Online Reverse Phase-High-Performance Liquid Chromatography

Jul 22, 2011 - High-resolution separation of homogeneous chitooligomers series from 2-mers to 7-mers by ion-exchange chromatography. Kecheng Li , Song...
0 downloads 0 Views 1MB Size
ARTICLE pubs.acs.org/ac

Online Reverse Phase-High-Performance Liquid Chromatography-Fluorescence Detection-Electrospray Ionization-Mass Spectrometry Separation and Characterization of Heparan Sulfate, Heparin, and Low-Molecular Weight-Heparin Disaccharides Derivatized with 2-Aminoacridone Fabio Galeotti and Nicola Volpi* Department of Biology, University of Modena and Reggio Emilia, Modena, Italy

bS Supporting Information ABSTRACT: A high-resolution online reverse-phase-highperformance liquid chromatography (RP-HPLC)-fluorescence detector (Fd)-electrospray ionization-mass spectrometry (ESIMS) separation and structural characterization of disaccharides prepared from heparin (Hep), heparan sulfate (HS), and various low-molecular-weight (LMW)-Hep using heparin lyases and derivatization with 2-aminoacridone (AMAC) are described. A total of 12 commercially available Hep/HS-derived unsaturated disaccharides were separated and unambiguously identified on the basis of their retention times and mass spectra. The constituent disaccharides of various samples, including unfractionated Hep/HS, fast-moving and slow-moving Hep components, and several marketed products, were characterized. Furthermore, for the first time, the saturated trisulfated disaccharide belonging to the nonreducing end of Heps was detected as being approximately 2% in unfractionated samples and ∼15 21% in LMW-Heps prepared by nitrous acid depolymerization. No desalting of the commercial products prior to enzymatic digestion or prepurification steps to eliminate any excess of AMAC reagent or interference from proteins, peptides, and other sample impurities before RP-HPLC-Fd-ESI-MS injection were necessary. This method has applicability for the rapid differentiation of pharmaceutical Heps and LMW-Heps prepared by means of different depolymerization processes and for compositional analysis of small amounts of samples derived from biological sources by using the highly sensitive fluorescence detector.

H

eparin (Hep) and heparan sulfate (HS) are structurally related linear, highly sulfated, long chain natural polysaccharides belonging to the family of glycosaminoglycans (GAGs)1 having many critical roles in pathophysiological processes.2 4 They consist of alternating 1,4-linked hexuronic acid (Hex), either D-glucuronic acid (GlcA) or L-iduronic acid (IdoA) both of which can be modified with 2-O-sulfo groups, and D-glucosamine (GlcN) residues with molecular weights ranging from 5 to 70 kDa.4,5 The GlcN residue can be modified with N-acetyl (GlcNAc) or N-sulfo (GlcNS) groups and can be substituted with 3- and/or 6-O-sulfo groups.4,5 HS possesses a more highly variable structure than Hep with less sulfo group substitution and is richer in GlcA and GlcNAc residues.4,5 Furthermore, Hep/HS sequence microheterogeneity also depends on the species, individual organism, organ, tissue, cell type, environment conditions, and developmental stage6 producing a large number of structural variations at the disaccharide level.4 9 r 2011 American Chemical Society

Hep possesses anticoagulant and antithrombotic properties used clinically over the last half-century as an anticoagulant drug.1,9 However, Hep presents several undesirable side effects including dangerous hemorrhagic complications.10,11 Thus, low-molecular-weight (LMW)-Heps (average 3 8 kDa) have been introduced as Hep substitutes with reduced side effects (a higher safety/efficacy ratio), a more predictable pharmacological action, sustained antithrombotic activity, and improved bioavailability.12,13 LMW-Hep is prepared, in low yield, by size fractionation of Hep, whereas it is produced in higher yield by the partial, controlled either chemical or enzymatic depolymerization of Hep by treatment with (1) nitrous acid, (2) heparinase, (3) by hydrolytic cleavage with hydrogen peroxide, or (4) by Received: June 8, 2011 Accepted: July 22, 2011 Published: July 22, 2011 6770

dx.doi.org/10.1021/ac201426e | Anal. Chem. 2011, 83, 6770–6777

Analytical Chemistry β-elimination.8,14 Because of the biological and pharmacological significance of Hep and derivatives and HS, there is a need for continued development of rapid and sensitive analytical techniques for their characterization at the level of disaccharide composition as well as for the structural elucidation of larger GAG sequences important for pharmaceutical applications and protein binding. Because of their high negative charge density and polydispersity along with their sequence complexity and heterogeneity, the structural characterization of Hep and HS polymers poses significant challenges for analytical chemists. Exhaustive enzymatic digestion with heparinases can reduce these GAGs to their disaccharide building blocks,15 which are easily separated and quantified by chromatography, electrophoresis, and mass spectrometry (MS). Fragmentation of GAGs to disaccharides is generally achieved using widely available bacterial lyases that introduce a C4 C5 double bond into the nonreducing terminal Hex residue of a released (oligo)disaccharide having a characteristic UV absorption at 232 nm.15 Nevertheless, unsaturated disaccharides are derivatized with fluorescent molecules to significantly increase detection sensitivity and specificity of highperformance liquid chromatography (HPLC)16 18 or other separation techniques. Furthermore, derivatization may also modify the disaccharide properties to enable or improve resolution. Strong anion exchange (SAX)-HPLC,19,20 largely used for the analytical separation of Hep/HS disaccharides, is based on their detection at 232 nm. However, it is unable to detect saturated disaccharides with high sensitivity, along with its difficulty to interface to MS due to the high ionic strength of the mobile phase. The development of reversed-phase ion-pairing (RPIP)HPLC or ultraperformance liquid chromatography (UPLC) separations of disaccharides with online MS detection is a promising technique for the analysis of related GAGs with minimal sample preparation.21 25 However, although several RPIP-LC MS techniques have been successfully used in Hep/HS analysis,21 25 there are some challenging analytical problems remaining for real biological samples, including the overlapping of disaccharide species and the broadening of peaks due to the deleterious effects from the proteins, salts, and other impurities contained in these samples. 2-Aminoacridone (AMAC) is a well-known fluorescent hydrophobic molecule that has been successfully used for the derivatization and separation of unsaturated (oligo)disaccharides.16 18 RP-HPLC offers a common sensitive chromatographic separation of AMAC-disaccharides sometimes providing a better resolution than other analytical techniques.16,17,26 However, UV or fluorescence detection do not provide specific structural information. This is a key point due to the highly complex nature of the Hep/HS disaccharide mixtures obtained by enzymatic digestion, in particular when novel macromolecules are purified from new sources. Furthermore, these disaccharide analyses require the use of standards and further separation runs to calibrate the system. As a consequence, for a specific determination of complex mixtures of Hep/HS (oligo)disaccharides from digests, analytical separation is usually required with evaluation by means of MS. In a previous study,26 we developed an online RP-HPLCelectrospray ionization-mass spectrometry (ESI-MS) separation and structural characterization of hyaluronic acid/chondroitin sulfate/dermatan sulfate disaccharides derivatized with AMAC providing a high-resolution system also applicable by using online a further fluorimetric detector (Fd) in addition to ESI-MS. The method described herein was used to determine the constituent

ARTICLE

disaccharides of various purified Hep/HS samples27,28 and several marketed Hep and LMW-Hep products. A total of 12 commercially available Hep/HS-derived unsaturated disaccharides were separated and quantified (currently, there are no Hep lyases capable of preparing a 3-O-sulfo group containing disaccharides) along with a saturated trisulfated disaccharide previously unreported in the HPLC(UPLC)-ESI-MS disaccharide analysis of these products.

’ EXPERIMENTAL SECTION Materials and Reagents. Hep with a molecular mass of ∼13.1 kDa and containing about 30 40% slow-moving (SM)Hep and 60 70% fast-moving (FM)-Hep moieties was purified from beef intestinal mucosa.27 30 The SM-Hep and FM-Hep components of Hep were purified as their barium salts at different temperatures, as previously reported.27 30 The first species has a mass of ∼14.9 kDa, while the second Hep component possesses a molecular mass of ∼7.9 kDa. HS with a mass of ∼14.0 kDa was prepared from beef spleen.27,28 Commercial samples were obtained from a variety of commercial suppliers and were from Calciparin, 12 500 UI/0.5 mL by Italfarmaco, from Emoklar, 12 500 UI/0.5 mL by IBN Savio srl, and from Ecafast, 12 500 UI/0.5 mL by Crinos. Idracemi heparin eyewash was from Farmigea, 300 000 UI/100 mL. LMW-Heps were dalteparin from Fragmin, 2500 UI/0.2 mL by Pfizer, reviparin from Clivarin, 1750 UI/0.25 mL by Schwarz Pharma, and enoxaparin from Clexane, 4000 UI/0.4 mL by Sanofi Aventis. A total of 12 unsaturated Hep/HS disaccharide standards (see Table S-1 in the Supporting Information) were obtained from Iduron Co (Manchester, U.K.). Heparin lyase I, heparinase, from Flavobacterium heparinum (EC 4.2.2.7), specific activity of 1.5 units/mg protein, and heparin lyase III, heparitinase, from Flavobacterium heparinum (EC 4.2.2.8), specific activity of 1.5 units/mg protein, were from Seikagaku Corporation (Tokyo City, Japan). 2-Aminoacridone (AMAC, >98%), glacial acetic acid, dimethylsulfoxide (DMSO, 99.9%), sodium cyanoborohydride (95%), methanol MS-grade, and all other reagents, of the purest grade available, were from Sigma-Aldrich. Enzymatic Treatment of Hep/HS Samples. Enzymatic digestions were carried out on 100 μg of Hep samples using 1 IU each of heparinase I (5 μL) and heparinase III (5 μL). The content of Hep/LMW-Hep in commercial samples was determined by uronic acid assay according to Cesaretti et al.31 A single international unit (IU) of enzyme is capable of producing 1 μmole of unsaturated uronic acid per minute at 37 °C, pH 7.0. The total reaction volume of 200 μL consisted of 10 mM sodium acetate (95 μL) and 10 mM calcium acetate (95 μL). The reaction solution was incubated for 8 h at 37 °C in a water bath. The reactions were stopped by boiling for 1 min. The extent of the enzymatic digestion was verified by typical SAX-HPLC19,20 separation of unsaturated disaccharides and larger oligosaccharides with UV detection. A percentage of unsaturated disaccharide production (w/w %) greater than ∼90% was generally determined for all Hep/HS samples. Derivatization of Unsaturated (Saturated) Disaccharides with AMAC. Derivatization of Δ-disaccharide standards or produced from the enzymatic treatment of Heps/HS with AMAC was performed as previously described by Volpi26 for chondroitin sulfate samples and hyaluronic acid with minor modifications. Treated heparinases samples were lyophilized 6771

dx.doi.org/10.1021/ac201426e |Anal. Chem. 2011, 83, 6770–6777

Analytical Chemistry

ARTICLE

and reconstituted with 5 μL of a 0.1 M AMAC solution in glacial acetic acid DMSO (3:17, v/v) and 5 μL of a freshly prepared solution of 1 M sodium cyanoborohydride in water. Then, the mixtures were centrifuged in a microfuge at 11 000g for 3 min. Derivatization was performed by incubating at 45 °C for 4 h. Finally, 15 μL of 50% v/v DMSO were added to the samples and aliquots were taken for RP-HPLC-Fd-ESI-MS analysis. RP-HPLC-Fd-ESI-MS. HPLC separation was performed on a 5 μm Discovery-C18 column (250 mm  4.6 mm) from SigmaAldrich. Eluent A was 60 mM ammonium acetate pH 5.6, and eluent B was methanol. The samples were injected, and a gradient from 5 to 10% eluent B in 5 min, from 10 to 20% B in 50 min, and from 20 to 50% B in 10 min, at a flow rate of 1.0 mL/min, was used. When used, the fluorescence detector (Fd) from Jasco model FP-1520, with an excitation wavelength of 425 nm and an emission wavelength of 520 nm, was connected online before the ESI-MS equipment. The column effluent entered the source of the ESI-MS for continuous detection by MS. ESI mass spectra were obtained using an Agilent 1100 VL series (Agilent Technologies, Inc.). The electrospray interface was set in the negative ionization mode with the capillary voltage at 3 500 V and a heat source of 350 °C in scanned spectra from 300 to 1500 Da (10 full scans/s) with a maximum accumulation time of 300 ms and an ICC target of 20 000. Nitrogen was used as a drying (12 L/min) and nebulizing gas (70 psi). Skimmer 1 was changed from 20.0 to 60.0 V for MS optimization and used at 20 V to obtain lower fragmentation and 60 V to acquire spectra having maximum intensity (see below). Accordingly, the capillary exit offset changed from 97.2 to 137.2 V. The trap drive was 42.0. Software versions were 4.0 LC/MSD trap control 4.2 and Data Analysis 2.2 (Agilent Technologies, Inc.). The disaccharides were quantified using the peak area of interest determined by retention time and m/z values from the total ion chromatogram (TIC).

’ RESULTS AND DISCUSSION Disaccharide Standards. Figure 1A shows the TIC obtained for the 12 commercially available Hep/HS-derived unsaturated disaccharides derivatized with AMAC. The separation and detection of the same standards by Fd coupled online before ESI-MS is illustrated in Figure S-1 in the Supporting Information. The retention time on the RP(C18)-column was inversely related to the number of sulfate groups in the disaccharide. The four disaccharides belonging to the set S were separated from each other and with respect to the other groups in ∼22 23 min. Disaccharides from set A were eluted from ∼22 46 min and those of group H from ∼24 38 min. Furthermore, N-acetylated disaccharides elute at longer times than other disaccharides having an N-sulfate group. Moreover, disaccharides sulfated on the C6 position of the GlcN elute at shorter times than those sulfated at the C2 position of the uronic acid. As it is evident, also the isomeric disaccharides are well resolved possibly for the presence of ammonium acetate able to contribute to their resolution, according to Jones et al.32 In fact, buffer solutions having lower molarity strongly affect the Hep/HS disaccharides separation (not shown). Additionally, using the current chromatography method, we were able to obtain very sharp peaks with no broad doublets due to the partial resolution of R- and β-anomers reported for other RP-HPLC ESI-MS approaches.23,25 In fact, the fluorotagged derivatization with AMAC18,26 eliminates the anomeric hydroxyl preventing

Figure 1. (A) Total ion chromatogram (TIC) of Hep/HS unsaturated disaccharides derivatized with AMAC in negative ion mode with skimmer 1 set at 60 V separated by means of RP-HPLC. (B) Total ion chromatogram (TIC) of N-acetylated and N-sulfo Hep/HS unsaturated disaccharides derivatized with AMAC in negative ion mode with skimmer 1 set at 60 V separated by means of RP-HPLC. The various disaccharide species illustrated in this figure are irrespective of their amounts. Peaks having retention times of ∼2 4 min are related to sample solvents.

mutorotation and thus anomers species. Finally, full separation was observed for virtually all disaccharides, in particular for those having the same number of sulfate groups with no need to use tandem MS.23 Besides the long time of the HPLC run, ∼50 min, useful for the optimal separation of standards (and saturated trisulfated disaccharide, see below), the variability in retention times was always found lower than 12%. Furthermore, once labeled, derivatized samples do not need to be analyzed immediately as they are stable for a long period of time if stored in the dark and frozen.16,17 The major disaccharide unit of Hep is IS with minor variously substituted disaccharide sequences (Table S-1 in the Supporting Information). The disaccharides denoted with an H which 6772

dx.doi.org/10.1021/ac201426e |Anal. Chem. 2011, 83, 6770–6777

Analytical Chemistry

ARTICLE

Figure 2. The ESI-MS spectrum in the negative mode with skimmer 1 set at Supporting Information) separated by means of RP-HPLC.

contain a free amino group are very rare particularly in Hep but also in HS.23 As a consequence, the main disaccharides useful for a Hep (and HS) characterization are related to groups S and A. The current RP-HPLC ESI-MS method is particularly suitable for their separation and quantitative evaluation as illustrated in Figure 1B.

20 V of each single AMAC-disaccharide species (see Table S-1 in the

In addition to excellent separation of the various disaccharides, ESI-MS provides the mass of each disaccharide generally with the related sodium adduct (Figure 2) and limited desulfated products24 with skimmer 1 set at 20 V. The m/z values observed in the negative ESI-MS spectra of unsaturated Hep/HS disaccharides 6773

dx.doi.org/10.1021/ac201426e |Anal. Chem. 2011, 83, 6770–6777

Analytical Chemistry

ARTICLE

Figure 3. Total ion chromatograms (TIC) of disaccharides produced by the action of heparin lyases on (A) beef mucosa Hep, (B) fast-moving Hep, (C) slow-moving Hep, and (D) beef spleen HS derivatized with AMAC and separated by means of RP-HPLC. Skimmer 1 was set at 60 V. SatIS corresponds to the saturated trisulfated disaccharide species.

derivatized with AMAC are in agreement with the calculated molecular mass values (Table S-2 in the Supporting Information). RP-HPLC separation of each disaccharide and ESI-MS detection with skimmer 1 set at 60 V of amounts from 1 to 10 000 ng (10 μg) produced increasing peak intensities and decreasing noise from which integrated peak areas could be accurately calculated from the TIC. In fact, the integrated disaccharide peak areas showed excellent linearity when plotted as a function of their amounts (Figure S-2 in the Supporting Information). In general, the coefficients of correlation for all disaccharides in a very large range of concentration were found to be very high, greater than 0.970, mainly due to the high efficiency of separation and to the sharp outline of the peaks. However, the efficiency of ionization decreased by ∼70% when skimmer 1 was set at 20 V to limit the production of desulfated products. Finally, by using Fd coupled online with ESI-MS, a sensitivity of ∼50 100 times greater (with gain set at 100) was observed for standard disaccharides (not shown). Disaccharide Composition of Heparins and Heparan Sulfate. Beef intestinal mucosa Hep and its two purified components, SM-Hep and FM-Hep,27 30 along with a sample of beef spleen HS were isolated from tissues and digested with heparinases and generated disaccharides derivatized with AMAC (Figure 3). The quantity of each disaccharide in these samples was calculated using the linear equations shown in Figure S-2 in the Supporting Information and normalized to 100% (Table 1). Fluorescence detection was also applied to determine disaccharide composition. However, because of the presence of several unknown peaks observed by means of fluorescence detection (see Figure S-3 in the Supporting Information as an example), the correct assignment

was performed by using the Fd connected online to ESI before MS acquisition. By correct and unambiguous peak assignment, the compositions of these Hep/HS samples obtained using Fd were found to be consistent with ESI-MS (Table 1). Furthermore, the clear identification of the saturated trisulfated disaccharide having a retention time close to the unsaturated species (Figure 3B and Figure S-3 in the Supporting Information) and a theoretical m/z 1 of 595.0 (788.2 labeled with AMAC) was obtained for the first time by this novel RP-HPLC separation and ESI-MS acquisition (Figure S-4 in the Supporting Information). The percentage of this disaccharide, clearly belonging to the nonreducing end (NRE) of Hep chains, was calculated to be ∼5 8% in FM-Hep by using the same calibration curve of IS assuming similar ionization properties of trisulfated species (because of lower sensitivity compared to Fd, the saturated trisulfated disaccharide was also detected in Hep by ESI-MS but quantified only by Fd). After the correct assignment, fluorescence detection produced fairly similar values (Table 1). Finally, the charge density value was found to be very high in SM-Hep according to its structure but higher in FM-Hep compared to previous studies.27 30 This last point may be due to the presence of a significant percentage of saturated trisulfated disaccharide detected by the present analysis compared to previous characterization, generally performed by UV detection at 232 nm, which is unable to detect this disaccharide species. Bovine spleen HS was found to be composed of a high percentage of IVA disaccharide, ∼67% (Table 1) with an overall charge density of ∼0.40 typical of HS samples. Our HS compositional results are in general agreement with previous studies,23,32 34 also considering the expected variability related to biologically 6774

dx.doi.org/10.1021/ac201426e |Anal. Chem. 2011, 83, 6770–6777

Analytical Chemistry

ARTICLE

Table 1. Quantity and Composition of Disaccharides of Hep/HS of Various Origins and Different Commercial Hep/LMW-Hep Samples Evaluated by Using ESI-MS (MS) or Fluorimetric Detection (Fd)a UFHep BM Hep MS

Fd

FM-Hep MS

Fd

SM-Hep MS

BS HS

Fd MS Fd

Calciparin MS

Fd

Ecafast

LMW-Hep

Emoklar

MS Fd MS

Fd

Idracemi MS

Fd

Clexane MS

Clivarin

Fd MS Fd

Fragmin MS

Fd

IS (ΔUA-2s-(1f4)-GlcNs,6s)

72.4 70.1 62.8 65.3 91.8 91.6 1.9 0.8

74.5 71.7 77.5 73.3 66.5 74.4 89.0 94.5 86.1 84.5 66.0 68.4 72.9 69.5

IIS (ΔUA-(1f4)-GlcNs,6s)

12.6 12.3 6.5

10.9 10.8 10.1 15.0 5.8

7.3

3.9

2.4

2.8

2.9

4.0 3.7

5.5

6.2

IIIS (ΔUA-2s-(1f4)-GlcNs) IVS (ΔUA-(1f4)-GlcNs)

5.2 1.9

6.1 2.3

15.3 11.3 7.3 4.2 2.6 0.0

7.0 1.2 0.6 1.2 0.0 22.0 24.1 2.6

2.9 2.4

1.7 3.9 0.0 3.3 3.5 9.4

0.0 5.4

3.8 0.0

1.8 0.0

1.7 1.5

2.4 1.5

2.4 1.9 0.0 0.0

2.1 0.7

2.8 0.8

IA (ΔUA-2s-(1f4)-GlcNAc,6s) 0.7

1.3

0.0

0.0

0.0

0.0 0.0 0.0

0.7

0.8

0.0 0.0 0.0

0.0

0.0

0.0

1.3

2.4

1.0 0.8

0.0

0.0

IIA (ΔUA-(1f4)-GlcNAc,6s)

2.3

1.5

0.0

0.0

0.0

0.0 0.0 0.0

2.0

3.4

0.0 0.0 0.0

0.0

0.0

0.0

0.0

0.0

2.9 1.6

0.0

0.0

IIIA (ΔUA- 2s-(1f4)-GlcNAc) 0.0

0.0

0.0

0.0

0.0

0.0 1.9 0.0

0.0

0.0

0.0 0.0 0.0

0.0

0.0

0.0

1.0

0.8

0.0 0.0

0.0

0.0

IVA (ΔUA-(1f4)-GlcNAc)

4.9

4.2

6.3

6.2

0.0

0.0 66.2 67.8 7.7

5.7

7.4 4.3 15.5 11.7 0.0

0.0

1.6

1.8

1.6 2.0

2.2

3.1

6.3

0.9

1.4 1.6 2.0

IH (ΔUA-2s-(1f4)-GlcNH,6s) 0.0

0.0

0.0

0.0

0.0

0.0 0.0 0.0

0.0

0.0

0.0 0.0 0.0

0.0

0.0

0.0

0.0

0.0

0.0 0.0

0.0

0.0

IIH (ΔUA- (1f4)-GlcNH,6s)

0.0

0.0

0.0

0.0

0.0

0.0 0.0 0.0

0.0

0.0

0.0 0.0 0.0

0.0

3.3

1.3

0.0

0.0

1.0 0.7

0.0

0.0

IIIH (ΔUA-2s-(1f4)-GlcNH) IVH (ΔUA-(1f4)-GlcNH)

0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0

0.0 5.2 4.7 0.0 0.0 0.0

0.0 0.0

0.0 0.0

0.0 0.0 2.8 0.0 0.0 0.0

1.2 0.0

0.0 0.0

0.0 0.0

4.0 0.0

3.7 0.0

0.0 0.0 0.0 0.0

1.7 0.0

1.6 0.0

saturated IS

0.0

2.2

4.9

8.3

0.0

0.0 0.0 0.0

0.0

2.3

0.0 0.0 0.0

0.0

0.0

0.0

0.0

0.0

21.1 20.9 14.9 16.0

charge densityb

2.58 2.60 2.51 2.59 2.92 2.92 0.40 0.36 2.54 2.57 2.59 2.61 2.23 2.44 2.82 2.92 2.76 2.75 2.79 2.82 2.81 2.77

a

The means of three different analyses are reported: BM, beef mucosa; FM-Hep, fast-moving Hep; SM-Hep, slow-moving Hep; BS, beef spleen; UFHep, unfractionated Heps. b Charge density (the sulfate-to-carboxyl ratio) was calculated considering the presence and the percentage of carboxyl and sulfate groups for each disaccharide.

derived samples and the possible additional errors introduced by the enzymatic depolymerization. Disaccharide Composition of Commercial Unfractionated Heparin Samples. Unfractionated Hep disaccharide composition was directly analyzed in four marketed samples, three intravenous formulations and one eyewash solution (Figure 4 and Table 1). According to previous studies,23,25,27 30,33 39 of the 12 common disaccharides, 5 (IS, IIS, IIIS, IVS, and IVA) were observed as being very common in virtually all Hep formulations, 3 (IA, IIA, and IIIA) in specific kinds of samples (Table 1), and the series of H disaccharides were observed in trace amounts. Overall, charge density values were found to be in the range of ∼2.4 2.6 with only Hep in the eyewash solution having a higher value, ∼2.8 2.9, fairly similar to SM-Hep also considering the disaccharide composition (Table 1). Finally, minor peaks were also observed to be common to all kinds of Hep species (but also present in disaccharide standards, see Figure 1) both in TIC (Figure 4) and fluorescence chromatograms. However, their m/z values were not found to correspond to any possible Hep derived molecular species. To date, the presence of saturated trisulfated disaccharide in calciparin has never been reported. In fact, we observed the presence of ∼2.5% of this species by fluorescence (Table 1) clearly due to the higher sensitivity of this detector (not shown). Also for this Hep formulation, the accurate identification and quantitation of constituent disaccharides is a key point for its correct structural characterization. Disaccharide Composition of Commercial LMW-Heparin Samples. Three LMW-heparins were directly analyzed in marketed samples, enoxaparin (Clexane) produced by β-eliminative cleavage by alkaline treatment, and dalteparin (Fragmin) and reviparin (Clivarin) both produced by nitrous acid degradation (Figure 4 and Table 1). Analysis of the disaccharides revealed an IS percentage of ∼85% for enoxaparin and a lower amount, ∼66 73% for nitrous acid produced LMW-heparins. However,

the present analytical approach was able to detect in these last two samples the presence of a high percentage of saturated trisulfated disaccharide, ∼20% for reviparin and ∼15% for dalteparin (Table 1). As a consequence, the sum of the unsaturated and saturated species produced comparable amounts, ∼85%, of trisulfated disaccharide with respect to enoxaparin, for overall fairly similar charge density values, 2.75 2.82. We should consider that enoxaparin does not show the presence of saturated NRE due to the β-eliminative process capable of introducing a double bond on the degraded chains. On the contrary, nitrous acid depolymerization is able to degrade native Hep without any modification of NRE.40 As a consequence, saturated trisulfated species in LWW-Hep products is higher than unfractionated Hep (2 8%, Table 1) due to their lower molecular mass. It is worthy of mention that disaccharide composition by HPLC and UV detection at 232 nm is a common analytical approach in quality control laboratories to evaluate LMW-Heps structure and quality during their production. Furthermore, a simple assay having the power to detect most contaminants in Hep samples is based on treatment with a cocktail of heparin lyases and UV spectrophotometric detection.41 As a consequence, the presence of important percentages of saturated trisulfated disaccharide in unfractionated and LMW-Hep samples is not determined by these very common analytical approaches. By considering the presence of saturated trisulfated species, the overall disaccharide composition of the three LMW-samples is in agreement with the disaccharide composition obtained by UPLC-MS.39 In fact, ∼85% IS, ∼5% IIS, ∼5% IIIS, ∼1% IVS, ∼2% IA, and