One-month sustained release microspheres of 125I-bovine calcitonin

Journal of Controlled Release 59 (1999) 55–62

One-month sustained release microspheres of 125 I-bovine calcitonin In vitro—in vivo studies ´ * ´ C. Evora R.V. Diaz, M. Llabres, ´ Quımica ´ ´ Farmaceutica ´ ´ , Av. Astrof ´ısico Francisco Sanchez s /n. Facultad de Farmacia, Departamento de Ingenierıa y Tecnologıa Universidad de La Laguna, 38200 La Laguna, Spain Received 14 April 1998; received in revised form 7 September 1998; accepted 5 October 1998

Abstract To obtain a 1-month release formulation of 125 I-bovine calcitonin, microspheres were prepared with three different PLA copolymers, PLGA I (mol. wt. [MW]530 000), polyethyleneglycol (PEG)–PLGA (MW534 000) and PLGA II (MW5 12 000) using the double emulsion method. The release of 125 I-bovine calcitonin was assayed in vitro using dialysis bags at 378C in isotonic phosphate buffer (pH 7.4). The in vitro release results indicated a very slow release rate for an optimal 1-month sustained release formulation. 125 I-bovine calcitonin microspheres were administered under the skin on the back of Wistar rats and the radioactivity at the injection site was subsequently measured over a 4-week period. The in vitro and in vivo profiles were affected by the weight average molecular weight of the copolymers. The 125 I-bovine calcitonin release rate was faster from microspheres prepared with PLGA II (MW512 000) than from microspheres prepared with higher molecular weight copolymers (PLGA I and PEG–PLGA). Microspheres prepared with PLGA II (MW512 000) release 100% of the dose in 1 month, in vivo release profiles presented two phases, during the first 2 weeks approximately 70% of the 125 I-bovine calcitonin injected was released, followed by a second slower phase.  1999 Published by Elsevier Science B.V. All rights reserved. Keywords: Calcitonin; Biodegradable polymers; One-month microspheres; Sustained peptide release; In vitro—in vivo release

1. Introduction Calcitonin (CT) is a calcium-regulating peptide hormone consisting of 32 amino acid residues with a disulphide bond between cysteine residues in positions 1 and 7. The entire amino acid sequence is essential for biological activity. CT has been found in different species, such as fish, mammals, birds and reptiles. There are several differences in amino acid *Corresponding author. Fax: 134-22-630-095. ´ E-mail address: [email protected] (C. Evora)

composition of CT from different sources and this is associated with different potencies. The bioavailability of CT via noninvasive routes such as when administered orally, is poor, simply because most of it destroyed by gastrointestinal enzymes. An alternative route of administration is intranasally which has again a limited absorption, in comparison with the invasive routes, such as, subcutaneous (SC) or intramuscular injection. However, the daily injection of CT has poor patient compliance for long-term therapy. For this reason, there has been a considerable interest in various research groups

0168-3659 / 99 / $ – see front matter  1999 Published by Elsevier Science B.V. All rights reserved. PII: S0168-3659( 98 )00179-5

56

R.V. Diaz et al. / Journal of Controlled Release 59 (1999) 55 – 62

around the world to develop new slow-release formulations [1–5]. Lee et al. [1] and Mehta et al. [5] showed that microspheres containing salmon CT (sCT) administered in rats resulted in sustained levels in the serum for over 5 and 5–9 days, respectively. Their results indicated that serum calcium levels are not well correlated to CT serum levels, because of the negative and positive feedback action provided by serum Ca 11 on CT and parathyroid hormone, respectively. In view of these results, we proposed [6] measurement of the radioactivity at the site of injection using an iodine-125-tracer ( 125 I-tracer) to study the release of bovine CT (bCT) from PLGA microspheres using surfactant agents in the preparation of the first emulsion. The results showed a very slow release rate despite the reduction of the initial burst effect. In this paper we report the method of obtaining 1-month sustained release microspheres of 125 I-bCT using biodegradable polymers, delivering CT at a release rate of 25 mg / day which is the recommended therapeutic dose, equivalent to 100 U / day of sCT (if bCT would have the same potency than sCT). The in vivo release rate evaluation is made by determining the radioactivity remaining at the site of injection.

2. Materials and methods

2.1. Materials bCT, low viscosity methyl cellulose and Stannous octoate were purchased from Sigma. Lactide was obtained from Aldrich, glycolide from Boehringer (Ingelheim) and polyethyleneglycol 6000 (PEG 6000) from J. Escuder. All other chemicals were obtained from Merck.

2.2. Methods 2.2.1. Polymer synthesis and characterization The polymers were synthesized by the ring opening reaction of DL-lactide and glycolide according to the method described by Kulkarni et al. [7] and Gref et al. [8], using Stannous octoate as catalyst. The molecular weight of the polymers were determined by the gel permeation method.

2.2.2. Microspheres preparation Microspheres were prepared as previously described [6]. In brief, bCT (1200 mg) was dissolved in 30% (v / v) acetic acid solution (500 ml or 1000 ml) containing 125 I-bCT (7 MBq), and emulsified in 5 ml of a polymer methylene chloride solution (150 mg / ml) by sonication at output 4 (50 W) for 30 s (Ultrasonic probe, TDI). The resulting emulsion was poured into 200 ml of an aqueous solution of polyvinyl alcohol (0.1%) and mixed vigorously using a homogenizer (Ultra-Turrax T 25, Jauke and Kunkel IKA-Werk) at 8000 rpm for 1 min. In order to evaporate the solvent, the mixture was allowed to stir for a further 2 h at 250 rpm at room temperature. The microspheres were collected after centrifugation, washing and freeze-drying. 2.2.3. Iodination of bCT Iodination of bCT by the chloramine-T method [9] and purification of the reaction mixtures were carried out as previously described [6], resulting in a labelled bCT with the specific activity of about 30?10 3 GBq / mol. In brief, 5 ml of Na 125 I (18.5 MBq) and 10 ml of chloramine-T (2 mg / ml) was added to 2 mg of bCT dissolved in 20 ml of phosphate buffer (0.01 M, pH 7.5). After mixing for 90 s, the reaction was stopped by addition of 50 ml of sodium metabisulphite (2.5 mg / ml) by reducing the excess chloramine-T. The separation of labelled bCT from unreacted iodide was achieved by passing the iodination reaction mixture through a prepacked column (Sephadex G-50 column 3030.9 cm). Labelled bCT was eluted from the column with 100 ml of 1 M acetic acid containing 1% bovine serum albumin (BSA). The radioactivity in all the fractions were measured and two or three fractions with the highest specific activities were pooled and passed through a second prepacked shorter column (Sephadex G-50 column 2730.7 cm), labelled bCT was eluted from the column with 10 mM acetic acid containing 1% BSA. 2.2.4. Peptide encapsulation efficiency The loading efficiency was calculated by referring the percentage of encapsulated labelled peptide to the total activity of labelled CT used to prepare the microspheres. The encapsulated labelled CT was determined measuring the radioactivity in the super-

R.V. Diaz et al. / Journal of Controlled Release 59 (1999) 55 – 62

natant and in an aliquot of freeze-dried microspheres in the g-counter (g-Counter Cobra Model, Packard).

2.2.5. In vitro release assays The release studies were carried out in triplicate in tubes and dialysis bags. 2.2.5.1. Tubes assays. Microspheres (2 mg) were dispensed into silanised tubes and incubated in 1 ml of isotonic phosphate-buffered saline (PBS) pH 7.4 containing 0.001% Tween-80 and 0.02% sodium azide at 378C for 6 weeks. 2.2.5.2. Dialysis bag assays. Microspheres (6 mg) were placed in a dialysis bag (Spectra / Por Membranes, MWCO: 12–14 000) containing 3 ml of isotonic PBS pH 7.4 with 0.001% Tween-80 and 0.02% sodium azide and suspended in an external medium (80 ml of the same buffer) at 378C. At each time 3 ml were withdrawn and replaced again with fresh buffer. 2.2.6. In vivo release assays The experiments were carried out in male Wistar rats (200–250 g), the rats were purchased from the Central University animal house. The animals had free food and water before and during the experimental time. The 125 I-bCT microspheres suspension were prepared in 250 ml sterile saline 0.5% methyl cellulose (low viscosity) solution. Three group of ten rats were administered (SC injection) to test the three different types of formulations, namely, PLGA I, PEG–PLGA and PLGA II.

57

2.2.7. In vivo 125 I-bCT measure The radioactivity at the site of injection was measured using an Iodine-detector (I-detector, Captus, Nuclear Iberica) with a collimator as previously described [6]. The release of the 125 I-bCT was followed periodically accumulating a 1 min spectrum at 27 KeV of the injection site over a maximum period of 30 days. The radioactivity measurement made at time zero is considered to be equivalent to 100% the injection dose. Then, the radioactivity was measured five times at different sampling times, and the median of five measurements were taken as the ‘‘true’’ value.

3. Results and discussion The release of 125 I-bCT from microspheres, prepared with three types of PLA copolymers (PLGA I, PEG–PLGA and PLGA II) using the double emulsion method, containing 0.2% of peptide, was determined. Polymer and microsphere characteristics are listed in Table 1. Our previous studies [6] suggested that 125 I-bCT inside the microspheres and in the PBS is stable after 6 weeks incubation in isotonic pH 7.4 at 378C.

3.1. Drug release in vitro Microspheres (batches A, B, C, D) prepared with PLGA I (MW530 000) and PEG–PLGA (MW5 34 000) showed a large initial burst, 20–40% of the dose was released during the first day of incubation followed by a very slow release rate (only 30–40%

Table 1 Polymer and microspheres characteristics Batch

Polymer

MW

Mn

LA–GA (mol%)

%IAP

d l,v (mm)

Percentile 25% (mm)

A B E F C D

PLGA I

30 000

22 000

60–40

PLGA II

12 000

9 500

45–55

PEG–PLGA

34 000

22 000

47–23 a

10 20 10 20 10 20

17.42 29.33 15.72 14.65 28.01 30.08

11.49 20.08 7.14 9.09 14.24 13.60

a

30% PEG. Weight average molecular weight (MW); number average molecular weight (Mn); molar percentage of lactide (LA) and glycolide (GA), determined by 1 H-NMR (Bruker, AMX-400); internal aqueous phase (IAP) and volume–length medium diameter (d l,v ) (Laser difraction, COULTER LS-100).

58

R.V. Diaz et al. / Journal of Controlled Release 59 (1999) 55 – 62

Fig. 1. Cumulative in vitro release of 125 I-bCT from PEG–PLGA (MW 34 000) microspheres in isotonic PBS (pH 7.4) at 378C using tubes (h batch C; s batch D) and dialysis bags (j batch C; d batch D).

was released in the following 29 days). The 125 I-bCT released percentage from PEG–PLGA obtained by the two in vitro release methods are shown in Fig. 1. The initial burst was always higher for formulations prepared with 20% of inner water phase, then the differences in the release rate were practically negligible. The in vitro release assays carried out in silanised tubes were abandoned in this study, since they did not offer any advantage over the dialysis bags. Dialysis bags as has been previously published by Park et al. [10] permitted a constant pH because water soluble oligomers, from polymer degradation, are removed avoiding protein and polymer degradation. On the other hand our previous studies showed that protein release rate in tubes was much slower than in vivo release. Dialysis bags simulate better the in vivo conditions and keep sink conditions. To increase the release rate, formulations E and F were prepared with PLGA II. This copolymer has a lower MW of 12 000 and the resulting microspheres have a smaller particle size and narrower particle size distribution, as shown in Table 1. The in vitro release profiles (Fig. 2), clearly showed that in 1

Fig. 2. Cumulative in vitro release of 125 I-bCT from PLGA II (MW 12 000) microspheres in isotonic PBS (pH 7.4) at 378C using dialysis bags (m batch E; ♦ batch F).

month approximately 50% of the dose was released, however, the release rate was too slow but constant. The initial burst is much lower than those obtained with microspheres prepared with the higher MW copolymers. The irregular release observed with formulations A, B, C and D could be partly due to the wider distribution size. In order to understand the in vitro release of 125 I-bCT from the microspheres the average molecular weights of polymers and the pH of the medium were monitored alongside the release assays. Polymer degradation was quantified by the degradation index (DI) [11], which is defined as the proportion of broken bonds in relation to the initial number of polymers molecules. The DI is calculated from the initial number-average molecular weight (Mn(0)) and the presented by the polymer after certain time (Mn(t)), trough the expression (Eq. (1)):

S

D

Mn(0) DI 5 ]] 2 1 Mn(t)

(1)

PLGA I and PEG–PLGA degradation profiles are very similar and show an increase in the DI (Fig. 3). As was expected, the pH did not change. The DI (Fig. 3) of both polymers increased from time zero

R.V. Diaz et al. / Journal of Controlled Release 59 (1999) 55 – 62

59

decrease. Therefore, as the amount of bCT used for microspheres preparation is very low it was assumed that adsorption of bCT could occur.

3.2.

Fig. 3. Degradation index of PLGA I (j), PLGA II (h) and PEG–PLGA (m), and pH (3) of the medium along the release assays.

to 24 h, indicating an initial degradation while the DI of PLGA was slightly reduced due to the solubilization of smallest chains, so the bulk molecular weight is practically unchanged along the release assay. Fig. 4 shows that PLGA II microspheres (batch F) after 1-month incubation are disintegrated and look like a polymer mass while the PLGA I (batch A) microspheres kept their integrity. The initial burst observed with formulations A, B, C and D, prepared with PLGA I and PEG–PLGA, was probably due to the initial polymer degradation that allowed the release of 125 I-bCT next to the surface. The microspheres prepared with those copolymers (PLGA I and PEG–PLGA) have a more compact core, evidenced by the increase in viscosity of the first water / oil emulsion and after 1-month incubation they still kept their integrity. Therefore, the diffusion of 125 I-bCT from the interior is much more difficult than from microspheres prepared with lower MW copolymers (formulation E and F) which could release the 125 I-bCT by diffusion, erosion and fragmentation (Fig. 4). The in vitro slow release rate of all these formulations could be due to the adsorption of 125 I-bCT to the polymers. Calis et al. [12] have shown adsorption of sCT to this type of biodegradable polymer. On the other hand Bodmer et al. [13] and Soriano et al. [14] have shown that the adsorption of BSA and bovine insulin increase when the MW of these polymers

125

I-bCT release in vivo

In vivo 125 I-bCT release was followed by measuring the radioactivity at the site of injection using the I-detector. In order to confirm that the observed release profile was that of the iodinated CT and not free iodide, Na 125 I and 125 I-bCT solutions were injected and the results showed that the elimination rate of Na 125 I was much faster than that of 125 I-bCT [6]. More than 90% of the Na 125 I dose was eliminated from the site of injection in 1 h compared to 125 I-bCT cleared after approximately 20 h. MacLean et al. [15] used a similar method for I-BSA in hydrogel, they also confirmed that the elimination rate from the site of injection of KI was faster than the I-BSA. In vivo release profiles of 125 I-bCT from formulations B (PLGA I), C and D (PEG–PLGA) in rats after SC injection are shown in Fig. 5. About 70– 80% of the dose is released in 1 month but the initial burst is high, even formulation C with 10% of inner water phase released more than 30% of the dose in the first 24 h. Therefore, these three formulations cannot be accepted as 1-month sustained release formulations. In the case of the PEG–PLGA microspheres, in vitro release correlated well with in vivo release, whereas PLGA I formulations did not have a good correlation (data not shown). The in vitro initial burst effects correlated well with in vivo initial release in all cases. Microspheres prepared with PLGA II (MW5 12 000) release 100% of the 125 I-bCT in 1 month (Fig. 6), in vivo release profiles presented two phases, during the first 2 weeks approximately 70% of the dose was released, followed by a second slower phase. As is shown in Fig. 6 very small and insignificant differences can be noticed in the in vivo release profile of formulations E and F. The in vitro and vivo correlation for 1-month controlled release microspheres is shown in Fig. 7. The in vivo release is much faster than the in vitro one, but presented a regular release rate in both

60

R.V. Diaz et al. / Journal of Controlled Release 59 (1999) 55 – 62

Fig. 4. Microphotograph of PLGA microspheres. Batch A: before (a) and after (b) 4 weeks incubation; Batch F: before (c) and after (d) 4 weeks incubation.

cases. The reason for these differences is not clear, these polymers are biocompatible, but might be that cellular immune responses are not completely avoided. Table 1 is a summary of particle characteristics, showing formulations E and F have smaller size than the others. It has been suggested that around 12 mm size could be the maximum phagocytic capability of cells [16]. As the 25% percentile (in volume) of formulations E and F are particles smaller than 10 mm, for these

particles the probability of being phagocytosed is higher than the other formulations.

4. Conclusion In summary, we have presented the results obtained from the study of three difference biodegradable polymers as carriers of 125 I-bCT, formulated in

R.V. Diaz et al. / Journal of Controlled Release 59 (1999) 55 – 62

61

Fig. 5. Percentage of dose remaining in the injection site vs. time after subcutaneous injection in rats (——— batch B; j batch C; d batch D).

Fig. 7. In vivo–in vitro correlation of 125 I-bCT released from PLGA II microspheres (m batch E; ♦ batch F).

microspheres, indicating that it is possible to obtain a 1-month release formulation. In rats, administration of a dose of 10 mg / kg of microspheres prepared using PLGA (45:55) with a MW of 12 000 resulted

in a release rate very close to the recommended therapeutic dose for long-term therapy.

Acknowledgements This work was supported by CITY as part of Project FAR 91-1106. R. V. Diaz acknowledges the support of the Direction General de Universidades ´ del Gobierno Autonomo de Canarias.

References

Fig. 6. Percentage of dose remaining in the injection site vs. time after subcutaneous injection in rats (m batch E; ♦ batch F).

[1] K.C. Lee, E.E. Soltis, P.S. Newman, K.W. Burton, R.C. Melita, P.P. Deluca, In vivo assessment of salmon calcitonin sustained release from biodegradable microspheres, J. Control. Release 17 (1991) 199–206. [2] M. Asano, M. Yoshida, H. Omichi, T. Mashimo, K. Okada, H. Yuasa, H. Yamanaka, S. Morimoto, H. Sakakibara, Biodegradable poly( DL-lactic acid) formulations in a calcitonin delivery system, Biomaterials 14 (1993) 797–799. [3] Tasset, N. Barette, S. Thysman, J.M. Ketelsleger, D. ´ Lemoine, V. Preat, Polyisobutylcyanoacrylate nanoparticles as sustained release system for calcitonin, J. Control. Release 33 (1995) 23–30. [4] J.L. Richardson, P.A. Ramires, M. Miglietta, M. Rochira, L.

62

[5]

[6]

[7]

[8]

[9]

[10]

R.V. Diaz et al. / Journal of Controlled Release 59 (1999) 55 – 62 Bacelle, L. Callegaro, L.M. Benedetti, Novel vaginal delivery systems for calcitonin: I Evaluation of HYAFF / calcitonin microspheres in rats, Int. J. Pharm. (1995) 9–15. R.C. Mehta, B.C. Thanoo, P.P. Deluca, Peptide-containing microspheres from low molecular weight and hydrophilic poly( D,L-lactide-co-glycolide), J. Control. Release 41 (1996) 249–257. ´ ´ C. Evora, R.V. Dıaz, I. Soriano, A. Delgado, M. Llabres, Effect of surfactant agents on the release of 125 I-bovine calcitonin from PLGA microspheres: in vitro—in vivo study, J. Control. Release 43 (1997) 59–64. R.C. Kulkarni, K.C. Pani, C. Newman, F. Leonard, Polylactic acid for surgical implants, Arch. Surg. 93 (1966) 839– 843. R. Gref, Y. Minamitake, M.T. Peracchia, V. Trubetskoy, V. Torchilin, R. Langer, Biodegradable long-circulating polymeric nanospheres, Science 263 (1994) 1600–1603. ¨ L. Sjodin, T. Nederman, M. Prahl, K. Montelius, Radioreceptor assay for formulations of salmon calcitonin, Int. J. Pharm. (1990) 135–142. T.G. Park, W. Lu, G. Crotts, Importance of in vitro experimental conditions on protein encapsulated poly( D,L-lactic acid-co-glycolic acid) microspheres, J. Control. Release 33 (1995) 211–222.

[11] P.A.R. Glynn, B.M.E. Var der Hoff, P.M. Reilly, A general model for prediction of molecular weight distributions of degraded polymers. Development and comparison with ultrasonic degradation experiments, J. Macromol. Sci. Chem. A6 (1976) 1653–1664. [12] S. Calis, R. Jejanthi, T. Tsai, R.C. Mehta, P.P. Deluca, Adsorption of salmon calcitonin to PLGA microspheres, Pharm. Res. 12 (1995) 1072–1076. [13] D. Bodmer, T. Kissel, E. Traechslin, Factors influencing the release of peptides and proteins from biodegradable parenteral depot systems, J. Control. Release 21 (1992) 129–138. ´ C. Evora, Release control of albumin [14] I. Soriano, M. Llabres, from polylactic acid microspheres, Int. J. Pharm. 125 (1995) 223–230. [15] D.S. MacLean, J.D. Robertson, M. Jay, D.J. Stalker, Noninvasive measurement of protein release from subcutaneous depot formulations in vivo using X-ray fluorescence, J. Control. Release 34 (1995) 167–173. [16] M. Kanke, I. Sniecinski, P. DeLuca, Interaction of microspheres with blood constituents. I. Uptake of polystyrene spheres by monocytes and granylocytes and effect on immune responsiveness of lymphocytes, J. Parent. Sci. Technol. 37 (1983) 210–217.