The present invention relates to polysaccharides and methods for their identification. In particular it relates to glycosaminoglycans (GAG _S ) and their structures. The present invention particularly relates to structures and structural attributes of heparins and other GAG _S with a high sulfation ratio and to methods for their identification and use for the qualification of products and methods.

Glycosaminoglycans (GAG _S ) are a complex group of linear polysaccharides constituted of different repeating disaccharide units. They are produced by many organisms and play an important role in physiological and physiopathologist processes.

Among various GAG _S , heparin is of particular interest for its function in biological processes such as the proliferation and growth of cells, in allergic and inflammatory reactions, in angiogenesis, in tumoral metastases and in coagulation.

Heparin extracted from animal cells is used as a drug as such or after conversion in Low Molecular Weight Heparin (LMWH). The main and widespread use of Heparin and of Low Molecular Weight Heparins (LMWH _S ) as a drug is as an anticoagulant; for this use, the world consumption is in the amount of hundreds of tons per year. Several studies have demonstrated that heparin and similar compounds can inhibit tumoral metastases. This represents an example of further potential use of heparin in the pharmaceutical field.

GAGs are characterized by a high degree of unhomogeneity in terms of molecular weight and composition of the repeating disaccharide units. Functional groups such as acetyl and sulfate groups are in fact present in different positions along the polysaccharide chain.

As a whole, such molecular variety hinders the precise structural characterization of GAG _S ; on the other hand it is evident the importance of properly analyze this class of compounds both to facilitate the understanding of their mechanisms of action (most of which are still unknown) and to properly evaluate their quality, in consideration of their widespread use in the pharmaceutical field for human use.

It is known that GAG _S are able to interact with a wide number of proteins and various biological and pharmacological activities derive from these interactions.

It has been demonstrated for GAG _S that particular structures characteristics, in particular specific sequences of sugars in the oligosaccharide chains, are responsible for their biological activities and in particular for the specificity of interaction between GAG and the corresponding protein. As regards heparin, the identification of the active pentasaccharide as a critical characteristic for the anticoagulant activity has allowed a considerable progress in the knowledge of the anticoagulant properties of this polysaccharide. The active pentasaccharide is made of a particular sequence of five saccharide units which is found in one third of the chains of sodium heparin and is able to bind Antitrombin III so interfering with the coagulation cascade. The structure of the active pentasaccharide is disclosed by Linhardt (Natural Product Report, 26, (2009), 305-452):

wherein R=-H; -S0 ₃ ^" and Ri = -COCH ₃ ; -S0 ₃ ^"

The discovery of this particular structure-activity relationship has encouraged and stimulated further efforts in the attempt to identify other sequences which are responsible of GAG/protein interactions and for this purpose new more powerful means will be useful, if not necessary, for the structural investigation of GAG _S .

As regards the characterization of GAG _S used for therapeutic purposes, it is important to be able to identify, in addition to the natural structural elements, also the structural elements which could derive from the isolation or production process of a particular GAG.

For example, one of the available methods for the bleaching of heparin provides an oxidation with potassium permanganate. Such treatment brings to the total or partial chemical modification of the linkage regions of heparin and to the partial oxidation of the terminal reducing glucosamine to "glucosaminic" acid.

For the production of LMWH _S enzymatic or chemical depolymerisation methods are used. Each LMWH has typical structural characteristics depending from the particular process used for the depolymerisation. The depolymerisation by chemical or enzymatic β-elimination introduces for example a double bond between the carbons 4 and 5 of uronic acid in the new non-reducing ends.

Moreover, the depolymerisation by chemical β-elimination can produce, in a more or less amount depending on the used reaction conditions, a 1 ,6-anhydro ring at the reducing end of the oligosaccharide chains.

The depolymerisation with nitrous acid, followed by reduction, modifies the reducing ends transforming them into 2,5-anhydro-mannitol.

The exact and complete determination of the composition of GAGs samples of natural or semi-synthetic origin of scientific and pharmacological interest is made difficult, if not even impracticable, by their complexity and unhomogeneity. Through spectroscopic techniques such as for example NMR spectroscopy, and in particular two-dimensional NMR spectroscopy, is instead possible to determine the presence and the mediate abundance of specific structural groups in GAGs samples. Apart from exceptions related to oligosaccharide mixtures with a very low polymerisation degree, it is not instead practically feasible the identification of the single constituent species of these GAGs samples; in order to know the sequence of each particular oligosaccharide chain that is present, it would be indeed necessary to have the correspondent samples with a high purity degree with the consequent need of a long complicated and often impracticable purification and isolation of the single species. It is instead more advantageous from an analytical point of view the determination of the composition of the mixtures deriving from the depolymerisation of GAG _S (building blocks analysis). In fact they are simpler mixtures compared to the original sample and their constituents have structures more easily to determine. The analysis of the resultant species allows to know the average composition in terms of constitutive units of the entire polymer. There are different enzymes and chemical depolymerisation methods able to specifically cut GAG _S, some of them can be used simultaneously or sequentially for structural studies.

The complete depolymerisation to monosaccharides (generally chemical) or disaccharides (generally enzymatic) can then give information regarding the composition of a particular GAG _S sample under examination without determining how such constitutive units were present in the sequences of original polysaccharide chains.

For the characterization of heparin or LMWHs the exhaustive depolymerisation with a mixture of enzymes (heparinase I, heparinase II, heparinase III) is frequently used. Under these conditions a mixture of building blocks essentially constituted by dimers, but also by small amounts of oligomers having dp>2 (where dp is the polymerisation degree measured as number of constituting monosaccharides) is formed.

Different analytical methods for the analysis and the individuation of the single building blocks in the depolymerisation mixture were developed; the most commonly used method consists of the separation of the building blocks by SAX-HPLC followed by UV or PAD detection; genuine standard of building blocks are today available for the identifications. However, many "building blocks" naturally present in GAGs as well as "non natural" (i.e. deriving from structural modifications due to isolation and production processes) are still lacking of available commercial standards.

One of the building blocks that is generally observed is that related to the bond region. No commercial standard is available for this building block.

Similarly, commercial standards of some tetrasaccharides of significant importance, obtained through enzymatic depolymerisation with heparinase I, II and III and which derive from the sequence of the active pentamer are not available.

There is a clear need of new, sensible and effective identification methods of such type of structures that in the current state of known art can be defined only through difficult and lengthy procedures comprising preparative isolation phases of high amounts of the material representative of the structures under study being this species are in small amount.

As regards heparin with low molecular weight (LMWH) there is the strong need to know in depth the structures that can be generated by the different processes used for their production starting from unfractionated heparin.

Potentially, mass spectrometry (MS), particularly if combined with techniques of chromatographic separation, could be a decisive tool to solve most of the above mentioned problems. However, the use of mass spectrometry (MS) in the structural GAGs investigation presents considerable difficulties because of the high charge density of GAGs and of the weakness of the sulfuric ester groups. In the attempt to detect GAGs through MS, various experts have observed the loss of sulfur trioxide (S0 ₃ ) molecules and the loss of water molecules also during the first ionization phase and transport of the ions to the detector for the detection of the molecular ions. The exact structure of the GAG under analysis cannot be determined from the fragments derived from the desufation process.

As regards the MS-MS studies, many researches have reported the difficulty in obtaining significant fragments of the skeleton of the molecule (fragmentation at the level of glycosidic bonds and cross ring fragmentations) without the simultaneous fragmentation of the sulfuric ester groups. The problem can be partially solved with a suitable choice of the ion to fragment; in fact to avoid the loss of S0 ₃ it has been demonstrated that the best condition is that in which the sulfuric group is deprotonated (Zaia,Costello, Anal. Chem. 2001 , 73, 6030-6039). Such condition is always more difficult to achieve with the increasing of the sulfation degree and with the polymerisation degree of the species under study. For this reason the application of MS to the study of the sequence of oligosaccharides has found a strong limitation. The examples referred to GAGs that can be found in the literature are often limited to a species with low sulfation degree (generally lower or equal to 1 , i.e. with the ratio between the number of sulfate groups in the molecule and its number of saccharide units not greater than one) or, when referred to species with higher sulfation degree, they have however a low polymerisation degree (mostly dimers). A further problem that is often underlined is the difficulty in obtaining a fragmentation without the loss of water molecules. Such type of fragmentation interferes with the work of structural attribution: in fact many of the possible structures of GAGs differ each other just for one water molecule.

Methods of derivatization of GAGs that is the substitution of sulfate groups with more stable groups under fragmentation conditions have been proposed to overcome the above mentioned problems. However they are cumbersome and lengthy procedures which inevitably lead to the loss of material available in very limited amounts.

Another proposed solution was to carry out the mass spectroscopy analysis in the presence of a "supercharging" species (such as for example sulfolane) or strong bases (such as for example sodium hydroxide) or metal ions (lithium, calcium) (Zaia et al. Anal. Chem. 201 1 , 83 (21 ), 8222-8229; Lindhardt et al. Anal. Chem. 2012, 84, 5475-5478; Zaia 2013 manuscript under publication).

Such processes imply the introduction of non volatile substances in the mass spectrometer and make necessary frequently disassembly and cleaning operations of the spectrometer.

Considering the advantages that MS detection may offer (high accuracy, high sensibility, speed performance) a further effort for the characterization of GAGs was made in the attempt to couple chromatographic separation techniques to MS detection.

Coupling the chromatography separation to the MS detection allows to save considerable time and offers the advantage of being applied also when the amount of the available sample is minimal through the miniaturization of the system by using for example capillary HPLC columns.

In most favorable cases, the chromatography-MS and CE/MS techniques can give a significant amount of information regarding the components under study. In fact in addition to the data regarding the molecular weight and the empirical formula, through LC-MS/MS or LC/MS ⁿ experiments, also information such as the determination of the exact structural formula and the sequence of the single components of the mixture under study can be derived.

In the specific case of GAGs, despite the great efforts made in the attempt to detect and identify them through the combination of a chromatographic technique with mass spectrometer, the results unfortunately still suffer from the limitations due to the difficulty of properly fragmenting GAGs without excessive loss of S0 ₃ and/or H ₂ 0 groups. The best results, reported in literature, of direct detection of the sample are still limited to oligomers with a low sulfation degree (generally with sulfation degree≤1 ).

In the state of art there are not reported examples of HPLC-MS/MS or HPLC-MS ⁿ with RP-IP-LC techniques and it was indeed observed that it is not possible to obtain fragmentation spectra from ion adducts with the ion-pairing agent and that also CID fragmentation spectra, obtained from the deprotonated pseudo-molecular ion in the presence of the ion-pairing agent, are very weak and then not significant (Naggi et al. J. Chromatogr. 647 (1993) 289-300). Furthermore, it has been reported that the Q-ToF type instruments with CID fragmentation are not suitable for the studies of fragmentation of GAGs as it is not possible to observe deprotonated ions of GAGs with a high charge: by using this type of instruments the desolvation process is so powerful that it produces excessive fragmentation of GAGs also in source (Zaia, Costello J. Am. Soc. Mass Spectrom. 2004, 15, 1534-1544). At the present time, the technique of CID fragmentation is considered not to be suitable for the structural study of GAGs as it is not efficient in creating cross ring fragmentation of highly sulfated oligosaccharides (Linhardt Mol. BioSyst. 2012, 8, 1613-1625). This type of fragmentation is on the contrary desirable in the structural study of GAGs, as the observed fragments allow to derive the exact position of the sulfate groups along the molecule and then the exact structure of the molecule. It is therefore evident how the request for more efficient methods of structural investigations for GAGs has not been yet satisfied till now.

We have now found an analytical method for the determination of chemical structures in samples of GAGs which consists in the identification and in the quantification of molecular species in samples of GAGs, pure or in admixture, carried out through mass spectrometry, wherein the ion species are generated and their related information of physico-chemical type are read and interpreted, leading to the identification of the structural molecular formulas of the species under study and to the attribution of the correspondence to specific chromatographic or electrophoretic signals of the separation systems optionally coupled to the mass spectrometer.

Therefore, the present invention relates to an analytical method for the structural determination of polysaccharides, preferably GAG _S , which comprises a mass spectrometer optionally coupled with a chromatographic or electrophoretic system.

A first object of the present invention is an analytical method for the structural determination of polysaccharides, pure or in admixture, preferably GAG _S , comprising a mass spectrometer optionally coupled with a chromatographic or electrophoretic system which comprises triple quadrupoles, ion traps (also linear), Q-ToF and ToF-ToF or FT-MS instruments.

Said mass spectrometer preferably uses Q-ToF with CID fragmentation.

In particular the Q-ToF instrument with CID fragmentation, used in combination with chromatographic or electrophoretic systems, allows to accurately determine the molecular masses of GAG _S oligomers, to sufficiently resolve the related isotopic patterns, to efficiently produce the glycosidic and cross ring fragmentations necessary for the structural determination keeping the desulfation and dehydration phenomena at sufficiently low levels, as it is known that they hinder the work of identification of molecular species. A mass spectrometer with a medium/high resolution (lower or equal than 20 ppm) and able to allow a suitable observation of the molecular isotopic pattern is preferably used. This is highly desirable especially when natural or semi-synthetic complex mixtures have to be analyzed.

The samples of GAG _S analyzed according to the method object of the present invention are monosaccharides, oligosaccharides, polysaccharides, or mixtures thereof, preferably mixtures of sulfated oligosaccharides or pure sulfated oligosaccharides, more preferably heparins, low molecular weight heparins (LMWH), ultra low molecular weight heparins (ULMWH).

Preferably the samples are sulfated oligosaccharides without being pre-treated or samples that are treated in advance, for example through chemical and/or enzymatic depolymerisation and/or in any way chemically modified.

A peculiar characteristic of the invention is that it allows the determination of the formulas of molecular structure also for pure or in admixture species having a sulfation degree higher than 1 or even≥ 1 .5 and a polymerization degree≥ 2 or even≥ 4.

In a preferred embodiment, the invention is applied to the analysis of sulfated oligosaccharide samples which have a sulfation degree higher than 1 or≥ 1 .5 and a polymerization degree≥ 2 or≥ 4.

A preferred practical embodiment of the invention is the identification of monosaccharides or sulfated polysaccharides or mixtures thereof, through the determination of their structural formulas (through MS/MS or MS ⁿ ).

ESI is preferably selected among the various available sources. The use of the ESI source does not need matrices and allows the on line coupling with LC or CE.

Preferably, when the analytical method object of the present invention is applied to isolated and purified oligomers or to very simple mixtures in which the components could be distinguished by molecular weight, it is not necessary to combine the mass spectrometer with chromatographic or electrophoretic systems.

Preferably, when the analytical method object of the present invention is applied to more complex mixtures, chromatographic or electrophoretic systems, preferably selected among SEC, HILIC, SAX, Grafititized carbon separations, RP-IP-LC, are used.

RP-IP-LC that allows the separation of oligomers with different lengths and in particular allows the separation of oligosaccharide isomers is more preferably used. The analytic separation of different positional isomers has proved to be useful as they cannot be separated by the mass detector.

Moreover, the RP-IP-LC chromatography is one of the chromatographic techniques which can be made directly MS compatible through the use of suitable solvents and through the choice of a suitable volatile ion-pairing agent without the need of desalinizations of the post column mobile phase.

Different volatile amines can be used as ion-pairing agent for the chromatographic separation of different molecular species of GAGs depending on the particular subclass of GAG to be analyzed.

Preferably the analytical method object of the present invention comprises the use of ion-pairing agents selected among volatile alkyl amines preferably selected among triethylamine, propylamine, tripropylamine, dibutylamine, butylamine, pentylamine, hexylamine, tripentylamine, more preferably dibutylamine.

Preferably the concentration of ion-pairing agent in mobile phase is lower than 20 millimol.

An embodiment of the present invention consists of the analysis of oligosaccharide mixtures using chromatographic or electrophoretic separative methods preferably RP-IP-LC type which allows the separation of each component in case of a mixture or the determination of the purity of the single oligosaccharide in case the sample is a pure oligosaccharide. In case of pure oligosaccharides it is possible to avoid the phase of chromatographic separation.

A particularly preferred embodiment of the invention comprises the optional post column and pre source addition of a sheath liquid with a suitably adjusted flow. The sheath liquid is preferably acetonitrile pure or optionally in admixture preferably with water or volatile organic solvents. The addition of non volatile additives such as sulfolane, acids, bases or salts is not needed.

The choice of the most suitable flow adjustment of the sheath liquid is within the knowledge of the skilled in the art.

We have surprisingly found that the addition of the above described sheath liquid (with a suitably regulated flow) is able to improve, despite the presence of the ion- pairing agent, the ionization of the sample, thus allowing to detect ions with a high charge status (z), in addition to produce an increase of the S/N ratio. A particularly preferred embodiment of the present invention consists of the choice of the operational ionization and ion transport parameters such that the fragmentation due to the loss of S0 ₃ and also, at the same time, due to the loss of water are minimized. It must be taken into account that the minimization of the fragmentations due to the loss of S0 ₃ and due to the loss of water is an essential condition for the exact determination of the molecular structural formulas. A fragmentation level due to the loss of S0 ₃ or water that can be considered acceptable is the one for which the signals corresponding to the fragmentation due to the loss of S0 ₃ or the loss of water do not exceed the intensity of 20%, or preferably of 10%, of the intensity of the parent signal.

The analytical method according to the present invention is characterized in that the fragmentations due to the loss of S0 ₃ or water do not exceed the intensity of 20%, more preferably they do not exceed 10%, of the intensity of the parent signal.

The sequence of the collecting and data elaboration phases that has been generally applied for the determination of GAG _S structure according to the method object of the present invention is described herein after:

Phase 1 : determination of the charge (z)

Phase 2: assumption of the possible empirical formulas

Phase 3: determination of the empirical formula

Phase 4: assumptions of the possible structural formula

Phase 5: choice of the ion for fragmentation

Phase 6: iterative detection of the key fragments

Phase 7: interpretation of the additional fragments.

In details:

PHASE 1 : DETERMINATION OF THE CHARGE (z)

The charge z is determined by the observation of the isotopic pattern of the examined signal and in particular by the differences of the m/z observed values among contiguous isotopic peaks.

PHASE 2: DETERMINATION OF THE POSSIBLE EMPIRICAL FORMULAS

By knowing z and from the experimental observation of the mass/charge ratio (m/z) the mass (m) is determined. Possible combinations of atoms compatible with m and with the observed m/z are determined.

Among the possible combinations that give a mass similar to that observed, only the possible combinations for which the difference among the observed m/z and the calculated m/z is not higher than an established limit depending from the used instrument are taken into account. Preferably the value of 20 ppm is selected.

PHASE 3: DETERMINATION OF THE EMPIRICAL FORMULA

The relative isotopic pattern is calculated for each different empirical formula assumed in the previous phase. The calculated isotopic pattern is compared with the observed isotopic pattern for the ion under analysis.

Among the possible empirical formula combinations that in which the calculated isotopic pattern is equal to the observed isotopic pattern is selected.

PHASE 4: ASSUMPTION OF THE POSSIBLE STRUCTURAL FORMULA

Using the knowledge about the molecule under study, the structural formulas reasonably compatible with the derived empirical formula are assumed.

PHASE 5: CHOICE OF THE ION FOR MS/MS FRAGMENTATION

The ion with negative charge (z) equal to the number of the present sulfate groups as determined according to the empirical formula is selected as ion for fragmentation.

If this ion is not detectable or its intensity is too low, the ion with one less negative charge unit is fragmented.

PHASE 6: ITERATIVE INDIVIDUATION OF THE KEY FRAGMENTS AND THEIR PREPARATION

The interpretation of the MS/MS spectrum proceeds with the gradually and iterative identification of the key fragments.

The key fragments are those fragments which allow the unique and undoubted identification of the structural characteristics of the ion species from which they are generated.

A key fragment is for example:

1 ) a fragment generally small in size, for which the structural formula is uniquely determined

2) a fragment having similar size to the ion from which it derives (that could be the molecular ion) such to allow the unique determination of structural details that are originally present from the difference with the ion from which it derives.

The individuation of key fragments can be considered iterative as every time one of this fragments is identified and then the structural details of the molecule under study are fixed, fragments which cannot be considered as "key" fragments in the previous reading of the spectrum, following the identified structural details can become themselves key fragments.

The individuation of the key fragments and their connection, that is compatible with the general knowledge on the type of chemical structures in the sample, allows to verify and identify the structural formula of the molecule under study.

The structural information deriving from a single key fragments have to be suitable combined with the general information relating to the chemical structures under study, to the empirical formula of the analyzed species, to the general knowledge of the fragmentation of sugars in mass spectrometry.

PHASE 7: INTERPRETATION OF FURTHER FRAGMENTS THAT ARE PRESENT IN THE SPECTRUM

The attribution of fragments with higher intensity that are present in the spectrum and the check of their coherence with the proposed structural formula allows to confirm that the hypothesis of structure is correct. Such elaboration can then be carried out not only on the fragments which are considered key fragments.

After the determination of the structure of the polysaccharides through the analytical method object of the present invention and having attributed the correspondence between the chromatographic signals and the structural formula, it is possible by applying the same chromatographic method, to use different detectors selected from example among an ultraviolet detector, diode array detector or evaporative light scattering detector (ELSD), by using them separately or simultaneously, to carry out further analysis recognizing and identifying the single detected species.

The analytical method object of the present invention allows to identify all the building blocks, up to the determination of their structural formula and irrespective of their specific chemical structure. Said method is characterized by the ability to fragment sulfated molecule also with a sulfation degree greater than one.

The method object of the present invention allows also the correct detection, fragmentation and identification of molecules with different chemical structure, for example devoid of sulfuric ester groups.

The method object of the present invention allows to separate the various building blocks and to proceed with their identification though a direct analysis of the sample under study without the need of further manipulation of said sample (for example to remove salts etc.) which could lead to the possible introduction of mistakes and/or the loss of sample.

It is just in the case of samples that are subjected to enzymatic digestion that the technique object of the present invention shows one of the most advantageous aspect. In fact, in many situations there could be a small amount of sample: the high sensibility of the mass detector together with the possibility to use also capillary HPLC columns create one of the techniques with the smallest consumption of sample up to now.

Commercially available HPLC or CE systems can be used for carrying out the invention and do not represent a limitation to its realization.

It is an object of the present invention the use of the described method within enzymatic mapping studies.

The analytical method object of the present invention allows to determine the structure of the molecule under study by also fragmenting the ion wherein the charge status is not equal to the number of the present sulfate groups.

A particular embodiment of the invention consists of the fragmentation, among all the observed ion, of the ion with negative charge (z) equal, or with one less, to the number of present sulfate groups by choosing the energy for fragmentation in function of the particular chemical species under study (in particular in function of its polymerization degree and its sulfation degree).

Then, the application of the analytical method object of the present invention allows to determine the structure of GAGs pure or in admixture.

Such individuations of the structures realized through the mass spectrometry together with chromatographic techniques according to the teachings of this invention has a great value in the field of analytical and regulatory activities related to the use of GAGs for pharmaceutical use.

The individuation of such structures can be in fact used for the qualification of GAGs as finished products, intermediates and raw materials for pharmaceutical use.

These qualifications can find their use for the approval of production batches and adopted in the field of comparison studies for regulatory purposes among production batches and reference samples of commercial drugs.

In the field of quality and regulatory control, the present invention is usefully applied in the individuation of the species that are present on GAGs deriving from possible adulteration or counterfeiting.

In the field of the identification of structure-activity relationships, this invention is very useful also in consideration of the extremely low amounts of sample that are necessary to carry out the identification of the molecular structures.

Such result is of great interest in the research of new substances in pharmaceutical field and in the individuation of new advantageous uses of already known substances.

The analytical method object of the present invention allowed the identification of new species that can have a valuable application also for their use as specific indicators of determined characteristics of GAGs mixtures. The identification of such specific indicators is possible by recording during the studies of mass spectrometry specific characteristics of the species in object such as for example their retention time during different chromatographic and/or electrophoretic conditions, their spectroscopic characteristics or their response factor with different detectors.

Therefore a further object of the present invention are the following compounds identified through the described analytical method:

d)

n) 4-deoxy-2-0-sulfo-enepyranosyluronic-(1 ,4)-2-deoxy-2-acetamido-pyranosyl- (1 ,4)-pyiranosyluronic-(1 ,4)-2-deoxy-2-sulfamido-3-0-sulfopyranosyl acid and salts thereof

o) 2-0-sulfo-pyranosyluronic-(1 ,4)-2-deoxy-2-sulfamido-6-0-sulfo-pyranosyl-(1 ,4)-2- 0-sulfopyranosyluronic-(1 ,4)-1 ,6-anhydrous-2-deoxy-2-sulfamido-pyranosyl acid and salts thereof

A correct study of the different sequences of the active pentamer in a sample of heparin or derivatives thereof and a correct quantification of all the building blocks which can be related to the active site, are desirable for the realization of a correct profile of the biological activity of the sample under study.

The above described tetramer (n), if present and recognized as building block in the enzymatic maps of heparin or derivatives thereof, can be related to a particular sequence of the active pentamer and then can represent an important element from a pharmacological point of view.

Analogously, with the method of the present invention, the tetrasaccharide (o) has been identified for the first time in samples of commercial enoxaparin. The concomitant presence of the group 1 ,6-anhydro at the reducing end and of the saturated non reducing end has been observed here for the first time. This represents then an important improvement in the characterization of a pharmaceutical product for human use.

Therefore, further objects of the present invention are the use of the compounds (a)- (o) for the structural determination of GAGs and the use of compounds (n)-(o) for the qualification of products for pharmaceutical use.

Some representative examples of the possible embodiments of the invention are reported herein below, without limiting it.

BRIEF DESCRIPTION OF THE FIGURES

A brief description of the figures related to the examples indicated in brackets is reported.

Figure 1 (Example 1 ): sodium heparin digested with heparinase I, II, and III, UV trace (232 nm).

Figure 2 (Example 1 ): MS spectrum of the peaks at RT 71 minutes.

Figure 3 (Example 1 ): isotopic pattern observed for the ion 247.50.

Figure 4 (Example 1 ): MS/MS analysis of the peak at 71 minutes. Complete spectrum and enlarged particular.

Figure 5 (Example 1 ): MS spectrum of the peak at RT 38 minutes.

Figure 6 (Example 1 ): isotopic pattern of the ion at m/z = 344.08.

Figure 7: MS/MS spectrum of the peak at RT 38 minutes. Complete spectrum and enlarged particular.

Figure 8 (Example 2): sodium dalteparin digested heparinase I, II, and III, UV trace (232 nm).

Figure 9 (Example 2): MS spectrum of the peak at RT 98 minutes.

Figure 10 (Example 2): isotopic pattern observed for the ion 287.48.

Figure 1 1 (Example 2): MS/MS analysis of the peak at RT 98 minutes. Complete spectrum and enlarged particular.

Figure 12 (Example 3): semi-preparative SAX-HPLC chromatogram with the indication (arrow) of the isolated peak.

Figure 13 (Example 3): RP-IP-HPLC-MS (TIC) analysis of the isolated peak by semi-preparative SAX chromatography.

Figure 14 (Example 3): MS spectrum of the peak at RT 53 minutes (RP-IP-HPLC- MS).

Figure 15 (Example 3): isotopic pattern observed for the ion 477.03

Figure 16 (Example 3): MS/MS analysis of the peak at RT 53 minutes (RP-IP-HPLC-

MS/MS).

Figure 17 (Example 4): oligosaccharide mixture isolated from sodium enoxaparin. Profile of UV (below) and MS (above) detection.

Figure 18 (Example 4): (600.6±0.2) Extracted Ion Chromatogram above and

(609.6±0.2) Extracted Ion Chromatogram below.

Figure 19 (Example 4): mass spectrum of peaks at RT=25÷31 minutes.

Figure 20 (Example 4): isotopic pattern observed for the ion 535.99.

Figure 21 (Example 4): isotopic pattern observed for the ion 544.99.

Figure 22 (Example 4): MS/MS spectrum of the ion 217.4 eluent among RT 25 and

30 minutes.

Figure 23 (Example 4): MS/MS spectrum of the eluent species type at RT between about 25 and 27 minutes.

Figure 24 (Example 4): MS/MS spectrum of the eluent species at RT between about 28 and 30 minutes.

Figure 25 (Example 5): oligosaccharide mixture isolated from sodium enoxaparin. Profile of UV (below) and MS (above) detection.

Figure 26 (Example 5): mass spectrum of the peak at RT=29 min.

Figure 27 (Example 5): ion at m/z 213.7938 (z=-5).

Figure 28 (Example 5): MS/MS spectrum of the ion 213.793, complete and enlarged.

Figure 29 (Example 6): oligosaccharide isolated from sodium enoxaparin. Profile of UV (below) and MS (above) detection.

Figure 30 (Example 6): mass spectrum of the peak at RT=42 min.

Figure 31 (Example 6): ion at m/z 383.6385 (z= -3).

Figure 32 (Example 6): ion at m/z 229.78 (z= -5).

Figure 33 (Example 6): MS/MS spectrum of the ion at m/z=229.8, complete and enlarged.

Abbreviations and definitions

Volatile alkyl amine = alkyl amine with a boiling temperature below or equal to 250 °C at atmospheric pressure.

ANDA = abbreviated new drug application

AT III = antitrombin III

CID = collision induced dissociation

dp = polymerization degree

ESI = electro spray ionization

FDA = food and drug administration

GAG = glycosaminoglycan

Sulfation degree = number of present sulfuric ester groups divided by its number of constitutive monosaccharide units

HILIC = hydrophilic interaction liquid chromatography

HPLC = high pressure liquid chromatography

LC = liquid chromatography

LMWH = low molecular weight heparin

MS = mass spectrometry

OSCS = oversulfated chondroitin sulfate

RP-IP-LC = reverse phase-ion pairing-high liquid chromatography

RP-IP-LC = RP-IP-LC wherein chromatography is of high pressure type

s = sulfur

SAX = strong anion exchange

SEC = size exclusion chromatography

S/N = signal to noise

Volatile organic solvent = organic solvent with a boiling temperature below or equal to 200 'C at atmospheric pressure,

z = ion charge

EXAMPLES EXAMPLE 1 Building blocks of sodium heparin

Identification of structures in a sample of sodium heparin exhaustively digested with heparinase l+ll+lll. In particular, identification of a dimer with sulfation degree equal to 1 and identification of the bond regions modified by the production process (sulfation degree = 0).

Preparation of the sample:

Sample: 5 mg/mL of sodium heparin depolymerised with heparinase I, II and III in buffer.

The enzymatic depolymerisation reaction was carried out at a temperature of 25 °C for 60 hours on a mixture prepared by mixing:

- 15 μΙ_ of a solution containing 20 mg/mL of sodium heparin under analysis,

- 60 μί of a buffered solution at pH 7.0 of AcONa 100 mM, containing 2 mM of Ca(OCH ₃ ) ₂ and 1 mg/mL of bovin serum albumin;

- 15 μί of a solution containing heparinase I, II and III each at a concentration of

0.5 UI/mL, in a 0.01 M buffer solution of KH ₂ P0 ₄ containing 2 mg/mL of bovin serum albumin

Chromatoghraphy:

Column: Thermo Hypersil C18, 3 μηι, 250 mm x 2.1 mm

Column temperature: 45 °C

Flow: 0.2 mL/min

Mobile phase A: 90% H ₂ 0, 10% MeOH, 10 Mm dibutylammoniumacetate pH 6.7 Mobile phase B: 10% H ₂ 0, 90% MeOH, 10 Mm dibutylammoniumacetate pH 6.7

Elution: gradient

Time % phase A % phase B

0 100 0

5 100 0

170 50 50

180 0 100

190 0 100

195 100 0

215 100 0 Sheath liquid: CH3CN, 0.1 mL/min

Injection volume: 20 Mi- Detector 1 : UV spectrophotometer, 232 nm

Detector 2: Bruker MicroTOF-Q mass spectrometer

Instrumental parameters of detector 2

Negative ion mode

Scan: 100-700 m/z (per analysis MS1 )

End Plat offset: -500V

Capillary Voltage: +3000V

Nebulizer gas pressure: 0.8 bar

Drying gas flow: 10 L/min

Drying temperature: 200 °C

Funnel 1 RF: 150 Vpp

Funnel 2 RF: 180 Vpp

Hexapole RF: 150 Vpp

ISCID: 0 eV

Quadrupole ion energy: 2.0 eV

Low mass: 100 m/z

Collision Energy: 0 eV

Collision RF: 200 Vpp

Transfer Time: 148 [is

Prepulse storage: 5 μ≤

MS/MS conditions Segment 1 : from 0 to 60 minutes isolation mass: 344.1 m/z

Isolation width: 8.0 m/z

Collision energy: -10eV

MSMS summation factor: 5x

Segment 2: from 60 to 150 minutes isolation mass: 247.5 m/z

Isolation width: 8.0 m/z

Collision energy: -10eV

MSMS summation factor: 5x

Results:

The depolymerised mixture analyzed by HPLC method showed the UV profile reported in figure 1 . The species corresponding to the UV signs visible at the retention time of 71 minutes and 38 minutes were detected.

Detection of the peak at RT 71 minutes

Phases 1 , 2 and 3

The MS spectrum of the peak at 71 minutes is shown in figure 2. For the determination of the empirical formula the pseudo molecular ion having two negative charges [M-2H ⁺ ] ² = 247.50 was used (Figure 3).

Assigned Empirical formula (for the neutral species M)

Error calculation in ppm

m/z calculated m/z measured difference in ppm

247.5000 247.5007 3

Calculated Isotopic Pattern

m/z Intensity

247.5000 100.00

248.0014 15.75

248.4993 13.49

249.0007 1 .92

249.4994 0.73

Phase 4

From the composition of the empirical formula, from the UV absorption and knowing the nature of the product under study it was defined that it was a dimer composed of uronic acid and glucosamine, having two sulfate groups as substituents, corresponding to the following general structure:

Phase 5

For the complete determination of the structural formula the MS/MS fragmentation of the ion [247.50] ^2" , for which z = 2 = number of sulfur atoms present in the empirical formula, was carried out. Phase 6

The following key fragments were identified:

1 ) m/z = [137.99] ^1-

Such fragment was referable to the positions 5 and 6 of the glucosamine. It was deduced that the second sulfate group was in position 6 of the amine. The glucosamine was then sulfated on the nitrogen and on the carbon in position 6. 3) m/z = [168.49] ² -

This fragment confirmed that both sulfate groups were on the amine. It was then possible to assign the sulfate groups on the nitrogen and in position 6 in view of the already identified key fragments 1 ) and 2). As a consequence it could be indirectly deduced that no sulfate group was on the uronic acid.

4) m/z = [157.01 ] ^1"

For the presence of the carboxyl group, this fragment was referable to the unsaturated uronic acid unit.

By combining the structural information that were acquired by the reading of the fragmentation spectrum, the structure of the examined molecule was determined:

Phase 7

The more intensive ion of the spectrum was the ion having two negative charges m/z=[189.49] ² -

2 groups R = S0 ₃ _; 2 groups R = H

that in view of the key fragments 1 ) and 2) had the following structure:

Peak at RT 38 minutes

Phases 1 , 2 and 3

The mass spectrum of the peak at 38 minutes is shown in figure 5.

For the determination of the empirical formula the pseudo molecular ion having two negative charges [M-2H ⁺ ] ² = 344.08 was used (figure 6). Assigned empirical formula (for the neutral species M)

Error calculation in ppm

m/z Calculated m/z Measured Difference in ppm

344.0855 344.0849 2

Calculated Isotopic Pattern

m/z Intensity

344.0855 100.00

344.5871 28.29

345.0882 8.38

345.5896 1 .61

346.0907 0.29

Phase 4

The empirical formula that was found to correspond to the empirical formula of the oxidized bond regions.

Phase 5

For the complete determination of the structural formula the MS/MS fragmentation of the ion [344.08] ^2" was carried out. The MS/MS spectrum of the peak at 38 minutes is shown in figure 7.

Phase 6 1 ) m/z = [157.01 ] ^1"

In the molecule there was an unsaturated uronic acid as confirmed also by the UV absorption at 232 nm.

2) m/z = [189.04] ¹ -

3) m/z = [207.05] ¹ -

The fragments 2) and 3) led to the hypothesis of the presence of a xylose ring bound to the -CH ₂ COOH sequence.

4) m/z = [369.10] ^1"

The position of the (1 ,4) bond was deduced from the knowledge of the nature of the product (in the bond regions the galactose ring was bound to the xylose ring with a 1 ,4 bond)

5) m/z = [531

The position of the (1 ,3 and 1 ,4) bonds was deduced from the knowledge of the nature of the product.

6) m/z = 601.16] ¹ -

By combining the structural information that were acquired by the reading of the fragmentation spectrum, the structure of the examined molecule was determined:

It was the sequence of the bond regions wherein the terminal serine was chemically modified by the production process.

EXAMPLE 2 Dalteparine building blocks with a high sulfation degree

Structural determination of a dimer with a sulfation degree = 1 .5 in samples of sodium dalteparin exhaustively digested with Heparinase l+ll+lll.

The preparation of the sample, the operative parameters used for chromatography, detector 1 , detector 2 were reported in example 1 . The settled parameters for the MS/MS were:

MS/MS conditions

isolation mass: 191 .3 m/z

Isolation width: 8.0 m/z

Collision energy: -7eV

MSMS summation factor: 5x

Results:

The UV profile of the depolymerised mixture analyzed by HPLC method was reported in figure 8.

Detection of the peak at RT 98 minutes

Phases 1 , 2 and 3

The mass spectrum of the peak at 98 minutes was shown in figure 9. For the determination of the empirical formula the pseudo molecular ion having two negative charges [M-2H ⁺ ] ² = 287.48 (Figure 10) was used.

Assigned Empirical formula (for the neutral species M)

Error calculation in ppm

m/z Calculated m/z Measured Difference in ppm

287.4784 287.4793 3

Calculated Isotopic Pattern

m/z Intensity

287.4784 100.00

287.9798 16.67

288.4775 18.77

288.9788 2.87

289.4770 1 .49

289.9782 0.21 Phase 4

From the composition of the empirical formula, from the UV absorption and knowing the nature of the product under study it was defined to be a dimer composed of uronic acid and glucosamine having three sulfate groups as substituents.

Phase 5

For the complete determination of the structural formula the MS/MS fragmentation of the ion [191 .31 ] ³ was carried out. The MS/MS spectrum of the peak at 98 minutes was shown in figure 1 1 .

Phase 6

The following key fragments were identified:

1 ) m/z = [117.98] ² -

O

the uronic acid had a sulfate group in position 2

2) m/z = [137.98] ¹ -

one of the sulfate groups was on the nitrogen of the amine

3) m/z = [138.97] ¹ -

One of the sulfate groups was in position 6 of the amine. The glucosamine was then sulfated at the nitrogen and at the carbon in position 6.

By combining the structural information that were acquired by the reading of the fragmentation spectrum the structure of the examined molecule was determined:

Example 3 Tetrasaccharide containing units of glucosamine 3-sulfate

The identification of the structure of a tetramer representative of the active site for the anti coagulant activity of heparin.

Experimental parameters:

Preparation of the sample:

Sample: 5 mg/mL of sodium enoxaparin depolymerised with heparinase I, II and III in buffer.

The reaction of enzymatic depolymerisation was carried out as reported in example 1 .

The resultant solution (8 ml total) was chromatographed on Waters Spherisorb SAX 250 mm L x 20 mm i.d. column, with sodium perchlorate as gradient, for the isolation of the unknown peak at RT 37 minutes (see figure 12).

The isolated solution containing the peak at RT= 37 minutes was partially desalted by osmosis using acetate cellulose membranes having a cut off of 100 Dalton and afterwards lyophilized.

The lyophilized product was dissolved in water at a concentration of about 1 mg/mL for the HPLC/MS and HPLC/MS/MS analysis.

Chromatography:

column: Kinetex C18, 2.6 m, 100 mm x 2.1 mm

column temperature: 30^ flow: 0.15 ml/min

mobile phase A: 95% H ₂ 0, 5% MeOH, 2.5 Mm dibutylammoniumacetate pH 6.7 mobile phase B: 10% H ₂ 0, 90% MeOH, 2.5 Mm dibutylammoniumacetate pH 6.7

Elution: gradient

Sheath liquid: CH ₃ CN, 0.1 mL/min

Volume of injection: 200 μΙ_

Sample: 0.02 mg/mL in water

Diverter valve: from 0 to 40 minutes the flow was sent to the discharge (elimination of the non-volatile salts in the sample), from 40 to 95 minutes the flow was sent to the ESI source.

Detector: Mass spectrometer Bruker MicroTOF-Q

The instrumental parameter of the detector were reported in example 1 .

MS/MS conditions

isolation mass: 317.7 m/z

Isolation width: 10.0 m/z

Collision energy: -10eV

MSMS summation factor: 3x

Results:

The chromatographic profile and the mass spectrum of the analyzed specie were shown in figures 13 and 14. The enlargement of the ion used for the obtainment of the empirical formula was reported in figure 15.

Phases 1 , 2 and 3 Assigned empirical formula (for the neutral species M)

Error calculation in ppm

m/z calculated m/z measured Difference in ppm

447.0341 4770.0332 -1.9

Calculated isotopic pattern

m/z Intensity

477.0341 100.0

477.5356 32.8

478.0341 25.0

478.5352 6.9

479.0343 2.7

479.5351 0.6

480.0348 0.2

Phase 4

From the composition of the empirical formula, from the UV absorption and knowing the nature of the product under study it was deduced to be a tetramer composed of two rings of uronic acid and two rings of glucosamine having three sulfate groups.

Phase 5

For the complete determination of the structural formula the MS/MS fragmentation of the ion [317.69] ^3" was carried out. The MS/MS spectrum of the peak at 53 minutes was shown in figure 16. Phase 6

The following key fragments were identified:

1 ) m/z = [137.98] ¹ -

One of the ammines had a S0 ₃ ^" substituent on the nitrogen

2) m/z= [123.47] ^2"

One of the amines was sulfated both on the carbon in position 6 and both on the nitrogen. A glucosamine ring had at least two sulfate groups.

3) m/z = [168.49] ² -

The amine was not sulfated in position 6. This fragment come from the same glucosamine ring from which also the fragment 2 derived.

4) m/z = [358.04] ^2"

This fragment showed that the amine sulfated on the nitrogen and in position 3 was preceded by non sulfated uronic acid and by non sulfated but N-acetyl glucosamine. It could be indirectly deduced that a sulfate group was in position 2 of the first ring of the uronic acid.

5) m/z = [458.06] ^1-

This fragment confirmed that the third sulfate group was on the first uronic acid. 6) m/z = [616.08] ¹ -

As fragment 5), fragment 6) confirmed that on the first three rings of the tetramer there was only one sulfate group. This fragment, together with the fragment 3), confirmed that the total count of the sulfate groups was correct and the analytic method did not involve desulfation.

By combining the structural information that were acquired by the reading of the fragmentation spectrum, the structure of the examined molecule was determined:

Example 4 Analysis of a fraction of enoxaparin tetrasaccharides

Identification of oligosaccharide molecules having a sulfation degree of 1 .25.

Sample:

Tetrasaccaride mixture obtained by fragmentation with the size exclusion chromatographic (SEC) technique from a commercial sample of sodium enoxaparin. The fraction obtained by SEC chromatography was further fractioned by ion pairing chromatography. The isolated fraction, after desalting and lyophilisation, dissolved in water at a concentration of 0.5 mg/mL.

Chromatography:

Column: Kinetex C18, 2.6 m, 100 mm x 2.1 mm

Column temperature: 30 °C

Flow: 0.15 ml/min

Mobile phase A: 90% H ₂ 0, MeOH, 5 Mm dibutylammoniumacetate pH 6.7 Mobile phase B: 10% H ₂ 0, 90% MeOH, 5 Mm dibutylammoniumacetate pH 6.7

Elution: gradient

Sheath liquid: CH ₃ CN, 0.1 mL/min

Volume of injection: 20 μΙ_

Detector 1 : UV spectrophotometer, 232 nm

Detector 2: Brucker MicroTOF-Q mass spectrometer

The instrumental parameters of detector 2 were reported in example 1 . MS/MS run 1 conditions

isolation mass: 213.8 m/z

Isolation width: 8.0 m/z

Collision energy: -5eV

MSMS summation factor: 10x

MS/MS run 2 conditions isolation mass: 217.4 m/z

Isolation width: 8.0 m/z

Collision energy: -5eV

MSMS summation factor: 10x

Results:

The chromatographic profile of the analyzed mixture was shown in figure 17.

It was denoted the presence of a family of oligosaccharide species with UV absorption at 232 nm (in particular two different species, chromatographically separated at RT 28 and 32 minutes, figure 17). From a careful analysis of the MS spectrum (figure 19) it was also noted the presence of a second family of oligosaccharide species different from the first family for 18 Dalton, i.e. an additional water molecule.

The extraction of the molecular ions of the two family of species was useful for the detection of the retention times at which the two families of species elute (figure 18). In the chromatogram more oligosaccharide species, some of these co-eluting, could be then detected.

Particularly complex was the RT zone from 25 to 31 minutes.

The scope of the present example was the identification of the species in the chromatographic zone from 25 to 31 and the determination of their structure.

Identification of the species showing UV absorption at 232 nm and having chromatographic maximum at RT = 28 minutes

Phases 1 , 2 and 3

Figures 19 and 20 Assigned empirical formula (for the neutral species M)

Error calculation in ppm

m/z Calculated m/z Measured Difference in ppm

535.9857 535.9843 -2.6

Calculated Isotopic Pattern

m/z Intensity

535.9856 100.00

536.4870 32.44

536.9851 34.89

537.4862 10.00

537.9848 5.52

538.4856 1 .41

538.9846 0.53

539.4853 1 .12

Phase 4

From the composition of the empirical formula, from the UV absorption and knowing the nature of the studied product it was deduced to be a tetramer composed of two rings of uronic acid and two rings of glucosamine, in the tetrasaccharide there are five sulfate groups.

Phase 5

For the complete determination of the structural formula the MS/MS fragmentation of the ion [213.79] ^5" was carried out. The MS/MS spectrum of the peak at 28 minutes was shown in figure 22. Phase 6

The following key fragments were identified:

1 ) m/z = [117.98] ² -

The first acid was unsaturated and sulfated in position 2.

2) m/z = [135.99] ² -

The second acid was sulfated in position 2

3) m/z = [187.01 ] ^2_

It was then deduced that the reducing amine was not sulfated in position 6 neither in position 3 but as highlighted by the fragment 4:

4) m/z = [170.67] ³ -

the reducing amine was N-sulfated

5) m/z = [138.97] ¹ - oso

As the structure of the amine was already determined, the non reducing amine was sulfated in position 6

6) m/z = [168.49] ² -

Fragment 6) confirmed that the non reducing amine was sulfated on the nitrogen and the carbon in position 6.

Phase 7

7) m/z = [203.74] ^4-

There were four sulfate groups in the first three tetramer rings, it was indirectly deduced that the last ring, that is the reducing amine, had only a sulfate group. 8) m/z = [232.74] ⁴ -

The structure of fragment 8 confirmed the presence of a sulfate group on the nitrogen of the reducing amine.

9) m/z = [207.99] ^4" NHSO3 OSO3 NHSO3

It was indirectly deduced that the sulfate group was in position 2 of the unsaturated acid.

By combining the structural information that were acquired by the reading of the fragmentation spectrum, the structure of the examined molecule was determined:

Identification of the species having a chromatographic maximum at RT minutes that did not show UV absorption

Phases 1 , 2 and 3

Figures 19 and 21

Assigned empirical formula (for the neutral species M)

Error calculation in ppm

m/z Calculated m/z Measured Difference in ppm

544.9909 544.9897 -2

Calculated Isotopic Pattern

m/z Intensita

544.9909 100.00

545.4923 32.51

545.9904 35.12

546.4915 10.09

546.9901 5.60

547.4909 1 .14

547.9899 0.54

548.4906 1 .12 Phase 4

From the composition of the empirical formula and knowing the nature of the studied product it was deduced to be a tetrasaccharide composed of two rings of uronic acid and two rings of glucosamine, in the tetrasaccharide there were five sulfate groups.

Phase 5

For the complete determination of the structural formula the MS/MS fragmentation of the ion [217.40] ^5" was carried out.

The analysis of the MS/MS spectrum showed that there were two different compounds having two different fragmentation spectra. At RT from about 25 to 27 minutes it was observed the spectrum reported in figure 23. Instead at RT from about 28 and 30 minutes it was observed the fragmentation spectrum reported in figure 24. At RT from 27 and 28 minutes it was observed a fragmentation spectrum that can be identified as the sum of the spectra reported in figures 23 and 24.

Identification of the sequence of the species eluting from RT 25 and 27 minutes

Phase 6

The following key fragments were identified:

1 ) m/z = [159.48] ² -

The inner amine was sulfated at the nitrogen and at the carbon in position 6.

2) m/z = [231.33] ³ -

The inner amine was followed by a 2-sulfated uronic acid and preceded by an uronic acid.

3) m/z = [203.74] ^4-

The inner amine was followed by 2-sulfated uronic acid and preceded by a saturated 2-sulfated uronic acid.

4) m/z = [187.01 ] ^2_

The fragment indicated the presence of a 2-sulfated uronic acid followed by an amine that was not sulfated in position 6.

It was then assumed that the four sugars were arranged in the sequence uronic acid-amine-uronic acid-amine, excluding that the non-reducing end could be an amine.

5) m/z = [232.74] ⁴ -

This fragment confirmed all the preceding observations and indirectly indicated that the sulfate group was on the last amine, in position N.

The fragment confirmed the presence of N-sulfated amines.

Phase 7

7) m/z = [170.67] ³ -

-ooc CH ₂ OH

By combining the structural information that were acquired by the reading of the fragmentation spectrum, the structure of the examined molecule was determined:

Identification of the sequence of the species eluting from RT 28 to 30 minutes Phase 6

1 ) m/z = [240.02] ¹ -

The inner amine was N-sulfated and had an OH group on the carbon in position 6. 2) m/z = [180.49] ² -

The inner amine was preceded by an uronic acid having a sulfate group in position 2 3) m/z = [245.33] ³ -

The inner amine was also followed by a saturated uronic acid having a sulfate group in position 2. Having assigned three sulfate groups to the first three sugars of the molecule it was deduced that the remaining two sulfate groups were on the glucosamine.

4) m/z = [138.97] ¹ -

The reducing amine had a sulfate group on the carbon in position 6. 5) m/z = [207.99] ^4-

This fragment confirmed what was indirectly assumed in the preceding phases. Moreover, this fragment, together with fragment 3, allowed to determine that the sequence of sugars was acid-amine-acid-amine.

Phase 7

6) m/z= [170.67 ² -

By combining the structural information that were acquired by the reading of the fragmentation spectrum, the structure of the examined molecule was determined:

EXAMPLE 5 Tetrasaccharide with 1 ,6-anhydro reducing end

Identification of a "non natural" species chemically modified by the production process, having a sulfation degree =1 .25, within a mixture of isolated tetramers from a commercial sample of sodium enoxaparin.

Sample:

Mixture of tetrasaccharides obtained by fractionation with size exclusion chromatography (SEC) technique from a commercial sample of sodium enoxaparin. The fraction resulting from SEC chromatography was further fractioned by ionic pairing chromatographic technique. The isolated fraction, after desalting and lyophilisation, was dissolved in water at a concentration of 0.5 mg/mL.

Chromatography:

Column: Kinetex C18, 2.6 m, 100 mm x 2.1 mm

Column temperature: 30 °C

Flow: 0.15 ml/min

Mobile phase A: 90% H ₂ 0, 10% MeOH, 5 Mm dibutylammoniumacetate pH 6.7 Mobile phase B: 10% H ₂ 0, 90% MeOH, 5 Mm dibutylammoniumacetate pH 6.7

Elution: gradient

Sheath liquid: CH ₃ CN, 0.1 mL/min

Volume of injection: 20 μΙ_

Sample: 0.5 mg/mL in water

Detector 1 : UV spectrophotometer, 232 nm

Detector 2: Brucker MicroTOF-Q mass spectrometer

The instrumental parameters of detector 2 were reported in example 1 .

MS/MS conditions

Isolation mass: 213.8 m/z

Isolation width: 8.0 m/z

Collision energy: -5eV

MSMS summation factor: 10x Results:

The chromatographic profile of the analyzed mixture was shown in figure 25.

The MS profile showed the presence of two main peaks: the first at the retention time of 29 minutes and the second at the retention time of 33 minutes.

The scope was to determine the structure of the species eluting at RT 29 minutes.

Phases 1 , 2 and 3

The MS analysis of the peak at RT 29 min was shown in figure 26.

The analysis of the mass spectrum showed the presence of a single species (M) that is detected in different charge status (from -2 to -5) also as adduct with the used ion pairing agent (DBA=dibutylammonium).

Ions with a significant intensity of fragmentation for the loss of S0 ₃ groups or for the loss of water molecules were not observed.

The penta-charged ion (figure 27) was used for the assignment of the correct empirical formula:

Assigned empirical formula (for the neutral species M)

Error calculation in ppm

m/z Calculated m/z Measured Difference in ppm

213.7899 213.7925 12

Calculated isotopic pattern

m/z Intensity

213.7899 100.00

213.9905 32.41

214.1897 34.88

214.3901 9.99

214.5895 5.52

214.7899 1 .40

214.9895 0.53

215.1897 0.12

Phase 4

1 ) From the number of nitrogen and carbon atoms it was deduced to be a tetramer formed by two uronic acid rings and two glucosamine rings 2) From the number of sulphur atoms it was assumed that the tetramer was pentasulfate;

3) From the UV information it was assumed that the non reducing end was saturated (there was no presence of the double bond on the uronic acid that is typical of enoxaparin)

4) It was a "non-natural" tetramer, i.e. that was not present in sodium heparin. In fact, from the empirical formula it was observed the lack of a water molecule (with respect to the empirical formula of the natural tetramer) and it could be assumed that it related to a species whose reducing end was protected in the form of 1 ,6- anhydro

Phase 5

For the complete determination of the structural formula the MS/MS fragmentation of the ion [213.79] ^5" was carried out. The fragmentation spectrum was shown in figure 28.

Phase 6

The following key fragments were identified:

ln the molecule there was the ion that characterized the species 1 ,6 anhyd terminal end of the molecule was then 1 ,6-anhydro-N-sulfate glucosamine.

2) m/z= [247.50] ^2- and m/z= [164.66] ^3-

The ring preceding the terminal amine was an uronic acid having a sulfate group on the carbon in position 2.

The inner amine was a N,6-sulfate amine.

4) m/z= [138.97] ^1"

This fragment confirmed the presence of the sulfated amine in position 6.

5) m/z= [203.49] ^4-

This fragment rom the empirical formula it was underlined that the examined species was a pentasulfate tetramer, with four sulfate groups assigned in the sequence of three rings of which the fragments were detected, then it could be deduced that a S0 ₃ H group was on the first ring of uronic acid, which was saturated as indicated by the UV spectrum wherein there was no absorption related to double bonds.

By combining the structural information that were acquired by the reading of the fragmentation spectrum, the structure of the examined molecule was determined:

EXAMPLE 6

Analysis of a purified oligosaccharide having a sulfation degree of 1 .5 isolated from a sample of commercial sodium enoxaparin. Example of fragmentation wherein the fragmented ion corresponds to the ion having z = S-1 .

Sample:

Mixture of tetrasaccharides obtained by fractionation with size exclusion chromatography (SEC) technique from a commercial sample of sodium enoxaparin. The fraction resulting from SEC chromatography was further fractioned by ionic pairing chromatography. The isolated fraction, after desalting and lyophilisation, was dissolved in water at a concentration of 0.5 mg/mL.

Chromatography:

The chromatographic conditions are reported in example 4.

Detector 1 : UV spectrophotometer, 232 nm

Detector 2: Brucker MicroTOF-Q mass spectrometer

The parameters of detector 2 are reported in example 1 .

Isolation mass: 229.8 m/z

Isolation width: 8.0 m/z

Collision energy: -5eV

MSMS summation factor: 10x

Results:

The chromatographic profile of the analyzed mixture was shown in figure 29.

Phases 1 , 2 and 3

The MS chromatographic profile showed the presence of the main peak at RT 42 minutes.

The MS analysis of the peak at RT 42 min was shown in figure 30.

The analysis of the mass spectrum showed the presence of a species (M) that is detected in different charge status (from -2 to -5) and as adduct with the used ion pairing agent (DBA=dibutylammonium).

Ions with a significant intensity of fragmentation for the loss of S0 ₃ groups or for the loss of water molecules were not observed.

The ion having three negative charges (figure 31 ) was used for the assignment of the correct empirical formula:

Assigned empirical formula (for the neutral species M)

Error calculation in ppm

m/z Calculated m/z Measured Difference in ppm

383.6403 383.6378 7

Calculated isotopic pattern

m/z Intensity

383.6403 100.00

383.9745 33.35

384.3065 40.32

384.6406 1 1 .99

384.9728 7.46

385.3067 1 .99

385.6393 0.85

385.9731 0,20

Phase 4

1 ) From the number of nitrogen and carbon atoms it was deduced to be a tetramer (presence of two uronic acid rings and two glucosamine rings)

2) From the number of sulphur atoms it was deduced that the tetramer is hexasulfate;

3) From the information of the UV spectrum it was deduced that the non reducing end was unsaturated (there was the presence of double bond on uronic acid that is typical of enoxaparin)

6 groups R = S0 ₃ H; 2 groups R = H

Phase 5

The MS/MS spectrum was obtained starting from the ion [M-5H] ^5" notwithstanding the examined species was an hexasulfate tetramer (figures 30 and 32).

Phase 6

1 ) m/z = [117.98] ² -

-ooc

The first acid of the tetramer was unsaturated and with a sulfate group in position 2. 2) m/z = [159.48] ² -

The inner amine was sulfated at the nitrogen and carbon in position 6.

3) m/z = [191.32] ³ -

The first dimer contained three sulfate groups overall which, on the basis of the information regarding fragments 1 ) and 2), were on the acid in position 2 and on the amine, at the nitrogen and at position 6. From this fragment it was deduced that three sulfate groups were present on the second dimer.

4) m/z = [135.99] ² -

-ooc

The second acid of the tetramer had a sulfate in position 2.

Being assigned four out of the six sulfate groups on the tetramer, it was deduced that the last amine had two sulfate groups.

The last amine had a sulfate group in position 6 and no sulfate group in position 3. The reducing amine was then sulfated at the nitrogen and at the carbon in position 6.

6) m/z = [197.32] ³ -

The fragment confirmed the presence of sulfate groups on the amine nitrogen and on the carbon in position 6.

The fragments 3) and 6) indicated that three sulfate groups were present on the first dimer and three sulfate groups on the second dimer.

Phase 7

The fragment confirmed the presence of N-sulfated amines.

8) m/z = [138.97] ¹ -

The fragment confirmed the presence of sulfated amines on the carbon in position 6.

9) m/z = [168.49] ² -

The fragment confirmed the presence of amines with only two sulfate groups as substituents.

By combining the structural information that were acquired by the reading of the fragmentation spectrum, the structure of the examined molecule was determined: