Evidence for a multivalent effect in inhibition of sulfatases involved in lysosomal storage disorders (LSDs)

Abstract

Two newly synthesized nonavalent polyhydroxylated pyrrolidine iminosugars are the first examples of multivalent inhibitors of GALNS and IDS lysosomal enzymes, whose deficiency leads to Morquio A syndrome and Hunter disease, respectively, and pave the way to a pharmacological chaperone or stabilizing enzyme therapy for these LSDs.

---

Lysosomal Storage Disorders (LSDs) are a group of about 50 rare metabolic diseases caused by genetic mutations resulting in a total or partial deficiency of the lysosomal proteins/enzymes, which leads to abnormal accumulation of not metabolized substrates inside the lysosomes. The resulting pathologic effects are multisystemic heterogeneous clinical manifestations, including neurological symptoms, which considerably reduce patients' life quality and expectation.¹ To date, there is no cure for the majority of LSDs, and the available treatments are only symptomatic and have several drawbacks.² For instance Enzyme Replacement Therapy (ERT), currently the most successful one yet limited to a few diseases, requires periodic infusions of the exogenous wild type enzyme, has a high cost and needs frequent hospitalization. In addition, patients can develop an anti-ERT response, and the therapy is not effective for LSDs with central nervous system (CNS) impairment.³ A promising new therapeutic approach to LSDs is based on Pharmacological Chaperones (PCs), that include Active Site Specific Chaperones (ASSCs).⁴ ASSCs are compounds which act as competitive inhibitors of the enzyme. When they are used in sub-inhibitory amounts (with consequent minimal side effects) they can correct the folding and/or stabilize the catalytic activity and half-life of the enzymes compromised by conformational changes. PC-ASSCs have several advantages, such as oral administration and reduced costs; moreover, thanks to their ability to recover endogenous enzyme activity, they are the only possible treatment to target the functional origin of LSDs at present.

Among LSDs, mucopolysaccharidoses Morquio A syndrome⁵ and Hunter disease⁶ are caused by the deficiency of N-acetylgalactosamine-6-sulfatase (GALNS) and iduronate-2-sulfatase (IDS) enzymes, respectively, resulting in accumulation of glycosaminoglycans (GAGs) in the lysosomes. This generates multisystemic symptoms ranging from skeletal and connective tissue abnormalities to profound cognitive impairment and developmental regression for Hunter disease.

Morquio A has an estimated incidence ranging from one in 76 000 births in Northern Ireland to one in 640 000 births in Western Australia.^5c ERT (elosulfase alfa, VIMIZIM®, BioMarin), has recently been approved by the FDA (Food and Drug Administration) and EMA (European Medicines Agency) as a treatment.⁷ Hunter syndrome has an incidence (probably underestimated) of 1.3 in 100 000 male newborns,⁸ and ERT (idursulfase, Elaprase®, Shire Human Genetic Therapies, Inc., Cambridge, MA) available since 2006, is now available in more than 40 countries but is not effective against the most severe form that involves the CNS.⁹ There is therefore a need for the identification of compounds which can act as PCs for these orphan diseases. In addition, it has been shown that exogenous enzymes infused in ERT suffer of a low (conformational) stability and, in the case of Fabry disease, it was recently demonstrated that a combined ERT therapy with use of PC is beneficial to the enzyme stability in time and allows to reduce the needed enzyme infusions.¹⁰

The most promising PCs for LSDs consist of iminosugars, nitrogenated glycomimetics with nitrogen replacing the endocyclic oxygen of carbohydrates, widely known as potent glycosidases inhibitors.¹¹ Recently, their inhibitory activity towards some glycosidases has been successfully enhanced in some instances by assembling them in multivalent architectures containing several bioactive units linked together to a common scaffold.¹² It is clear from the first reported example of a multivalent DNJ (deoxynojirimycin) inhibitor¹³ that the multivalency approach can turn a weak and poorly selective inhibitor into a more potent and selective one, although a rationale for this behavior is still not available. The majority of studies have focused on the interaction of DNJ and DMJ (deoxymannojirimycin) derivatives with commercially available jack-bean α-mannosidase, an enzyme for which a dimeric structure has been postulated,¹⁴ suggesting that the multimeric nature of the enzyme plays a key role.¹² However, a rationalization of the inhibitor valency role is not trivial, since the multivalent binding modes are of different type and far to be unequivocally elucidated. Indeed, when dealing with enzymes possessing a multimeric form, not only statistical rebinding, chelate and subsite binding effects are involved, but also clustering effects or formation of cross-linked networks are possible.^12a,c No correlation of the valency number with the degree of the enzyme aggregation has been established so far, albeit it is apparent that multimeric glycosidases are better responding to multivalent inhibitors.

Taking into consideration the dimeric nature of GALNS as revealed by its X-ray structure,¹⁵ we envisaged that this lysosomal enzyme might also accept multivalent inhibitors. It is worth noting that only a few examples regarding the action of multivalent inhibitors on therapeutically relevant enzymes have been described and none on sulfatases.¹⁶ In particular, the large and less specific substrate-binding pockets found in lysosomal sulfatases are able in principle to accept a wide range of substrates, but also makes the design of inhibitors more complex. The interaction of iminosugar based compounds with these enzymes has never been investigated to date.

We therefore decided to evaluate the inhibitory activity of a series of tri-, tetra- and nonavalent compounds already available in our laboratories based on piperidine or pyrrolizidine iminosugars towards GALNS in comparison with appropriate monomeric counterparts.¹⁷ However, inhibitory activities higher than 60% were never achieved,^17b and only moving to pyrrolidine iminosugars much more interesting results (around 90% inhibition) were obtained and are reported herein. We also extended the investigation to the related sulfatase iduronate-2-sulfatase (IDS). The naturally occurring 1,4-dideoxy-1,4-imino-D-arabinitol (DAB-1, 7, Scheme 1) has never been employed to build multivalent architectures, apart from our earlier attempts on calixarene derivatives where difficulties were encountered in the final deprotection step.¹⁸ DAB-1 is known to be an active compound with a broad inhibitory spectrum towards mammalian glycosidases.^11,19 In our hands, N-alkyated derivatives of 7 showed nanomolar inhibition of GCase (glucocerebrosidase), the deficient enzyme in Gaucher disease.^20,21 Conversely, the synthesis of new pyrrolidine iminosugars as potential PCs⁴ has been poorly investigated to date either in monovalent or multivalent presentation.²² Here we report, as a proof of this concept, that the simultaneous presentation of new pyrrolidine type iminosugars affords micromolar inhibitors of GALNS and IDS enzymes, thus opening the way to the identification of PCs for the related diseases. The syntheses of the nonavalent pyrrolidine iminosugars 10 and 14 are reported in Scheme 1. Starting from D-arabinose derived nitrone 1a,²³ a two-step reduction with NaBH₄ followed by Zn in AcOH afforded the amine 2, recently described using a different strategy,²⁴ in 98% overall yield N-alkylation using 1-azido-6-bromohexane 3 (ref. 25) in THF in basic conditions under microwave (MW) irradiation and afforded the azide 4 in 88% yield. Copper catalyzed cycloaddition with 3-butyn-1-ol in THF/H₂O with CuSO₄ and sodium ascorbate under MW yielded 5 (93%), that upon catalytic hydrogenation with H₂ and Pd/C in acidic MeOH led to the monovalent reference compound 6 in 74% yield.

Scheme 1 (a) 1a, NaBH₄, EtOH, rt, 16 h; (b) Zn, AcOH/H₂O, rt, 1.5 h, 98% over two steps; (c) TEA, THF, MW 150 °C, 4 h, 88%; (d) 3-butyn-1-ol (1.1 equiv.), CuSO₄ (30 mol%), sodium ascorbate (60 mol%), THF/H₂O 2 : 1, MW 80 °C, 45 min, 93%; (e) H₂, Pd/C, MeOH, HCl conc., rt, 2 d, 74%; (f) K₂CO₃, CH₃CN/H₂O 5 : 1, MW 120 °C, 2 h, 92% for 8, 66% for 12; (g) 8 or 12 (9.3 equiv.), CuSO₄ (30 mol%), sodium ascorbate (60 mol%), THF/H₂O 2 : 1, MW 80 °C, 45 min, 81% for 10, 71% for 14; (h) 3-butyn-1-ol (1.2 equiv.), CuSO₄ (30 mol%), sodium ascorbate (60 mol%), THF/H₂O 2 : 1, MW 80 °C, 45 min, 89%.

To reduce the overall number of steps and to avoid a debenzylation reaction on the nonavalent derivatives, that proved to be difficult and not reproducible in our previous studies,¹⁷ deprotected DAB-1 (7)²⁶ was reacted with 3 in CH₃CN/H₂O in the presence of K₂CO₃ under MW irradiation at 120 °C. These conditions allowed to achieve a selective N-alkylation leading to the one-step synthesis of the deprotected azide 8 in 92% yield after careful optimization. Indeed, previous reactions with different solvents and bases (e.g., CH₃CN, K₂CO₃, MW, 150 °C, 3 h, or DMF, TBAI, K₂CO₃, rt, 3 d) gave much lower yields and scarcely reproducible results.

Copper catalyzed cycloaddition with nonadiyne scaffold 9 (ref. 27) using 9.3 equiv. of 8 afforded the nonavalent compound 10 (81%), that was purified upon FCC and size exclusion chromatography. We also took into account a different configuration of the bioactive pyrrolidine by preparing a nonavalent derivative of 1,4-dideoxy-1,4-imino-D-ribitol (11), in turn obtained from D-ribose derived nitrone 1b as reported.²⁰ N-Alkylation with 3 employing the same conditions as described for compound 8 afforded 12 in 66% yield. Cycloaddition with 3-butyn-1-ol in THF/H₂O with CuSO₄ and sodium ascorbate under MW gave the monovalent reference compound 13 in 89% yield, while the same reaction with the nonavalent scaffold 9 (9.3 equiv. of 12) afforded the ribose configured nonavalent compound 14 in 71% yield, purified upon FCC and size exclusion chromatography (see ESI† for experimental details).

The inhibitory activities of nonavalent compounds 10 and 14 (together with their reference monovalent counterparts 6 and 13), towards GALNS and IDS enzymes were evaluated in extracts from a pool of human leukocytes isolated from healthy donors (1 mM concentration of inhibitor, 37 °C and optimal pH conditions, see ESI†).^28,29 The obtained results (Table 1) show a remarkable inhibitory activity of 10 and 14 for both GALNS and IDS enzymes, with IC₅₀ in the micromolar range, while the monovalent reference compounds give negligible (if any) inhibition. An opposite trend in terms of IC₅₀ is observed for the two nonavalent compounds versus the two enzymes (compound 10 is a better GALNS inhibitor, while compound 14 inhibits more efficiently IDS, see Table 1) suggesting that a change in the absolute configuration of just one stereogenic centre in the pyrrolidine bioactive moiety does have a role in affecting their affinity. Our results clearly indicate that the simultaneous presentation of multiple binding units generates more potent inhibitors, with relative potency (rp, calculated as the IC₅₀ ratio of the reference compound vs. that of the corresponding multivalent compound) ranging from 23 to 177. Moreover, also a positive multivalent effect occurs in all cases (rp/n > 1, where n is the compound valency) suggesting that the inhibitory activity enhancement is not simply due to a concentration effect. In particular, the best multivalent effect is observed with compound 14 towards IDS, for which a rp/n of 19.7 was calculated. It is also worth noting that, at least to our knowledge, compounds 10 and 14 represent the first iminosugar-based inhibitors of GALNS and IDS enzymes identified to date. Compounds 10 and 14 gave the best inhibitory activities with respect to more flexible tri- and tetravalent compounds bearing the same pyrrolidine units (detailed results will be reported in due course). Therefore, in our hands the combination of the quite rigid aromatic scaffold with the presence of the nine bioactive units results in the best multivalent effect observed with the reported sulfatases. The synthesis and biological evaluation of other related multivalent pyrrolidines based on different scaffolds are currently ongoing.

Table 1 Inhibition of human GALNS and IDS in leukocytes extract percentage of inhibition at 1 mM, IC₅₀ (in parenthesis, μM)

Compound	GALNS	rp^a	rp/n^b	IDS	rp^a	rp/n^b
a Relative potency, calculated as the IC50 ratio of the reference compound vs. that of the corresponding multivalent compound.b Relative potency per binding unit, where n is the compound valency (9 in our case).c These data have been obtained by a fitting of the measured values, since at 5 mM concentration 100% inhibition was not reached.
6	15% (3900 ± 200)^c	1		7% (3200 ± 160)^c	1
10	94% (47 ± 5)	83	9.2	96% (140 ± 5)	23	2.5
13	6% (5000 ± 200)^c	1		11% (5500 ± 300)^c	1
14	85% (85 ± 8)	59	6.5	94% (31 ± 2)	177	19.7

---

Conclusions

In conclusion, we report how the multimerization of pyrrolidine iminosugar DAB-1 and one of its epimers (both found in nature as monomers) affords multivalent ligands able to impart strong inhibition of GALNS and IDS enzymes, while the monovalent reference compounds are only negligible inhibitors. We therefore demonstrated that a good multivalent effect can be achieved with pyrrolidine-based clusters towards therapeutically relevant enzymes deficient in LSDs for which no PCs have been yet identified. The reported findings^30,31 could open the way to the development of the first pharmacological chaperones (PCs) for Morquio A syndrome and Hunter disease to be used alone or as stabilizers of the recombinant wild type enzymes in a combined PC/ERT therapy.^10,32 More detailed kinetic studies are ongoing and will be reported in due course.

Acknowledgements

This work was supported by the Italian Ministry of Health (grant Ricerca Finalizzata RF-2011-02347694 to AM and FC) and by MIUR-Italy (PRIN 2010–2011, 2010L9SH3K 006 to AG). Ente Cassa di Risparmio di Firenze is acknowledged for a fellowship to CM (grant No 2013/0366) and to GD (grant No 2014/0303). AM and SC also thank AIMPS (Associazione Italiana Mucopolisaccaridosi e malattie affini), AMMeC (Associazione Malattie Metaboliche e Congenite, Italia) and Fondazione Meyer ONLUS, Firenze for their technical support in computer assistance. CP also thanks the European Research Council (ERC Grant agreement 291349).

Notes and references

1. (a) A. Mehta and B. Winchester, Lysosomal Storage Disorders: a Practical Guide, Wiley-Blackwell, 2013; (b) F. M. Platt, Nature, 2014, 510, 68.
2. S. Ortolano, I. Vieitez, C. Navarro and C. Spuch, Recent Pat. Endocr. Metab. Immune Drug Discov., 2014, 8, 9.
3. M. Rohrbach and J. Clarke, Drugs, 2007, 67, 2697.
4. (a) R. E. Boyd, G. Lee, P. Rybczynski, E. R. Benjamin, R. Khanna, B. A. Wustman and K. J. Valenzano, J. Med. Chem., 2013, 56, 2705; (b) E. M. Sánchez-Fernández, J. M. García Fernández and C. Ortiz Mellet, Chem. Commun., 2016, 52, 5497–5515.
5. (a) L. Morquio, Arch. Med. Enf., 1929, 32, 129; (b) M. F. Algahim and G. Hossein Almassi, Ther. Clin. Risk Manage., 2013, 9, 45; (c) G. A. Solanki, K. W. Martin, M. C. Theroux, C. Lampe, K. K. White, R. Shediac, C. G. Lampe, M. Beck, W. G. Mackenzie, C. J. Hendriksz and P. R. Harmatz, J. Inherited Metab. Dis., 2013, 36, 339.
6. B. K. Burton and R. Giugliani, Eur. J. Pediatr., 2012, 171, 631.
7. M. Sanford and J. H. Lo, Drugs, 2014, 74, 713.
8. M. Beck, F. A. Wijburg and A. Gal, Eur. J. Hum. Genet., 2012, 20 DOI:10.1038/ejhg.2011.143.
9. (a) M. Scarpa, Expert Opin. Orphan Drugs, 2013, 1, 89; (b) R. Tomanin, A. Zanetti, F. D'Avanzo, A. Rampazzo, N. Gasparotto, R. Parini, A. Pascarella, D. Concolino, E. Procopio, A. Fiumara, A. Borgo, A. C. Frigo and M. Scarpa, Orph. J. Rare Dis., 2014, 9, 129.
10. K. Valenzano, Abstracts of Papers, 251^st ACS National Meeting & Exposition, USA, San Diego, CA, March 13-17, 2016.
11. P. Compain and O. R. Martin, Iminosugars: from Synthesis to Therapeutic Applications, Wiley VCH, New York, 2007.
12. For reviews, see: (a) P. Compain and A. Bodlenner, ChemBioChem, 2014, 15, 1239; (b) A. Marra, P. Dumy, R. Zelli and J.-F. Longevial, New J. Chem., 2015, 39, 5050; (c) S. G. Gouin, Chem.–Eur. J., 2014, 20, 11616; (d) N. Kanfar, E. Bartolami, R. Zelli, A. Marra, J.-Y. Winum, S. Ulrich and P. Dumy, Org. Biomol. Chem., 2015, 13, 9894.
13. J. Diot, M. I. García-Moreno, S. G. Gouin, C. Ortiz Mellet, K. Haupt and J. Kovensky, Org. Biomol. Chem., 2009, 7, 357.
14. A. Kumara and S. M. Gaikwad, Int. J. Biol. Macromol., 2011, 49, 1066.
15. Y. Rivera Colón, E. K. Schutsky, A. K. Zita and S. C. Garman, J. Mol. Biol., 2012, 423, 736.
16. (a) C. Decroocq, D. Rodríguez-Lucena, K. Ikeda, N. Asano and P. Compain, ChemBioChem, 2012, 13, 661; (b) P. Compain, C. Decroocq, A. Joosten, J. de Sousa, D. Rodríguez-Lucena, T. D. Butters, J. Bertrand, R. Clément, C. Boinot, F. Becq and C. Norez, ChemBioChem, 2013, 14, 2050; (c) A. Joosten, C. Decroocq, J. de Sousa, J. P. Schneider, E. Etamé, A. Bodlenner, T. D. Butters and P. Compain, ChemBioChem, 2014, 15, 309.
17. (a) G. D'Adamio, C. Parmeggiani, A. Goti, A. J. Moreno-Vargas, E. Moreno-Clavijo, I. Robina and F. Cardona, Org. Biomol. Chem., 2014, 12, 6250; (b) C. Matassini, S. Mirabella, A. Goti, I. Robina, A. J. Moreno-Vargas and F. Cardona, Beilstein J. Org. Chem., 2015, 11, 2631.
18. F. Cardona, G. Isoldi, F. Sansone, A. Casnati and A. Goti, J. Org. Chem., 2012, 77, 6980.
19. N. Asano, K. Oseki, H. Kizu and K. Matsui, J. Med. Chem., 1994, 37, 3701.
20. C. Parmeggiani, S. Catarzi, C. Matassini, G. D'Adamio, A. Morrone, A. Goti, P. Paoli and F. Cardona, ChemBioChem, 2015, 16, 2054.
21. For a similar study on pyrrolidine based iminosugars, see: A. Kato, I. Nakagome, K. Sato, A. Yamamoto, I. Adachi, R. J. Nash, G. W. J. Fleet, Y. Natori, Y. Watanabe, T. Imahori, Y. Yoshimura, H. Takahata and S. Hirono, Org. Biomol. Chem., 2016, 14, 1039.
22. For recent reports of divalent and trivalent pyrrolidine derivatives inhibitors of α-L-fucosidase see: (a) E. Moreno-Clavijo, A. T. Carmona, A. J. Moreno-Vargas, L. Molina, D. W. Wright, G. J. Davies and I. Robina, Eur. J. Org. Chem., 2013, 32, 7328; (b) A. Hottin, S. Carrión-Jiménez, E. Moreno-Clavijo, A. J. Moreno-Vargas, A. T. Carmona, I. Robina and J.-B. Behr, Org. Biomol. Chem., 2016, 14, 3212.
23. (a) F. Cardona, E. Faggi, F. Liguori, M. Cacciarini and A. Goti, Tetrahedron Lett., 2003, 44, 2315; (b) A. T. Carmona, R. H. Wightman, I. Robina and P. Vogel, Helv. Chim. Acta, 2003, 86, 3066; (c) D. Martella, G. D'Adamio, C. Parmeggiani, F. Cardona, E. Moreno-Clavijo, I. Robina and A. Goti, Eur. J. Org. Chem., 2016, 1588.
24. W.-C. Cheng, C.-Y. Wenga, W.-Y. Yun, S.-Y. Chang, Y.-C. Lin, F.-J. Tsai, F.-Y. Huang and Y.-R. Chen, Bioorg. Med. Chem., 2013, 21, 5021.
25. F. Coutrot and E. Busseron, Chem.–Eur. J., 2009, 15, 5186.
26. P. Merino, I. Delso, T. Tejero, F. Cardona, M. Marradi, E. Faggi, C. Parmeggiani and A. Goti, Eur. J. Org. Chem., 2008, 2929.
27. Y. M. Chabre, D. Giguère, B. Blanchard, J. Rodrigue, S. Rocheleau, M. Neault, S. Rauthu, A. Papadopoulos, A. A. Arnold, A. Imberty and R. Roy, Chem.–Eur. J., 2011, 17, 6545.
28. O. P. Van Diggelen, H. Zhao, W. J. Kleijer, H. C. Janse, B. J. H. M. Poorthuis, J. Van Pelt, J. P. Kamerling and H. Galjaard, Clin. Chim. Acta, 1990, 187, 131.
29. Y. V. Voznyi, J. L. Keulemans and O. P. van Diggelen, J. Inherited Metab. Dis., 2001, 24, 675.
30. The following patent has been deposited: F. Cardona, C. Parmeggiani, C. Matassini, A. Goti, A. Morrone, S. Catarzi and G. D'Adamio, Italian Patent no. 102016000013155, Università di Firenze, AOU Meyer, 2016.
31. Elaboration of the biological data reported in the present manuscript was facilitated by the contribution of P. Paoli.
32. (a) J.-S. Shen, N. J. Edwards, Y. B. Hong and G. J. Murray, Biochem. Biophys. Res. Commun., 2008, 369, 1071; (b) C. Porto, M. Cardone, F. Fontana, B. Rossi, M. R. Tuzzi, A. Tarallo, M. V. Barone, G. Andria and G. Parenti, Mol. Ther., 2009, 17, 964; (c) E. R. Benjamin, R. Khanna, A. Schilling, J. J. Flannagan, L. J. Pellegrino, N. Brignol, Y. Lun, D. Guillen, B. E. Ranes, M. Frascella, R. Soska, J. Feng, L. Dungan, B. Young, D. J. Lockhart and K. J. Valenzano, Mol. Ther., 2012, 20, 717.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis, characterization and ¹H and ¹³C NMR spectra of new compounds, biochemical characterization on human GALNS and IDS of multivalent compounds. See DOI: 10.1039/c6ra15806d

---