4 and 5, respectively.
For the rhamnose operon assays, it was very important to
establish that there was no contamination of 6-deoxy-d-gulose
(6) with rhamnose arising from incomplete separation of 2 and
22 in the reduction of ketone 21. Accordingly, an alternative
approach to 6-deoxy-gulonolactone 22 by a double inversion
of C4 and C5 of the talono-lactone 26 was investigated. Se-
quential formation of the mesylate 27 from 26, followed by
treatment with aqueous potassium, hydroxide gave 22 in 76%
which was identical to that prepared from the reduction of the
ketone 21. A sample of 6-deoxy-d-gulose 6 by this route gave
an identical response in the rhamnose operon experiments to
that derived from the ketone;[12] this established that there
was no contamination with rhamnose in the synthesis via 21.
1
(100%) with an identical H NMR spectrum to that of an au-
thentic sample of d-quinovose (see the Supporting Informa-
tion).
Treatment of the triflate 10 with lithium iodide hydrate in
butanone gave the iodide 31 by rapid SN2 displacement of the
triflate which underwent further attack by iodide to give
iodine[31] and the deoxygenated lactone 32 (74%). Reduction
of 32 by DIBALH in dichloromethane afforded the lactols 33
(96%) which on hydrolysis by acid ion exchange resin gave l-
olivose 13[32] (100%).
In a number of cases it has been established that azide dis-
placement of a triflate at C2 of a lactone may result in isolation
of the azide with either retention or inversion of configura-
tion.[25,33] For the preparation of C2 nitrogen derivatives of
rhamnose and quinovose, high yield SN2 substitution of the tri-
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
3
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Scheme 3. i) (CF3SO2)2O, pyridine, THF, À208C, 95%, ii) CF3COOCs, butanone, 608C, 4 h, 90%, iii) TBDMSCl, imidazole, DMF, 71%, iv) DIBALH, CH2Cl2, À788C,
1 h, 88%, v) DOWEXꢄ 50WX8-200 (H+ form), H2O, RT, 15 h, 100%, vi) LiI·3H2O, butanone, 608C, 16 h, 74%, vii) DIBALH, CH2Cl2, À788C, 3 h, 96%, viii) DOWEXꢄ
50WX8-200 (H+ form), H2O, RT, 18 h, 100%.
flate 10 with sodium azide in DMF gave the kinetic gluco-azide
34 with inversion or the thermodynamic manno-azide 36 with
retention, respectively (Scheme 4). Reaction of the stable tri-
flate 10 with sodium azide in DMF at À408C formed the in-
verted gluco-azide 34 (75%). After column purification, the
gluco-azide 34 was reduced by DIBALH in dichloromethane to
the lactols 35 (84%). When the SN2 reaction was conducted at
room temperature, 10 formed the thermodynamic manno-
azide 36 (85%). The initially formed 34 was converted to the
more stable all cis-azide 36 under the reaction conditions. The
two azides were readily distinguishable by TLC and easily sepa-
rable by column chromatography. The triflate 11 derived from
the gluco-alcohol 28 was less stable than the manno-triflate
10. When crude triflate 11 in DMF was treated with sodium
azide at À208C, the manno-azide 36 (70%) was isolated to-
gether with the less stable gluco-azide 34 (24%); there is com-
petition between azide displacement to give 36 (70%) and ep-
imerization of triflate 11 to the more stable triflate 10 which
then forms the less stable azide 34 (24%). Reduction of 36
with DIBALH in dichloromethane gave the manno-lactols 37
(87%). After the reduction, there was no indication of any epi-
merization of the gluco-35 to the manno-37 lactols.
Hydrogenation of 38 in the presence of palladium on carbon
and hydrochloride acid afforded the hydrochloride salt of 2-
amino-2,6-dideoxy-l-glucose 39[34] (98%). Hydrogenation of
the azide 35 in the presence of acetic anhydride and palladium
in ethyl acetate/1,4-dioxane gave the protected gluco-acet-
amide 40 (74%). Removal of the benzylidene acetal in 40 by
DOWEX-catalyzed hydrolysis in aqueous ethanol afforded acet-
amido-quinovose 41[34] (90%). Identical procedures on the
rhamno-azide 37 gave the azido-42 (95%), amino-43[34] (100%)
and acetamido-45 (80% from 37) rhamnose analogues in
equally high yields.
3. 2-Deoxy-2-fluoro-l-rhamnose 16 and 2-deoxy-2-fluoro-l-
quinovose (17) from 3,4-di-O-acetyl rhamnal
It was not possible either to displace the triflates in 10 by any
of a number of nucleophilic sources of fluoride (including TASF,
TBAF and CsF) or to introduce fluorine directly by reaction of
the alcohol 9 with DAST[35] or Xtalfluor.[36]
Accordingly, treatment of diacetylrhamnal 15 with Select-
Fluor in aqueous acetonitrile[37] followed by acetylation with
acetic anhydride in pyridine during the workup caused electro-
philic fluorination to give a mixture of the four sugar triace-
tates 46a/b and 47 (Scheme 6) with little stereoselectivity. One
Hydrolysis of the protected gluco-lactols 35 by acid ion ex-
change resin gave the quinovose azide 38 (99%; Scheme 5).
Scheme 4. i) (CF3SO2)2O, pyridine, THF, À208C, 95%, ii) NaN3, DMF, À408C, 75%, iii) DIBALH, CH2Cl2, À788C, 3 h, 84%, iv) see text; 85% 36 from 10,
v) (CF3SO2)2O, pyridine, THF, À208C, vi) NaN3, DMF, À408C, 70% 36 and 24% 34 (over two steps), vii) DIBALH, CH2Cl2, À788C, 3 h, 87%.
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
4
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 5. i) DOWEXꢄ 50WX8-200 (H+ form), H2O, RT, ii) H2, 10% Pd/C, HCl, iii) H2, 10% Pd/C, Ac2O, iv) DOWEXꢄ 50WX8-200 (H+ form), H2O.
of the four triacetates, the b-fluoro-rhamnose 46b, was isolat-
ed as a pure crystalline solid. Hydrolysis of 46b gave a pure
sample of previously unknown 2-deoxy-2-fluoro-rhamnose (16;
27% overall yield) for biological evaluation. Hydrolysis of the
mixture of 46a and 47a/b gave a mixture of fluoro-rhamnose
16 and fluoro-quinovose 17 (38% yield); a pure sample of
fluoro-quinovose 17 was obtained by preparative HPLC.
Scheme 7. i) H2, Raney Ni, H2O, ii) IK7, iii) RI, iv) DTE, v) AI.
NMR analysis of synthetic free 6-deoxy-hexoses
1. 6-Deoxy-d-gulose (6)
The synthesis and characterization, but not the NMR analysis,
of 6-deoxy-d-gulose (6) has been reported.[23] 1H and 13C NMR
spectra of 6 in D2O, pH 8.0, showed two spin-systems (Table 1),
confirmed as pyranose forms by the HMBC correlations be-
tween C1 and H5 (d-gulose also exists mainly in the pyranose
form[42]). The a and b pyranose anomers could be identified by
the magnitude of coupling constant of anomeric proton (J1,2 of
b anomer was 8.3 Hz because of the trans-diaxial configura-
tion; J1,2 of a anomer was 3.0 Hz) and were present in an a-
pyranose/b-pyranose ratio of 1:10.
Scheme 6. (i) SelectFluor, water, MeCN; then Ac2O in pyridine, 27% of pure
46b, ii) CF3COOH, H2O, 100%, iii) CF3COOH, H2O, 100% (mixture of free
sugars), then pure 17 was isolated by preparative HPLC.
4. 1-Deoxy-l-fructose and 6-deoxy-l-altrose from rhamnose
by biotechnology
The biotechnology of Izumoring[38] allows access to substantial
amounts of rare deoxy sugars from rhamnose in water without
the need for any protection. Hydrogenation of rhamnose 1 in
the presence of Raney nickel gave l-rhamnitol 48 (96%;
Scheme 7). Oxidation of the (4S,5S) diol in 48 by Enterobacter
aerogenes IK7 (IK7) afforded 1-deoxy-l-fructose 19; 37 g of 19
were obtained by direct crystallization of the crude oxidation
product from 50 g of rhamnose 1.[39] Longer times of the reac-
tion of 48 with IK7 gave oxidation of the (2S,3S) diol to form
rhamnulose (6-deoxy-l-fructose) 49 (10%);[41] 49 was also be
obtained by l-rhamnose isomerase (RI) equilibration with 1.[30]
d-Tagatose-3-epimerase (DTE) equilibrates all C3 epimeric pairs
of deoxyketoses[40] and equilibrated 49 at C3 to give 6-deoxy-
l-psicose 50.[41] Reaction of 50 with l-arabinose isomerase (AI)
afforded 6-deoxy-l-altrose 18 allowing the ready preparation
of 10 g of 18.
2. 6-Deoxy-l-talose (8)
1H and 13C NMR spectra of 6-deoxy-l-talose (8) in D2O, pH 7.8,
showed four spin-systems (Table 2), confirmed as the pyranose
and furanose forms by the HMBC correlations between C1 and
H5 (pyranose) and H4 (furanose). The a- and b-anomers of the
pyranose forms of 8 were identified from the ROESY spectra,
the H1 of the b-anomer showing strong correlations with H3
and H5 whereas the H1 of the a-anomer only showed a strong
correlation with H2 (Figure 1). The furanose forms were more
difficult to analyze. The b-anomer was tentatively assigned on
the basis of a stronger H1 to H2 ROESY correlation (see the
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
5
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 1. 13C NMR data of 6-deoxy-d-gulose 6 (D2O), a-py/b-py=1:10.
C1
C2
C3
C4
C5
C6
a pyranose
b pyranose
1H[a]/J[b]
13C
1H/J
13C
5.11/3.0
91.9
4.85/8.3
92.8
3.88/3.0/3.3
63.8
3.57/8.3/3.2
68.0
4.01/3.3/4.0
70.0
4.07/3.2/3.2
70.4
3.75/4.0/1.0
70.6
3.62/3.2/0.6
71.0
4.37/1.0/6.6
61.5
4.10/0.6/6.5
68.4
1.21/6.6
13.8
1.22/6.5
14.3
[a] ppm. [b] Hz.
Supporting Information). In addition, the 13C data in this study
agree with a previous report.[43]
the large trans-diaxial coupling between H1 and H2 and
a strong ROESY correlation between H1 and H5 (Figure 2). The
analysis of furanose forms was more difficult. The b-anomer
was tentatively assigned on the basis of a weak H1 to H4
ROESY correlation, not observed for the a-anomer.
Since 6-deoxy-b-d-allofuranose has an identical ring confor-
mation to 6-deoxy-a-l-talofuranose (they only differ by the
configuration at C5; Figure 2), the NMR parameters of the a-
and b-furanose forms of 7 should be very similar to the b- and
a-furanose forms 8, respectively. This provides additional sup-
port for the assignment of the furanose a- and b-anomers in 7
and 8.
Figure 1. Key ROESY correlations of 6-deoxy-a-l-talopyranose and 6-deoxy-
b-l-talopyranose.
3. 6-Deoxy-d-allose (7)
The NMR spectroscopy data of 6-deoxy-b-d-allopyranose has
been previously reported although in a different solvent.[44] 1H
and 13C NMR spectra of 6-deoxy-d-allose 7 in D2O, pH 6.1,
showed four spin-systems (Table 3), confirmed as the pyranose
and furanose forms by the HMBC correlations between H1 and
C5 (pyranose) and C4 (furanose; see the Supporting Informa-
tion). The b-anomer of the pyranose forms was identified by
4
Figure 2. 6-Deoxy-b-d-allopyranose in a C1 conformation and 6-deoxy-a-l-
talofuranose and 6-deoxy-b-d-allofuranose.
Table 2. 13C NMR data of 6-deoxy-l-talose 8 (D2O), a-py/b-py/a-fu/b-fu=1.00:0.72:0.36:0.14.
C1
C2
C3
C4
C5
C6
a pyranose
b pyranose
a furanose
b furanose
1H[a]/J[b]
13C
5.21/1.3
95.3
4.77/1.1
94.6
5.23/1.6
101.3
5.35/2.6/1.3
96.8
3.80/1.3/3.2
71.0
3.85/1.1/3.3
71.8
3.97/1.6/4.9
76.0
4.08/n.a.[c]
71.6
3.91/3.2/3.2
66.0
3.75/3.3/3.3
69.3
4.20/4.9/6.4
71.7
4.08/n.a.
71.3
3.74/3.2/1.0
72.8
3.66/3.3/1.0
71.7
3.71/6.4/6.4
86.4
3.90/n.a.
87.1
4.17/1.0/6.7
67.6
3.69/1.0/6.5
71.9
3.83/6.4/6.3
69.5
3.84/n.a.
67.9
1.24/6.7
16.4
1.27/6.5
16.2
1.23/6.3
18.7
1.22/6.4
19.0
1H/J
13C
1H/J
13C
1H/J
13C
[a] ppm. [b] Hz. [c] Cannot be assigned due to overlaps or weak signals.
Table 3. 13C NMR data of 6-deoxy-d-allose 7 (D2O), a-py/b-py/a-fu/b-fu=0.11:1.00:0.03:0.04.
C1
C2
C3
C4
C5
C6
a pyranose
b pyranose
a furanose
b furanose
1H[a]/J[b]
13C
5.08/3.4
93.4
4.85/8.3
93.9
5.35/4.2
96.6
5.22/2.3
101.1
3.70/3.4/3.4
68.0
3.40/8.3/3.0
72.1
4.04/n.a.
71.8
3.94/n.a.
76.1
4.12/n.a.[c]
72.1
4.12/3.0/2.8
71.7
4.16/5.9/3.9
69.4
4.30/5.2/5.2
70.7
3.36/n.a.
72.3
3.36/2.8/9.8
72.9
4.00/3.9/3.8
87.8
3.82/n.a.
86.3
4.04/9.4/6.4
63.5
3.82/9.8/6.3
70.1
3.93/n.a.
67.6
3.93/n.a.
68.1
1.26/6.4
17.2
1.24/6.3
17.6
1.20/6.7
17.9
1.22/6.7
17.9
1H/J
13C
1H/J
13C
1H/J
13C
[a] ppm. [b] Hz. [c] Cannot be assigned due to overlaps or weak signals.
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
6
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
4. l-Gluco-analogues and l-manno-analogues
of the trans-diaxial configuration (Figure 3). In contrast, in the
l-manno-series, the magnitude of the coupling constants of
anomeric protons were similar due to the trans-diequatorial or
cis-equatorial-axial configurations (J1,2 =0–2 Hz; Figure 3), while
the 1H peaks of a-anomers were in lower field than their
b counterparts, which was consistent with NMR data of l-rham-
nose.[45c]
The ten l-gluco-series (Table 4) and l-manno-series targets syn-
thesized from benzylidene protected l-rhamnonolactone 9,
have similar NMR spectra to those of glucose and mannose.[45]
Table 5 shows the NMR comparison of anomeric protons of
1
l-manno-analogues and the anomeric H, 13C shifts and their
coupling constants, in D2O. The pyranose forms were the pre-
dominant (>95% NMR ratio) in all analogues as judged by the
intensities of the 13C NMR peaks in the region associated with
pyranose forms (d=93–97 ppm) compared to the intensities in
the region associated with furanose forms (the anomeric car-
bons of which are generally at lower field). In the l-gluco-series
Conclusions
This paper describes the efficient production of 15 6-deoxy-
sugars from l-rhamnose by a synergy of chemical synthesis
and biotechnology as part of a project to find alternative non-
metabolizable inducers to rhamnose for the l-rhamnose
operon. Several inducers with different activities were identi-
fied: 1-deoxy-l-fructose 19 and 6-deoxy-l-altrose 18 were
1
(12, 38, 39, 41, 17) and l-olivose 13, the H NMR peaks of b-
anomeric protons were showed larger coupling constants
(J1,2 =8–9 Hz) than their a counterparts (J1,2 =3–4 Hz) because
Table 4. NMR comparison of anomeric protons of l-gluco-analogues.
l-gluco-series
12
5.10[a]/3.8[b]
93.2
38
5.29/3.7
91.6
39
5.38/3.5
89.5
41
5.06/3.5
91.0
17
5.38/4.0
89.7
H1a/J1,2
C1a
H1b/J1,2
C1b
4.55/7.9
96.9
4.67/8.1
95.5
4.91/8.4
93.1
4.59/8.5
95.4
4.87/7.9
92.5
[a] ppm. [b] Hz.
Table 5. NMR comparison of anomeric protons of l-manno-analogues.
l-manno-series
13
5.31/3.7
1.0
42
5.20/1.1
n.a.[c]
92.7
43
5.32/1.2
n.a.
45
4.85/1.1
n.a.
16
5.30/1.8
n.a.
[b]
H1a[a]/J1,2
J1,2’
C1a
91.7
90.8
94.8
91.8
H1b/J1,2
J1,2’
4.91/9.8
2.0
5.01/1.1
n.a.
5.17/1.5
n.a.
4.72/1.5
n.a.
4.98/0
n.a.
C1b
93.8
93.4
91.3
95.0
92.7
[a] ppm. [b] Hz. [c] Not available.
Figure 3. Conformations of 2-substituted l-quinovose and l-rhamnose.
Chem. Eur. J. 2016, 22, 1 – 10 www.chemeurj.org
7
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
G. W. J. Fleet, J. Chem. Soc. Perkin Trans. 1 1999, 2735–2745; c) R. Lucas,
P. Balbuena, J. C. Errey, M. A. Squire, S. S. Gurcha, M. McNeil, G. S. Besra,
B. G. Davis, ChemBioChem 2008, 9, 2197–2199; d) A. E. Hꢅkansson, J. va-
n Ameijde, G. Horne, R. J. Nash, M. R. Wormald, A. Kato, G. S. Besra, S.
Gurcha, G. W. J. Fleet, Tetrahedron Lett. 2008, 49, 179–184; e) S. Dharu-
man, Y. Wang, D. Crich, Carbohydr. Res. 2016, 419, 29–32.
[12] C. L. Kelly, Z. Liu, A. Yoshihara, S. F. Jenkinson, M. R. Wormald, J. Otero,
A. Estevez, A. Kato, M. H. S. Marqvorsen, G. W. J. Fleet, R. J. Estevez, K.
Izumori, J. T. Heap, ACS Synth. Biol. 2016, DOI: 10.1021/acssyn-
bio.6b00030.
[13] a) T. G. Frihed, M. Bols, C. M. Pedersen, Chem. Rev. 2015, 115, 3615–
3676; b) R. Bielski, Z. J. Witczak, Current Org. Chem. 2014, 18, 1897–
1912.
[14] a) K. J. Hale, L. Hough, S. Manaviazar, A. Calabrese, Org. Lett. 2015, 17,
1738–1741; b) K. J. Hale, L. Hough, S. Manaviazar, A. Calabrese, Org.
Lett. 2014, 16, 4838–4841.
weak inducers; l-olivose 13 and 6-deoxy-2-fluoro-l-rhamnose
16 were moderate inducers. Further studies explored their in-
duction kinetics and confirmed l-olivose 13, 6-deoxy-2-fluoro-
l-rhamnose 16 and 6-deoxy-l-altrose 18 were poorly metabo-
lized in E. coli and gave sustained induction.[12] Such inducers
with different strength could be used for different purposes
for example, production of toxic protein, production of certain
protein in constant rate. Rhamnonolactones have rarely been
used as chirons[46] but are ideal intermediates for the chemical
modification of rhamnose. Full NMR characterization of several
6-deoxyhexoses has been provided for the first time.
Experimental Section
[15] a) J. N. Dominguez, L. N. Owen, Carbohydr. Res. 1979, 75, 101–106;
b) S. J. Mantell, P. S. Ford, D. J. Watkin, G. W. J. Fleet, D. Brown, Tetrahe-
dron 1993, 49, 3343–3358; c) S. J. Mantell, G. W. J. Fleet, D. Brown, J.
Chem. Soc. Perkin Trans. 1 1992, 3023–3028.
Full details can be found in the Supporting Information includ-
ing copies of the NMR spectra (1H, 13C and 19F).
[16] S. W. Johnson, D. Angus, C. Taillefumier, J. H. Jones, D. J. Watkin, E.
Floyd, J. G. Buchanan, G. W. J. Fleet, Tetrahedron: Asymmetry 2000, 11,
4113–4125.
[17] S. R. Sanapala, S. S. Kulkarni, J. Am. Chem. Soc. 2016, 138, 4938–4947.
[18] M. P. Watterson, A. Martin, T. M. Krulle, J. C. Estevez, G. W. J. Fleet, Tetra-
hedron: Asymmetry 1997, 8, 4111–4120.
[19] B. G. Davis, A. Hull, C. Smith, R. J. Nash, A. A. Watson, D. A. Winkler, R. C.
Griffiths, G. W. J. Fleet, Tetrahedron: Asymmetry 1998, 9, 2947–2960.
[20] J. S. Brimacombe, J. Minshall, L. C. N. Tucker, Carbohydr. Res. 1973, 31,
146–150.
Acknowledgements
This work was supported by the Biotechnology and Biological
Sciences Research Council (BB/M011321/1 to J.T.H.), the Lever-
hulme Trust (G.W.J.F.), Xunta de Galicia (project GRC2014/040)
and Galchimia S.A.
[21] A. J. Fairbanks, N. C. Carpenter, G. W. J. Fleet, N. G. Ramsden, I. C.
de Belo, B. G. Winchester, S. S. Al-Daher, G. Nagahashi, Tetrahedron
1992, 48, 3365–3376.
Keywords: 6-deoxyhexoses
expression · rhamnonolactone · rhamnose operon
·
carbohydrates
·
gene
[22] a) C. F. Masaguer, Y. Bleriot, J. Charlwood, B. G. Winchester, G. W. J. Fleet,
Tetrahedron 1997, 53, 15147–15156; b) Y. Blꢁriot, C. F. Masaguer, J.
Charlwood, B. G. Winchester, A. L. Lane, S. Crook, D. J. Watkin, G. W. J.
Fleet, Tetrahedron 1997, 53, 15135–15146; c) Z. Liu, A. Yoshihara, M. R.
Wormald, S. F. Jenkinson, V. Gibson, K. Izumori, G. W. J. Fleet, Org. Lett.
2014, 16, 5663–5665.
[1] a) Z. Zheng, L. L. Chen, F. L. Zuo, W. Zhao, S. W. Chen, Zhongguo Sheng-
wu Gongcheng Zazhi 2013, 33, 131–137; b) T. Mascher, S. L. Zimmer,
T. A. Smith, J. D. Helmann, Antimicrob. Agents Chemother. 2004, 48,
2888–2896; c) R. S. Gupta, J. Ahnn, Korean J. Biol. Sci. 2003, 46, 173–
186; d) H. Kamaladini, S. N. A. Abdullah, M. Abdul Aziz, J. Biosci. Bioeng.
2011, 111, 217–225.
[23] L. M. Lerner, Carbohydr. Res. 1975, 44, 116–120.
[24] a) K. Bock, I. Lundt, C. Pedersen, Carbohydr. Res. 1988, 179, 87–96; b) H.
Kold, I. Lundt, C. Pedersen, Acta Chem. Scand. 1994, 48, 675–678.
[25] A. F. G. Glawar, S. F. Jenkinson, S. J. Newberry, A. L. Thompson, S. Naka-
gawa, A. Yoshihara, K. Akimitsu, K. Izumori, T. D. Butters, A. Kato, G. W. J.
Fleet, J. Org. Biomol. Chem. 2013, 11, 6886–6899.
[26] P. A. Levene, J. Compton, J. Biol. Chem. 1936, 116, 169.
[27] P. M. Collins, W. G. Overend, Chem. Ind. 1963, 375–376.
[28] S. W. Johnson, S. F. Jenkinson, D. Angus, C. Taillefumier, J. H. Jones,
G. W. J. Fleet, Tetrahedron: Asymmetry 2004, 15, 2681–2686.
[29] A. A. Bell, L. Pickering, M. Finn, C. de La Fuente, T. M. Krulle, B. G. Davis,
G. W. J. Fleet, Synlett 1997, 1077–1078.
[30] S. Shompoosang, A. Yoshihara, K. Uechi, Y. Asada, K. Morimoto, Biosci.
Biotechnol. Biochem. 2014, 78, 317–325.
[31] R. P. Elliott, G. W. J. Fleet, Y. S. Gyoung, N. G. Ramsden, C. Smith, Tetrahe-
dron Lett. 1990, 31, 3785–3788.
[32] M. Bols, I. Lundt, E. R. Ottosen, Carbohydr. Res. 1991, 222, 141–149.
[33] J. C. Dho, G. W. J. Fleet, J. M. Peach, K. Prout, P. W. Smith, Tetrahedron
Lett. 1986, 27, 3203–3204.
[34] M. B. Perry, V. Daoust, Carbohydr. Res. 1973, 27, 460–3.
[35] R. Albert, K. Dax, U. Katzenbeisser, H. Sterk, A. E. Stuetz, J. Carbohydr.
Chem. 1985, 4, 521–528.
[36] a) T. Liang, C. N. Neumann, T. Ritter, Angew. Chem. Int. Ed. 2013, 52,
8214–8264; Angew. Chem. 2013, 125, 8372–8423; b) O. Mahꢁ, A.
L’Heureux, M. Couturier, C. Bennett, S. Clayton, D. Tovell, F. Beaulieu, J.
Paquin, J. Fluorine Chem. 2013, 153, 57–60.
[37] I. Fokt, S. Szymanski, S. Skora, M. Cybulski, T. Madden, W. Priebe, Carbo-
hydr. Res. 2009, 334, 1464–1473.
[38] K. Izumori, J. Biotechnol. 2006, 124, 717–722.
[39] P. Gullapalli, T. Shiji, D. Rao, A. Yoshihara, K. Morimoto, G. Takata, G. W. J.
Fleet, K. Izumori, Tetrahedron: Asymmetry 2007, 18, 1995–2000.
[2] A. Boronat, J. Aguilar, J. Bacteriol. 1979, 140, 320–326.
[3] a) J. F. Tobin, R. F. Schleif, J. Mol. Biol. 1990, 211, 1–4; b) J. F. Tobin, R. F.
Schleif, J. Mol. Biol. 1987, 196, 789–799; c) S. M. Egan, R. F. Schleif, J.
Mol. Biol. 1993, 234, 87–98.
[4] a) T. Stumpp, B. Wilms, J. Altenbuchner, BIOspektrum 2000, 6, 33–36;
b) H. Beuttler, J. Hoffmann, M. Jeske, B. Hauer, R. D. Schmid, J. Alten-
buchner, V. B. Urlacher, Appl. Microbiol. Biotechnol. 2011, 89, 1137–1147.
[5] a) K. Terpe, Appl. Microbiol. Biotechnol. 2006, 72, 211–222; b) A. Haldi-
mann, L. L. Daniels, B. L. Wanner, J. Bacteriol. 1998, 180, 1277–1286;
c) M. J. Giacalone, A. M. Gentile, B. T. Lovitt, N. L. Berkley, C. W. Gunder-
son, M. W. Surber, Biotechniques 2006, 40, 355–64.
[6] J. R. Wickstrum, J. M. Skredenske, V. Balasubramaniam, K. Jones, S. M.
Egan, J. Bacteriol. 2010, 192, 225–232.
[7] a) J. Monod, G. Cohen-Bazire, M. Cohn, Biochem. Biophys. Acta 1951, 7,
585–99; b) W. Boos, K. Wallenfels, Eur. J. Biochem. 1968, 3, 360–363;
c) T. Lederer, M. Kintrup, M. Takahashi, P. E. Sum, G. A. Ellestad, W.
Hellen, Biochemistry 1996, 35, 7439–7446.
[8] a) S. L. Chiang, J. J. Mekalanos, Infect. Immun. 1999, 67, 976–980; b) Y.
Yamashita, K. Tomihisa, Y. Nakano, Y. Shimazaki, T. Oho, T. Koga, Infect.
Immun. 1999, 67, 3693–3697; c) K. A. Kantardjieff, C. Y. Kim, C. Naranjo,
G. S. Waldo, T. Lekin, B. W. Segelke, A. Zemla, M. S. Park, T. C. Terwilliger,
B. Rupp, Acta Crystallogr. Sect. D 2004, 60, 895–902.
[9] M. Y. Mistou, I. C. Sutcliffe, N. M. van Sorge, FEMS Microbiol. Rev. 2016;
DOI: 10.1093/femsre/fuw006.
[10] S. T. Cardona, C. L. Mueller, M. A. Valvano, Appl. Environ. Microbiol. 2006,
72, 2547–2555.
[11] a) J. P. Shilvock, J. R. Wheatley, B. Davis, R. J. Nash, R. C. Griffiths, M. G.
Jones, M. Mueller, S. Crook, D. J. Watkin, C. Smith, Tetrahedron Lett.
1996, 37, 8569–8572; b) J. P. Shilvock, J. R. Wheatley, R. J. Nash, A. A.
Watson, R. C. Griffiths, T. D. Butters, M. Muller, D. J. Watkin, D. A. Winkler,
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
8
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[40] a) A. Yoshihara, S. Haraguchi, P. Gullapalli, D. Rao, K. Morimoto, G.
Takata, N. A. Jones, S. F. Jenkinson, M. R. Wormald, R. A. Dwek, G. W. J.
Fleet, K. Izumori, Tetrahedron: Asymmetry 2008, 19, 739–745; b) N. A.
Jones, S. F. Jenkinson, R. Soengas, M. Fanefjord, M. R. Wormald, R. A.
Dwek, G. P. Kiran, R. Devendar, G. Takata, K. Morimoto, K. Izumori,
G. W. J. Fleet, Tetrahedron: Asymmetry 2007, 18, 774–786.
[41] P. Gullapalli, A. Yoshihara, K. Morimoto, D. Rao, S. F. Jenkinson, M. R.
Wormald, G. W. J. Fleet, K. Izumori, Tetrahedron Lett. 2010, 51, 895–898.
[42] J. B. Lambert, G. Lu, S. R. Singer, V. M. Kolb, J. Am. Chem. Soc. 2004, 126,
9611–9625.
de Bruyn, M. Antennas, R. de Gussem, G. G. S. Dutton, Carbohydr. Res.
1976, 47, 158–163.
[46] a) S. J. Mantell, G. W. J. Fleet, D. Brown, Tetrahedron Lett. 1992, 33,
4503–4506; b) S. J. Mantell, G. W. J. Fleet, D. Brown, J. Chem. Soc. Chem.
Commun. 1991, 1563–1564; c) S. W. Johnson, G. W. J. Fleet, J. H. Jones,
J. Pept. Sci. 2006, 12, 559–561; d) S. W. Johnson, S. F. Jenkinson, D.
Angus, I. Perez-Victoria, A. A. Edwards, T. D. W. Claridge, G. E. Tranter,
G. W. J. Fleet, J. H. Jones, J. Pept. Sci. 2005, 11, 517–524; e) S. W. John-
son, S. F. Jenkinson, D. Angus, I. Perez-Victoria, T. D. W. Claridge, G. W. J.
Fleet, J. H. Jones, J. Pept. Sci. 2005, 11, 303–318.
[43] S. Shompoosang, A. Yoshihara, K. Uechi, Y. Asada, K. Morimoto, J. Biosci.
Bioeng. 2016, 121, 1–6.
[44] S. Kobayashi, S. Onozawa, T. Mukaiyama, Chem. Lett. 1992, 2419–2422.
[45] a) D. Lafont, P. Boullanger, Tetrahedron: Asymmetry 2006, 17, 3368–
3379; b) W. A. Bubb, Concepts Magn. Reson. Part A 2003, 19, 1–19; c) A.
Received: May 24, 2016
Published online on && &&, 0000
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
9
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
Rhamnonolactone Chiron Perfection
[from Team Effort]: The efficient pro-
duction of fifteen 6-deoxy-sugars from
l-rhamnose by a synergy of chemical
synthesis and biotechnology to find al-
ternative non-metabolizable inducers to
rhamnose for the l-rhamnose operon is
described. Several inducers with differ-
ent levels of activity were identified.
&
Chemical Biology
Z. Liu, A. Yoshihara,* C. Kelly, J. T. Heap,
M. H. S. Marqvorsen, S. F. Jenkinson,
M. R. Wormald, J. M. Otero, A. Estꢀvez,
A. Kato, G. W. J. Fleet,* R. J. Estꢀvez,*
K. Izumori
&& – &&
6-Deoxyhexoses from l-Rhamnose in
the Search for Inducers of the
Rhamnose Operon: Synergy of
Chemistry and Biotechnology
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
10
ꢃ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!