Synthesis of a fluorogenic substrate for α-L -iduronidase

An alternative reaction pathway towards the preparation of an L -idopyranose derivative and its application to the synthesis of the α-L -iduronidase fluorogenic substrate 4-methylcoumarin-7-yl- α-L -iduronic acid as well as its 3-undecyl derivative are described. The L - ido sugar was prepared by converting the commercially available diacetone-α-D -glucose to methyl 1,2,3,4-tetra-O - acetyl-L -idopyranuronate via oxidation, esterification, and regioselective acetylation of the key intermediate 1,2:3,5-di-O -isopropylidene-  - L -idofuranose. Mitsunobu-type glycosylation was employed in the coupling of the L - ido donor with the 4-methylcoumarin acceptors. This newly developed route reduced the difficulties previously encountered in the synthesis of the α-L - iduronidase fluorogenic substrate.


Introduction
Mucopolysaccharidosis I (MPS I) is a genetic disorder caused by the body's inability to make an enzyme called α-L-iduronidase (IDUA), 1 one of the enzymes involved in the degradation of glycosaminoglycans.Failure of IDUA to hydrolyze dermatan sulfate and heparan sulfate can lead to the accumulation of partially degraded products in lysosomes, which can result to cell, tissue, and organ dysfunction.Generally, the disease can manifest as milder condition but there are severe cases that can lead to death during the first decade of life. 2 Currently, both hematopoietic stem cell transplantation and enzyme replacement therapy are available for the treatment of MPS I, 3 but early detection is essential to avoid the severe progression of the disease. 4ne well-known technique used for the diagnosis of all clinical types of MPS I is the measurement of IDUA activity, 5 wherein 4-methylcoumarin-7-yl-α-L-iduronic acid (4) is used as fluorogenic substrate. 6However, this particular substrate is very expensive (>US$100 per mg) primarily because of the relative unavailability and difficult synthesis of the L-idose moiety.In addition, the 4-methylcoumarin group has poor water solubility, and IDUA activity assays require a wide range of substrate concentration. 7Although several synthetic approaches have been reported on the synthesis of L-ido sugars and their coumarin derivatives, the strategy still involves tedious steps and remains low yielding at the glycosylation stage. 6,8Thus, we explored and developed alternative methods of preparing L-ido sugars, 8c,d,9 such as compound 3, from different D-glucose derivatives and used these materials as key intermediates for the synthesis of the fluorogenic substrate 4. Scheme 1.Previous synthetic pathways for the synthesis of the L-ido sugar 3 and compound 4.
Scheme 1 outlines the synthetic pathways we established for the preparation of compounds 3 and 4 from either 3,5-O-benzylidene-1,2-O-isopropylidene-α-D-glucopyranose 8c (1) or diacetoneα-D-glucose 8d (2).Although the overall yields were relatively good, most reactions are difficult to perform and each step requires tedious purification.Furthermore, the glucopyranose derivative 1 is more expensive than other D-glucose derivatives.This led us to explore other reaction pathways that would provide the best synthetic route for the synthesis of L-ido derivatives and the fluorogenic substrate 4 from the commercially available and cheap compound 2. We report herein the progress of our work.

Results and Discussion
Our vision for the preparation of the L-ido sugars is based on the double ketal fixation of the 1,2 and 3,5-hydroxy groups of D-glucose to form a cis-anti-cis-fused tricyclic D-glucofuranosyl derivative, which could undergo elimination to form a 5-exo-double bond followed by electrophilic addition to give the desired products.8d,9a Accordingly, further studies were carried out to determine an alternative and efficient synthetic route for the L-ido synthesis using the cheaper compound 2 as the starting material (Scheme 2).Scheme 2. Preparation of 2,3,4-tri-O-acetyl-L-idopyranuronate methyl ester (8).
The L-idofuranose derivative 5 was readily acquired from the highly selective hydroboration of the enol ether generated after treatment of the 3-alcohol 2 with triphenylphosphine, Nbromosuccinimide (NBS), and freshly distilled 1,8-diaza-bicyclo[5.4.0]undec-7-ene (DBU).8d In our previous reports, exposure of this L-ido derivative with ethanolic-HCl solution furnished compound 3. 8d, 9 We opted not to follow this synthetic route to avoid the more complicated steps upon coupling of the sugar with the coumarin moiety 8d and pursued an alternative approach in preparing the L-ido sugar donor.Oxidation of compound 5 was performed using 2,2,6,6tetramethyl-1-piperidinyloxy free radical (TEMPO) to produce the corresponding carboxylic acid which underwent methylation (CH3I, KHCO3) to form the ester 6 in a two-step yield of 88%.Hydrolysis of the isopropylidene protecting groups in 6 using trifluoroacetic acid (TFA) immediately followed, without further purification, by peracetylation (Ac2O, pyridine) unfortunately delivered the tetraacetate 7 in only 17% yield.We found that lactone formation occurred as a major side-reaction upon treatment of 6 with TFA.The yield was improved to 38% when the products after acid hydrolysis were subjected to NaOMe prior to peracetylation.Bromination of the anomeric carbon using HBr in acetic acid followed by AgOTf-promoted hydration successfully converted 7 into the desired L-ido hemiacetal 8 (79% in two steps).Scheme 3. Formation of an orthodiester with AgOTf as coupling agent.
We have already tried to synthesize the target molecule 4 through the coupling of the donor 9 with 7-hydroxy-4-methylcoumarin (10) in the presence AgOTf as promoter (Scheme 3).8d However, the orthoester 11 was the major product isolated instead of the desired α-adduct.Using this knowledge, we decided to explore a Mitsunobu-type glycosylation reaction for the coupling of the hemiacetal 8 with the acceptor 10.Scheme 4. Preparation of compound 4 and its 3-undecyl derivative 15.
The coumarin 10 was first prepared via Pechmann condensation of resorcinol and ethyl acetoacetate. 10With the donor and acceptor blocks in hand, the glycosylation was performed using PPh3 and diethyl azodicarboxylate (DEAD).This reaction produced compound 13 in 73% yield (α/β 1.7/1).Saponification of the α-anomer with LiOH supplied the target material 4 in excellent yield.To enable surface attachment of the fluorogenic substrate, we also synthesized the alkyl-substituted coumarin 12 using the same procedure utilized for the preparation of 10.The Mitsunobu-type glycosylation of 12 with the hemiacetal 8 afforded compound 14 in 68% yield (α/β 1.2/1).Subsequent alkaline hydrolysis of the α-adduct generated the desired product 15.

Conclusions
We have successfully developed an alternative method for the synthesis of an L-idofuranose derivative from the cheaper and more common diacetone-α-D-glucose.Although the yield was lower compared to the previous methods we reported, this newly established method involved reaction steps that were less tedious and more convenient to perform.Further, employing the Lido sugar synthesized from this new method for the synthesis of 4-methylcoumarin-7-yl-α-Diduronic acid and its 3-undecyl derivative via Mitsunobu-type glycosylation avoided problems and difficulties previously encountered.Thus, this synthetic route is a promising development in the search for a highly efficient synthesis of the fluorogenic substrate for IDUA.

Experimental Section
General.The reaction organic solvents were purified and dried from a safe purification system.Flash column chromatography was carried out as recommended with Silica Gel 60 (230-400 mesh, E. Merck).TLC was performed on pre-coated glass plates of Silica Gel 60 F254 (0.25 mm, E. Merck); detection was executed by spraying with a solution of Ce(NH4)2(NO3)6, (NH4)6Mo7O24, and H2SO4 in water and subsequent heating on a hot plate. 1 H and 13 C NMR spectra were recorded with Bruker AMX400, AV400 and AV500 MHz instruments.Chemical shifts are measured in ppm and calibrated using the solvent peaks as reference.Mass spectra were obtained with a VG 70-250S mass spectrometer in the FAB mode.Elemental analyses were measured with a Perkin-Elmer 2400 CHN instrument.