Microspheres of hydrolysed starch with endogenous, charged ligands

Abstract

Biodegradable microspheres having a diameter of 10-2000 m having cross-linked hydrolysed starch onto which at least one type of ligand has been coupled via a carboxylic ester bond. The ligand shall be an endogenous, charged molecule with a molecular mass of less than 1000 Da having at least one additional carboxylic acid function in addition to the one utilised for coupling the ligand to the microsphere and/or at least one amine function. On average 0.05-1.5 ligands are coupled to each glucose moiety in the hydrolysed starch.

Claims

1. A biodegradable microsphere having a diameter of 10 to 2000 m comprising cross-linked hydrolysed starch onto which a di-methyl glycine ligand has been directly coupled via a carboxylic ester bond formed between the carboxylic acid group of the di-methyl glycine ligand and a hydroxyl group of a glucose residue in the cross-linked hydrolysed starch, and wherein on average 0.05 to 1.5 di-methyl glycine ligands have been coupled to each glucose moiety in the hydrolysed starch, and wherein the di-methyl glycine ligands are biodetachable in vivo.

2. The microsphere according to claim 1, wherein the di-methyl glycine ligand has a physiologically active counter ion.

3. The microsphere according to claim 2 for use in hemostasis, wherein the microsphere has a mean diameter of 10-200 m.

4. The microsphere according to claim 3, wherein the counter ion is ellagic acid.

5. A material for use in connection with wound healing comprising microspheres according to claim 1 in powder form, wherein the microspheres, upon administration to a wound, form a three-dimensional structure comprising voids between the microspheres, wherein the voids are at least 30 m.

6. The material according to claim 5, wherein each microsphere is part of a homogenous size fraction with maximum variation of the mean diameter of 15%.

7. The material according to claim 5, wherein the microspheres have a mean diameter of from 200 m to 2000 m.

8. The microsphere according to claim 1 for use in in vitro cell culture, wherein the microspheres have a mean diameter of from 200 m to 1000 m.

9. A wound dressing comprising the material according claim 5.

10. A method of carrying out hemostasis by topically or intraoperatively administering an effective amount of a microsphere according to claim 1 to a mammal suffering from a bleeding wound.

11. A method for carrying out wound healing by topically or intraoperatively adding an effective amount of a material according to claim 5 to a mammal suffering from a wound.

12. A method for in vitro cultivation of cells, wherein at least one microsphere according to claim 1 is added to a culture medium to which also the cells to be cultivated are added, the cells are then allowed to propagate.

13. A new biodegradable wound healing composition comprising: a plurality of biodegradable microspheres, each microsphere of the wound healing composition having a diameter of 10 to 2000 m and cross-linked hydrolysed starch onto which a di-methyl glycine ligand has been directly coupled via a carboxylic ester bond formed between the carboxylic acid group of the di-methyl glycine ligand and a hydroxyl group of a glucose residue in the cross-linked hydrolysed starch, and wherein on average 0.05 to 1.5 di-methyl glycine ligands are coupled to each glucose moiety in the hydrolysed starch, and wherein the di-methyl glycine ligands are biodetachable in vivo.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention is described in more detail below in the Examples, which refer to the appended drawings on which:

(2) FIG. 1 is a schematic picture of a degradable starch microsphere (DSM) and the chemical modifications performed in this study.

(3) FIG. 2 illustrates that the swelling of the microspheres may be assumed to follow Fick's diffusion with an initial rapid swelling rate that declines exponentially:
Y=Y.sub.(1e.sup.kt)
wherein k=the first-order swelling constant, and Y=the volume increase at maximum swelling.

(4) FIG. 3 illustrates platelet adhesion. FIG. 3 A shows phase contrast and fluorescent micrographs showing the DSM and DSM-adhered platelets according to the different modified batches. FIG. 3. B shows close-ups of the junction between two aggregated DSM (batch 4) and the platelet aggregates attached to the DSM. Imaged using differential interference contrast (DIC) microscopy.

(5) FIG. 4 illustrates an in vivo study of three of the DSM batches. Batches 5, 6 and 9 were evaluated in an experimental bleeding model (renal trauma) in anti coagulated rats. All animals treated with batch 9 obtained primary hemostasis, 29% re-bled within 20 min observation. The other batches demonstrated significantly less hemostatic efficiency with few animals achieving primary hemostasis.

(6) FIG. 5 illustrates blood loss according to treatment batch in the experimental in vivo study. Blood loss was measured by weighing the excessive blood collected in gauze. There was a significant difference in blood loss between the different batches (p=0.001), where batch 5 was unmodified DSM, batch 6 proved activation of the coagulation and DSM in batch 9 adsorbed platelets.

EXAMPLES

(7) The degradable starch microspheres (DSM) were prepared by emulsion, cross-linking of hydrolysed starch with epichlorohydrin in toluene. The DSM are subsequently washed repeatedly with ethanol followed by distilled water and finally successively dehydrated with increasing concentrations of ethanol and finally dried over night at 60 C.

(8) Details on Preparation of the DSM

(9) 2 g of sodium hydroxide is dissolved in 280 mL purified water and 2 g sodium borohydride is added and dissolved. 153 g of hydrolysed starch is dissolved by slow stirring for at least 2 hours. 20 g of surfactant (Rhodafac PA17) is dissolved in 450 g toluene. The starch solution is then added and emulsified in the toluene solution, the temperature is increased to 70 C. and the emulsion is stirred until the desired droplet size distribution has been attained. 22 g of epichlorohydrin is added and crosslinking is performed for 5 hours. The mixture is cooled to room-temperature and allowed to sediment whereafter the supernatant is decanted. The DSM are given three washes with 95% ethanol, one wash with 0.8% acetic acid, followed by 4 washes with purified water and finally dyhydrated with absolute ethanol before drying at 60 C. in a ventilated drying cabinet.

(10) Determination of Degree of Substitution (DS)

(11) The degree of substitution is defined as the average number of substitutes per glucose monomer.

(12) The method of alkali saponification, followed by titration of the excess of alkali was employed for the determination of the degree of substitution. To a sample of 250 mg of DSM 10 mL of 0.50 M NaOH was added and this was allowed to stand at room temperature for 72 h with occasional shaking. The excess of NaOH was titrated with 0.50 M HCl using phenolphthalein as indicator.

(13) Determination of Degradability with Amylase

(14) A sample of DSM (3-6 mg) was diluted with phosphate buffer, pH 7 (5 ml) and then 400 l human saliva was added, followed by incubation at 37 C. for 4 h. The sample was allowed to stand for 20 min or was centrifuged and then a small sample was taken from the bottom and analysed by microscope to determine the presence or absence of microspheres.

(15) General Procedure for Substitution of DSM with Dioic Acids (Examples Listed in Table 1)

(16) DSM (1 g) was suspended in DMF (10 ml), to this mixture succinic anhydride (154 mg, 1.54 mmol) and pyridine (124 l, 1.60 mmol) were added. The mixture was stirred and heated to 90 C. over night and then the material was washed three timed with 40 ml of ethanol followed with 5 ml saturated NaHCO.sub.3 and then three times with 30 ml of water. The material was dehydrated with ethanol and dried in an oven at 60 C. The material was analysed with FTIR showing ester carbonyl at 1730 cm.sup.1.

(17) DS: 0.25 (determined as described above).

(18) Degradable by -amylase (determined as described above).

(19) General Procedure for Substitution of DSM with Esters

(20) Modification with Betaine

(21) Betaine (1.66 g, 10.8 mmol) and CU (1.75 g, 10.8 mmol) were mixed with 50 ml of DMF and heated to 80 C. for 2 h. Then DSM (5 g) was added and the temperature was raised to 90 C. and the mixture was stirred over night. The mixture was washed with ethanol (250 ml) two times, diluted hydrogen chloride (250 ml) and two times with water (250 ml). The material was dehydrated with ethanol and dried over night at 60 C.

(22) FTIR showing ester carbonyl at 1751 cm.sup.1.

(23) DS: 0.23 (determined as described above).

(24) Degradable by -amylase (determined as described above).

(25) Modification with Dimethyl-Glycine

(26) As in the example with betaine above, but DSM (2 g), N,N-Dimethylglycine hydrochloride (430 mg, 3.1 mmol) and CDI (500 mg, 3.1 mmol) were used

(27) FTIR showing ester carbonyl at 1753 cm.sup.1.

(28) DS: 0.24 (determined as described above).

(29) Degradable by -amylase (determined as described above).

(30) Modification with N.sub.-Acetyl-L-Arginine

(31) As in the example with betaine above, but DSM (2 g), N.sub.-Acetyl-Larginine (623 mg, 2.5 mmol), CDI (400 mg, 2.5 mmol) were used.

(32) FTIR showing ester carbonyl at 1748 cm.sup.1.

(33) DS: 0.24 (determined as described above).

(34) Degradable by -amylase (determined as described above).

(35) Modification with Proline

(36) As in the example with betaine above, but DSM (1 g), Boc-Pro-OH (266 mg, 1.2 mmol), CDI (200 mg) were used followed by deprotecting of the tert-butoxycarbonyl with TFA.

(37) FTIR showing ester carbonyl at 1743 cm.sup.1.

(38) Degradable by -amylase (determined as described above).

(39) Modification with Glycine

(40) As in the example with betaine above, but DSM (1 g), Boc-Gly-OH (216 mg, 1.2 mmol), CDI (200 mg) were used followed by deprotecting of the tert-butoxycarbonyl with TFA.

(41) FTIR showing ester carbonyl at 1748 cm.sup.1.

(42) Degradable by -amylase (determined as described above).

(43) Modification with Phenylalanine

(44) As in the example with betaine above, but DSM (1 g), Boc-Phe-OH (327 mg, 1.2 mmol), CDI (200 mg) were used followed by deprotecting of the tert-butoxycarbonyl with TFA.

(45) FTIR showing ester carbonyl at 1743 cm.sup.1.

(46) Degradable by -amylase (determined as described above).

(47) Non-Detatchable Surface Modifications Used in Investigation of Charge Effects

(48) The surface modifications are illustrated in FIG. 1.

(49) Octenylsuccinate (Negative and Hydrophobic)

(50) 80 g of DSM were suspended in purified water, N-octenyl succinic anhydride (Pentagon) was added to 0.08 g/g dry DSM and the reaction was continued for 3 h. A pH above 7.4 was maintained by additions of 0.75 M NaOH. The resulting material was washed 8 times with 2000 mL of purified water and thereafter dehydrated with increasing concentrations of ethanol and finally dried over night at 60 C. (Hui Rea. Preparation and properties of octenyl succinic anhydride modified potato starch. Food Chemistry 2009; 114:81-6).

(51) Carboxymethylation (Negative)

(52) 50 g of DSM were suspended in purified water; chloroacetic acid was added to 0.1 g/g dry DSM and the reaction were continued for 5 h at 70 C. Before adding the chloroacetic acid it was dissolved in water and neutralised with 1 M NaOH. The resulting material was washed 6 times with 2000 mL of purified water and thereafter dehydrated with increasing concentrations of ethanol and finally dried over night at 60 C. (Tomaski P, Schilling, C. H. Chemical modification of starch. Adv Carbohydr Chem Biochem 2004; 59:175-403).

(53) Acetylation (Hydrophobic)

(54) 50 g of DSM were suspended in purified water, acetic anhydride was added to 0.05 g/g dry DSM. Acetic anhydride was added drop by drop and a pH between 7.3 and 7.8 was maintained by additions of 0.75 M NaOH. The resulting material was washed 7 times with 2000 mL of purified water and thereafter dehydrated with increasing concentrations of ethanol and finally dried over night at 60 C. (Sathe S K, Salunkhe, D. K. Isolation, Partial Characterisation and Modification of the Great Northern Bean (Phaseolus vulgaris L.) Starch. J Food Sci 1981; 46:617-21).

(55) Diethylaminoethyl Chloride, Aldrich (Positive)

(56) 50 g of DSM were suspended in purified water, 0.375 mol of DEAE hydrochloride was added and the temperature was increased to 60 C. 250 ml of 3 M sodium hydroxide solutions was added and the reaction was maintained at 60 C. for one hours. The DSM was than washed with 20 L of purified water in a Bchner funnel. The DSM was then dehydrated and dried as above (Manousos M, Ahmed M, Torchio C, Wolff J, Shibley G, Stephens R, et al. Feasibility studies of oncornavirus production in microcarrier cultures. In Vitro 1980 June; 16(6):507-15).

(57) Ellagic Acid (Adsorbed/Absorbed Negative)

(58) Ellagic acid (Alfa Aesar) was passive adsorbed using two different methods. Method 1: 0.1 mM ellagic acid was dissolved in water and then mixed with the DSM. Method 2: 0.1 mM ellagic acid was dissolved in ethanol and then mixed with the DSM (Ratnoff O D, Saito H. Interactions among Hageman factor, plasma prekallikrein, high molecular weight kininogen, and plasma thromboplastin antecedent. Proc Natl Acad Sci USA 1979 February; 76(2):958-61). Washing and drying as above. The ellagic acid was passively absorbed/adsorbed and was not applicable for measurement of charges.

(59) The different surface modifications were produced with standard modification protocols (not optimised). The modifications were selected for proving the concept of a hemostatic effect in vitro and in vivo, and were not assessed for being toxicologically acceptable in humans.

(60) Surface Charge

(61) The degree of surface charge was measured by a PCD 02, Particle Charge Detector (Mtek).

(62) Design

(63) The nine different modified DSM were randomised and blinded. No information about the modifications was sent to the performers of the studies.

(64) Characterisation of DSM

(65) The morphology of the starch microspheres was determined by observation in microscope (AxioObserver Z1, Zeiss), and sphere diameters were measured for a minimum of five spheres in each of the nine batches. Absorption was determined by measurement of diameter before and at fixed time intervals (1, 3, 9, 15 and 30 s) after addition of 100 L phosphate buffer. A minimum of five spheres from each batch were measured and their volume was then calculated, assuming the DSM were completely spherical. Swelling of the microspheres occurs by diffusion of water into and hydration of the polymer, a process that continues towards equilibrium at maximum relaxation of the cross-linked starch chains. Consequently it may be assumed that the process follow Fick's diffusion with an initial rapid swelling rate that declines exponentially. The data may thus be explained by:
Y=Y.sub.(1e.sup.kt)
wherein k is the first-order swelling constant and Y.sub. is the volume increase at maximum swelling.
In-vitro Platelet Adhesion

(66) To study the possible affinity/interaction between the various DSM batches and factors of known importance to the coagulation process, platelet adhesion to the different DSM batches was investigated. 450 l of heparinised platelet-rich plasma was added to test tubes containing 1 g DSM and thereafter agitated in an orbital shaker for 20 minutes at 500 rpm. Thereafter the DSM were thoroughly washed in PBS by repeatedly letting the DSM sediment to the bottom and exchange the supernatant with fresh PBS and thereafter vortex the tube. DSM-adhered platelets were then fixed with 3.7 PFA in PBS and permeabilised using 0.1% Triton-X in PBS, and finally fluorescently stained with Alexa 546-Phalloidin. Thorough rinsing was performed between each step in the procedure. Images of DSM and fluorescent platelets were acquired with an AxioObserver Z1 (Zeiss) fluorescence microscope and AxioVision (Zeiss) imaging software.

(67) In vivo Pilot Study in an Experimental Renal Bleeding Model

(68) The study was performed in accordance to the guidelines of good laboratory practice and approved by the Local University Ethics Committee for Animal Experiments. Three different batches of DSM were chosen based on the outcome of the in-vitro studies described above. One neutral batch, one that activated the coagulation and finally one batch with platelet adhesion properties were chosen for the in-vivo testing. The batches were blinded and randomised to the investigator performing the study. Twenty-one adult acclimatised male Sprauge-Dawley rats (median weight 342 g, iqr: 314-360) with free supply of food and water were anaesthetised (Hynorm, Janssen Pharma, Belgium and Midazolam Hameln, Pharma Hameln, GmbH). After catheterisation of the jugular vein (for IV injections) a transversal laparotomy was performed. The left kidney was dissected and the renal vessels were clamped two minutes after IV administration of Unfractionated Heparin (U H, LEO Pharma N S, Denmark) 200 IU/kg. The lateral one third of the kidney was then resected and 1 mL of randomised DSM applied on the raw kidney surface, manual compression started (with a gauze compress between the starch powder and the investigators finger) and the vessel clamp was removed. Compression remained for 2 minutes, then released for control of hemostasis. If bleeding occurred compression continued with hemostatic controls each minute. Primary hemostasis was defined as no visible bleeding within 20 minutes from renal resection. Animals obtaining hemostasis were observed another 20 minutes for possible re-bleeding. All animals were euthanised with an IV injection of phenobarbiturate acid and ethanol. Blood loss was collected and weighed. Study endpoints were: ability to obtain primary hemostasis, time to hemostasis, frequency of re-bleeding and blood loss.

(69) Statistics

(70) Descriptive data are presented with median values and individual or inter quartile range (iqr). Non-parametric test were performed, since the distribution of data was skewed. .sup.2 tests were performed for contingency tables and Kruskal-Wallis analysis of variance was used when unpaired data were compared. A p value of <0.05 was considered significant.

(71) The software SPSS 17.0 for Mac and Windows (www.spss.com) was used.

(72) Results

(73) Modifications of DSM

(74) Surface charges are given in table 2. The synthetic procedure was not optimised and carboxymethylation did not result in appreciable surface charge. Acetylation is not expected to change surface charge whereas the other methods should lead to significant positive and negative surface charges.

(75) TABLE-US-00002 TABLE 2 The chemical modifications of the DSM and the outcome in measured charges. Size Batch: Modification of DSM: inclusion: Charge: 1 N-octenyl succinic 11.8 equ/g anionic anhydride 2 Chloroacetic acid 0.7 equ/g anionic 3 Acetic anhydride 0.3 equ/g anionic 4 Diethylaminoethyl >80 m 459 equ/g cationic chloride 5 No surface modification 0.5 equ/g cationic 6 Ellagic acid.sup.1 NA 7 Diethylaminoethyl <80 m Not measured chloride 8 Ellagic acid.sup.2 NA 9 Diethylaminoethyl >150 m 100 equ/g cationic chloride.sup.3 .sup.1Dissolved in water .sup.2Dissolved in ethanol .sup.3More extensive crosslinking
Characterisation of Starch Spheres

(76) There was a significant difference in dry diameter between the batches (p=0.006), batch 6 having the smallest size spheres (median diameter 54 m, iqr: 38-58) and batch 2 the largest (median diameter 72 m, iqr: 67-76). After addition of phosphate buffer all batches increased rapidly in volume (FIG. 2), and after 30 s they had expanded between 5 and 25 times their dry volume (table 3). The amount of swelling was significantly different between the batches (p=0.001).

(77) TABLE-US-00003 TABLE 3 DSM dry volume and after 30 seconds in phosphate buffer solution. Dry volume Volume pL* Volume DSM Batch pL* 30 s increase % 1 157 1123 700 (110-165) (943-1150) (647-1000) 2 195 1047 600 (158-231) (871-1629) (512-715) 3 102 775 760 (79-128) (653-760) 4 128 1838 1308 (120-180) (1551-2187) (1001-1559) 5 88 998 1071 (60-126) (688-1110) (853-1264) 6 82 750 1058 (28-105) (439-1114) (714-1311) 7 166 829 538 (130-219) (760-1083) (371-751) 8 113 659 619 (95-163) (599-875) (521-666) 9 125 3368 2593 (92-249) (1697-3368) (2040-2593) *Median values (iqr), volume in pikoliter (1 pL = 1 ml.sup.9)
Platelet Stimulation

(78) There was an evident adherence of platelets to DSM in three of the modified batches (No. 4, 7 and 9), whereas the rest of the DSM-batches did not affect platelets at all (FIG. 3). The results were confirmed using PRP from three different donors.

(79) The Randomised, Blinded In vivo Pilot Experimental Study

(80) All animals treated with batch 9 obtained primary hemostasis, compared to 14-43% primary hemostasis with the other batches (FIG. 4). Time to hemostasis also differed between the groups (p=0.044), batch 9 treated animals were fastest (median 2 min: iqr: 2-3:20) whereas batch 6 required median 6 min (n=3) and batch 5 10 min (n=1) before they ceased to bleed. Two batch 9 treated animals were the only re-bleeders (p=NS, compared to the other batches). Batch 9 treated animals had less blood loss (median 1 g, iqr: 0.4-1.2) compared to the other batches (batch 5: 5 g, 4.3-6.7, batch 6: 5.3 g, 2.2-8.6), p=0.001 (FIG. 5).

(81) The postulated hemostatic effect of DSM by absorption of fluid (and small molecules) from the blood and concentrating endogenous coagulation factors on the spheres, may be dependent of a fast and considerable swelling of the microspheres. All batches in this study increased their volume rapidly after addition of phosphate buffer, but both the velocity and the total amount of swelling differed between the batches. Swelling depends on relaxation of the poly-glucose chains as they are hydrated. This is restricted by many cross links and facilitated by charge repulsion of the ligands. We could find no clear correlations with the measured characteristics (e.g. charge), though. Low cross linking and high and fast swelling implicate rapid degradation and therefore the increase in volume will not be hazardous even if applied intra operatively in locations where space may become limited at the end of the procedure. In this study the rapid absorption of fluid and swelling of the DSM was not sufficient for hemostasis in vivo, only 1 of 7 animals obtained primary hemostasis treated with non-modified microspheres.

(82) The DSM with superior hemostatic capacity in vivo proved to be those with platelet stimulating properties. Platelet adhered to the positively charged DSM, the diethylaminoethyl (DEAE) prepared batches (4, 7 and 9), which is in accordance with reported platelet adherence to surfaces exposing positive charged groups (Lee J H, Khang G, Lee J W, Lee H B. Platelet adhesion onto chargeable functional group gradient surfaces. J Biomed Mater Res 1998 May; 40(2):180-6). No objective quantification of amount of platelets that adhered to respective DEAE-modified batch was performed, but by ocular assessment there was no obvious difference in the amount of platelet adherence between batch 4, 7 and 9 , even if there was a measured difference in charge between batch 4 and 9. DEAE-chloride reacts with the hydroxylgroups on the DSM surface, generating DEAE groups that are positive at physical pH. DEAE ligands render microspheres that are non-biodegradable and probably unsuitable for human use. However, as a proof-of concept to distinguish if the spheres can become platelet-adherent and whether this has any clinical hemostatic significance, the DEAE modification was valuable. A fast and efficient stimulation of platelets is crucial for instant hemostasis produced by a physical plug of aggregated platelets. The platelets are also required for efficient amplification and propagation of thrombin generation, a process strongly catalysed by the stimulated platelet surface, resulting in a fibrin network that stabilises the primary platelet plug.