Polyglycerol derivative and a method for manufacturing the same

Abstract

It is provided a polyglycerol derivative, comprising a dendritic polyglycerol backbone and at least one substituent in the nature of a covalently bound negatively charged group chosen from the group consisting of sulfates, sulfonates, phosphates, phosphonates, bisphosphonates, carboxylates and combinations thereof. The substituent is bound to the polyglycerol backbone via a linker, the linker being chosen from the group consisting of moieties being or comprising a carbamate group, an ester group, an orthoester group, an amide group, a disulfide bridge group, an acetal group, an imine group and combinations thereof.

Claims

1. A polyglycerol derivative, comprising a dendritic polyglycerol backbone and at least one negatively charged substituent selected from the group consisting of sulfates, sulfonates, phosphates, phosphonates, bisphosphonates, carboxylates, and a combination thereof, wherein the at least one negatively charged substituent is bonded to the polyglycerol backbone via a linker, wherein the linker is selected from the group consisting of a moiety comprising a carbamate group, a moiety comprising an ester group, a moiety comprising an orthoester group, a moiety comprising an amide group, a moiety comprising a disulfide bridge group, a moiety comprising an acetal group, a moiety comprising an imine group, a carbamate group, an ester group, an orthoester group, an amide group, a disulfide bridge group, an acetal group, an imine group, and a combination thereof, wherein the linker comprises 1 to 10 carbon atoms, and wherein a plurality of hydroxyl groups of the polyglycerol backbone is substituted by at least one of the following substituents R, the counter-ion being optionally different from Na+: ##STR00002##

2. The polyglycerol derivative according to claim 1, wherein the at least one negatively charged substituent is a sulfate.

3. The polyglycerol derivative according to claim 1, wherein the polyglycerol backbone has a degree of substitution between 10 and 100%.

4. The polyglycerol derivative according to claim 1, wherein the linker is an ester group or the linker comprises an ester group.

5. The polyglycerol derivative according to claim 1, wherein the linker is a substituted or non-substituted hydrocarbon residue that is optionally interrupted by at least one N, O and/or S atom and that comprises a carbamate group, an ester group, an orthoester group, an amide group, a disulfide bridge group, an acetal group, an imine group, or a combination thereof.

6. The polyglycerol derivative according to claim 5, wherein the hydrocarbon residue is a substituted or non-substituted C.sub.1-C.sub.10 alkyl that comprises a carbamate group, an ester group, an orthoester group, an amide group, a disulfide bridge group, an acetal group, an imine group, and a combination thereof, and wherein the hydrocarbon residue is optionally interrupted by at least one N, O and/or S atom.

7. A medicament comprising a pharmaceutically active substance and the polyglycerol derivative according to claim 1 as a carrier for the pharmaceutically active substance.

8. A medicament according to claim 7 for inhibiting the complement system of an organism and/or for inhibiting L-selectin binding to its natural receptor.

9. A medicament according to claim 7 for treating an inflammatory disease.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Further aspects and details of the invention will be explained with respect to figures and exemplary embodiments.

(2) FIG. 1A shows the chemical structure of the polyglycerol backbone of an exemplary dendritic polyglycerol compound.

(3) FIG. 1B shows a second possibility to depict the chemical structure of the polyglycerol of FIG. 1A.

(4) FIG. 2A shows a reaction scheme of the reaction of a polyglycerol to a polyglycerol sulfate according to prior art.

(5) FIG. 2B shows the general reaction scheme of the reaction of polyglycerol to a substituted polyglycerol derivative.

(6) FIG. 3A shows a first residue R that can be used as substituent for substituting polyglycerol (cf. FIG. 2B).

(7) FIG. 3B shows a second residue R that can be used as substituent for substituting polyglycerol (cf. FIG. 2B).

(8) FIG. 3C shows a third residue R that can be used as substituent for substituting polyglycerol (cf. FIG. 2B).

(9) FIG. 4A shows a reaction scheme for the reaction of polyglycerol to a polyglycerol derivative bearing residue R of FIG. 3A;

(10) FIG. 4B shows a reaction scheme for the reaction of polyglycerol to a polyglycerol derivative bearing residue R of FIG. 3B.

(11) FIG. 4C shows a reaction scheme for the reaction of polyglycerol to a polyglycerol derivative bearing residue R of FIG. 3C.

(12) FIG. 5 shows concentration dependent coagulation times of different polyglycerol compounds.

(13) FIG. 6 shows levels of concentration dependent complement activation of different polyglycerol compounds.

(14) FIG. 7A shows a reaction scheme of the degradation of a polyglycerol derivative bearing residue R of FIG. 3A.

(15) FIG. 7B shows a reaction scheme of the degradation of a polyglycerol derivative bearing residue R of FIG. 3B.

(16) FIG. 7C shows a reaction scheme of the degradation of a polyglycerol derivative bearing residue R of FIG. 3C.

(17) FIG. 8A shows a first degradation profile of different polyglycerol derivatives.

(18) FIG. 8B shows a second degradation profile of different polyglycerol derivatives.

(19) FIG. 8C shows a third degradation profile of different polyglycerol derivatives.

(20) FIG. 8D shows a forth degradation profile of polyglycerol sulfate.

(21) FIG. 9 shows the chemical structure of dPG-DMPTACN-TPS.

(22) FIG. 10 shows the chemical structure of dPGS-DMPTACN.

DETAILED DESCRIPTION

(23) FIG. 1A is a schematic depiction of the polyglycerol backbone of an exemplary (idealized) dendritic polyglycerol (dPG). Such dPG can be used as starting point for preparing polyglycerol compounds.

(24) FIG. 1B shows an abbreviated possibility to depict the polyglycerol structure of FIG. 1A. For sake of simplification only, the abbreviated depiction of FIG. 1B shows only two hydroxyl groups, although the polyglycerol bears significantly more hydroxyl groups (cf. FIG. 1A).

(25) FIG. 2A shows the reaction scheme of sulfation of dendritic polyglycerol (dPG) 1 to dendritic polyglycerol sulfate (dPGS) 2. dPGS can also be denoted as stable polyglycerol sulfate. The sulfation was carried out by sulfur trioxide pyridine (SO.sub.3.Py) as sulfation agent. This reaction is well known from prior art.

(26) FIG. 2B shows the reaction scheme of an exemplary embodiment for producing polyglycerol compounds that bear a linker are which differs from substitutes known from prior art. This reaction can be carried out, e.g., in 3 different ways A), B), C) that will be explained in more detail with respect to FIGS. 4A to 4C. The reaction products can be denoted as biodegradable dPGS.

(27) FIG. 3A shows a residue R that can be the substituent of the polyglycerol compound produced according to the reaction scheme of FIG. 2B, wherein the resulting polyglycerol compound can be denoted as dPG-amidoglyceryl succinyl sulfate (dPG-ASuS).

(28) FIG. 3B shows a residue R that can be the substituent of the polyglycerol compound produced according to the reaction scheme of FIG. 2B, wherein the resulting polyglycerol compound can be denoted as dPG-thioglyceryl pentanoatyl sulfate (dPG-TPS).

(29) FIG. 3C shows a residue R that can be the substituent of the polyglycerol compound produced according to the reaction scheme of FIG. 2B, wherein the resulting polyglycerol compound can be denoted as dPG-thioglyceryl methyl-propanoatyl sulfate (dPG-TMPS).

(30) FIGS. 4A to 8D will now be explained in more detail referring to exemplary embodiments.

(31) Example: Synthesis and Characterization of Degradable Dendritic Polysulfates

(32) With its multiple hydroxyl groups, dendritic polyglycerol (dPG) allows a versatile derivatization and the introduction of different functional groups. By implementation of cleavable linkers and subsequent sulfation, shell degradable compounds with a high anti-inflammatory activity could be developed, which lead to neutral dPG and multiple sulfated linkers of low molecular weight after full cleavage. To develop compounds with different degradation patterns, linkers varying in hydrophobicity, length and flexibility with hydrolytically or enzymatically cleavable groups were introduced to the dPG-backbone and investigated. Since the molecular weight of the polymer and the number of sulfate groups have a significant influence on the L-selectin binding [2] and hence anti-inflammatory activity, three biodegradable compounds with comparable molecular weights and numbers of sulfate groups were synthesized (cf. FIG. 2B in connection to FIGS. 3A, 3B and 3B; the individual reaction schemes are depicted in FIGS. 4A, 4B and 4C).

(33) Dendritic polyglycerol as the core scaffold was synthesized with a molecular weight of 5,100 g mol.sup.1, low PDI (1.6) and degree of branching of 60% via a controlled living anionic ring-opening multibranching polymerization (ROMBP) of glycidol by slow monomer addition on partially deprotonated polyvalent 1,1,1-tris(hydroxyl-methyl) propane (TMP) as the starter [9].

(34) dPG-Amidoglyceryl Succinyl Sulfate (dPG-ASuS)

(35) dPG-amidoglyceryl succinyl sulfate (dPG-ASuS) (3) (see FIG. 3A) which contains a polar but comparably short and rigid linker including an ester and an amide functionality, offers the opportunity of acidic as well as enzymatic cleavage and was synthesized over three steps (FIG. 4A). Dendritic polyglycerol (1) was reacted with succinic anhydride for two days at room temperature in pyridine, acting as solvent and base to deprotonate dPG and initiate the ring opening of succinic anhydride. Ultrafiltration under addition of sodium chloride to avoid the precipitation of dPG-succinic acid gave dPG sodium succinate (dPG-Su, 6) in 75% yield with a full conversion (fully functionalized). dPG-amidoglyceryl succinate (dPG-ASu) (7) was prepared by a modified peptide coupling procedure from Pickaert et al. [10] with aminoglycerol, 1-ethyl-3-(3-dimethylamino-propyl)carbodiimide hydrochloride (EDC.HCl), and a catalytic amount of 4-(dimethyl-amino)pyridine (DMAP). Dialysis yielded the amide 7 in 85% with a degree of functionalization (dF) of 95%. Subsequent sulfation with SO.sub.3.pyridine complex [3] gave dPG-amidoglyceryl succinyl sulfate (ASuS, 3) in 71% yield with a dF=63% as a highly water soluble colourless product.

(36) dPG-Thioglyceryl Pentanoatyl Sulfate (dPG-TPS)

(37) dPG-thioglyceryl pentanoatyl sulfate (dPG-TPS, 4) (see FIG. 3B) containing a very long, flexible and very hydrophobic linker including an ester functionality was synthesized over three steps (FIG. 4B). Dendritic polyglycerol (1) was reacted with an equimolar amount of pentenoyl chloride and an excess of triethylamine as base in DMF. dPG-pentenoate (8) was obtained in 92% yield with a dF=84%, as determined by .sup.1H-NMR. Subsequent radical thiolene reaction of 8 with thioglycerol was performed under UV-radiation for 90 minutes at room temperature with 2,2-dimethoxy-2-phenylacetophenone (DMPA) as the photoinitiator. dPG-thioglyceryl pentanoate (dPG-TP, 9) was obtained in 77% yield with complete conversion as determined by .sup.1H-NMR due to absent signals of the allyl group at 4.71-5.27 ppm and 5.54-5.96 ppm. Further sulfation of 9 with SO.sub.3.pyridine complex gave dPG-TPS (4) in 92% yield with a dF=67%.

(38) dPG-Thioglyceryl Methylpropanoatyl Sulfate (dPG-TMPS)

(39) The implementation of a long, flexible and relatively hydrophobic linker containing a carbamate and an ester functionality was accomplished by the synthesis of dPG-thioglyceryl methylpropanoatyl sulfate (dPG-TMPS, 5) (see FIG. 3C). Whereas both, the ester and carbamate group represent acid labile moieties, the ester functionality could also be cleaved by enzymes like carboxylesterases. dPG-TMPS was synthesized over three steps with 50% of the linker (FIG. 4C). The carbamate 10 was prepared following a modified procedure from Bryant et al. [11]. In analogy dendritic polyglycerol (1) was reacted with 0.9 eq. of 2-isocyanato-ethyl methacrylate to give dPG-methacrylate (10) with a dF=50%, as determined by .sup.1H-NMR. To avoid gel formation butyl-hydroxytoluene (BHT) was added and direct conversion of 10 to the corresponding glycidyl thioether by Michael-addition of thioglycerol was realized. dPG-thioglyceryl methyl propanoate (dPG-TMP, 11) was isolated in 62% yield over two steps due to partial polymerization of the intermediate 10. Complete conversion was determined by .sup.1H-NMR due to the vanished signals of the methacrylate group at 5.55-5.57 ppm and 6.08-6.10 ppm. Subsequent sulfation with SO.sub.3.pyridine gave dPG-thioglyceryl methylpropanoatyl sulfate (dPG-TMPS, 5) in 84% yield with a dF=78%.

(40) Dendritic Polyglycerol Sulfate (dPGS)

(41) Dendritic polyglycerol sulfate (dPGS, 2) was prepared as non-degradable analog to compare its biocompatibility and anti-inflammatory activity with the synthesized shell degradable compounds. Sulfation of dendritic polyglycerol with SO.sub.3 pyridine gave dPGS with a dF=91% in 68% yield.

(42) .sup.1H-NMR analysis confirmed the intact structure of the degradable compounds after each reaction. Analytic data of the prepared polysulfates are summarized in Table 1.

(43) TABLE-US-00001 TABLE 1 Specification of the prepared polysulfates 2-5 and the dPG scaffold 1. M.sub.n dS d.sub.h SD -potential SD IC.sub.50 Polymer # [g mol.sup.1] NS [%] [nm] PDI [mV] [nM] dPG 1 5,100 4.0 0.4 0.22 dPGS 2 11,500 63 91 5.8 0.6 0.28 18.6 2.3 0.65 dPG-ASuS 3 25,500 84 63 6.4 0.6 0.37 35.4 3.0 1.3 dPG-TPS 4 25,900 85 67 7.4 0.7 0.23 27.4 4.6 0.65 dPG-TMPS 5 22,600 81 78 5.8 0.6 0.21 25.7 4.7 0.2 M.sub.n = Number average molecular weight calculated from the dF. NS = Number of sulfate groups per polymer. DS = Degree of sulfation determined by .sup.1H-NMR and elemental analysis. d.sub.h = Hydrodynamic diameter (mean standard deviation (SD)) by DLS in PBS (pH 7.4) from the size distribution by volume. PDI = Polydispersity index (DLS). -potential (mean SD) in 10 mM phosphate buffer (pH 7.4). IC.sub.50 values describe the compound concentration required to inhibit ligand binding of L-selectin functionalized Au nanoparticles.

(44) Sulfated shell degradable polymers were synthesized with comparable molecular weights from 22,600 g mol.sup.1 up to 25,900 g mol.sup.1 and sizes between 5.8 nm and 7.4 nm. Sulfation of the precursors yielded highly water soluble anionic polymers with a degree of functionalization over 63% and similar numbers of sulfate groups ranging from 81 to 85 as determined by .sup.1H-NMR and elemental analysis. -potential measurements showed surface charges between 26 mV and 35 mV. The slightly lower -potential of dPG-ASuS (3) is probably caused by not fully functionalized succinic acid residues of the precursor, since dPG carboxylates were shown to exhibit a more negative surface charge compared to sulfates [3]. Using the same polyglycerol scaffold, non-degradable dPGS (2) in contrast was prepared with a lower molecular weight of 11,600 g mol.sup.1 and a size of 5.8 nm. With a degree of sulfation of 91%, dPGS contained 63 functional groups and a higher -potential of 18.6 mV compared to the shell degradable compounds.

(45) L-Selectin Inhibition

(46) The anti-inflammatory potential of the prepared polysulfates was estimated by the quantification of L-selectin inhibition, determined by a concentration dependent, competitive SPR-based binding assay, which was previously described in detail [12]. In short, L-selectin-IgG chimeras were coated on Au nanoparticles to imitate L-selectin expressing leukocytes. L-selectin ligands including Sialyl-Lewis.sup.X (sLe.sup.X, 20 mol %) and sulfated tyrosine (sTyr, 5 mol %), were immobilized on the surface of a BIAcore sensor chip to mimic the endothelium surface. Then the L-selectin coated Au nanoparticles were passed over the sensor chip resulting in a binding signal, which was set to 100% and served as reference. Subsequently, L-selectin functionalized nanoparticles were pre-incubated with a potential inhibitor and passed over the

(47) surface of the chip. The inhibitor concentration dependent reduction of the binding signal was recorded and calculated as relative binding of the reference. The concentration that caused a 50% reduced binding was quoted as IC.sub.50 value. Each inhibitor concentration was applied at least in triplicate. This reproducible in vitro experiment mimics the leukocyte binding to endothelial surface in vivo in the presence of a potential inhibitor in good approximation. IC.sub.50 values of the prepared polysulfates are shown in Table 1.

(48) For all degradable compounds IC.sub.50 values in the low nano to high picomolar range were found, comparable to the non-degradable dendritic polyglycerol sulfate, which confirms their high anti-inflammatory activity. Whereas dPG-TPS (4) and dPGS (2) showed similar IC.sub.50 values of 0.65 nM, dPG-ASuS (3) binds L-selectin one order of magnitude less (1.3 nM). However, the lower (but still very suited) IC.sub.50 value of dPG-TMPS (5) of 0.2 nM demonstrates that the L-selectin binding affinity is not only dependent on the number of sulfate groups and molecular weight but also depends on the nature of the linker, since all degradable compounds were prepared with comparable characteristics regarding their molecular weight and number of functional groups.

(49) Blood Coagulation

(50) To confirm the biocompatibility of the shell degradable polysulfates their influence on blood coagulation and complement activation was investigated and compared to non-degradable dPGS. The anticoagulant activity of the prepared polymers was evaluated by activated partial thromboplastin time (APTT) clotting assay using citrated human platelet poor plasma (PPP). Fresh PPP was incubated with the sulfates 2-5 or neutral dPG (1), in concentrations between 100 nM and 1 M. Heparin was used as standard compound and was added in concentrations of 10 nM and 50 nM. After incubation, the clotting time was determined and compared to untreated control which was set to 100%. As depicted in FIG. 5, prepared polysulfates showed a negligible influence on the coagulation time at a concentration of 100 nM compared to heparin, which already exhibits a slightly higher anti-coagulant activity at 10 nM.

(51) However, with increasing concentrations all polysulfates were found to prolong the APTT, whereas the neutral dendritic polyglycerol did not influence the blood coagulation at all. At 500 nM the blood clot times were almost doubled whereas a concentration of 1 M lead to a threefold prolonged APTT. The comparable clotting time patterns of all polysulfates indicate that the linker itself does not have a significant influence on the blood coagulation. Still, the prepared compounds showed a much lower anticoagulant activity even at the highest concentration (1 M) compared to heparin which lead to a 3.5-fold prolonged coagulation time at a concentration of 50 nM.

(52) Complement Activation

(53) Since the activation of the complement is known to cause severe problems including multiple organ dysfunctions or septic shocks due to the release of pro-inflammatory complement proteins, the anaphylatoxins C3a and C5a [13], interactions of the prepared polysulfates with the complement system have to be taken into account. The complement activation in human serum was tested for the immunoglobin dependent classical pathway with an ELISA-based method. Serum samples were incubated with the test compounds in concentrations of 50 nM to 2.5 M.

(54) The level of complement activity is stated as percentage of the untreated serum control in FIG. 6. Heparin was used as reference. Complement activity is given as percent of the untreated control. All tested compounds were found to decrease the complement activation. Whereas neutral dPG (1) showed only slightly reduced activity even at high concentrations of around 80% of the control, the sulfated compounds exhibited a much stronger reduction of the complement activation in a concentration dependent manner. dPGS (2) showed similar activities compared to heparin up to a concentration of 1 M and a higher complement activation at 2.5 M. For dPG-ASuS (3) a comparable activity was found concentrations of 50 nM and 250 nM, but enhanced reduction to 30% and 3% activity at concentrations of 500 nM and 1 respectively. Almost full inhibition of complement was observed at a concentration of 2.5 M. Surprisingly, dPG-TPS (4) and dPG-TMPS (5) performed even better, only 2.5% complement activity was already found at a concentration of 250 nM and almost total inhibition at 500 nM. These findings might be related to the longer and more flexible linker between the sulfate groups and the polyglycerol backbone of the degradable polysulfates compared to dPGS and heparin, which could increase the probability of interactions with proteins of the complement cascade. Moreover, both dPG-TMPS (5) and dPG-TPS (4) contain a thioether moiety within their linker which also seems to play an important role in protein targeting. The results clearly indicate that the prepared biodegrable polysulfates can be used as potent complement inhibitors of the classical pathway for the treatment of inflammation associated diseases.

(55) In Vitro Degradation Studies

(56) Degradation of the new polysulfates was investigated over a period of 4 weeks in PBS buffer under neutral (pH 7.4) and acidic (pH 5.0) conditions. Enzymatic degradation was analyzed with carboxylesterase I (CES I). Latter is found in high concentrations in the liver and was used due to the known accumulation of dPGS in hepatic Kupffer cells [7, 8]. Since the synthesized shell degradable polymers contain linkers that exhibit esters and additional amide or carbamate moieties, hydrolytic or enzymatic cleavage of different functional groups could take place.

(57) However, as evident from .sup.13C-NMR analysis only ester hydrolysis proceeded, leading to sulfated carboxylic acids of low molecular weight and respective high molecular weight alcohols.

(58) FIGS. 7A to 7C show reaction schemes of the degradation of the three polyglycerol compounds dPG-ASuS (3), dPG-TPS (4) and dPG-TMPS (5).

(59) Degradation of the polysulfates was monitored by .sup.1H-NMR or by elemental analysis in case of dPGS. Occurring spikes in the .sup.1H-NMR spectra in the area of broad polymer and linker signals as well as the presence of new peaks indicated hydrolysis of the compounds. In case of dPG-ASuS (3) sharp peaks arised over time between 3.59 ppm and 4.30 ppm while a new signal at 2.77 ppm appeared due to the -CH.sub.2 group of the formed acid. For dPG-TMPS (5) spikes between 3.19 ppm and 4.71 ppm were observed as well as broadening of the signal between 2.53 ppm and 3.04 ppm, latter peaks caused by the methyl-neighboring CH-group and the two methylen bridges next to the thioether. The appearance of multiple CH.sub.3 signals indicates the presence of different methyl species due to the cleavage of the ester. In comparison, degradation of the carbamate function would show much less influence on the methyl group. For dPG-TPS (4) sharp signals between 2.18 ppm and 4.75 ppm were found over time as well as shrinking of the peak between 4.13 ppm and 4.75 ppm. Ester cleavage was also quantified by .sup.1H-NMR.

(60) In case of dPG-TMPS (5) and dPG-TPS (4) the integral ratios of areas that remained the same were compared to that of signals which decreased over time. For dPG-TMPS (5) the integral of the methyl group at 1.24 ppm was determined relative to that from 4.50 ppm to 3.28 ppm. In case of dPG-TPS (4) the integral between 2.07 ppm and 1.21 ppm was compared to that of the diminishing peak between 4.57 ppm and 4.13 ppm. Since dPG-ASuS (3) showed no significant decrease of any signal, the integral of the arising peak at 2.77 ppm was subtracted from that of the peak caused by the CH.sub.2 groups between the ester and amide moiety. The integral ratio of the respective non-treated polymer was set to 100%, and changes in the ratio were calculated as %, directly giving the content of ester. Because the cleavage of sulfate groups in case of dPGS (2) would not show significant differences in the .sup.1H-NMR spectra and hence no precise quantification of degradation would be possible, the sulfur content was determined by elemental analysis after purification via GPC column.

(61) The stability of the polysulfates at different conditions is depicted in FIGS. 8A to 8D. As determined by elemental analysis after purification by size exclusion chromatography (SEC), dPGS (2) is stable over 4 weeks (FIG. 8D). In contrast, for dPG-ASuS (3) slow degradation was observed, whereas dPG-TMPS (5) and dPG-TPS (4) were found to undergo much faster decomposition. In case of dPG-ASuS (3) around 80% of the esters remained over 4 weeks, while approximately half of the linkers were already cleaved within 1 week in case of dPG-TMPS (5) and dPG-TPS (4). However, complete decomposition of the ester linkages was not observed for any of the cleavable polysulfates. Surprisingly, no relevant influence of the pH value on the degradation process was obvious, although esters are known to underlie acidic rather than neutral cleavage. Moreover, the data shown indicate that carboxylesterase I does not contribute to the degradation of the compounds which might be due to electrostatic interactions of the enzyme with sulfate groups of the substrate or might be caused by the tightly packed structure of the polymers that may shield the enzyme from acting. Nevertheless, the prepared shell cleavable polysulfates can be considered as potent degradable dPGS-analogs and hence are interesting scaffolds for long term treatment of inflammation-related diseases.

(62) Summarizing, the synthesis of highly water soluble shell degradable polysulfates via implementation of hydrolytically or enzymatically cleavable linkers into a biocompatible dendritic polyglycerol (dPG) backbone and further sulfation is disclosed. The compounds were prepared with similar molecular weights as well as numbers of sulfate groups and contained either only ester groups (dPG-thioglyceryl pentanoatyl sulfate, dPG-TPS) or additional amide (dPG-amidoglyceryl succinyl sulfate, dPG-ASuS) and carbamate functionalities (dPG-thioglyceryl methylpropanoatyl sulfate, dPG-TMPS). All polymers were investigated regarding their degradation behavior, blood coagulation properties, complement activation and L-selectin binding in vitro. Dendritic polyglycerol sulfate (dPGS) was used as stable analog for comparison. Very slow degradation was found for dPG-ASuS whereas in case of dPG-TPS and dPG-TMPS a much faster decomposition was observed under all test conditions compared to the cleavage-resistant dPGS. As determined by an APTT assay, all prepared polysulfates showed a comparable clotting time pattern similar to dPGS and only a slight influence on blood coagulation up to a concentration of 100 nM. In marked contrast complement activation was strongly influenced by dPG-TMPS and dPG-TPS. Total inhibition was observed at nanomolar concentrations. A further anti-inflammatory activity was determined via a competitive SPR-based L-selectin binding assay. For all degradable compounds IC.sub.50 values in the low nano to high picomolar range were found, comparable to dPGS. The binding inhibition increased in the order dPG-ASuS<dPG-TPS=dPGS<dPG-TMPS confirming the high anti-inflammatory activity of the newly prepared compounds. The studies show that the shell degradable polysulfates are potent cleavable dPGS-analogs and hence are suited compounds for the long term treatment of chronic inflammations or can be applicable in tissue engineering due to their low anti-coagulant and high anti-inflammatory properties. As remarkably strong inhibitors of the complement, theses scaffolds can also be considered as a new class of anti-complement therapeutics with desirable pharmacologic properties to prevent the progress of tissue damage within inflammatory diseases.

(63) In the following, more details on the performed experiments are disclosed.

(64) Reactions including air or moisture sensitive substances were carried out under argon atmosphere using flame-dried glassware and anhydrous solvents. Chemicals were reagent grade and were used without further purification. Dialysis was performed in benzoylated cellulose tubings (molecular weight cut off (MWCO): 2000 g mol.sup.1) changing the solvent at least 8 times over a period of 72 h. Ultrafiltration was performed in solvent-resistant stirred cells with PLAC regenerated cellulose membranes (MWCO 1000 g mol.sup.1). .sup.1H- and .sup.13C-NMR spectra were recorded on a Jeol ECX 400 spectrometer operating at 400 and 101 MHz or on a Bruker Biospin Avance 700 spectrometer operating at 700 and 176 MHz. Chemical shifts were reported in ppm using the deuterated solvent peak as the internal standard (CDCl.sub.3: (.sup.1H)=7.26 ppm, (.sup.13C)=77.16 ppm; CD.sub.3OD: (.sup.1H)=3.31 ppm, (.sup.13C)=49.00 ppm; D.sub.2O: (.sup.1H)=4.79 ppm). .sup.13C-Spectra were broadband proton decoupled. Multiplicity of NMR-signals is listed as s (singlet) or m (multiplet). Signal assignment was partially performed by two-dimensional NMR spectra (COSY, HMQC, HMBC). IR measurements were recorded on a Nicolet Avatar 320 FT-IR equipped with a DTGS detector from 4000 to 650 cm.sup.1 and evaluated with the software EZ OMNIC ESP. Wavenumbers .sub.max were reported in cm.sup.1, the intensities of absorption bands were assigned as strong (s), medium (m), and weak (w). Elemental analysis was performed on a VARIO EL III instrument using sulfanilic acid as standard. DLS and -potential measurements were carried out on a Zetasizer Nano ZS equipped with a 4 mW HeNe laser (, =633 nm, NIBS) operating with a 173 scattering angle (backscatter). Particle size was measured in UV-transparent disposable cuvettes (8.5 mm) at 25 C. Samples were dissolved in Dulbecco's PBS (DPBS, 150 mM, lx, without Ca.sup.2+, Mg.sup.2+, pH=7.4) at a concentration of 2 mg ml.sup.1. The solutions were filtered once trough a 0.45 m PTFE syringe filter and twice trough a 0.2 m PTFE syringe filter. Samples were equilibrated for 60 seconds at 25 C.; subsequently, the measurement was performed with 15 scans per sample. The stated values are the mean of at least 15 independent measurements. For -potential measurements samples were dissolved in phosphate buffer (PB, 10 mM, pH 7.4) at a concentration of 1 mg ml.sup.1. The solutions were filtered once through a 0.2 m Cellulose acetate syringe filter and measured by applying an electric field across the polymer at 25 C. in folded DTS 1060 capillary cells. Data evaluation was performed with Malvern Zetasizer Software 6.12. The stated values and standard deviations are the mean of at least five independent measurements with 15 scans each and are based on the Smoluchowski model. UV irradiation was performed using an USHIO super high pressure mercury lamp (USH 102d; 100 W, 0.12 Amps) without a filter.

(65) Degradation Studies

(66) Degradation studies were performed in PBS buffer with pH 5.0 or pH 7.4, respectively over 4 weeks at 37 C. with shaking at 200 rpm using total sample concentrations of 2 mol ml.sup.1. pH values were kept constant over time by addition of NaOH if needed. For enzymatic degradation studies samples were dissolved in PBS buffer with pH 7.4 and Carboxylesterase I (CES I, isoform b, human, 1000 U ml.sup.1) was added to a total concentration of 1.86 units ml.sup.1. Every 4 days 50% of the enzyme activity was added since 50% of the activity is lost after 4 days as shown in an enzyme activity assay. After each time point (7, 14, 21, 28 days) an aliquot of 500 L was removed, the pH value was adjusted to 7.0 and the solvent was evaporated. The residue was dissolved in D.sub.2O and the solution was filtered through a 0.2 m PTFE syringe filter (VWR). In case of enzymatic degradation additionally bis(4-nitrophenyl)phosphoric acid (BNPP) was added to a total concentration of 1 mM to inhibit the enzyme. The progress of degradation was either determined by .sup.1H-NMR spectra as content of esters or by elemental analysis of the sulfur content (see supporting information). Degradation studies were performed as duplicates and samples were stored at 10 C.

(67) Synthesis

(68) Dendritic Polyglycerol (dPG) with a molecular weight of M.sub.n=5,100 g mol.sup.1 bearing in average 69 hydroxyl groups per molecule, a PDI<1.8 and a degree of branching of 60% was synthesized according to literature via anionic ring opening polymerization of glycidol using 1,1,1-tris(hydroxymethyl)propane (TMP) as the starter [9]. Molecular weights of further derivatives were calculated from the particular conversion as determined by .sup.1H-NMR spectroscopy or from the sulfur content obtained by elemental analysis.

(69) General Procedure for Sulfation

(70) Sulfated polymers were prepared according to Turk et al. [1]. In brief, the polyol was dissolved in dry DMF, heated to 60 C. and a solution of SO.sub.3.pyridine complex (1.2 eq. per OH group) in dry DMF was added dropwise over 2 hours. After addition the mixture was stirred over night (o.n.) at 60 C. Then aq. NaOH (2 M, 0.9 eq. per SO.sub.3.pyridine complex) and water were added, and the pH was adjusted to 7 by addition of aq. NaOH (10%). The solvent was evaporated and the crude product was subjected to ultrafiltration in water. After freeze-drying the product was obtained as a colorless solid.

(71) Dendritic Polyglycerol Sulfate (dPGS) (2)

(72) dPG sulfate was synthesized according to the general procedure for sulfation by applying dendritic polyglycerol (5,100 g mol.sup.1, 69 OH groups). The compound was obtained as colorless solid after freeze-drying. M.sub.n (dPG-core)=5,100 g mol.sup.1, M.sub.n=11,600 g mol.sup.1, dF=91%, yield: 74%. .sup.1H-NMR (400 MHz, D.sub.2O): =3.45-4.15 (m, 5H, PG-backbone); 4.15-4.48 (m, 2H, C.sub.prim. H.sub.2OSO.sub.3); 4.59-4.89 (m, 1H, C.sub.sec.HOSO.sub.3) ppm.

(73) Dendritic Polyglycerol Sodium Succinate (dPG-Su) (6)

(74) To a solution of dPG (5.00 g, 0.98 mmol, 67.62 mmol OH groups) in pyridine (30 ml) succinic anhydride (7.44 g, 74.36 mmol, 1.1 eq. per OH group) was added at room temperature. After stirring for 2 d at room temperature the solvent was evaporated. The residue was dissolved in water and the pH value was adjusted to 7 by addition of aq. NaOH (10%). The crude product was subjected to ultrafiltration in water and addition of small amounts of NaCl for the first three cycles to avoid precipitation of dPG-succinic acid. Freeze drying yielded the compound as a colorless solid. M.sub.n (dPG-core)=5,100 g mol.sup.1, M.sub.n=13,600 g mol.sup.1, dF=100%, yield: 75%

(75) .sup.1H-NMR (400 MHz, D.sub.2O): =2.43-2.85 (m, 4H, CH.sub.2CH.sub.2COOR, CH.sub.2CH.sub.2COOR); 3.26-4.48 (m, 5H, PG-backbone); 4.99-5.35 (m, 0.5H, PG-backbone) ppm. .sup.13C-NMR (176 MHz, D.sub.2O): =30.0 (CH.sub.2CH.sub.2COOR); 31.0 (CH.sub.2CH.sub.2COONa); 63.0, 63.9, 65.7, 67.8, 68.1, 69.3, 69.8, 70.7, 71.0, 71.8, 77.2, 78.3 (PG-backbone); 174.6, 175.1 (COOR); 179.3 (COONa) ppm. IR max: 807 (m), 844 (m), 873 (m), 999 (s), 1079 (s), 1150 (s), 1202 (s), 1254 (s), 1312 (m), 1368 (s), 1403 (s), 1574 (s), 1727 (s), 2352 (w), 2879 (m), 2920 (m), 3398 (m) cm.sup.1.

(76) Dendritic Polyglycerol Amidoglyceryl Succinate (dPG-ASu) (7)

(77) 1-Aminoglycerol (5.54 ml, 6.59 g, 70.00 mmol, 2.0 eq. per acid group) was dissolved in water (50 ml) and the pH value of the resulting solution was adjusted to 8 by addition of HCl (1 M). Then DMAP (0.93 g, 7.61 mmol, 0.25 eq. per acid group) was dissolved in water (20 ml), the pH value was adjusted to 8 by means of HCl (1 M), and dPG succinate sodium salt (6.00 g, 0.44 mmol, 30.50 mmol acid groups) was added. After addition of EDC.HCl (5.85 g, 30.50 mmol, 1.0 eq. per acid group) 1-Aminoglycerol in water was added immediately and the mixture was stirred over night at room temperature. Dialysis in water yielded the product as a colorless honey like oil. M.sub.n (dPG-core)=5,100 g mol.sup.1, M.sub.n=17,000 g mol.sup.1, dF=95%, yield: 85%

(78) .sup.1H-NMR (400 MHz, D.sub.2O): =2.36-2.77 (m, 4H, CH.sub.2CH.sub.2COOR, CH.sub.2CH.sub.2COOR); 3.05-4.41 (m, 10H, PG-backbone, aminoglycerol), 4.86-5.37 (m, 1H, PG-backbone) ppm. .sup.13C-NMR (176 MHz, D.sub.2O): =29.3 (CONHCH.sub.2CH.sub.2); 30.1 (CH.sub.2CH.sub.2COOR, CH.sub.2CH.sub.2COONa); 41.8, 42.7 (CONHCH.sub.2CHCH.sub.2); 63.3 (CONHCH.sub.2CHCH.sub.2); 65.8, 67.8, 68.0, 69.6 (PG-backbone); 70.3 (CONHCH.sub.2CHCH.sub.2); 71.8, 72.2, 73.7, 77.0, 78.2, 79.5 (PG-backbone); 174.5 (CONHR, (COOR); 179.4 (COONa) ppm. IR .sub.max: 800 (w), 869 (w), 932 (w), 1000 (m), 1083 (s), 1161 (s), 1206 (m), 1240 (m), 1384 (m), 1408 (m), 1552 (m), 1552 (m), 1650 (s), 1698 (m), 1729 (s), 2876 (m), 2926 (m), 3323 (m) cm.sup.1.

(79) Dendritic Polyglycerol Amidoglyceryl Succinyl Sulfate (dPG-ASuS) (3)

(80) dPG-amidoglyceryl succinyl sulfate was synthesized according to the general procedure for sulfation by applying dPG-amidoglyceryl succinate (17,000 g mol.sup.1, 132 OH groups). The compound was obtained as colorless solid after freeze-drying. M.sub.n (dPG-core)=5,100 g mol.sup.1, M.sub.n=25,500 g mol.sup.1, dF=63%, yield: 71%

(81) .sup.1H-NMR (400 MHz, D.sub.2O): =2.46-2.88 (m, 4H, CH.sub.2CH.sub.2COOR, CH.sub.2CH.sub.2COOR); 3.16-4.75 (m, 10H, PG-backbone, aminoglycerol sulfate), 5.14-5.43 (m, 0.5H, PG-backbone) ppm. .sup.13C-NMR (176 MHz, D.sub.2O): =29.4 (CONHCH.sub.2CH.sub.2); 30.3 (CH.sub.2CH.sub.2COOR); 31.6 (CH.sub.2CH.sub.2COONa); 39.7, 43.0 (CONHCH.sub.2CHCH.sub.2); 63.4 (CONHCH.sub.2CHCH.sub.2); 66.8, 67.4, 68.4 (PG-backbone); 69.9 (CONHCH.sub.2CHCH.sub.2); 71.9, 75.3, 77.0, 78.3 (PG-backbone); 174.2 (CONHR); 174.5 (COOR); 180.5 (COONa) ppm. IR .sub.max: 771 (m), 936 (m), 1008 (m), 1034 (m), 1071 (m), 1215 (s), 1408 (w), 1462 (w), 1557 (w), 1652 (m), 1698 (m), 1729 (m), 2883 (w), 2931 (w), 3479 (w) cm.sup.1.

(82) Dendritic Polyglycerol Pentenoate (8)

(83) To a solution of dPG (2.08 g, 0.39 mmol, 27.06 mmol OH groups) in DMF (20 ml) was added NEt.sub.3 (4.8 ml) and BHT (0.33 g, 10 wt % of acid chloride). After cooling to 0 C., pentenoyl chloride (3.0 ml, 3.21 g, 27.06 mmol) was added over 30 min. The resulting solution was stirred over night in the melting ice bath, methanol (2.0 ml) was added and stirred for additional 5 min. Afterwards the solvent was removed, the residue was dissolved in chloroform, filtered and dialyzed. Evaporation yielded the product as a brown oil. M.sub.n (dPG-core)=5,100 g mol.sup.1, M.sub.n=9,900 g mol.sup.1, dF=84%, yield: 92%.

(84) .sup.1H-NMR (400 MHz, D.sub.2O): =1.96-2.96 (m, 4H, COCH.sub.2CH.sub.2, COCH.sub.2CH.sub.2); 2.96-4.47 (m, 6H, PG-backbone); 4.71-5.27 (m, 2H, CHCH.sub.2); 5.54-5.96 (m, 1H, CHCH.sub.2) ppm. .sup.13C-NMR (176 MHz, D.sub.2O): =27.3, 28.7, 28.8 (COCH.sub.2CH.sub.2); 31.2, 32.1, 33.2, 33.3, 33.4, 33.5 (COCH.sub.2CH.sub.2); 58.2, 62.7, 63.6, 65.3, 68.6, 69.6, 70.0, 71.2, 71.4, 71.6, 72.1, 72.5, 72.7, 78.7 (PG-backbone); 115.6, 117.6 (CHCH.sub.2); 134.1, 136.5, 136.6 (CHCH.sub.2); 168.5, 168.7, 172.2, 172.5, 172.8, 173.0 (COCH.sub.2CH.sub.2) ppm.

(85) Dendritic Polyglycerol Thioglyceryl Pentanoate (dPG-TP) (9)

(86) To a solution of dPG pentenoate (3.43 g, 0.35 mmol, 20.19 mmol ene groups) in chloroform (80 ml) was added thioglycerol (2.1 ml, 2.60 g, 24.23 mmol 1.2 eq. of ene groups) and 2,2-dimethoxy-2-phenylacetophenone (51.5 mg, 1.5 wt % of dPG pentenoate). The solution was degassed 3 times (freeze-pump-thaw) and irradiated with UV light for 90 min at room temperature. The solvent was removed, the residue was dissolved in methanol and dialyzed in methanol. Evaporation yielded the product as brown honey-like gel. M.sub.n (dPG-core)=5,100 g mol.sup.1, M.sub.n=17,200 g mol.sup.1, dF=100%, yield: 77%.

(87) .sup.1H-NMR (400 MHz, D.sub.2O): =1.24-2.13 (m, 4H, COCH.sub.2CH.sub.2, COCH.sub.2CH.sub.2CH.sub.2); 2.17-3.04 (m, 6H, COCH.sub.2CH.sub.2, SCH.sub.2CH.sub.2, SCH.sub.2CH); 3.42-4.62 (m, 9H, PG-backbone, SCH.sub.2CHOH, SCH.sub.2CHCH.sub.2) ppm. .sup.13C-NMR (176 MHz, D.sub.2O): =23.7, 25.1 (COCH.sub.2CH.sub.2); 28.4, 30.1 (COCHCH.sub.2CH.sub.2); 33.2 (SCH.sub.2CH.sub.2); 34.6 (COCH.sub.2CH.sub.2); 36.3 (SCH.sub.2CH); 64.0, 65.0 (PG-backbone); 66.0 (SCH.sub.2CHCH.sub.2); 66.8, 69.5, 69.7, 70.6, 71.0, 71.2 (PG-backbone); 72.8 (SCH.sub.2CH); 73.7, 74.0, 78.7, 80.0 (PG-backbone); 170.8, 174.6, 174.9, 175.1, 175.4 (COCH.sub.2CH.sub.2) ppm. IR .sub.max: 752 (w), 880 (m), 927 (m), 1028 (s), 1069 (s), 1173 (m), 1256 (m), 1343 (w), 1411 (s), 1453 (w), 1730 (s), 2521 (w), 2871 (m), 2918 (m), 3379 (m) cm.sup.1.

(88) Dendritic Polyglycerol Thioglyceryl Pentanotyl Sulfate (dPG-TPS) (4)

(89) dPG-thioglyceryl pentanoatyl sulfate was synthesized according to the general procedure for sulfation by applying dPG-thioglyceryl pentanoate (17,200 g mol.sup.1, 127 OH groups). The compound was obtained as colorless solid after freeze-drying. M.sub.n (dPG-core)=5,100 g mol.sup.1, M.sub.n=25,900 g mol.sup.1, dF=67%, yield: 92%.

(90) .sup.1H-NMR (400 MHz, D.sub.2O): =1.22-2.16 (m, 4H, COCH.sub.2CH.sub.2, COCH.sub.2CH.sub.2CH.sub.2); 2.16-3.15 (m, 6H, COCH.sub.2CH.sub.2, SCH.sub.2CH.sub.2, SCH.sub.2CH); 3.19-4.75 (m, 9H, PG-backbone, SCH.sub.2CHOSO.sub.3, SCH.sub.2CHCH.sub.2) ppm. .sup.13C-NMR (176 MHz, D.sub.2O): =22.3, 23.5 (COCH.sub.2CH.sub.2); 26.7, 28.3 (COCHCH.sub.2CH.sub.2); 30.9, 31.7 (SCH.sub.2CH.sub.2); 33.5 (SCH.sub.2CH); 34.8 (COCH.sub.2CH.sub.2); 58.0, 61.9, 63.2 (PG-backbone); 67.7 (SCH.sub.2CHCH.sub.2); 68.3, 68.6, 69.9, 70.7, 71.9, 75.6 (PG-backbone); 76.5 (SCH.sub.2CH); 78.3 (PG-backbone); 174.8, 175.5 (COCH.sub.2CH.sub.2) ppm. IR .sub.max: 685 (w), 805 (w), 864 (m), 927 (m), 997 (s), 1045 (s), 1170 (s), 1198 (s), 1411 (w), 1455 (w), 1489 (w), 1634 (w), 1709 (m), 2871 (m), 2929 (m), 3381 (m) cm.sup.1.

(91) Dendritic Polyglycerol Methacrylate (10)

(92) To a solution of dPG (2.69 g, 0.53 mmol, 36.39 mmol OH groups) and BHT (0.50 g, 2.27 mmol, 10 wt % of isocaynatoethyl methacrylate) in DMF (15 ml) was added 2-isocyanato-ethyl methacrylate (4.62 ml, 5.08 g, 32.75 mmol, 0.9 eq. per OH group) over 30 min at 60 C. The resulting mixture was stirred over night at room temperature and was directly used in the next step without purification to avoid uncontrolled polymerization the product. M.sub.n (dPG-core)=5,100 g mol.sup.1, M.sub.n=10,500 g mol.sup.1, dF=50%, yield: n.d.

(93) .sup.1H-NMR (400 MHz, CDCl.sub.3): =1.90 (s, 3H, CH.sub.3); 3.09-4.27 (m, 14H, dPG-backbone, NHCH.sub.2CH.sub.2, NHCH.sub.2CH.sub.2), 5.56 (s, 1H, CCH.sub.2), 6.09 (s, 1H, CCH2) ppm. .sup.13C-NMR (176 MHz, CCl.sub.3): =18.3 (CH.sub.3); 40.1 (NHCH.sub.2); 62.0 (dPG-backbone); 63.7 (NHCH.sub.2CH.sub.2); 66.1, 68.9, 69.5, 71.3, 72.7, 73.8, 78.7, 80.1 (PG-backbone); 126.2 (CH.sub.3CCH.sub.2); 136.0 (CH.sub.3CCH.sub.2); 156.5, 156.9 (NHCO); 167.4 (CH.sub.3CCO) ppm. IR .sub.max: 666 (w), 731 (m), 753 (m), 816 (w), 863 (w), 912 (m), 944 (m), 1045 (s), 1095 (s), 1159 (s), 1255 (m), 1297 (m), 1388 (w), 1405 (w), 1454 (m), 1533 (m), 1637 (w), 1667 (m), 1709 (s), 2245 (w), 2342 (w), 2360 (w), 2875 (w), 2926 (w), 3355 (m) cm.sup.1.

(94) Dendritic Polyglycerol Thioglyceryl Methylpropanoate (dPG-TMP) (11)

(95) To a solution of dPG-methacrylate (0.53 mmol, 18.55 mmol methacrylate groups, 50% linker) in DMF was added NEt.sub.3 (0.6 ml) and thioglycerol (6.4 ml, 8.03 g, 74.20 mmol, 4.0 eq. per methacrylate group). After stirring over night at room temperature the solvent was evaporated and the residue was dissolved in methanol. After filtration the filtrate was dialyzed in methanol. Evaporation yielded the product as a colorless gel. M.sub.n (dPG-core)=5,100 g mol.sup.1, M.sub.n=14,300 g mol.sup.1, dF=100%, yield: 62% over two steps.

(96) .sup.1H-NMR (400 MHz, CD.sub.3OD): =1.25 (s, br, 3H, CH.sub.3); 2.32-2.95 (m, 5H, COCHCH.sub.2, COCHCH.sub.2, SCH.sub.2CHOH); 3.37-4.42 (m, 16H, PG-backbone, NHCH.sub.2CH.sub.2, NHCH.sub.2CH.sub.2, SCH.sub.2CHOH, CH.sub.2OHCHOH) ppm. .sup.13C-NMR (176 MHz, CD.sub.3OD): =17.2 (CH.sub.3); 36.7 (SCH.sub.2CHCH.sub.3); 37.0 (SCH.sub.2CHOH); 40.9 (NHCH.sub.2); 41.5 (CH.sub.3CH); 62.8 (PG-backbone); 64.5 (NHCH.sub.2CH.sub.2); 65.9 (CH.sub.2OHCHOH); 67.2, 69.8, 70.1, 70.6, 70.9, 71.2, 71.4 (PG-backbone); 72.4 (SCH.sub.2CHOH); 73.6, 73.9, 79.8, 80.1, 81.4, 81.5 (PG-backbone); 158.5; 158.8 (NHCO); 176.7 (CHCO) ppm. IR .sub.max: 775 (m), 876 (m), 930 (m), 1035 (s), 1067 (s), 1156 (s), 1251 (s), 1342 (m), 1377 (m), 1411 (m), 1457 (m), 1538 (m), 1704 (s), 2357 (w), 2876 (m), 2923 (m), 3361 (m) cm.sup.1.

(97) Dendritic Polyglycerol Thioglyceryl Methylpropanoatyl Sulfate (dPG-TMPS) (5)

(98) dPG-thioglyceryl methylpropanoatyl sulfate was synthesized according to the general procedure for sulfation by applying dPG-thioglyceryl methylpropanoate (14,300 g mol.sup.1, 105 OH groups). The compound was obtained as colorless solid after freeze-drying. M.sub.n (dPG-core)=5,100 g.sup.1 M.sub.n=22,600 g mol.sup.1, dF=78%, yield: 84%.

(99) .sup.1H-NMR (400 MHz, D.sub.2O): =1.28 (s, br, 3H, CH.sub.3); 2.59-3.09 (m, 5H, COCHCH.sub.2, COCHCH.sub.2, SCH.sub.2CHOSO.sub.3); 3.22-4.74 (m, 16H, PG-backbone, NHCH.sub.2CH.sub.2, NHCH.sub.2CH.sub.2, SCH.sub.2CHOH, CH.sub.2OHCHOH) ppm. .sup.13C-NMR (176 MHz, D.sub.2O): =16.2 (CH.sub.3); 32.1 (SCH.sub.2CHCH.sub.3); 35.1, 35.3 (SCH.sub.2CHOSO.sub.3); 39.4 (NHCH.sub.2); 40.1 (CH.sub.3CH); 60.5 (PG-backbone); 63.9 (NHCH.sub.2CH.sub.2); 66.8 (CH.sub.2OHCHOH); 67.6, 68.5, 68.5, 70.1 (PG-backbone); 75.9, 76.5 (SCH.sub.2CHOSO3); 77.2, 78.2 (PG-backbone); 158.0, 158.6 (NHCO); 177.6 (CHCO) ppm. IR .sub.max: 774 (m), 810 (w), 922 (m), 998 (s), 1037 (s), 1064 (s), 1102 (s), 1159 (s), 1216 (s), 1346 (w), 1376 (w), 1415 (w), 1457 (m), 1538 (m), 1705 (s), 2881 (w), 2936 (w), 3370 (m) cm.sup.1.

(100) Clotting Assay

(101) The clotting assays were performed for activated partial thromboplastin time (APTT) at individual concentrations using an Amelung coagulometer (Type 410A4MD). The measurements refer to the clotting time [s] of the untreated plasma control which was set to 100%. To determine the APTT, 100 l plasma and 100 l Actin FS were mixed and incubated (3 min, 37 C.) with 4 l test compound (final concentrations 0-1000 nM). The reaction was started by the addition of 100 l of pre-warmed (37 C.) clotting activator CaCl.sub.2. The data are presented as meanS.D. of two independent experiments.

(102) Complement Activation Assay

(103) Normal pooled human serum from six donors was obtained by centrifugation (20 min, 4 C., 3400g) of coagulated blood samples. The supernatant was stored in 100 l aliquots at 20 C. until use and then thawed for one minute at 37 C. Complement activation was tested for the classical pathway with an ELISA-based assay. Briefly, a multi-well plate setup with IgM coated wells was used for activation of the classical complement pathway. Therefore, 100 l of freshly thawed and diluted (1:101) serum samples were incubated (60 min, 37 C.) with 2 l test compound (final concentrations 0-2500 nM). The formation of the specific complement membrane attack complex (MAC), the membrane pore with components C5b-9 was detected with an alkaline phosphatase (AP)-labeled antibody and substrate para-nitrophenylphosphate (PNPP). Absorbance at 405 nm was recorded with a SpectraMax 340PC. The stated values for any concentration are the meanS.D. of 2-3 independent measurements.

(104) The biodistribution of embodiments of polyglycerol derivatives according to the claimed invention was tested in comparison to the biodistribution of polyglycerol derivatives according to prior art. By these experiments, it could be shown that the linker being present in the polyglycerol derivatives according to the claimed invention serves for much more favorable biodistribution of the polyglycerol derivatives. Some of these experiments will be explained in the following.

(105) The compound dPG-DMPTACN-TPS was tested as exemplary embodiment in comparison to dPGS-DMPTACN. Both dPG-DMPTACN-TPS and dPGS-DMPTACN comprise a 1,4-bis(2-pyridinylmethyl)-1,4,7-triazacyclononane (DMPTACN) residue. The chemical structure of dPG-DMPTACN-TPS is depicted in FIG. 9, and the chemical structure of dPGS-DMPTACN is shown in FIG. 10. The linker used for dPG-DMPTACN-TPS comprises eight carbon atoms and one disulfide bridge. It is connected to the polyglycerol core by an ester linkage. Two sulfate groups are linked to the linker. In case of dPGS-DMPTACN, the sulfate group is directly coupled to the polyglycerol core.

(106) The biodistribution of dPG-DMPTACN-TPS and dPGS-DMPTACN was tested 4 hours and 24 hours after administration. Thereby, the recovered percentage of the administered dose in different organs and in urine has been determined. The results are summarized in the following Table 2.

(107) TABLE-US-00002 TABLE 2 Results of biodistribution experiments. Biodistribution after Biodistribution after 4 hours (% of 24 hours (% of administered dose) administered dose) dPG- dPG- DMPTACN- dPGS- DMPTACN- dPGS- Organ TPS DMPTACN TPS DMPTACN Spleen 0.43 1.20 0.47 1.21 Kidney 4.68 4.28 1.79 3.04 Liver 3.58 5.55 1.53 3.11 Duodenum 3.54 12.71 9.01 4.44 Colon 0.80 0.53 6.54 0.29 Urine 1.94 20.33 31.67 31.05

(108) dPG-DMPTACN-TPS accumulates to a significantly lower extent than dPGS-DMPTACN in spleen both after 4 hours and after 24 hours.

(109) In kidney, the accumulation of dPG-DMPTACN-TPS after 4 hours is comparable to the accumulation of dPGS-DMPTACN (considering the standard deviation of the performed experiments). After 24 hours, the accumulation of dPG-DMPTACN-TPS is significantly lower than the accumulation of dPGS-DMPTACN.

(110) In liver, the accumulation of dPG-DMPTACN-TPS is again significantly lower than the accumulation of dPGS-DMPTACN after 4 hours and after 24 hours.

(111) In the duodenum, accumulation of dPG-DMPTACN-TPS is significantly lower than the accumulation of dPGS-DMPTACN after 4 hours, but higher after 24 hours. This indicates that the excretion pathway via the duodenum is taken later in case of dPG-DMPTACN-TPS.

(112) Interestingly, the recovery rate of dPG-DMPTACN-TPS in the colon is comparable to that of dPGS-DMPTACN after 4 hours, but an order of magnitude higher after 24 hours. It appears thatalthough dPGS-DMPTACN is accumulated in the duodenumno significant excretion of dPGS-DMPTACN via the colon takes place. Rather, it appears that an absorption in the duodenum of dPGS-DMPTACN might occur. In contrast, the obtained data clearly shows that dPG-DMPTACN-TPS is excreted via the colon after 24 hours.

(113) In urine, a lower occurrence of dPG-DMPTACN-TPS than of dPGS-DMPTACN is observed after 4 hours. However, after 24 hours, the recovery rate of dPG-DMPTACN-TPS equals the recovery rate of dPGS-DMPTACN.

(114) Summarizing, the obtained data clearly shows that dPG-DMPTACN-TPS is less accumulated than dPGS-DMPTACN in spleen, kidney and liver either after 24 hours or after 4 hours and after 24 hours. The data further shows that the excretion of dPG-DMPTACN-TPS via the duodenum and the colon is significantly better than that of dPGS-DMPTACN. In addition, the excretion of dPG-DMPTACN-TPS via urine equals the excretion of dPGS-DMPTACN after 24 hours.

(115) Therewith, the linker being present in dPG-DMPTACN-TPS provides this compound with a favorable biodistribution behavior with respect to dPGS-DMPTACN having no such linker.

LIST OF REFERENCES CITED IN THE PRECEDING SECTIONS

(116) [1] H. Trk, R. Haag, S. Alban, Bioconjugate Chemistry 2004, 15, 162-167. [2] M. Weinhart, D. Grger, S. Enders, S. B. Riese, J. Dernedde, R. K. Kainthan, D. E. Brooks, R. Haag, Macromolecular Bioscience 2011, 11, 1088-1098. [3] M. Weinhart, D. Grger, S. Enders, J. Dernedde, R. Haag, Biomacromolecules 2011, 12, 2502-2511. [4] J. Dernedde, A. Rausch, M. Weinhart, S. Enders, R. Tauber, K. Licha, M. Schirner, U. Zgel, A. von Bonin, R. Haag, Proceedings of the National Academy of Sciences 2010, 107, 19679-19684. [5] K. Oishi, Y. Hamaguchi, T. Matsushita, M. Hasegawa, N. Okiyama, J. Dernedde, M. Weinhart, R. Haag, T. F. Tedder, K. Takehara, H. Kohsaka, M. Fujimoto, Arthritis & Rheumatology 2014, 66, 1864-1871. [6] K. Licha, P. Welker, M. Weinhart, N. Wegner, S. Kern, S. Reichert, I. Gemeinhardt, C. Weissbach, B. Ebert, R. Haag, M. Schirner, Bioconjugate Chemistry 2011, 22, 2453-2460. [7] D. Grger, F. Paulus, K. Licha, P. Welker, M. Weinhart, C. Holzhausen, L. Mundhenk, A. D. Gruber, U. Abram, R. Haag, Bioconjugate Chemistry 2013, 24, 1507-1514. [8] C. Holzhausen, D. Grger, L. Mundhenk, P. Welker, R. Haag, A. D. Gruber, Nanomedicine: Nanotechnology, Biology and Medicine 2013, 9, 465-468. [9] A. Sunder, R. Hanselmann, H. Frey, R. Mlhaupt, Macromolecules 1999, 32, 4240-4246. [10] G. Pickaert, M. Cesario, R. Ziessel, The Journal of Organic Chemistry 2004, 69, 5335-5341. [11] S. J. Bryant, K. A. Davis-Arehart, N. Luo, R. K. Shoemaker, J. A. Arthur, K. S. Anseth, Macromolecules 2004, 37, 6726-6733. [12] S. Enders, G. Bernhard, A. Zakrzewicz, R. Tauber, Biochimica et Biophysica Acta (BBA)General Subjects 2007, 1770, 1441-1449. [13] M. Kirschfink, Immunopharmacology 1997, 38, 51-62.