Current Synthetic Approaches to the Synthesis of Carbasugars from Non-Carbohydrate Sources

Carbasugars are a group of carbohydrate derivatives in which the ring oxygen is replaced by a methylene group, producing a molecule with a nearly identical structure but highly different behavior. Over time, this definition has been extended to include other unsaturated cyclohexenols and carba-, di-, and polysaccharides. Such molecules can be found in bacterial strains and the human body, acting as neurotransmitters (e.g., inositol trisphosphate). In science, there are a wide range of research areas that are affected by, and involve, carbasugars, such as studies on enzyme inhibition, lectin-binding, and even HIV and cancer treatment. In this review article, different methods for synthesizing carbasugars, their derivatives, and similar cyclohexanes presenting comparable characteristics are summarized and evaluated, utilizing diverse starting materials and synthetic procedures.

Various strategies have been employed for the synthesis and isolation of carbasugars, mostly starting from simple monosaccharides such as glucosamine and resulting in the corresponding carbasugars [17][18][19]. Other approaches work from simple hexopyranoses to more complex carbafuranoses such as carbaarabinofuranosides [20,21] or even more complex carbasugar derivatives such as fluorinated carbasugars [22]. All these methods start from simple monosaccharidic compounds and require numerous synthetic steps to produce the final product [23][24][25][26][27][28][29]. In past years, scientists have developed and investigated possibilities for easily accessible alternative approaches. Starting with more basic chemical structures, a wide range of different strategies have been employed to achieve the goal of synthesizing carbasugars from non-carbohydrate sources.
Natural products containing carbasugar subunits fall beyond the scope of this review and will not be dealt with here, though their application in medicinal chemistry is of course important [11]. The chemical synthesis of carbasugars and their derivatives comprises the topic of this review. There are different approaches to these compounds, which are classified broadly into two groups: synthetic methods that use non-carbohydrates as starting materials, and procedures that make use of carbohydrates as precursors [12]. In pharmaceutical production, carbasugar synthesis is not being used, as it is based most dominantly on carbohydrate sources. Due to the complexity and inefficiency of the synthesis pathways, mass production of carbasugar-based pharmaceuticals is not feasible. Thus, a new type of carbomimetic synthesis is desired in order to ensure an efficient access to carbasugar production. Therefore, in this review, we concentrate on strategies that start from non-carbohydrate sources.

Synthesis Methods
The following sections will review six methods for synthesizing carbasugars from non-carbohydrate sources. The methods discussed are based on cylohexadiendiol, Norbornene, the Diels-Alder reaction, iodobenzene, methyl benzoate, or benzoquinone.

Synthesis Based on Cyclohexadiendiol
In this section, we analyze the synthesis of pseudosugars based on cyclohexadiendiol. The pericosines A (1), B (3), and C (4), shown in Scheme 2, occur naturally and were isolated originally from the fungus Periconia byssoides. This fungus can be found in the gastrointestinal tract of the sea hare Aplysia kurodai [30,31]. The initial synthesis of pericosine was accomplished by Donohoe in 1998 [32]. Since then, interest in synthesizing pericosines has increased steadily.
Periscosines are known for their cytotoxicity against P388 lymphocytic leukaemia cells, antitumor activity against murine P388 cells, and selective growth inhibition against human cancer cell lines HBC-5 and SNB-75 [33]. By fulfilling these functions, pericosines may be important in treating cancer. In addition, the structural similarities between pericosines (polyhydroxylated cyclohexenes) and pseudosugars (polyhydroxylated cyclohexanes) suggest that both could be classified as carbasugars [34].
Based on the reactions shown in Scheme 3, the dihydroxylation of cis-dihydrodiol 5, using the Donohoe procedure [35], gave a 4:1 mixture of two cis,cis-tetraol diastereoisomers. These diastereoisomers sequentially resulted from an oxidative attack on the same face at the 5,6 and 3,4 double bonds. Following the reaction in Scheme 3, the authors were able to isolate, after chromatography, regioisomer 6 with a 70% yield [34].
Furthermore, the authors theorized that the hydroxyl group at the C6 position of compound 6 would be less reactive in the subsequent acetal reaction because the C6 hydroxyl was hydrogen-bonded to the adjacent carbomethoxy group. Following  the reaction, tetraol 6 was treated at room temperature with acetone under acidic conditions, which resulted in a 7:4:9-mixture of acetonides 7, 8, and 9. There was no acetal formation between the hydroxyl groups at C4 and C5. Acetonide 7 (74% overall yield) was produced by separating the mono-and bis-acetonides, repeatedly recycling the unwanted acetals 8 and 9 via hydrolysis, and reacetalizing the recovered tetraol 6 [34]. The authors also achieved synthesis of (+)-pericosine B (3) after hypothesizing that this form could be synthesized from tetraol 6. Synthesis was achieved after selective protection of the three hydroxyl groups on carbons C3, C4, and C5. To achieve the stated protection, the authors used the bulky TBS protecting group. However, the remaining hydroxyl group at C6 proved resistant to methylation under a broad range of conditions. Another way to accomplish this synthesis was to employ (1S,2R)-3-cyanocyclohex-3-ene-1,2-diol 11 bearing a CN group that was less bulky than a CO 2 Me group (Scheme 4). By following the same steps used for the reaction of cis-dihydrodiols 5 and 19, the authors produced a 4:1 mixture of cis,cis-and cis,trans-tetraols in the dihydroxylation of diol 11. From this, the authors isolated the major cis,cis isomer 12 with a 54% yield based on column chromatography. The reaction of tetraol 12 with tert-butyldimethylsilyl triflate produced the tri-TBS derivative 13 as the major product with 85% overall yield, together with small amounts of other unidentified inseparable isomers. The free C6 hydroxyl group of the coarse sample of silyl derivative 13 was then methylated, under mild conditions, and a purified sample of methyl ether 14 was isolated easily from the product mixture. The nitrile group in compound 14 was partially reduced to the aldehyde 15, followed by oxidation with sodium chlorite, resulting in 16. The authors then attempted to form the methyl ester 18 by reaction of carboxylic acid 16 with excess diazomethane. Instead of achieving the desired compound 18, the crystalline pyrazoline cycloadduct 17 resulted from this reaction. Thus, synthesis of compound 18 is achieved by base-mediated methylation of carboxylic acid 16, followed by acid-catalyzed removal of the TBS protecting groups, which produced (+)-pericosine B (3) through seven steps, with an overall yield of 12% [34].
In conclusion, it was demonstrated that cis-dihydrodiols derived from methyl benzoate, iodobenzene, and cyanobenzene are versatile complementary intermediates for the rapid synthesis of pericosines A, C, and B, respectively.
Additionally, the laboratories used mutant strains (e.g., UV4, 39D) of the soil bacterium Pseudomonas putida and Escherichia coli, each containing toluene dioxygenase. This factor provided access to a vast range of over 400 metabolites [34].
The strategic approach for strains to produce 2,3-trans-CHD 28 can be compared to the work performed using Klebsiella pneumoniae strains. Non-pathogenic E. coli strains are common hosts for genetic modification because of the availability of potent E. coli mutants and their well-established fermentation conditions. In this way, previous studies acknowledged another host for synthesizing carbasugars [34,36].

Synthesis Based on Norbornene
In this section, we analyze carbasugar synthesis based on norbornene, hereby covering only one paper reporting this approach. The structural entities of cyclitols and the polyhydroxylated cyclohexanoids compose important segments of a wide range of natural products, like antibiotics, and they exhibit biological activity profiles extending from glycosidase inhibitors to antidiabetes and anticancer agents.
To achieve the synthesis pathway for the carbasugars, the authors first used a 'bottom-to-top' Grob-like fragmentation process in an easily accessible 2,7-disubstituted norbornane derivative 33 to cleave the C1-C2 bond 34. Furthermore, the process included extraction of the five-membered ring 35 from the bridged bicyclic frame while retaining full functionalization (see Scheme 7). Taking the paper under consideration, the authors developed a beneficial method for polyhydroxylated cyclohexenoid synthesis. In the following, polyhydroxylated cyclohexenoids are obtained from the same starting materials and subsequently interchanged with functional groups, thereby showing the possibility of extracting either five-or sixmembered rings from the norbornyl system [37].
In the cited paper, the authors tested a new approach that involved switching the functionalities in 33 to those in 36. By doing so, the authors arranged a "top-tobottom" sequence involving C7-C1 bond cleavage (see 37, Scheme 8) to deliver a functionally adorned cyclohexanoid 38 in a regio-and stereoselective manner [37].
To execute the approach shown in Scheme 8, clear access to 36 is needed. To ensure passage, the process in Scheme 9 was followed. Bicyclic alcohol 39, available from 5,5-dimethoxy-1,2,3,4-tetrachlorocyclopentadiene and vinyl acetate, was tosylated and exposed to OsO 4 -mediated catalytic dihydroxylation to yield exo,exodiol 40. By designing an amberlyst mediated single-pot protection-deprotection in 40, the authors achieved the 7-norbornanone derivative 41. By exposing 41 to

Scheme 7 Synthesis of acetonide compound 35
NaOMe, the authors achieved smooth top-to-bottom fragmentation, yielding the cyclohexene methyl ester as a single product. The secured stereochemistry of the six ring carbons in cyclohexenoid 42 can be further elaborated to carbasugars [37].
Dihydroxylation across the C5-C6 double bond in cyclohexenoid ester 42 occurred through addition of osmium tetroxide, producing a large yield of cis-diol 43 in a stereoselective manner (Scheme 10). Lithium aluminium hydride (LAH) reduction of the ester group of 43, followed by acetonide deprotection provided the naturally occurring carbasugar pseudo-α-galactose (29). This carbasugar was characterized as penta-acetate 44. Another method would involve the reduction of the ester group in 42 with LAH and acetylation, resulting in 45 (Scheme 10). To create 46, the carbon-carbon double bond of 45 must be converted to an oxirane (Scheme 10).
Finally, pseudo-α-fucopyranose (50), which was synthesized from ester 42, showed a possible use as an inhibitor of fucosyltransferases. Due to this potential, the authors found another method for synthesizing 50. Instead of using 39 as a starting point, the authors used the diol ester 43 obtained from 42. This structure was converted into bis-acetonide and subjected to LAH reduction, yielding 51. Next, 51 was tosylated, and reductive detosylation using sodium borohydride was carried out in DMSO. This led to installation of the β-methyl group and bis-acetonide 52. Deprotection of 52 provided pseudo-α-fucopyranose (50), which presented spectroscopic characteristics identical to those noted in previous research [37] (Scheme 11).

Synthesis Based on Diels-Alder Reaction
This reaction, named after its discoverers Diels and Alder from Kiel [38], involves the formation of a ring of six carbon atoms, wherein a conjugated diene and dienophile are linked. The importance of the Diels-Alder reaction is that C-C bonds can be created with high stereoselectivity. Diels-Alder reactions play an important role in the synthesis of natural substances, such as the production of steroids (e.g., the female sex hormone estradiol). In this instance, the racemic cyclohexadiene derivative 53 reacted selectively in a catalytic enantioselective Nitroso-Diels-Alder reaction to give the two main products ent-anti-54 and anti-55, thereby showing the selective formation of two products from eight possible isomers (Scheme 12) [39].
The dienes used for cycloaddition were noted by the authors to be easily accessible, and the products were highlighted as valuable starting materials for the synthesis of carbasugars, such as peracetylated 2-epi-validamine (58, Scheme 15) [40]. It was hypothesized that divergent reactions from racemates are also possible in other Diels-Alder reactions of unsymmetrical dienophiles with racemic cyclic dienes. This approach could lead to a new concept in the field of stereoselective cycloadditions [39]. A Nitroso-Diels-Alder reaction with the highly enantiomerically enriched diene 53a (R = (S)-CHPhOTBDPS, 98% ee) was then examined. The authors found that 53a could be produced easily by desymmetrizing 1,4-cyclohexadiene. At the same time, the ent-compounds in Scheme 12 were not considered. The formation of adducts 54a and 55a occurred in CH 2 Cl 2 in the presence of [Cu(MeCN) 4 ] PF6 (10 mol%), a chiral diphosphine (10 mol%), and 2-nitrosopyridine (6 h at −78 °C and then 12 h at −20 °C). The resulting two products, entanti-54a and ent-anti-55a, were produced in excellent yields (column chromatography, SiO 2 ) and isolated with high enantiomeric excesses (see Scheme 13) [39]. Next, a Nitroso-Diels-Alder reaction was carried out with the racemic dienes 53b-f (Scheme 14). The implementation of 53b was successful and presented excellent anti/syn-selectivity. The adducts ent-anti-54b and anti-55b were developed with high enantioselectivities. The diene 53c showed poorer diastereoselectivity (anti/syn = 7:1) since the less sterically demanding CH 2 OTBDPS group represents a less efficient shield for the syn-side than the alpha-branched substituents. Similar results were observed for the benzyl-substituted diene 53d and the diene 53e, which carries an acetoxymethyl group. The best result was shown for the Ph-substituted diene 53f (see Scheme 14) [39].
Another approach for the synthesis of a racemate corresponding to validamine 65 was developed by Suami et al. [41,42], whose approach was based on the Diels-Alder addition of furan to acrylic acid, resulting in oxanorbornene 59 (Scheme 16). Through hydrogenation in ethyl acetate and oxidation utilizing H 2 O 2 , an intermediate compound was achieved in 76% yield. Reduction and acetylation gave 69% of oxanorbornane 60, followed by acetolysis with AcOH and H 2 SO 4 , giving protected 5a-carba-β-dl-glucopyranose 61 (20% yield). Deprotection of 61 resulted in 5a-carba-β-dl-glucopyranose (62). Reaction of compound 62 with DMP in dimethylformamide (DMF) in the presence of p-TSA and subsequent crystallization resulted in an inseparable 2:3 crystalline sulfonate mixture of compound 63 and 3-p-toluenesulfonate. Sodium azide in DMF caused creation of a new hexane compound from 63, while preserving the 3-p-toluenesulfonate (11%). Evaporation gave 67% of the new azide compound 64. Reduction with H 2 in the presence of Raney nickel in ethanol, followed by acid hydrolysis of the protective groups resulted in the racemate corresponding to validamine 65. The racemate was isolated as penta-N,O-acetyl-dl-validamine (28%, Scheme 16) [41,42].

Synthesis of Carba-β-d-Altropyranose and Carba-α-l-Galactopyranose
The reaction processes for all four carbasugars are based on iodobenzene cisdihydrodiol 67 (Scheme 18). The synthesis of iodobenzene cis-dihydrodiol from iodobenzene through the use of Pseudomonas putida was clearly reported by Derek et al. [46] and will not be elaborated. The first reaction step involved the with sodium acetate and methanol, the (3aR,6R,7R,7aS)-α,β-unsaturated ester 74 was created with a yield of 81% after being catalyzed through palladium(II) acetate [43].

Alternate Synthesis of Carba-β-l-Glucopyranose
As noted earlier, the previously presented synthesis pathway for carba-β-lglucopyranose did not yield a suitable amount of the desired carbasugar. Thus, a longer yet more efficient synthesis pathway was developed.

Synthesis of (−)-Gabosine A
Based on iodobenzene (1S,2S)-cis-dihydrodiol 67, Banwell et al. [48] presented a pathway that provides (−)-gabosine A (116) over six steps (Scheme 23). The reaction started by protecting compound 67 at the less sterically hindered hydroxyl group at C1 under a nitrogen atmosphere, thus forming TBDPS-ether 111. Utilizing UpJohn dihydroxylation conditions [49], two hydroxyl groups were formed at the non-halogenated carbon-carbon double bond, yielding triol 112. The addition of an acetonide occurred under known conditions in the presence of triethylamine, thereby forming acetonide compound 113. This compound was oxidized under Swern conditions, yielding the ketone 114, followed by replacement of the iodine with a methyl group, utilizing iron-catalyzed reactions developed by Cahiez and Avedissian [50].

Synthesis Based on Methyl Benzoate
As stated in the previous section, various different types of benzene derivatives are used for carbasugar synthesis. This section explores methyl benzoate as a reactant. Similarly, methyl benzoate was converted into its cis-dihydrodiol metabolite, thereby ensuring the same skeletal structure as mono carbasugars along with the desired relative configuration at C2 and C3 to synthesize three more sugar analogues.

Synthesis of Carba-β-l-Talopyranose
The next reactions started with the previously synthesized cis-dihydrodiol methyl benzoate derivative 120 (Scheme 26). The addition of osmium tetroxide with trimethylamine-N-oxide in a dichloromethane solution resulted in a 70% yield of the tetraol compound 127. Catalytic hydrogenation resulted primarily in the saturated tetraol compound 128, while also yielding 30% of the achiral meso-triol 129. This mixture was inseparable under charcoal/Celite chromatography and was treated with 2,2-DMP and p-TSA, resulting in a mono acetonide 131/bis-acetonide 130 mixture, which was then separated using flash-column chromatography to yield 59% of the bis-acetonide compound 130.
An alternative synthesis route was also explored. This route began with compound 127 and results in the same saturated bis-acetonide compound 130. This synthesis was achieved by changing the order of the two reactions (hydrogenation and protecting). This process started by protecting all four hydroxyl groups with an acetonide using 2,2-DMP and p-TSA to yield 80% of another bis-acetonide compound CO  134 and was followed by catalytic hydrogenation utilizing the previously presented conditions. This process afforded 91% of the desired saturated bis-acetonide while completely suppressing the competing hydrogenation reaction (i.e., 127 → 128/129). Both remaining steps were similar to all previous reactions, starting with reduction of the carbomethoxy group function to a hydroxymethyl group (Scheme 26) using LAH in Et 2 O to give alcohol compound 132 (76% yield) and deprotection using TFA in aqueous THF to yield 86% of carba-β-l-talopyranose (118). The carbasugar was then converted to its penta-acetate derivative 133 using A c2 O and pyridine (85% yield) for characterization [51].

Synthesis of Carba-α-l-Talopyranose
Finally, carba-α-l-talopyranose (119) was synthesized in seven steps (Scheme 27). The reaction started with the installation of an epoxide group at C4 and C5 of the cis-dihydrodiol derivative of methyl benzoate 120. This was achieved via the addition of MCPBA in dichloromethane to give 82% cis-diol epoxide 135. Compound 135 was protected by an acetonide (98% yield), followed by opening the epoxide ring using tert-butanol in water (pH 8 buffer) to produce 68% of cyclohexene transdiol 137. The two hydroxyl groups were then protected by acetylation, yielding 98% triester 138. Through catalytic hydrogenation similar to previous reactions, compound 138 was transformed into the saturated ester 139 (83% yield). LAH was added to a Et 2 O solution to afford the trihydroxy acetonide-protected carba-α-ltalopyranose (140, 71% yield). The compound was deprotected using TFA and THF in water, yielding 88% of the desired carbasugar 119. Furthermore, the compound

Based on Benzoquinone
The final section covers the synthetic pathways for various unsaturated carbasugars, including streptol and gabosine, resulting from masked p-benzoquinone. Leermann et al. [16] explored the synthesis of several unsaturated carbasugars used in lectinbinding studies to determine the influence of sugar derivatives on enzyme inhibition and cancer therapy.

Synthesis of Dibromo Acetate Intermediates
The starting materials for all pathways in this section were the two intermediates 145 and 149 formed from 2-acetoxymethyl benzoquinone (142) and 2-methyl benzoquinone (146), respectively (Schemes 28, 29). 2-Acetoxymethyl benzoquinone (142) was dibrominated regioselectively at the unsubstituted double bond, yielding 98% of dibromo acetate 143. This compound was further reduced using sodium borohydride in diethyl ether and water (82% yield) and acetylated under the conditions described in previous sections to form the racemic dibromo acetoxymethyl diacetate intermediate 145 [16].

Synthesis of Streptol
Streptol (164) can be obtained from intermediate compound 145 in four steps, starting with the formation of an epoxide from the trans bromide and acetoxy group via addition of lithium hydroxide in Et 2 O and methanol, preserving relative configuration at C4 and C5 and yielding epoxide 158 (83% yield). Nucleophilic ring opening in water, followed by acetylation, yielded 51% of bromide diacetate 160. Inversion and acetylation of the bromide group resulted in penta-acetate 162. Deacetylation of 162 with methanolic sodium provided the desired rac-streptol (164) with an 81% yield (Scheme 32) [16].

Synthesis of Unsaturated Pentaol 165
Similar to Scheme 32, the synthesis from compound 149 began with epoxidation to 159 (Scheme 33). Ring opening occurred through the addition of carbon tetrabromide in water, followed by acetylation, yielding 37% of the bromide triacetate compound 161. Through bromide displacement by acetate, the authors obtained the methyl tetraol 165 (Scheme 33) with the same relative configuration as streptol (164) [16].

Synthesis of Unsaturated Pentaol 171
The last pentaol compound was obtained in six steps (Scheme 34). Initially, intermediate 145 was deacetylated using potassium carbonate in methanol, resulting in dibromide diol 166, whose vicinal hydroxyl groups were protected using 2,2-DMP and p-TSA in acetone, ultimately yielding 100% acetonide 167. Epoxidation occurred via the addition of sodium hydroxide in Et2O and water with a 52% yield of epoxide 168, followed by deprotection and acetylation comparable to those described in earlier pathways, yielding 52% of bromide tetra acetate 169. The Woodward reaction conditions [47]

Summary
The present review highlights the success of carbasugar synthesis from non-sugar compounds. Due to their structural and configuration similarities with the desired products, the reactants used were able to deliver the products in few synthetic steps with comparatively high efficiency. These results suggest that future experiments may permit generation of the same products even faster and with higher yields. Due to the great diversity in the structures already known, the road is now paved for successful research in chemical and clinical medicine, including HIV and tumor treatment [14,53,54]. Due to the growing population and prevalence of common diseases, the importance of pseudo-sugar synthesis is greater than ever before. Ultimately, the synthetic routes described in this review offer information on already successful research and provide an impetus for the development of new discoveries in the field of carbasugars.