IChO 1999 · Theory problem 5

Glycoside A from Rosaceae seeds

Glycoside A (\(\ce{C20H27NO11}\)), found in seeds of Rosaceae gives a negative test with Benedicts’ or Fehling’s solutions. Enzymatic hydrolysis of A yields (-) B, \(\ce{C8H7NO}\) and C, \(\ce{C12H22O11}\), but complete acid hydrolysis gives as organic products, (+) D, \(\ce{C6H12O6}\) and (-) E, \(\ce{C8H8O3}\).

C has a \(\beta\)-glycosidic linkage and gives positive test with Benedicts’ or Fehling’s solution. Methylation of C with \(\ce{MeI/Ag2O}\) gives \(\ce{C20H38O11}\), which upon acidic hydrolysis gives 2,3,4-tri-O-methyl-D-glucopyranose and 2,3,4,6-tetra-O-methyl-D-glucopyranose.

(±)B can be prepared from benzaldehyde and \(\ce{NaHSO3}\) followed by \(\ce{NaCN}\). Acidic hydrolysis of (±)B gives (±)E, \(\ce{C8H8O3}\).

5.1 Write structures of A – D with appropriate stereochemistry in Haworth projection, except for B.

Glycoside A is found to be toxic and believed to be due to extremely toxic compound F, liberated under the hydrolytic conditions. Detoxification of compound F in plant may be accompanied by the reactions (stereochemistry not shown).

Figure 1

A small amount of compound F in human being is believed to be detoxified by a direct reaction with cystine giving L-cysteine and compound I, \(\ce{C4H6N2O2S}\) which is excreted in urine (stereochemistry not shown).

Figure 2

Compound I shows no absorption at 2150-2250 \(\mathrm{cm^{-1}}\) in its IR spectrum but a band at 1640 \(\mathrm{cm^{-1}}\) and the bands of carboxyl group are observed.

5.2 Write molecular formula for compounds F and G, and structural formula for compounds H and I and indicate stereochemistry of H. (Table 5.1 may be useful for structure identification.)

(-)1-Phenylethane-1-d, \(\ce{C6H5CHDCH3}\) can be prepared in optically active form and the magnitude of its rotation has the relatively high value, \([\alpha]_{\mathrm D}\) is equal to -0.6.

Figure 3

The absolute configuration of (-)1-phenylethane-1-d is related to (-) E according to the following reactions.

Figure 4

Compound (-) M can also be obtained from compound N as follows.

Figure 5

5.3 Deduce the absolute configuration of (-) E and the structure with configuration of each intermediate (J – O) in the sequence with the proper \(R,S\)-assignment as indicated in the answer sheet.

5.4 Choose the mechanism involved in the conversion of compound O to 1-phenylethane-1-d.

Table 5.1 Characteristic Infrared Absorption

Stretching Vibration Region (\(\mathrm{cm^{-1}}\)) Stretching Vibration Region (\(\mathrm{cm^{-1}}\))
\(\ce{C-H}\) (alkane) 2850-2960 \(\ce{O-H}\) (free alcohol) 3400-3600
\(\ce{C-H}\) (alkene) 3020-3100 \(\ce{O-H}\) (H-bonded alcohol) 3300-3500
\(\ce{C=C}\) 1650-1670 \(\ce{O-H}\) (acid) 2500-3100
\(\ce{C-H}\) (alkyne) 3300 \(\ce{C-O}\) 1030-1150
\(\ce{C#C}\) 2100-2260 \(\ce{NH}\), \(\ce{NH2}\) 3310-3550
\(\ce{C-H}\) (aromatics) 3030 \(\ce{C-N}\) 1030, 1230
\(\ce{C=C}\) (aromatics) 1500-1600 \(\ce{C=N}\) 1600-1700
\(\ce{C-H}\) (aldehyde) 2700-2775, 2820-2900 \(\ce{C#N}\) 2210-2260
\(\ce{C=O}\) 1670-1780

Official English text of the problem.

Answers

  • 5.1: A is conventionally identified as amygdalin, the β-D-gentiobioside of \((R)\)-mandelonitrile. B is \((R)\)-mandelonitrile. C is gentiobiose, β-D-glucopyranosyl-\((1\to6)\)-D-glucopyranose. D is D-glucose.
  • 5.2: F is hydrogen cyanide, \(\ce{HCN}\); G is hydrogen sulfide, \(\ce{H2S}\). H is β-cyano-L-alanine, configuration \(S\). I is 2-amino-4,5-dihydro-1,3-thiazole-4-carboxylic acid, equivalently its 2-iminothiazolidine tautomer.
  • 5.3: The configurations are \(\boxed{E:R,\ J:R,\ K:R,\ L:R,\ M:S,\ N:S,\ O:S}\). Their structures are specified below.
  • 5.4: \(S_\mathrm{N}2\) displacement by deuteride, with inversion at the benzylic carbon, gives \((R)\)-1-phenylethane-1-d.

The following SMILES specify the structures. For the reducing sugars C and D, a β reducing-end anomer is shown; their free anomeric centres can mutarotate.

Compound Name and stereochemical SMILES
A Amygdalin, \((R)\)-mandelonitrile β-D-gentiobioside: N#C[C@H](O[C@@H]1O[C@H](CO[C@@H]2O[C@H](CO)[C@@H](O)[C@H](O)[C@H]2O)[C@@H](O)[C@H](O)[C@H]1O)c1ccccc1
B \((R)\)-Mandelonitrile: N#C[C@H](O)c1ccccc1
C Gentiobiose: O[C@@H]1O[C@H](CO[C@@H]2O[C@H](CO)[C@@H](O)[C@H](O)[C@H]2O)[C@@H](O)[C@H](O)[C@H]1O
D D-Glucose, β-D-glucopyranose shown: O[C@@H]1O[C@H](CO)[C@@H](O)[C@H](O)[C@H]1O
E \((R)\)-Mandelic acid: O=C(O)[C@H](O)c1ccccc1
F Hydrogen cyanide: C#N
G Hydrogen sulfide: S
H β-Cyano-L-alanine, \((S)\)-2-amino-3-cyanopropanoic acid: N[C@@H](CC#N)C(=O)O
I \((4R)\)-2-Amino-4,5-dihydro-1,3-thiazole-4-carboxylic acid: NC1=N[C@@H](CS1)C(=O)O
J Ethyl \((R)\)-2-ethoxy-2-phenylacetate: CCOC(=O)[C@H](OCC)c1ccccc1
K \((R)\)-2-Ethoxy-2-phenylethan-1-ol: OC[C@H](OCC)c1ccccc1
L \((R)\)-2-Ethoxy-2-phenylethyl benzenesulfonate: CCO[C@@H](COS(=O)(=O)c1ccccc1)c1ccccc1
M \((S)\)-1-Ethoxy-1-phenylethane: C[C@H](OCC)c1ccccc1
N \((S)\)-1-Phenylethanol: C[C@H](O)c1ccccc1
O \((S)\)-1-Phenylethyl benzenesulfonate: C[C@H](OS(=O)(=O)c1ccccc1)c1ccccc1
Isotopic product \((R)\)-1-Phenylethane-1-d: [2H][C@H](C)c1ccccc1

Solution

5.1 — Identification of the aglycone and sugars

Benzaldehyde, O=Cc1ccccc1, forms a bisulfite addition compound, which is converted into its cyanohydrin:

\[ \ce{C6H5CHO + NaHSO3 -> C6H5CH(OH)SO3Na} \]

\[ \ce{C6H5CH(OH)SO3Na + NaCN -> C6H5CH(OH)CN + Na2SO3}. \]

The species used here are sodium hydrogen sulfite [Na+].[O-]S(=O)O, sodium cyanide [Na+].[C-]#N, sodium hydroxy(phenyl)methanesulfonate [Na+].[O-]S(=O)(=O)C(O)c1ccccc1, and sodium sulfite [Na+].[Na+].[O-]S(=O)[O-].

Attack on either face of the initially planar carbonyl gives racemic mandelonitrile, N#CC(O)c1ccccc1. Thus B is mandelonitrile. Hydrolysis of its nitrile group gives mandelic acid:

\[ \ce{C6H5CH(OH)CN + 2H2O + H+ -> C6H5CH(OH)COOH + NH4+}. \]

This converts racemic B into racemic mandelic acid, O=C(O)C(O)c1ccccc1. The absolute configuration of the naturally derived B and E is established in 5.3.

Both products of the methylation analysis are derivatives of D-glucopyranose. Therefore C contains two D-glucose units, and D is D-glucose.

During methylation, each free hydroxyl becomes a methyl ether. The formula change is

\[ \ce{C12H22O11 -> C20H38O11}, \]

corresponding to eight substitutions of \(\ce{OH}\) by \(\ce{OCH3}\), each adding \(\ce{CH2}\). The balanced methylation is

\[ \ce{C12H22O11 + 8CH3I + 4Ag2O -> C20H38O11 + 8AgI + 4H2O}. \]

The methylated product, octa-O-methylgentiobiose, represented with a β methyl-glycoside terminus, is:

CO[C@@H]1O[C@H](CO[C@@H]2O[C@H](COC)[C@@H](OC)[C@H](OC)[C@H]2OC)[C@@H](OC)[C@H](OC)[C@H]1OC

Acid hydrolysis gives:

  • 2,3,4-Tri-O-methyl-D-glucopyranose, β anomer shown: O[C@@H]1O[C@H](CO)[C@@H](OC)[C@H](OC)[C@H]1OC.
  • 2,3,4,6-Tetra-O-methyl-D-glucopyranose, β anomer shown: O[C@@H]1O[C@H](COC)[C@@H](OC)[C@H](OC)[C@H]1OC.

The first product lacks methylation at position 6 because that oxygen originally participated in the intersugar linkage. The second glucose had free hydroxyls at positions 2, 3, 4 and 6. Therefore the linkage is \(1\to6\); its stated β configuration identifies C as gentiobiose.

Hydrolysis also removes the methyl group installed at the originally free anomeric hydroxyl, releasing methanol:

\[ \ce{C20H38O11 + 2H2O -> C9H18O6 + C10H20O6 + CH3OH}. \]

The reducing behaviour of C shows that one anomeric centre remains a hemiacetal. In A, this centre is converted into an acetal by attachment to the hydroxyl oxygen of mandelonitrile. Consequently A is nonreducing. Its formula agrees with condensation of B and C:

\[ \ce{C8H7NO + C12H22O11 - H2O = C20H27NO11}. \]

The hydrolyses balance as follows:

\[ \underbrace{\ce{C20H27NO11 + H2O -> C8H7NO + C12H22O11}} _{\text{enzymatic hydrolysis: A gives B and C}}, \]

\[ \ce{C12H22O11 + H2O -> 2C6H12O6}, \]

and, including nitrile hydrolysis,

\[ \underbrace{\ce{C20H27NO11 + 4H2O + H+ -> 2C6H12O6 + C8H8O3 + NH4+}} _{\text{complete acid hydrolysis: A gives two D and one E}}. \]

Haworth descriptions. Draw each glucopyranose ring with its ring oxygen at the upper right and C1 at the right. For each D-glucose unit, the C5 hydroxymethyl substituent points up, while the substituents at C2, C3 and C4 point down, up and down, respectively.

  • In C, the nonreducing glucose’s C1–O bond points up and connects to O6 of the reducing glucose. The reducing glucose’s C1–OH can point down, α, or up, β.
  • In the conventional structure of A, the reducing unit’s C1–O bond also points up, and this oxygen connects to the \((R)\)-mandelonitrile carbon.
  • In D, C1–OH can point down or up. The α-D-glucopyranose alternative is O[C@H]1O[C@H](CO)[C@@H](O)[C@H](O)[C@H]1O.
  • For B, use a Fischer projection with \(\ce{CN}\) at the top, phenyl at the bottom, H on the left and OH on the right. This is the \(R\) configuration.

5.2 — Toxic compound and detoxification products

Mandelonitrile can dissociate into benzaldehyde and hydrogen cyanide:

\[ \ce{C6H5CH(OH)CN <=> C6H5CHO + HCN}. \]

Hence \(\boxed{\mathbf F=\ce{HCN}}\). Combining glycoside hydrolysis with this dissociation gives

\[ \ce{C20H27NO11 + 2H2O -> 2C6H12O6 + C6H5CHO + HCN}. \]

The plant reaction replaces the sulfhydryl group of L-cysteine with a cyano group:

\[ \ce{HCN + HSCH2CH(NH2)COOH -> H2S + NCCH2CH(NH2)COOH}. \]

Here \(\ce{NCCH2}\) denotes \(\ce{N#C-CH2}\). Thus \(\boxed{\mathbf G=\ce{H2S}}\), and H is β-cyano-L-alanine.

L-Cysteine is N[C@@H](CS)C(=O)O. Its α-carbon retains its spatial arrangement during the side-chain substitution. H is consequently the L, or \(S\), amino acid. Its Fischer projection has \(\ce{COOH}\) at the top, \(\ce{CH2CN}\) at the bottom, \(\ce{NH2}\) on the left and H on the right.

Hydration of the nitrile gives L-asparagine, N[C@@H](CC(N)=O)C(=O)O:

\[ \ce{NCCH2CH(NH2)COOH + H2O -> NH2COCH2CH(NH2)COOH}. \]

For the human detoxification reaction, L-cystine is

N[C@@H](CSSC[C@H](N)C(=O)O)C(=O)O.

The overall balance is

\[ \ce{C6H12N2O4S2 + HCN -> C3H7NO2S + C4H6N2O2S}. \]

Cyanide attacks a disulfide sulfur, producing cysteine and S-cyano-L-cysteine, N[C@@H](CSC#N)C(=O)O. The amino group can then attack the cyano carbon intramolecularly, forming a five-membered ring.

For I,

\[ \mathrm{DBE}=\frac{2(4)+2+2-6}{2}=3. \]

These three degrees of unsaturation are accounted for by one carboxyl carbonyl, one ring and one \(\ce{C=N}\) bond. The absence of a nitrile absorption and the band at \(1640\ \mathrm{cm^{-1}}\), within the given \(1600\text{–}1700\ \mathrm{cm^{-1}}\) imine range, support this cyclic amidine.

The ring sequence is sulfur–amidine carbon–nitrogen–carboxyl-bearing carbon–methylene. I can be written as either:

  • 2-Amino-4,5-dihydro-1,3-thiazole-4-carboxylic acid: NC1=N[C@@H](CS1)C(=O)O.
  • Its tautomer, 2-iminothiazolidine-4-carboxylic acid: N=C1N[C@@H](CS1)C(=O)O.

The retained cysteine stereocentre is \(R\), although the question explicitly requires stereochemistry only for H.

5.3 — Absolute configurations and intermediates

First assign the configuration of the displayed deuterated product. Its CIP priorities are

\[ \ce{C6H5}>\ce{CH3}>\ce{^{2}H}>\ce{^{1}H}. \]

Phenyl outranks methyl because its attached carbon is compared with carbon neighbours, whereas methyl has hydrogen neighbours. Deuterium outranks protium by isotope mass.

In the supplied Fischer projection, methyl is at the top, phenyl at the bottom, D on the left and H on the right. Since the lowest-priority H points toward the observer, reversing the apparent sense gives \(R\).

Deuteride displaces the sulfonate group of O with inversion. Therefore O is \(S\). Sulfonylation of N modifies oxygen without breaking the stereogenic carbon–oxygen bond, so N is also \(S\).

The route from N to M is oxygen alkylation. It likewise preserves the stereocentre:

\[ \ce{2C6H5CH(OH)CH3 + 2K + 2C2H5I -> 2C6H5CH(OC2H5)CH3 + 2KI + H2}. \]

Thus M is \(S\).

Now follow the other route.

  1. E → J: ethylation of both acidic O–H groups

The alcohol becomes an ethyl ether and the carboxylic acid becomes an ethyl ester:

\[ \ce{C6H5CH(OH)COOH + 2C2H5I + Ag2O -> C6H5CH(OC2H5)COOC2H5 + 2AgI + H2O}. \]

No bond at the stereogenic carbon is displaced.

  1. J → K: ester reduction

The ester group becomes \(\ce{CH2OH}\), while the ethyl ether remains intact. A balanced net equation for reduction and work-up is

\[ \ce{C6H5CH(OC2H5)COOC2H5 + 2H- + 2H+ -> C6H5CH(OC2H5)CH2OH + C2H5OH}. \]

Hydride and acid are supplied in separate stages.

  1. K → L: sulfonylation of the primary alcohol

\[ \ce{C6H5CH(OC2H5)CH2OH + C6H5SO2Cl + C5H5N -> C6H5CH(OC2H5)CH2OSO2C6H5 + C5H6N+ + Cl-}. \]

  1. L → M: replacement of the primary sulfonate by hydrogen

\[ \ce{C6H5CH(OC2H5)CH2OSO2C6H5 + H- -> C6H5CH(OC2H5)CH3 + C6H5SO3-}. \]

Substitution occurs at the neighbouring primary carbon, so the spatial arrangement at the original stereocentre remains unchanged.

The configuration labels nevertheless change between L and M, because the priorities change:

Compounds CIP order at the stereogenic carbon
E \(\ce{OH}>\ce{COOH}>\ce{C6H5}>\ce{H}\)
J \(\ce{OC2H5}>\ce{COOC2H5}>\ce{C6H5}>\ce{H}\)
K \(\ce{OC2H5}>\ce{CH2OH}>\ce{C6H5}>\ce{H}\)
L \(\ce{OC2H5}>\ce{CH2OSO2C6H5}>\ce{C6H5}>\ce{H}\)
M \(\ce{OC2H5}>\ce{C6H5}>\ce{CH3}>\ce{H}\)

The second and third priorities exchange when the substituted hydroxymethyl group becomes methyl. Consequently E, J, K and L are \(R\), although M is \(S\). This label change does not represent inversion of the stereocentre.

For precise Fischer drawings, put phenyl at the bottom and H on the left in every structure below. Place the listed groups at the top and right; horizontal bonds project toward the observer.

Compound Top group Right group Configuration
E \(\ce{COOH}\) \(\ce{OH}\) \(R\)
J \(\ce{COOC2H5}\) \(\ce{OC2H5}\) \(R\)
K \(\ce{CH2OH}\) \(\ce{OC2H5}\) \(R\)
L \(\ce{CH2OSO2C6H5}\) \(\ce{OC2H5}\) \(R\)
M \(\ce{CH3}\) \(\ce{OC2H5}\) \(S\)
N \(\ce{CH3}\) \(\ce{OH}\) \(S\)
O \(\ce{CH3}\) \(\ce{OSO2C6H5}\) \(S\)

Formation of O balances as

\[ \ce{C6H5CH(OH)CH3 + C6H5SO2Cl + C5H5N -> C6H5CH(OSO2C6H5)CH3 + C5H6N+ + Cl-}. \]

Finally, nitrile hydrolysis of B does not alter its stereogenic carbon. Its priority sequence is \(\ce{OH}>\ce{CN}>\ce{C6H5}>\ce{H}\), corresponding to the same ordering as in E. Therefore B, and the mandelonitrile centre of A, are also \(R\).

The signs of optical rotation alone do not determine any of these \(R/S\) assignments.

5.4 — Mechanism of O → 1-phenylethane-1-d

The mechanism is a concerted \(S_\mathrm{N}2\) substitution:

\[ \ce{C6H5CH(OSO2C6H5)CH3 + ^2H- -> C6H5CHDCH3 + C6H5SO3-}. \]

Deuteride attacks the benzylic carbon from the side opposite the carbon–oxygen bond. Simultaneously, that bond’s electron pair moves onto oxygen, releasing benzenesulfonate. The tetrahedral centre undergoes Walden inversion, giving the displayed \(R\) product from \(S\)-O.

The replacement changes the priority order from oxygen–phenyl–methyl to phenyl–methyl–deuterium. This is an even cyclic permutation, so the physical inversion also changes the descriptor from \(S\) to \(R\).

For completeness, the remaining reagents and small products appearing in the equations are:

Species SMILES
Water; methanol; ethanol O; CO; CCO
Proton; ammonium; hydrogen [H+]; [NH4+]; [H][H]
Hydride; deuteride [H-]; [2H-]
Methyl iodide; ethyl iodide CI; CCI
Silver(I) oxide; silver iodide [Ag+].[Ag+].[O-2]; [Ag+].[I-]
Potassium; potassium iodide [K]; [K+].[I-]
Pyridine; pyridinium; chloride n1ccccc1; [nH+]1ccccc1; [Cl-]
Benzenesulfonyl chloride; benzenesulfonate O=S(=O)(Cl)c1ccccc1; O=S(=O)([O-])c1ccccc1
Lithium aluminium hydride [Li+].[AlH4-]
Lithium aluminium deuteride [Li+].[2H][Al-]([2H])([2H])[2H]
Diethyl ether; hydronium CCOCC; [OH3+]

The solution was checked against the official solution.