Investigation of Action Pattern of a Novel Chondroitin Sulfate/Dermatan Sulfate 4- O -Endosulfatase

Recently, a novel CS/DS 4- O -endosulfatase was identified from a marine bacterium and its catalytic mechanism was investigated further (Wang, W., et.al (2015) J. Biol. Chem. 290, 7823-7832; Wang, S., et.al (2019) Front. Microbiol. 10:1309). In the study herein, we provide new insight about the structural characteristics of substrate which determine the activity of this enzyme. The substrate specificities of the 4- O -endosulfatase were probed by using libraries of structure-defined CS/DS oligosacccharides issued from synthetic and enzymatic sources. We found that this 4- O -endosulfatase GlcUA(3S)β1-3GalNAc(4S) unit) the reducing end rather than the non-reducing end of CS/DS chain remains to be investigated further. The results showed that although 4- O -endosulfatase is an endo-type enzyme, it degraded CS/DS chains gradually from the reducing end to the non-reducing end, and the sulfate groups on nearby GlcUA affect the enzyme hydrolysis of 4- O -sulfate on GalNAc. More complex and diverse oligosaccharides, which will be useful to complete this description of the enzyme action pattern, are currently under preparation. In conclusion, this study provides a more comprehensive insight into the desulfation pattern of the 4- O endosulfatse from Vibrio sp. FC509, which will benefit the application of this enzyme in editing the sulfation patterns of CS/DS oligo-/polysaccharides, as well as their structural and functional studies. 2S, 6S, 2- O -sulfate, 4- O -sulfate and 6- O -sulfate, respectively; N-acetyl-D-galactosamine;

Sulfatases are a large enzyme family and can catalyze hydrolysis of sulfate esters in a wide range of substrates from small steroids to complex carbohydrates on cell surface, such as glycosaminoglycans (GAGs) (22). Increasing evidences show that sulfatases with different desulfation activities on specific sulfate esters of CS/DS chains are suitable and important tools for exploring the roles of CS/DS in various biological processes (23). CS/DS sulfatases are characterized by the desulfation of CS/DS saccharides and can be generally designated based on the desulfated position. For example, GalNAc-4-O-sulfatase and GalNAc-6-O-sulfatase catalyze the hydrolysis of the sulfate ester at the C4 and C6 position of GalNAc residues, respectively, and Δ 4,5 HexUA-2-O-sulfatase hydrolyzes the sulfate ester on C2 position of the unsaturated hexuronate at the non-reducing end of CS/DS or heparin/heparan sulfate oligosaccharides generated by GAGs lyases (24)(25)(26). It is worth noting that most of the identified GAG sulfatases are exosulfatases, which attacks only the sulfate groups of saccharide residues at the end of GAG chains, in particular of short oligosaccharides from the digestion of GAGs by GAG-degrading enzymes (21,26). By contrast, only two CS/DS endosulfatases, which catalyze sulfate hydrolysis within the GAGs chains, have been reported so far. They have been individually identified in Bacteroides thetaiotaomicron (27) and Vibrio sp. FC509 (28), and both are 4-O-endosulfatase. Compared to exosulfatase, endosulfatases are more potent tools for structure-function-studies of CS/DS, but their action patterns remain to be characterized in detail.
In our previous study, the 4-O-endosulfatase from Vibrio sp. FC509 could efficiently remove the 4-Osulfate from various CS/DS polysaccharides, with an efficiency of 17-65%. In the study herein, we investigated whether structural determinants in the CS/DS polysaccharides may block the action of the endosulfatase, explaining why 4-O-sulfate groups were not fully removed in these polysaccharides. For that purpose, we have used a serie of structure-homogenous hexasaccharides (dp6s) either from synthetic source or prepared from the polysaccharide substrates CS-A and CS-E and sequenced thorough an enzymatic method (29). By using these aforementioned hexasaccharides with other reported CS oligosaccharides as substrates, some important action features of the 4-O-endosulfatase were revealed. These delineated characteristic features will be very helpful for the application of this enzyme in the structural and functional studies of CS/DS. described by Bigge et al (35). Free 2-AB was removed by extraction with chloroform. An aliquot (5 pmol) 2-AB-labeled dp6s of CS-A were individually analyzed by anion-exchange HPLC on a YMC-Pack PA-G column eluted with a linear gradient from 16 to 460 mM NaH2PO4 over 60 min at a flow rate of 1.0 ml/min at room temperature, and an aliquot (5 pmol) 2-AB-labeled dp6s of CS-E were individually analyzed by anion-exchange HPLC on a YMC-Pack PA-G column eluted with a linear gradient from 16 to 690 mM NaH2PO4 over 90 min. The samples were monitored using a fluorescence detector with excitation and emission wavelengths of 330 and 420 nm, respectively.

Determination of the desulfated products of 4-O-endosulfatase by mass spectrometry
Desulfation of synthetic CS oligosaccharides by 4-O-endosulfatase was carried out as previously described (28). Briefly, 10 μl of the enzyme preparation was mixed with 5 μL of CS oligosaccharides (100 μM final) in 35 μL of 72 mM phosphate buffer, pH 8.1 (50 mM final), and incubated at 30 °C, for 1, 2, 3, 4, 5, 6 and 24 h. After incubation, the resulting oligosaccharides were isolated and desalted using a C18 + Carbon-SPE TopTip cartridge (Glygen, Columbia, MD, USA) as previously described (36). Each elution fraction was pooled, freeze-dried, and dissolved in 10 μl water for MS analysis. MALDI-time-of-flight (TOF) mass spectrometer (MS) experiments were performed using a PerSeptive Biosystems Voyager-DE Pro STR mass spectrometer (Applied Biosystems/MDS SCIEX, Foster City, CA, USA) equipped with a nitrogen UV laser (337 nm wavelength) pulsed at 20 Hz frequency. The MS was operated in the negative ion reflector mode with an accelerating potential of -20 kV and a grid percentage equal to 70%. Mass spectra were recorded with the laser intensity set just above the ionization threshold (2900-3000 in arbitrary units, on our instrument) with a pulse width of 3 ns. An ion extraction delay was set to 150 and 300 ns for analyze with dp2, dp4 and dp6, respectively. Typically, mass spectra were obtained by accumulation of 100-200 laser shots for each analysis and processed using Data Explorer 4.0 software (Applied Biosystems). The HABA/TMG2 ionic liquid, used as the matrix, was prepared by following a literature procedure (37).
Nano-electrospray (ESI)-MS and MSn of CS oligosaccharides were carried out using a static probe coupled to LTQ-Orbitrap XL from Thermo Scientific (San Jose, CA, USA) and operated in negative ionization mode, with a spray voltage at -0.8 to -1.0 kV. Two microliters of each aqueous sample was continuously infused through a fused silica tip with a 360-μm outer diameter, 20 μm inner diameter with a nominal tip end inner diameter of 10 ± 1.0 μm (Pico-tip, FS360-50-15-CE-20-C10.5, New Objective, Woburn, MA, USA) which provided a consistent signal during 15 min. Applied voltages were -35and -113 V for the ion transfer capillary and the tube lens, respectively. The ion transfer capillary was held at 200°C. Resolution was set to 30,000 (at m/z 400) for all studies, and the m/z ranges were set to 150-2000 in profile mode and in the normal mass range. Spectra were analyzed using the acquisition software XCalibur 2.0.7 (Thermo Scientific, San Jose, CA, USA), without smoothing or background subtracts. During MS/MS scans, collision-induced dissociation (CID) fragmentation occurred in the linear ion-trap analyzer and detection in Orbitrap with centroid mode. For CID fragmentation, an activation Q value of 0.25 and an activation time equal to 30 ms were used. Normalized collision energy (NCE) was set at 25% for MS2/MS experiments and 15-25% for MSn experiments (with n>2). The automatic gain control (AGC) allowed accumulation up to 1.10 6 ions for Fourier-transform MS (FTMS) scans, 2.10 5 ions for FTMSn scans and 1.10 4 ions for ion-trap MS (ITMSn) scans. Maximum injection time was set to 500 ms for both FTMS and FTMSn scans and 100 ms for ITMSn scans. For all scan modes, 1 μs can was acquired. The precursor selection window was 2 Da and 3 Da during MS 2 and MSn (with n>2), respectively. Each ion with z ≥2 present in the full-scan MS and exhibiting a signal intensity of >500 arbitrary units was fragmented by iterative ESI-MSn steps (with n≥2). Fragments resulting from fragmentation steps were annotated according to the nomenclature described by Domon and Costello (38).

Sequencing of CS structure-homogenous dp6s
The sequencing strategy was based on previous paper (29). Firstly, the disaccharide composition of each CS dp6 (5 pmol) was determined by complete digestion with 5 mU commercial CSase ABC (EC 4.2.2.20) for 2h and labeled by 2-AB (35) followed by anion exchange HPLC. Then, the disaccharide at the non-reducing end of each CS dp6 was determined by digestion of 2-AB-labeled dp6 with CSase ABC and relabeling with 2-AB followed by analysis with anion exchange HPLC (28). Briefly, an aliquot (5 pmol) of 2-AB-labeled dp6 were degraded with 5 mU CSase ABC, and then the digest was relabeled with 2-AB. After removing free 2-AB by extraction with chloroform, the resulted sample was analyzed by anion exchange HPLC. Finally, the disaccharide located at the reducing end of dp6 was identified by digestion of 2-AB-labeled dp6 with HCDLase, because this enzyme is a novel exolyase that can remove 2-AB labeled sulfated but not unsulfated disaccharides from the reducing ends of 2-AB-labeled CS oligosaccharides (29). All these preparations were individually analyzed by anion-exchange HPLC on a YMC-Pack PA-G column eluted with a linear gradient from 16 to 460 mM NaH2PO4 over 60 min at a flow rate of 1.0 ml/min at room temperature, and the samples were monitored using a fluorescence detector with excitation and emission wavelengths of 330 and 420 nm, respectively (39). Identification and quantification of the resulting disaccharide units were achieved by comparison with CS-derived authentic unsaturated disaccharide units (40).

Substrate-degrading characteristic analysis of 4-O-endosulfatase
An aliquot (10 pmol) of CS dp6 and dp8 fractions were treated with 20 mU 4-O-endosulfatase in 20 μl 50 mM NaH2PO4-Na2HPO4, pH 8.0, at 30°C for 72 h, respectively (28). Enzyme-treated products were heated in boiling water for 10 min and then cooled in ice-cold water for 10 min. After centrifugation at 15,000×g for 15 min, the supernatants were collected. Half amounts of the samples were further degraded with 5 mU CSase ABC for disaccharide composition determination. Then all samples were labeled by 2-AB and individually analyzed by anion exchange HPLC on a YMC-Pack PA-G column at a flow rate of 1.0 ml/min at room temperature. The resultants of CS-A dp6, CS-E dp6 and dp8 ΔC-A-D-C treated with 4-Oendosulfatase were eluted with a linear gradient from 16 to 460 mM NaH2PO4 over 60 min, from 16 to 690 mM NaH2PO4 over 90 min and from 300 to 800 mM NaH2PO4 over 60 min, respectively. Disaccharide composition analysis of these dp6 and dp8 samples treated with or without 4-O-endosulfatase was carried out by anion exchange HPLC on a YMC-Pack PA-G column eluted with a linear gradient from 16 to 460 mM NaH2PO4 over 60 min. And all the samples were monitored using a fluorescence detector with excitation and emission wavelengths of 330 and 420 nm, respectively. To illustrate the effect of the CS-A length on the activity of the 4-O-sulfatase, size-defined CS-A oligosaccharides (1 μg) with different lengths were treated with 4-O-endosulfatase (1 mU) in 20 μl 50 mM NaH2PO4-Na2HPO4, pH 8.0, at 30°C for 10 min, and then the disaccharide composition of each reaction was analyzed as described above.

Isolation of CS unsaturated dp6s
A commercial preparation (200 mg) of CS-A and CS-E were partially digested with 0.1 U endolyase HCLase, and the resulting oligosaccharides were fractionated by a Superdex TM Peptide 10/300 GL column ( Figure S1). Then dp6s from CS-A ( Figure S2A) and CS-E ( Figure S2B) were subfractionated by HPLC on anion exchange column of YMC-Pack Polyamine and YMC-Pack PA-G, respectively. Thirteen main dp6 fractions were obtained. Purified dp6s were labeled by 2-AB and analyzed by fluorescent HPLC for purity. Each purified fraction gave a single symmetrical peak, confirming its high purity. All thirteen pure dp6 fractions were obtained for further sequencing and the desulfation pattern studies of 4-O-endosulfatase.
To identify the disaccharide units at the reducing ends of remained non-sequenced dp6 fractions, the 2-AB-labeled non-sequenced dp6 samples were further degraded with HCDLase followed by HPLC analysis using a fluorescence detector. As shown in Figure S4, CS-A3 and CS-A5 partially degraded by HCDLase due to the resistant of the middle O unit in the 2-AB-labeled dp6 to the digestion by HCDLase (29), and thus the released A unit at the reducing end and its undigested parental dp6 were appeared together. Consequently, the sequences of CS-A3 and CS-A5 can be inferred as ΔC-O-A, ΔA-O-A, respectively. The disaccharide units from reducing ends of other dp6 were ΔC unit of CS-A10-1, CS-A10-2, CS-E7 and CS-E13 ( Figure S4C, S4D, S4F and S4G), and ΔA unit of CS-A10-3 and CS-E15 ( Figure  S4E and S4H), respectively. Therefore, based on the results above, we can deduce that the structures of CS-A10-1, CS-A10-2, CS-A10-3, CS-E7, CS-E13 and CS-E15 were ΔA-A-C, ΔC-A-C, ΔC-C-A, ΔE-A-C, ΔE-E-C and ΔE-E-A, respectively ( Table 2).

Influence of neighbour disaccharides on the digestion pattern by 4-O-endosulfatase
To investigate the digestion specificity of the 4-O-endosulfatase, thirteen structure-defined CS unsaturated dp6 with different sulfation patterns and a CS dp8 ΔC-A-D-C (31) were individually treated with 4-O-endosulfatase exhaustively. The reaction mixtures were directly analyzed by anion exchange HPLC to detect desulfated oligosaccharides. As shown in Figure 2, all samples except CS-A7 and ΔC-A-D-C were eluted at lower salt concentration than their corresponding parental oligosaccharides, suggesting that some sulfate groups were removed from these oligosaccharides.
To further identify the positions of the sulfate groups hydrolyzed by 4-O-endosulfatase, the enzymetreated oligosaccharides were degraded with CSase ABC and then labeled with 2-AB for disaccharide composition analysis by anion exchange HPLC as described under "Experimental Procedures". As shown in Figure 3 and Table 3, all the A and E units in dp6 oligosaccharides were transformed into the unsulfated disaccharide O units and monosulfated disaccharide C units, respectively. Based on the structures of all the test dp6 oligosaccharides, we can conclude that this 4-O-endosulfatase can completely remove the 4-Osulfate group of A unit or E unit no matter they are at the reducing end, non-reducing end, or internal of dp6, and this desulfation of A unit or E unit in dp6 does not be inhibited by their adjacent O unit or C unit. However, when the dp8 ΔC-A-D-C was treated with 4-O-endosulfatase, the 4-O-sulfate group of A unit could not be removed ( Figure 3 and Table 3), suggesting that the HexUA(2S)1-3GalNAc(6S) unit connected to A unit might inhibit the action of 4-O-endosulfatase.

Effect of substrate size on the activity of 4-O-endosulfatase
Although the optimal substrate CS-A is composed of major component A units and minor component O and C units, only 65% of A units were desulfated by the 4-O-endosulfatase even under a very strong reaction condition (28). We suppose that the enzymatic activity may be affected by the size of substrate. To investigate this hypothesis, the hydrolysis capacity of the endosulfatase on oligosaccharides saturated A of increasing size (dp2), A-A (dp4) and A-A-A (dp6) was determined by MALDI-TOF mass spectrometry at various incubation times. Under limited enzymatic activity the kinetics analysis showed a rapid and total decrease of intact A-A and A-A-A, indicating that the enzyme is able to catalyze the desulfation of these longer substrate molecules in few hours ( Figure S5A). The initial decrease of the totally 4-O-sulfated species was faster with dp6, then dp4 and dp2, suggesting that the first 4-O-sulfate group on the longer substrate is easier to be removed. However, the total desulfation of A-A or A-A-A is much more time-consuming than that of A unit, indicating that chain length of substrates affects the enzymatic activity ( Figure. S5B). Then, the removal percentage of 4-O-sulfates from substrates with different length were analyzed by using a series of unsaturated CS-A oligosaccharides from dp2 to dp12 as substrates digested with 4-O-endosulfatase within a certain time (10 min). As shown in Table 4, the 4-O-desulfation ratio of A units in the prepared oligosaccharides were significantly decreased with the length increase of oligosaccharide substrates, thus explaining why the desulfation of CS-A polysaccharides by this 4-

Effect of saturated CS disaccharides on the action of 4-O-endosulfatase
In previous studies, the specificity and action pattern of the 4-O-endosulfatase were investigated on substrates with limited structural diversity by using common unsaturated CS oligosaccharides produced by CS/DS lyases (23). Here, the action of 4-O-endosulfatase was studied by using as substrate a larger library of synthetic CS based-saturated disaccharides including monosulfated A unit and C unit, disulfated E unit, D unit, K unit and L unit, and trisulfated M unit. MALDI-TOF MS analysis of the enzyme reaction products show that saturated A unit and E unit can be completely converted to O unit (m/z 396.11) and C unit (m/z 498.05), respectively, by the treatment with this endosulfatase (Figure S6), indicating that this enzyme shows the same specific activity of 4-O-desulfation whether the substrates are saturated or unsaturated (23). However, the 4-O-sulfates of K unit (GlcUA(3S)β1-3GalNAc(4S)), L unit (GlcUA(3S) β1-3GalNAc(6S)) and M unit (GlcUA(3S)β1-3GalNAc(4,6S)) could not be removed. It may be due to the presence of the 3-O-sulfate of GlcUA that affects the conformation of the enzyme, or inhibits the binding of substrates to enzyme, thereby resulting the inability of the endosulfatase on these substrates ( Figure S7).

Action pattern of the 4-O-endosulfatase on saturated CS oligosaccharide
In previous studies, we have shown that 4-O-endosulfatase can remove 4-O-sulfate groups starting from the reducing end to the non-reducing end of an unsaturated dp6 ΔA-A-A (41). To further investigate if the unique unsaturated uronic acid produced by the cleavage of CS/DS lyases is a key factor to affect the action direction of the endosulfatase, the desulfated products formed from a synthetic saturated dp6 A-A-A at various times of incubation with this enzyme have been sequenced by multistage mass spectrometry. The MS spectrum showed a major species at m/z 689.5937 after 3 h incubation, which corresponded to a disulfated dp6 product. The serial fragmentations from MS 2 to MS 5 of this disulfated product indicated that the desulfated position was on the reducing end residue GalNAc ( Figure 4A-D). In details, MS 2 of the doubly charged ion at m/z 689.5937 yielded a major C5 2ion at m/z 588.0510 corresponding to a disulfated pentasaccharide. It resulted from a glycosidic rupture releasing the terminal GalNAc unit at the reducing end of the parent dp6, thereby forming the C5 2ion [dp6-GlNAc-2Na] 2-( Figure 4A). Subsequent fragmentation (MS 3 ) of this C5 2ion yielded through glycosidic break a series of the monocharged ions C2 (m/z 498.0518), C3 (m/z 696.0432) and C4 (m/z 1001.0842) ( Figure 4B), corresponding to di-, tri-and dp4 respectively, which located the two sulfate groups on the non-reducing part of the dp6 product. Finally, further dissociation by MS 4 and MS 5 confirmed the location of the sulfate groups on the GalNAc residue within the central (Z3at m/z 300.0375, Figure 4C) and the terminal non-reducing disaccharides (C2at m/z 498.0513, Figure 4C, and Z5 -/ 0,1 X6at m/z 282.0270/312.0374, respectively, Figure 4D) of the disulfated dp6 product. Similar conclusions can be deducted with CS-dp4 analysis ( Figure 5). After 24 h incubation, except a major species at m/z 587.6564 (2-) corresponding to the fully unsulfated dp6, a small peak at m/z 632.6847 was detected on mass spectrum, which corresponded to traces of monosulfated dp6 product. Iterative fragmentations from MS 2 to MS 6 were performed to determine the sulfation site along the dp6. Several relevant serial glycosidic bond cleavages were obtained, which allowed the full sequencing: the C5 ion at m/z 537.0843 and 1075.0843 singly and doubly charged, respectively, ( Figure 6) from MS 2 , the C4 ion at m/z 438.0773 and 899.1452 singly and doubly charged, respectively, ( Figure 6) from MS 3 , the C3 ion at m/z 336.5373 and 674.0839 singly and doubly charged, respectively, ( Figure 6) from MS 4 , and an intense C2ion at m/z 498.0575 from MS 5 , evidenced the presence of the sulfate group on the non-reducing end disaccharide unit ( Figure 6). MS 6 of the last disaccharide fragment C2yielded the following fragments C1 -, Z5and 0,1 X6at m/z 193.0344, m/z 282.0270 and m/z 312.0376, respectively, which unambiguously confirmed that the remaining sulfate to remove is carried by the GalNAc residue at the non-reducing end of the monosulfated dp6 product ( Figure 6). The results showed that, same as unsaturated oligosaccharide, the 4-O-endosulfatase removed 4-O-sulfate groups initially from the reducing end to the non-reducing end of saturated substrates.

Discussion
CS/DS 4-O-endosulfatase from Vibrio sp. FC509, the first CS/DS sulfatase identified from marine bacteria, was characterized for basic enzymatic properties and substrate-degrading specificity in our previous study (28). The results show that the endosulfatase can specifically remove the 4-O-sulfate from 4-O-sulfated GalNAc residues in various CS and DS polysaccharide chains, but the desulfation extent of the 4-O-sulfate varies from 17 to 65%, indicating that in addition to sulfate position some other structural factors have a significant influence on the action of this enzyme. Recently, a structural study has shown that the residues at the active site of the 4-O-endosulfatase adopt a more favorable conformation to interact with longer CS/DS chains compared with the strict exolytic 4-O-sulfatase exoPB4SF (sharing 83% sequence identity with the 4-O-endosulfatase), so that hexasaccharides could easily dock into the catalytic pocket of 4-O-endosulfatase but not in exoPB4SF (41). However, these structural data do not explain why the 4-Oendosulfatase cannot remove all 4-O-sulfate from the polysaccharide. Libraries of enzymatic and synthetic CS/DS oligosaccharides used in the study herein allowed gaining more insight about factors regulating the 4-O-endosulfatase activity.
CS/DS hexasaccharide with internal 4-O-sulfated disaccharide is the minimal size oligosaccharide to demonstrate the endo-activity of the 4-O-endosulfatase. In this study, a series of structure-defined unsaturated hexasaccharides with various sulfation patterns were prepared from the polysaccharide substrates CS-A and CS-E through partial digestion by the endolyase HCLase (35) followed by fractionation using gel filtration and anion-exchange chromatography. Surprisingly, the 4-O-endosulfatase can completely remove the 4-O-sulfate group of A unit or E unit whatever their location in hexasaccharide (reducing end, non-reducing end, or internal), and this desulfation of A unit or E unit by this enzyme is not inhibited by the adjacent O unit or C unit. However, in the case of octasaccharide ΔC-A-D-C in which the adjacent D unit is sulfated on uronic residue, the 4-O-sulfate group of the internal A unit could not be removed by this enzyme, suggesting that the D unit connected to A unit should inhibit the action of 4-Oendosulfatase. In the same way, we found that the 4-O-endosulfatase cannot remove 4-O-sulfate groups from K unit (GlcUA(3S)β1-3GalNAc(4S)) and M unit (GlcUA(3S)β1-3GalNAc(4,6S)), indicating that that 3-O-sulfate on GlcUA residue hinders the hydrolysis of 4-O-sulfate from K or M unit. Taken together, these results show the 4-O-endosulfatase action requires non-sulfated adjacent uronic residues. However, it should be noted desulfation of CS-A from bovine trachea, which usually contains no D unit, reached only 65% of A units, indicating that some other structural factors affect the effective action of this enzyme which need further investigated. As mentioned above, our recent study has shown that the binding capacity of the 4-O-endosulfatase to long substrate sequences, larger than dp2, plays a key role for its endolytic activity. Accordingly, we found that the hydrolysis of the first 4-O-sulfate group on the longer substrate A-A-A is faster than that of equimolar A or A-A. To further investigate the effect of substrate length on the enzymatic activity, a series of size-defined unsaturated oligosaccharides from dp2 to dp12 were prepared from CS-A and used to compare the activity of the 4-O-endosulfatase against each of them. The results clearly show that in a short incubation time (10 min) the removal percentage of 4-O-sulfates significantly decreases with the increase of substrate size. We assume that removal of the first sulfate groups along the substrate sequence leads to a decrease in substrate affinity, which may explain the low activity of the 4-Oendosulfatase against CS-A polysaccharide compared to oligosaccharides.
The 4-O-endosulfatase has initially been found with several GAGs lyases in the genome of a marinederived CS/DS-degrading bacterium Vibrio sp. FC509 (28), thus suggesting that unsaturated CS/DS oligosaccharides produced by GAGs lyases must be its natural substrates. In our previous study, we have shown that this enzyme sequentially remove 4-O-sulfate in an oriented way from the reducing end of unsaturated dp6 ΔA-A-A. To investigate if the unsaturated double bond of ΔA located at the non-reducing end affects the action direction of the enzyme, we analysed by NanoESI-MSn the enzyme desulfation of synthetic saturated hexasaccharide A-A-A, and the results clearly showed that this enzyme also sequentially removes 4-O-sulfates from the reducing end to non-reducing end of A-A-A, indicating that ΔA is not a decisive factor to the action direction of the enzyme. The preference for the terminal disaccharide unit from which the endosulfatase action is initiated, is likely due to the freedom or flexibility of terminal area of a polysaccharide is higher than internal area (42). This situation occurs due to the freedom or flexibility of terminal area of a polysaccharide is higher than internal area (42). However, the reason why this enzyme acts from the reducing end rather than the non-reducing end of CS/DS chain remains to be investigated further.
The results showed that although 4-O-endosulfatase is an endo-type enzyme, it degraded CS/DS chains gradually from the reducing end to the non-reducing end, and the sulfate groups on nearby GlcUA affect the enzyme hydrolysis of 4-O-sulfate on GalNAc. More complex and diverse oligosaccharides, which will be useful to complete this description of the enzyme action pattern, are currently under preparation. In conclusion, this study provides a more comprehensive insight into the desulfation pattern of the 4-Oendosulfatse from Vibrio sp. FC509, which will benefit the application of this enzyme in editing the sulfation patterns of CS/DS oligo-/polysaccharides, as well as their structural and functional studies.