IChO 2006 · Theory problem 10

Natural Products - Structural Analysis

Figure 1

Licorice (Glycyrrhizia. Uralensis)

Figure 2

Licorice Root

The flavour extracted from the licorice root is 50 – 150 times sweeter than table sugar. The most important and abundant compound responsible for the sweetness and medicinal effects of licorice is glycyrrhizin (\(\ce{C42H62O16}\)).

Glycyrrhizin requires three equivalents of \(\ce{NaOH}\) to effect neutralization. When glycyrrhizin was subjected to acid hydrolysis, Glycyrrhizinic acid (A (\(\ce{C30H46O4}\))) and B (\(\ce{C6H10O7}\)) were obtainned in a 1:2 molar ratio (Scheme 1).

Figure 3

Scheme 1

When glycyrrhizin was methylated with methyl iodide (\(\ce{MeI}\)) at every possible site before hydrolysis, hydrolysis produced A’ (methyl glycyrrhizinate), and C and D (Scheme 2).

B, C and D exist as mixtures of anomers.

Figure 4

Scheme 2

Methylation of C and D with \(\ce{MeI}\) produced the same isomeric mixture of compounds, J (Scheme 3).

Figure 5

Scheme 3

C was reduced with \(\ce{LiAlH4}\) to give K, and L was produced by the reduction of K. Oxidative cleavage of vicinal diol of L with \(\ce{NaIO4}\) produced M and two equivalents of formaldehyde. Reduction of M produced N. The structure and stereochemistry of N was confirmed by the synthesis of N from D-(-)-tartaric acid through methylation followed by reduction (Scheme 4). A \(\ce{^{1}H}\)-NMR spectrum of L showed two distinct peaks for methyl groups. (There is no symmetry in L)

Figure 6

Scheme 4

10.1 Complete structures for L, M, and N in the answer sheet.

10.2 How many structures for C are possible? Complete possible structures for C.

To determine the correct structure of C, following set of reactions were performed.

J was reduced to E, and acid hydrolysis of E produced F. Reduction of F generated G, and G was oxidized with \(\ce{NaIO4}\) to H with formation of one equivalent of formaldehyde. I was obtained from H through reduction. Among all compounds from A to I, only I was optically inactive (Scheme 5).

Figure 7

Scheme 5

10.3 Complete structures for G and I.

10.4 Which one is the correct structure for C among ones you have drawn in 10-2?

10.5 Complete structures for B, D, and J.

10.6 Complete the structure for Glycyrrhizin.

Official English text of the problem.

Answers

  • 10.1.

L: 3,4-di-O-methyl-D-glucitol, \(\ce{C8H18O6}\).
SMILES: OC[C@H](O)[C@@H](OC)[C@H](OC)[C@H](O)CO

M: (2S,3S)-2,3-dimethoxybutane-1,4-dial, \(\ce{C6H10O4}\).
SMILES: O=C[C@@H](OC)[C@H](OC)C=O

N: (2R,3R)-2,3-dimethoxybutane-1,4-diol, \(\ce{C6H14O4}\).
SMILES: OC[C@@H](OC)[C@H](OC)CO

  • 10.2. There are two possible sugar configurations, each occurring as two anomers: thus two anomeric pairs, or four individual cyclic stereoisomers.

Candidate 1: methyl 3,4-di-O-methyl-D-glucopyranuronate.

α: COC(=O)[C@H]1O[C@H](O)[C@H](O)[C@@H](OC)[C@@H]1OC

β: COC(=O)[C@H]1O[C@@H](O)[C@H](O)[C@@H](OC)[C@@H]1OC

Candidate 2: methyl 3,4-di-O-methyl-L-gulopyranuronate.

α: COC(=O)[C@@H]1O[C@@H](O)[C@@H](O)[C@@H](OC)[C@@H]1OC

β: COC(=O)[C@@H]1O[C@H](O)[C@@H](O)[C@@H](OC)[C@@H]1OC

  • 10.3.

G: 2,3,4-tri-O-methyl-D-glucitol, \(\ce{C9H20O6}\).
SMILES: OC[C@H](OC)[C@@H](OC)[C@H](OC)[C@H](O)CO

I: meso-2,3,4-tri-O-methylxylitol, \(\ce{C8H18O5}\).
SMILES: OC[C@H](OC)[C@@H](OC)[C@H](OC)CO

  • 10.4. Candidate 1 is correct: C is methyl 3,4-di-O-methyl-D-glucopyranuronate, as the α/β pair specified in 10.2.

  • 10.5.

B: D-glucopyranuronic acid, \(\ce{C6H10O7}\).

α: O=C(O)[C@H]1O[C@H](O)[C@H](O)[C@@H](O)[C@@H]1O

β: O=C(O)[C@H]1O[C@@H](O)[C@H](O)[C@@H](O)[C@@H]1O

D: methyl 2,3,4-tri-O-methyl-D-glucopyranuronate, \(\ce{C10H18O7}\).

α: COC(=O)[C@H]1O[C@H](O)[C@H](OC)[C@@H](OC)[C@@H]1OC

β: COC(=O)[C@H]1O[C@@H](O)[C@H](OC)[C@@H](OC)[C@@H]1OC

J: methyl 1,2,3,4-tetra-O-methyl-D-glucopyranuronate, \(\ce{C11H20O7}\); equivalently, the methyl ester bearing a methyl glycoside at C1 and methoxy groups at C2, C3 and C4.

α: COC(=O)[C@H]1O[C@H](OC)[C@H](OC)[C@@H](OC)[C@@H]1OC

β: COC(=O)[C@H]1O[C@@H](OC)[C@H](OC)[C@@H](OC)[C@@H]1OC

  • 10.6. The established connectivity is A 3-O-[D-glucopyranuronosyl-(1→2)-D-glucopyranuronoside], with all three carboxyl groups free.

The conventional β,β structure of glycyrrhizin, \(\ce{C42H62O16}\), is:

CC1(C)[C@@H](O[C@@H]6O[C@H](C(=O)O)[C@@H](O)[C@H](O)[C@H]6O[C@@H]7O[C@H](C(=O)O)[C@@H](O)[C@H](O)[C@H]7O)CC[C@]2(C)[C@H]3C(=O)C=C4[C@@H]5C[C@@](C)(C(=O)O)CC[C@]5(C)CC[C@@]4(C)[C@]3(C)CC[C@@H]12

Qualification: the supplied degradation experiments determine the connectivity and sugar configurations, but do not independently establish the two original β anomeric configurations.

Solution

In the Fischer projections described below, the carbon chain is vertical, horizontal bonds point toward the viewer, and hydrogen occupies the position opposite the stated oxygen substituent. Sugar positions refer to the original six-carbon chain.

For the reduction balances, \(\ce{[H]}\) denotes a formal reducing hydrogen equivalent. Ester reduction is written as

\[ \ce{RCOOCH3 + 4[H] -> RCH2OH + CH3OH}. \]

This expresses the net organic transformation after work-up.

10.1. Structures of L, M and N

Hydrolysis has the balanced stoichiometry

\[ \ce{C42H62O16 + 2H2O ->[HCl] C30H46O4 + 2C6H10O7}. \]

The aglycone contains one carboxyl group. Since glycyrrhizin consumes three equivalents of base and yields two identical molecules of B, each B supplies one further carboxyl group.

Replacing an O–H hydrogen by methyl adds \(\ce{CH2}\). Relative to B, C contains three such increments:

\[ \ce{C9H16O7}=\ce{C6H10O7}+3\ce{CH2}. \]

One methyl group forms the carboxylic methyl ester; the other two are methyl ethers. Ester reduction removes the ester methyl carbon from the isolated carbohydrate product as methanol, giving K, followed by reduction to the alditol L:

\[ \ce{C9H16O7 + 4[H] ->[LiAlH4] C8H16O6 + CH3OH} \qquad(\mathbf C\rightarrow\mathbf K), \]

\[ \ce{C8H16O6 + H2 ->[Raney-Ni] C8H18O6} \qquad(\mathbf K\rightarrow\mathbf L). \]

The production of two equivalents of formaldehyde requires a terminal vicinal diol at each end of L. Its constitution must therefore be

\[ \ce{HOCH2-CH(OH)-CH(OCH3)-CH(OCH3)-CH(OH)-CH2OH}. \]

Periodate cleaves the C1–C2 and C5–C6 bonds, leaving

\[ \ce{OHC-CH(OCH3)-CH(OCH3)-CHO} \]

as M. Reduction of its two aldehyde groups gives N:

\[ \ce{HOCH2-CH(OCH3)-CH(OCH3)-CH2OH}. \]

The balanced reactions are

\[ \ce{C8H18O6 + 2NaIO4 -> C6H10O4 + 2HCHO + 2NaIO3 + 2H2O}, \]

\[ \ce{C6H10O4 + 2H2 ->[Raney-Ni] C6H14O4}. \]

The tartaric-acid experiment fixes the absolute configurations of the two central carbons.

D-(-)-tartaric acid is (2S,3S)-2,3-dihydroxybutanedioic acid, SMILES
O=C(O)[C@@H](O)[C@H](O)C(=O)O.

Its fully methylated derivative is dimethyl (2S,3S)-2,3-dimethoxybutanedioate, SMILES
COC(=O)[C@@H](OC)[C@H](OC)C(=O)OC.

The balances are

\[ \ce{C4H6O6 + 4CH3I + 2Ag2O -> C8H14O6 + 4AgI + 2H2O}, \]

\[ \ce{C8H14O6 + 8[H] ->[LiAlH4] C6H14O4 + 2CH3OH}. \]

Neither transformation changes the spatial configuration at the two existing stereocentres. The Fischer projections of M and N, with identical terminal groups at the top and bottom, consequently have methoxy left, then methoxy right.

Thus M is \((2S,3S)\), whereas N is \((2R,3R)\). This change of descriptors results from changed CIP priorities upon reduction; it does not indicate inversion.

For L, the central methoxy groups likewise lie left at C3 and right at C4. The remaining possibilities concern the hydroxyl groups at C2 and C5:

C2 hydroxyl C5 hydroxyl Consequence
Right Left End-exchange symmetry makes the methoxy groups equivalent
Left Right End-exchange symmetry makes the methoxy groups equivalent
Right Right Methoxy groups are inequivalent
Left Left Same alditol as the preceding row after rotating the Fischer projection by \(180^\circ\)

The two distinct methyl resonances therefore select the single asymmetric alditol L given in the Answers.

To draw it, place \(\ce{CH2OH}\) at both ends and arrange the oxygen substituents from C2 through C5 as:

\[ \boxed{\text{OH right;\ OMe left;\ OMe right;\ OH right}.} \]

10.2. Possible structures of C

Reduction to L makes both ends of the original sugar into \(\ce{CH2OH}\). It therefore erases the distinction between the original reducing end and the original carboxyl end.

Assigning those ends in the two possible ways gives the following Fischer configurations. The top is the original C1 aldehyde position and the bottom is the C6 methyl ester.

Candidate for C C2–OH C3–OMe C4–OMe C5–OH
D-glucuronic configuration Right Left Right Right
L-guluronic configuration Left Left Right Left

In the cyclic structures, the C5 oxygen forms the pyranose ring, and C1 is a hemiacetal bearing a free hydroxyl group. Each configuration therefore gives an α and a β anomer.

Hence there are two configurational candidates, represented by four individual anomers.

For a conventional Haworth drawing with the ring oxygen at the upper right:

  • D-glucuronic candidate: C5–\(\ce{COOCH3}\) up, C2–OH down, C3–OMe up, C4–OMe down; C1–OH down for α and up for β.
  • L-guluronic candidate: C5–\(\ce{COOCH3}\) down, C2–OH up, C3–OMe up, C4–OMe down; C1–OH up for α and down for β.

These are the structures encoded by the four SMILES in 10.2.

10.3. Structures of G and I

Exhaustive methylation gives J with methyl groups at the carboxyl oxygen and at O1, O2, O3 and O4.

Reduction of the ester produces a C6 primary alcohol. Acid hydrolysis then removes the anomeric methyl group, while retaining the ordinary methyl ethers at C2, C3 and C4. Reduction of the reducing sugar consequently gives an alditol of constitution

\[ \ce{HOCH2-CH(OCH3)-CH(OCH3)-CH(OCH3)-CH(OH)-CH2OH}. \]

For the D-glucuronic candidate, this is G, 2,3,4-tri-O-methyl-D-glucitol. Its Fischer oxygen-substituent sequence is

\[ \boxed{\text{OMe right;\ OMe left;\ OMe right;\ OH right}.} \]

The intermediate structures for this candidate are:

Compound Identification Representative β-anomer SMILES
E Methyl 2,3,4-tri-O-methyl-D-glucopyranoside OC[C@H]1O[C@@H](OC)[C@H](OC)[C@@H](OC)[C@@H]1OC
F 2,3,4-Tri-O-methyl-D-glucopyranose OC[C@H]1O[C@@H](O)[C@H](OC)[C@@H](OC)[C@@H]1OC

The corresponding α forms differ only at C1. The balanced transformations are

\[ \ce{C11H20O7 + 4[H] ->[LiAlH4] C10H20O6 + CH3OH}, \]

\[ \ce{C10H20O6 + H2O ->[HCl] C9H18O6 + CH3OH}, \]

\[ \ce{C9H18O6 + H2 ->[Raney-Ni] C9H20O6}. \]

Only C5–C6 in G is a free vicinal diol. Its cleavage produces one molecule of formaldehyde and the aldehyde H:

\[ \ce{HOCH2-CH(OCH3)-CH(OCH3)-CH(OCH3)-CHO}. \]

H is open-chain 2,3,4-tri-O-methyl-L-xylose, SMILES
OC[C@H](OC)[C@@H](OC)[C@H](OC)C=O.

Reduction gives I, with \(\ce{CH2OH}\) at both ends and the Fischer methoxy sequence right–left–right:

\[ \ce{HOCH2-CH(OCH3)-CH(OCH3)-CH(OCH3)-CH2OH}. \]

The balanced reactions are

\[ \ce{C9H20O6 + NaIO4 -> C8H16O5 + HCHO + NaIO3 + H2O}, \]

\[ \ce{C8H16O5 + H2 ->[Raney-Ni] C8H18O5}. \]

The two ends of I are identical. Its internal mirror symmetry exchanges C2 with C4, making it a meso compound and therefore optically inactive. This gives the required structure of I.

10.4. Selection of C

The L-guluronic candidate would instead give a pentitol with the Fischer methoxy sequence left–left–right.

That alternative is 2,3,4-tri-O-methyl-D-lyxitol, SMILES
OC[C@@H](OC)[C@@H](OC)[C@H](OC)CO.

It is chiral and would remain optically active. The stated optical inactivity of I therefore selects the D-glucuronic candidate:

\[ \boxed{\mathbf C=\text{methyl 3,4-di-O-methyl-D-glucopyranuronate}.} \]

The terminal aldehyde reduction does not generate a new stereocentre, so the optical inactivity is explained by the meso structure, without invoking racemization.

With this selection, K is 3,4-di-O-methyl-D-glucopyranose; a representative β-anomer SMILES is
OC[C@H]1O[C@@H](O)[C@H](O)[C@@H](OC)[C@@H]1OC.

10.5. Structures of B, D and J

Removing the methyl groups from the identified C gives B, D-glucuronic acid.

The substitution patterns are:

Compound C1 substituent C2 substituent C3 substituent C4 substituent C6 group
B \(\ce{OH}\) \(\ce{OH}\) \(\ce{OH}\) \(\ce{OH}\) \(\ce{COOH}\)
C \(\ce{OH}\) \(\ce{OH}\) \(\ce{OCH3}\) \(\ce{OCH3}\) \(\ce{COOCH3}\)
D \(\ce{OH}\) \(\ce{OCH3}\) \(\ce{OCH3}\) \(\ce{OCH3}\) \(\ce{COOCH3}\)
J \(\ce{OCH3}\) \(\ce{OCH3}\) \(\ce{OCH3}\) \(\ce{OCH3}\) \(\ce{COOCH3}\)

All retain the D-gluco configuration at C2–C5.

The formula differences confirm these assignments:

\[ \begin{aligned} \mathbf C &: \quad \ce{C6H10O7}+3\ce{CH2}=\ce{C9H16O7},\\ \mathbf D &: \quad \ce{C6H10O7}+4\ce{CH2}=\ce{C10H18O7},\\ \mathbf J &: \quad \ce{C6H10O7}+5\ce{CH2}=\ce{C11H20O7}. \end{aligned} \]

To reach J, C requires methylation of O1 and O2, whereas D requires methylation only of O1:

\[ \ce{C9H16O7 + 2CH3I + Ag2O -> C11H20O7 + 2AgI + H2O}, \]

\[ \ce{2C10H18O7 + 2CH3I + Ag2O -> 2C11H20O7 + 2AgI + H2O}. \]

Thus both give the same pair of methyl-glycoside anomers of J, as required.

10.6. Structure of glycyrrhizin

Methylation analysis identifies the original glycosidic attachment positions: hydroxyl groups exposed only by hydrolysis were unavailable for methylation in intact glycyrrhizin.

Attachment to the aglycone. The depicted aglycone A is commonly called glycyrrhetinic acid, although the problem labels it “Glycyrrhizinic acid.” Its stereochemical SMILES is

CC1(C)[C@@H](O)CC[C@]2(C)[C@H]3C(=O)C=C4[C@@H]5C[C@@](C)(C(=O)O)CC[C@]5(C)CC[C@@]4(C)[C@]3(C)CC[C@@H]12

After methylation and hydrolysis, A’ contains only one added methyl group. It is methyl glycyrrhetinate, SMILES

CC1(C)[C@@H](O)CC[C@]2(C)[C@H]3C(=O)C=C4[C@@H]5C[C@@](C)(C(=O)OC)CC[C@]5(C)CC[C@@]4(C)[C@]3(C)CC[C@@H]12

Its carboxyl group was methylated, but its C3 hydroxyl was protected by a glycosidic linkage during methylation. Hydrolysis restores that hydroxyl. Consequently, the sugar chain is attached through the aglycone’s C3 oxygen.

Attachment between the sugars. The inner sugar gives C, which has an unmethylated hydroxyl at C2 after hydrolysis. Its O2 must therefore have participated in the intersugar glycosidic bond. The terminal sugar gives D, methylated at all three non-anomeric hydroxyl positions.

The connectivity is therefore

\[ \boxed{ \text{terminal D-GlcpA} \xrightarrow{\;1\to2\;} \text{inner D-GlcpA} \xrightarrow{\;1\to O3\;} \mathbf A } \]

where GlcpA denotes a glucopyranuronic-acid residue.

For the drawing, copy the aglycone in Scheme 1 and:

  1. Replace the hydrogen of its C3 hydroxyl by the anomeric carbon of the inner D-glucopyranuronic-acid residue.
  2. Attach the terminal residue’s anomeric carbon through the inner residue’s O2.
  3. Retain the carboxyl group on the aglycone and the C6 carboxyl group of each sugar as free acids.
  4. For the conventional β,β glycyrrhizin structure, draw both anomeric glycosidic substituents on the same side as the corresponding C5 carboxyl substituent.

The three free carboxyl groups explain the neutralization:

\[ \ce{C42H62O16 + 3NaOH -> C42H59Na3O16 + 3H2O}. \]

The composition also agrees with condensation of A and two molecules of B:

\[ \ce{C30H46O4 + 2C6H10O7 -> C42H62O16 + 2H2O}. \]

There are five free alcoholic hydroxyl groups and three free carboxyl groups in glycyrrhizin, giving eight methylations. The combined balance for exhaustive methylation followed by hydrolysis is

\[ \ce{C42H62O16 + 8CH3I + 4Ag2O -> C31H48O4 + C9H16O7 + C10H18O7 + 8AgI + 2H2O}, \]

whose organic products are A’, C and D, respectively.

For completeness, the small species used in these balances have the following SMILES: formaldehyde C=O; methanol CO; water O; hydrogen [H][H]; methyl iodide CI; hydrogen chloride Cl; lithium aluminium hydride [Li+].[AlH4-]; silver(I) oxide [Ag+].[Ag+].[O-2]; silver iodide [Ag+].[I-]; sodium periodate [Na+].[O-]I(=O)(=O)=O; sodium iodate [Na+].[O-]I(=O)=O; sodium hydroxide [Na+].[OH-]. Raney nickel is a porous solid catalyst, for which [Ni] represents the element but not the catalyst morphology.

The solution was checked against the official solution.