In vitro and in vivo preclinical evaluation of a minisphere emulsion-based R of salmon calcitonin formulation (SmPill ) Tanira A.S. Aguirre, M´onica Rosa, Ivan Coulter, David J. Brayden PII: DOI: Reference:
S0928-0987(15)30009-9 doi: 10.1016/j.ejps.2015.09.001 PHASCI 3352
To appear in: Received date: Accepted date:
9 June 2015 1 September 2015
Please cite this article as: Aguirre, Tanira A.S., Rosa, M´ onica, Coulter, Ivan, Brayden, David J., In vitro and in vivo preclinical evaluation of a minisphere emulsion-based R of salmon calcitonin, (2015), doi: 10.1016/j.ejps.2015.09.001 formulation (SmPill)
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
T
ACCEPTED MANUSCRIPT
IP
In vitro and in vivo preclinical evaluation of a minisphere emulsion-based
MA
NU
SC R
formulation (SmPill®) of salmon calcitonin
UCD School of Veterinary Medicine and UCD Conway Institute, University College Dublin,
CE P
a
TE
D
Tanira A. S. Aguirre a,b, Mónica Rosab, Ivan Coulterb, David J. Braydena*
Belfield, Dublin 4; bSigmoid Pharma, Dublin City University, Invent Centre, Dublin 9,
AC
Ireland
*Corresponding author. Tel: +353 1 7166013; fax: +353 1 7166104 E-mail address: [email protected]
1
ACCEPTED MANUSCRIPT Abstract Salmon calcitonin (sCT, MW 3432 Da) is a benchmark molecule for an oral peptide delivery
T
system because it is degraded and has low intestinal epithelial permeability. Four dry
IP
emulsion minisphere prototypes (SmPill®) containing sCT were co-formulated with
SC R
permeation enhancers (PEs): sodium taurodeoxycholate (NaTDC), sodium caprate (C10) or coco-glucoside (CG), or with a pH acidifier, citric acid (CA). Minispheres protected sCT from thermal degradation and the released sCT retained high bioactivity, as determined by
NU
cyclic AMP generation in T47D cells. Pre-minisphere emulsions of PEs combined with sCT
MA
increased absolute bioavailability (F) compared to native sCT following rat intra-jejunal (i.j.) and intra-colonic (i.c.) loop instillations, an effect that was more pronounced in colon.
D
Minispheres corresponding to ~2000 I.U. (~390 µg) sCT /kg were instilled by i.j. or i.c.
TE
instillations and hypocalcaemia resulted from all prototypes. The absolute F (i.j.) of sCT was 11.0, 4.8, and 1.4 % for minispheres containing NaTDC (10 µmol/kg), CG (12 µmol/kg) or
CE P
CA (32 µmol/kg) respectively. For i.c. instillations, the largest absolute F (22% in each case) was achieved for minispheres containing either C10 (284 µmol/kg) or CG (12 µmol/kg), while
AC
the absolute F was 8.2% for minispheres loaded with CA (32 µmol/kg). In terms of relative F, the best data were obtained for minispheres containing NaTDC (i.j.), a 4 fold increase over sCT solution, and also for either C10 or CG (i.c.), where there was a 3 fold increase over sCT solution. Histology of instilled intestinal loops indicated that neither the minispheres nor components thereof caused major perturbation. In conclusion, selected SmPill® minisphere formulations may have the potential to be used as oral peptide delivery systems when delivered to jejunum or colon.
2
ACCEPTED MANUSCRIPT Keywords: salmon calcitonin, oral bioavailability, intestinal permeation enhancers, oral
IP
T
peptide and protein delivery, oil-in-water emulsions.
SC R
1 Introduction
One of the most common approaches involving lipid emulsions for oral drug delivery is use
NU
of self-emulsifying systems (SES). They are composed of oil, surfactant, co-surfactant, and co-solvents in different ratios in an absence of a water phase. SES form transparent isotropic
MA
mixtures, which spontaneously form fine oil-in-water (O/W) emulsions in aqueous GI fluids using the agitation provided by gastric motility (Kohli et al., 2010). Improved bioavailability
D
of drugs formulated with self-emulsifying techniques is attributed in part to creation of small
TE
oil droplets, which provide a large interfacial area for pancreatic lipase-mediated hydrolysis of triglycerides to promote rapid release of the drug or formation of mixed micelles with bile
CE P
salts containing the active pharmaceutical ingredient (API) (Kommuru et al., 2001). Lipidbased drug delivery systems have been used successfully for the formulation of small
AC
molecules with solubility issues from the Biopharmaceutical Classification System’s Class II and IV (Shaikhn et al., 2012). SES may also be applied to Class III molecules as well as for peptides, since they can potentially overcome enzymatic degradation and low epithelial permeability (Kohli et al., 2010). Solid oral dosage forms using self-emulsifying O/W approaches include soft gelatin capsules, pellets, microspheres and minispheres. For example, porous polystyrene minispheres were used as carriers to aid solidification of SES containing Captex 200® (oil), Cremophor EL® (surfactant) and Capmul MCM® (co-surfactant) to formulate laratidine (Patil and Paradkar, 2006). Among formulation approaches to obtain biocompatible minispheres are the use of calcium pectinate gels (Günter et al., 2014), chitosan (Shu and Zhu, 2000) and hydrogels (Mohamadnia et al., 2007). These minispheres 3
ACCEPTED MANUSCRIPT are typically formed by ionotropic gelation of a solution containing the drug and the gelforming agent, which is dropped into a crosslinking solution containing coating compounds.
T
However, a problem associated with minispheres is low mechanical strength leading to sub-
SC R
IP
optimal release profiles (Shu and Zhu, 2000).
A novel oral system, SmPill® (Sigmoid Pharma, Ireland), contains emulsions formulated as minispheres, which are subsequently loaded into a gelatin capsule as the final dosage form.
NU
SmPill® presents the drug in a dissolved form and the release profile can in theory be
MA
designed to target the optimal intestinal region for absorption (Coulter, 2010; Moodley and Coulter, 2008; Sigmoid P., 2014). Importantly, the combination of components in the
D
formulation provides sufficient mechanical strength and, pre-solubilisation in the emulsion
TE
enables adaptation for hydrophilic or lipophilic molecules. An additional feature for delivery of poorly permeable molecules is the potential to incorporate intestinal permeation enhancers
CE P
(PEs), of which C10 (Maher et al., 2009), sodium taurodeoxycholate (NaTDC) (Mrestani et al., 2003) and coco glucoside (CG) (Aguirre et al., 2014) were selected here. Acidifying
AC
excipients, including citric acid (CA), can also be added in order to protect peptides from serine proteases (Welling et al., 2014). In the final aspect of production, minispheres can be coated with polymers and then presented in capsules designed to target small intestinal or colonic regions of the GI tract, in order to either to respectively maximise systemic absorption or to locally target the drug. A related approach has been used in marketed products in the latter case with coated tablets of budesonide that dissolve when the pH increases to >7.0 in the colon of patients with mild-to-moderate ulcerative colitis (Sandborn et al., 2015). Colonic targeting is the principle behind the SmPill®-enabled cyclosporine A formulation, CyCol®, which is designed to provide local topical delivery of solubilized cyclosporine, but to avoid systemic exposure in the treatment of moderate-severe ulcerative
4
ACCEPTED MANUSCRIPT colitis [55]. It has completed two Phase II studies in ulcerative colitis (NCT01033305, 2012;
T
NCT02130414, 2014).
IP
To investigate the potential for SmPill® to enhance systemic bioavailability of a hydrophilic
SC R
peptide, we used salmon calcitonin (sCT, MW 3432) as a model because, due to its sensitivity to pancreatic serine proteases and poor intestinal permeability, it is useful in evaluating delivery technologies (Maher and Brayden, 2012)
ar ete nasal ersions of
NU
pepti es inclu in sC are estimate to ha e absolute bioa ailability of
1% (Grant and
MA
Leone-Bay, 2012), and the most advanced oral formulation of sCT, OSTORATM (Tarsa Therapeutics, USA), recently completed Phase III to give comparable pharmacokinetics to
D
the nasal comparator (Binkley et al., 2012). OSTORA TM is a Eudragit® L 30 D-55-coated
TE
tablet in which sCT and CA are co-localised in vesicles in the core, so it lacked a recognised PE in order to facilitate the regulatory path, although this may partially account for the 1%. We demonstrate that sCT bioactivity was not reduced during minisphere manufacturing and that the inclusion of selected PEs enable significant systemic bioavailability enhancement in both regions. 2.1 Material and methods
5
ACCEPTED MANUSCRIPT All chemicals were obtained from Sigma Aldrich, Ireland, except for CG, which was supplied as Plantacare® 818 UP (Cognis, Germany). Media, buffers and supplements were obtained
T
from GIBCO®, Ireland. Synthetic sCT was purchased from Polypeptide Laboratories
IP
(Copenhagen, Denmark); it had an average activity of 5100 I.U./mg. Transcutol HP®,
SC R
Cremophor EL®, and Miglyol 818® were obtained from Gattefossé (France), BASF (Germany) and Sasol (UK) respectively. The ParameterTM cAMP ELISA was obtained from
MA
2.2 sCT-SmPill® minispheres preparation
NU
R&D Systems (UK).
sCT minispheres were prepared following a standard operation procedure with minor modifications (Coulter et al., 2010). Approximately 1700mg of oil phase composed of
TE
D
Transcutol HP®, Cremophor EL®, and Miglyol 818® at a ratio of 57: 24: 19 (w/w) respectively, was mixed with approximately 24 mL of aqueous phase, comprising gelatin, PE
CE P
(NaTDC, C10, or CG), or CA to form a homogeneous pre-minisphere emulsion. 1 mL of a 20 mg/mL sCT solution in water was added to the emulsion and mixed at 65°C for 5 min. The
AC
sCT-containing final emulsion was then extruded into cold oil to form minispheres. Minispheres were kept in oil at 2-8°C for 30 min, separated with a sieve and allowed to dry at 2-8°C. The theoretical percentage of each component in dried minispheres is shown (Table 1).
6
ACCEPTED MANUSCRIPT 2.3 Rotational viscometer
Rheological characteristics of sCT-SmPill® pre-minisphere emulsions were determined using
IP
T
a rotational viscometer (DV-II + PRO Di ital Viscometer, Broo fiel Instruments, UK) with
SC R
an UL adapter and a LV1 spindle. Analysis was carried out at 60°C, varying the speed from 0.1 to 0.6 RPM in order to maintain the torque lower than 100%.
NU
2.4 Multiple light scattering
MA
The physical stability of sCT-SmPill® pre-minisphere emulsions were analysed by multiple light scattering using Turbiscan® Lab equipment (Formulaction Ltd, France). Samples were
D
dispensed into the equipment glass cell and were analysed at 50°C with scans every 5 min for
TE
20 min. The detection is made for the whole height of the sample from the bottom to the top
CE P
of the cell every 40µm. Data was acquired with synchronous detectors which receives transmitted light (T) through the sample (at 180°) and backscattered light (BS) by the sample (at 45°)
he li ht source was an electro luminescent io e in the near infrare (λ=880 nm)
AC
This technique is useful for identification of particle migration (sedimentation and creaming) and particle size variation (coalescence and flocculation).
2.5 Scanning electron microscopy of sCT-SmPill® minispheres
Shape and surface of minispheres were examined with a scanning electron microscopy (SEM) (Jeol Scanning Microscope, JSM-5800, Tokyo, Japan). The minispheres were carbon and gold sputtered (Jeol Jee 4B SVG-IN, Tokyo, Japan) before analysis.
2.6 sCT in vitro cyclic AMP bioactivity assay in T47D cells 7
ACCEPTED MANUSCRIPT A concern over formulating sCT in SmPill® was maintenance of sCT bioactivity after incorporation into emulsions at 65°C, followed by release. Two prototype sCT minispheres
T
were prepared with gelatin, sorbitol and the three oil phase components, but without any PE
IP
or CA. The influence of temperature was assessed by examining the effects of incubating
SC R
sCT in the emulsion at 65°C for either 5 min or 15 h. Negative control minispheres without sCT were prepared under parallel conditions. Dried minispheres were dissolved in PBS at 37°C using an orbital shaker prior adding to cells. T47D cell culture was carried out as
NU
described elsewhere (Ryan et al., 2009), with the density of seeded cells altered to 2.5x105
MA
cells/well. The theoretical loading of the two prototype formulations was approximately 0.4 mg sCT/100 mg minispheres (Table 1), and individual calculations were carried for
D
prototypes in order to allow the same concentration of minisphere-released sCT to be
TE
exposed to T47D cells. Dilutions were carried in PBS supplemented with 0.2mM IBMX to obtain a final concentration of 10 nM sCT to be incubated with cells. Native sCT solution
CE P
was prepared in PBS supplemented with 0.2 mM IBMX and added to T47D cells as a positive control. 0.2 mM IBMX in PBS was used as a control for basal levels of cAMP.
AC
Further experiments to assess the effect of heat exposure duration on free sCT bioactivity were carried by incubating a solution of sCT in water at 65°C for 1, 5, 20 and 900 min. In addition, selected PEs were added to solutions of native sCT in concentrations that mimicked similar ratios to those in minispheres (Table 1), and these combinations were incubated for 15 h at 65°C. The cyclic AMP ELISA on T47D cells was carried out according to previous descriptions (Fowler et al., 2005).
2.7 Rat intra-intestinal instillations
Animal experimental procedures were performed in compliance with the Irish Department of Health and Children licence number, B100/4193, in compliance with EC Directive 8
ACCEPTED MANUSCRIPT 86/609/EEC. Male Wistar rats weighing 280-340 g were housed under controlled environmental conditions with a 12:12 h light/dark cycle. Anaesthesia was induced with
T
isoflurane (Iso-Vet, 1000 mg/g isoflurane liquid for inhalation, Piramal Healthcare, UK)
IP
using a vaporising unit with suitable delivery mask (Blease Medical Equipment Ltd., UK), at
SC R
a rate of 4 L/min mixed with 1L/min O2. It was maintained by reducing the rate of isoflurane to 2 L/min. Rats were euthanized at the end of experiments by overdosing with 0.5ml EuthatalTM (Merial Animal Health Ltd., UK). Absorption studies were carried out using an in
NU
situ instillation method as previously described (Cheng et al., 2010), but with minor
MA
modifications. Rats were given water supplemented with 5% glucose two days prior to experiments to avoid hypoglycaemia during fasting period and surgery. Animals were fasted
D
for approximately 18 h before anaesthesia. The proximal jejunum or descending colon were
TE
exposed after a midline laparotomy. Approximately 5 cm jejunum or colon were isolated by tying the extremities up with size 4 braided silk suture (Mersilk®, Ethicon Ltd., UK), taking
CE P
care to avoid damage to blood vessels. Pre-minisphere emulsions with a concentration of 195 µg sCT/ml (1000 I.U./ml) were prepared maintaining component ratios (Table 1). Emulsions
AC
were warmed to 37°C before instilling 2000 I.U. sCT/kg (test) or PBS (control) using a 1ml syringe fitted with a 30G needle.
For instillation of sCT-SmPill® minispheres, the proximal extremity of the jejunal or colonic loop was tightened with silk and a small cut was made distally. A 2mm diameter cannula was inserted 10 mm into the hole, and sCT-SmPill® minispheres corresponding to ~2000 I.U. sCT/kg (~390 µg sCT/kg) were instilled with a forceps. The segment was then tightened as far as possible at the hole in order to avoid the absorption through damaged tissue; 500µl of water-for-injection was then instilled. Exposed segments were inspected to make sure there was no leakage of solutions or bleeding, and were then returned to the abdominal cavity and
9
ACCEPTED MANUSCRIPT the outer skin layer was sutured using Mersilk® W505 sutures with a cutting needle. Blood samples totalling 350 µl were taken from anaesthetised animals via retro orbital venepuncture
T
using a glass capillary tube of 1.35 mm internal diameter cut in two (75 mm total length) at
IP
pre-determined intervals. Samples were kept on ice before being centrifuged (6500 g, 5 min,
SC R
4ºC). The serum was transferred into fresh tubes and kept at -20ºC before analysis.
NU
2.8 Serum calcium analysis: pharmacodynamics (PD)
Serum samples were thawed, vortexed and centrifuged at 5,000 g at 4ºC for 5 min to remove
MA
excess lipid layers. 60-100 μl serum samples were transferre to micro-tubes fitted with 85 mm tubes. In some cases, 0.9% NaCl was used as a diluent to expand the sample sizes when
D
less than 60 μl serum was a ailable A fully automate clinical chemical analyser (Ran ox,
TE
UK) was used for total calcium measurement (Reagent CA 3871, Randox) by a colorimetric
CE P
method (Ryan et al., 2009). Decreases in calcium serum were calculated by setting the initial basal concentration of calcium at 100 %.
AC
2.9 Serum sCT analysis: pharmacokinetics (PK)
sCT was measured in rat serum using an extraction-free sCT ELISA (Cat. S-1155, Peninsula Laboratories, USA) with detection limit of 25 ng/ml (Ryan et al., 2009). The OD of each well was determined within 30 min, using a colorimeter microplate reader set to 450 nm. The standard curve had a sigmoid shape revealing an inverse relationship between sCT concentration and absorbance. Results were calculated from the standard curve using the software GraphPad® Prism 5.0. According to a validation carried, the assay is highly sensitive and reproducible regardless of source of sCT or rat serum (Aguirre, 2013). Absolute sCT bioavailability (F) was calculated according to Equation 1, which compares the 10
ACCEPTED MANUSCRIPT bioavailability (estimated as area under the curve (AUC)) obtained from extravascular (intestinal regional instillation) delivery of sCT-SmPill® with the bioavailability following
IP
T
intravenous (i.v.) dosing of sCT (Gabrielsson and Weiner, 2000).
SC R
Equation 1
NU
where AUCe.v. and AUCi.v. denote the area under the curve for extravascular and intravenous concentration-time profiles, respectively; Dosee.v. and Dosei.v. are the respective doses. The
MA
relative F was calculated according to Equation 2, and measures the bioavailability of sCTSmPill® formulation compared to the bioavailability obtained for native sCT delivered in the
TE
D
same intestinal segment (Gabrielsson and Weiner, 2000).
CE P
Equation 2
Where
denotes the absolute bioavailability calculated for the extravascular
AC
dosing of the selected sCT-SmPill® formulation and
denotes the absolute
bioavailability calculated for the extravascular dosing of native sCT.
2.10 Histology
Following intra-intestinal instillation, tissues were immersed in 10% (v/v) buffered formalin for at least 48 h an embe e in paraffin wax 5 μm sections were cut on a microtome (Leitz 1512; GMI, USA), mounted on adhesive coated slides, and stained with haematoxylin and eosin (H & E) or alcian blue. Slides were visualized under a light microscope (NanoZoomer 2.0-HT light microscopy, Hamamatsu) and images were taken with high-resolution camera 11
ACCEPTED MANUSCRIPT (Micropublisher 3.3 RTV; Q Imaging, Canada) and Image-Pro® Plus version 6.3 acquisition
T
software (Media Cybernetics Inc., USA).
IP
2.11 Statistics
SC R
Statistical analysis was carried out using Prism-5® software (GraphPad®, USA). Unpaired Stu ent’s t-tests or ANOVA with Dunnett's post-test were used as stated. Results were
NU
expressed as the mean ± SEM. A significant difference was considered to be present if P
S0928-0987(15)30009-9 doi: 10.1016/j.ejps.2015.09.001 PHASCI 3352
To appear in: Received date: Accepted date:
9 June 2015 1 September 2015
Please cite this article as: Aguirre, Tanira A.S., Rosa, M´ onica, Coulter, Ivan, Brayden, David J., In vitro and in vivo preclinical evaluation of a minisphere emulsion-based R of salmon calcitonin, (2015), doi: 10.1016/j.ejps.2015.09.001 formulation (SmPill)
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
T
ACCEPTED MANUSCRIPT
IP
In vitro and in vivo preclinical evaluation of a minisphere emulsion-based
MA
NU
SC R
formulation (SmPill®) of salmon calcitonin
UCD School of Veterinary Medicine and UCD Conway Institute, University College Dublin,
CE P
a
TE
D
Tanira A. S. Aguirre a,b, Mónica Rosab, Ivan Coulterb, David J. Braydena*
Belfield, Dublin 4; bSigmoid Pharma, Dublin City University, Invent Centre, Dublin 9,
AC
Ireland
*Corresponding author. Tel: +353 1 7166013; fax: +353 1 7166104 E-mail address: [email protected]
1
ACCEPTED MANUSCRIPT Abstract Salmon calcitonin (sCT, MW 3432 Da) is a benchmark molecule for an oral peptide delivery
T
system because it is degraded and has low intestinal epithelial permeability. Four dry
IP
emulsion minisphere prototypes (SmPill®) containing sCT were co-formulated with
SC R
permeation enhancers (PEs): sodium taurodeoxycholate (NaTDC), sodium caprate (C10) or coco-glucoside (CG), or with a pH acidifier, citric acid (CA). Minispheres protected sCT from thermal degradation and the released sCT retained high bioactivity, as determined by
NU
cyclic AMP generation in T47D cells. Pre-minisphere emulsions of PEs combined with sCT
MA
increased absolute bioavailability (F) compared to native sCT following rat intra-jejunal (i.j.) and intra-colonic (i.c.) loop instillations, an effect that was more pronounced in colon.
D
Minispheres corresponding to ~2000 I.U. (~390 µg) sCT /kg were instilled by i.j. or i.c.
TE
instillations and hypocalcaemia resulted from all prototypes. The absolute F (i.j.) of sCT was 11.0, 4.8, and 1.4 % for minispheres containing NaTDC (10 µmol/kg), CG (12 µmol/kg) or
CE P
CA (32 µmol/kg) respectively. For i.c. instillations, the largest absolute F (22% in each case) was achieved for minispheres containing either C10 (284 µmol/kg) or CG (12 µmol/kg), while
AC
the absolute F was 8.2% for minispheres loaded with CA (32 µmol/kg). In terms of relative F, the best data were obtained for minispheres containing NaTDC (i.j.), a 4 fold increase over sCT solution, and also for either C10 or CG (i.c.), where there was a 3 fold increase over sCT solution. Histology of instilled intestinal loops indicated that neither the minispheres nor components thereof caused major perturbation. In conclusion, selected SmPill® minisphere formulations may have the potential to be used as oral peptide delivery systems when delivered to jejunum or colon.
2
ACCEPTED MANUSCRIPT Keywords: salmon calcitonin, oral bioavailability, intestinal permeation enhancers, oral
IP
T
peptide and protein delivery, oil-in-water emulsions.
SC R
1 Introduction
One of the most common approaches involving lipid emulsions for oral drug delivery is use
NU
of self-emulsifying systems (SES). They are composed of oil, surfactant, co-surfactant, and co-solvents in different ratios in an absence of a water phase. SES form transparent isotropic
MA
mixtures, which spontaneously form fine oil-in-water (O/W) emulsions in aqueous GI fluids using the agitation provided by gastric motility (Kohli et al., 2010). Improved bioavailability
D
of drugs formulated with self-emulsifying techniques is attributed in part to creation of small
TE
oil droplets, which provide a large interfacial area for pancreatic lipase-mediated hydrolysis of triglycerides to promote rapid release of the drug or formation of mixed micelles with bile
CE P
salts containing the active pharmaceutical ingredient (API) (Kommuru et al., 2001). Lipidbased drug delivery systems have been used successfully for the formulation of small
AC
molecules with solubility issues from the Biopharmaceutical Classification System’s Class II and IV (Shaikhn et al., 2012). SES may also be applied to Class III molecules as well as for peptides, since they can potentially overcome enzymatic degradation and low epithelial permeability (Kohli et al., 2010). Solid oral dosage forms using self-emulsifying O/W approaches include soft gelatin capsules, pellets, microspheres and minispheres. For example, porous polystyrene minispheres were used as carriers to aid solidification of SES containing Captex 200® (oil), Cremophor EL® (surfactant) and Capmul MCM® (co-surfactant) to formulate laratidine (Patil and Paradkar, 2006). Among formulation approaches to obtain biocompatible minispheres are the use of calcium pectinate gels (Günter et al., 2014), chitosan (Shu and Zhu, 2000) and hydrogels (Mohamadnia et al., 2007). These minispheres 3
ACCEPTED MANUSCRIPT are typically formed by ionotropic gelation of a solution containing the drug and the gelforming agent, which is dropped into a crosslinking solution containing coating compounds.
T
However, a problem associated with minispheres is low mechanical strength leading to sub-
SC R
IP
optimal release profiles (Shu and Zhu, 2000).
A novel oral system, SmPill® (Sigmoid Pharma, Ireland), contains emulsions formulated as minispheres, which are subsequently loaded into a gelatin capsule as the final dosage form.
NU
SmPill® presents the drug in a dissolved form and the release profile can in theory be
MA
designed to target the optimal intestinal region for absorption (Coulter, 2010; Moodley and Coulter, 2008; Sigmoid P., 2014). Importantly, the combination of components in the
D
formulation provides sufficient mechanical strength and, pre-solubilisation in the emulsion
TE
enables adaptation for hydrophilic or lipophilic molecules. An additional feature for delivery of poorly permeable molecules is the potential to incorporate intestinal permeation enhancers
CE P
(PEs), of which C10 (Maher et al., 2009), sodium taurodeoxycholate (NaTDC) (Mrestani et al., 2003) and coco glucoside (CG) (Aguirre et al., 2014) were selected here. Acidifying
AC
excipients, including citric acid (CA), can also be added in order to protect peptides from serine proteases (Welling et al., 2014). In the final aspect of production, minispheres can be coated with polymers and then presented in capsules designed to target small intestinal or colonic regions of the GI tract, in order to either to respectively maximise systemic absorption or to locally target the drug. A related approach has been used in marketed products in the latter case with coated tablets of budesonide that dissolve when the pH increases to >7.0 in the colon of patients with mild-to-moderate ulcerative colitis (Sandborn et al., 2015). Colonic targeting is the principle behind the SmPill®-enabled cyclosporine A formulation, CyCol®, which is designed to provide local topical delivery of solubilized cyclosporine, but to avoid systemic exposure in the treatment of moderate-severe ulcerative
4
ACCEPTED MANUSCRIPT colitis [55]. It has completed two Phase II studies in ulcerative colitis (NCT01033305, 2012;
T
NCT02130414, 2014).
IP
To investigate the potential for SmPill® to enhance systemic bioavailability of a hydrophilic
SC R
peptide, we used salmon calcitonin (sCT, MW 3432) as a model because, due to its sensitivity to pancreatic serine proteases and poor intestinal permeability, it is useful in evaluating delivery technologies (Maher and Brayden, 2012)
ar ete nasal ersions of
NU
pepti es inclu in sC are estimate to ha e absolute bioa ailability of
1% (Grant and
MA
Leone-Bay, 2012), and the most advanced oral formulation of sCT, OSTORATM (Tarsa Therapeutics, USA), recently completed Phase III to give comparable pharmacokinetics to
D
the nasal comparator (Binkley et al., 2012). OSTORA TM is a Eudragit® L 30 D-55-coated
TE
tablet in which sCT and CA are co-localised in vesicles in the core, so it lacked a recognised PE in order to facilitate the regulatory path, although this may partially account for the 1%. We demonstrate that sCT bioactivity was not reduced during minisphere manufacturing and that the inclusion of selected PEs enable significant systemic bioavailability enhancement in both regions. 2.1 Material and methods
5
ACCEPTED MANUSCRIPT All chemicals were obtained from Sigma Aldrich, Ireland, except for CG, which was supplied as Plantacare® 818 UP (Cognis, Germany). Media, buffers and supplements were obtained
T
from GIBCO®, Ireland. Synthetic sCT was purchased from Polypeptide Laboratories
IP
(Copenhagen, Denmark); it had an average activity of 5100 I.U./mg. Transcutol HP®,
SC R
Cremophor EL®, and Miglyol 818® were obtained from Gattefossé (France), BASF (Germany) and Sasol (UK) respectively. The ParameterTM cAMP ELISA was obtained from
MA
2.2 sCT-SmPill® minispheres preparation
NU
R&D Systems (UK).
sCT minispheres were prepared following a standard operation procedure with minor modifications (Coulter et al., 2010). Approximately 1700mg of oil phase composed of
TE
D
Transcutol HP®, Cremophor EL®, and Miglyol 818® at a ratio of 57: 24: 19 (w/w) respectively, was mixed with approximately 24 mL of aqueous phase, comprising gelatin, PE
CE P
(NaTDC, C10, or CG), or CA to form a homogeneous pre-minisphere emulsion. 1 mL of a 20 mg/mL sCT solution in water was added to the emulsion and mixed at 65°C for 5 min. The
AC
sCT-containing final emulsion was then extruded into cold oil to form minispheres. Minispheres were kept in oil at 2-8°C for 30 min, separated with a sieve and allowed to dry at 2-8°C. The theoretical percentage of each component in dried minispheres is shown (Table 1).
6
ACCEPTED MANUSCRIPT 2.3 Rotational viscometer
Rheological characteristics of sCT-SmPill® pre-minisphere emulsions were determined using
IP
T
a rotational viscometer (DV-II + PRO Di ital Viscometer, Broo fiel Instruments, UK) with
SC R
an UL adapter and a LV1 spindle. Analysis was carried out at 60°C, varying the speed from 0.1 to 0.6 RPM in order to maintain the torque lower than 100%.
NU
2.4 Multiple light scattering
MA
The physical stability of sCT-SmPill® pre-minisphere emulsions were analysed by multiple light scattering using Turbiscan® Lab equipment (Formulaction Ltd, France). Samples were
D
dispensed into the equipment glass cell and were analysed at 50°C with scans every 5 min for
TE
20 min. The detection is made for the whole height of the sample from the bottom to the top
CE P
of the cell every 40µm. Data was acquired with synchronous detectors which receives transmitted light (T) through the sample (at 180°) and backscattered light (BS) by the sample (at 45°)
he li ht source was an electro luminescent io e in the near infrare (λ=880 nm)
AC
This technique is useful for identification of particle migration (sedimentation and creaming) and particle size variation (coalescence and flocculation).
2.5 Scanning electron microscopy of sCT-SmPill® minispheres
Shape and surface of minispheres were examined with a scanning electron microscopy (SEM) (Jeol Scanning Microscope, JSM-5800, Tokyo, Japan). The minispheres were carbon and gold sputtered (Jeol Jee 4B SVG-IN, Tokyo, Japan) before analysis.
2.6 sCT in vitro cyclic AMP bioactivity assay in T47D cells 7
ACCEPTED MANUSCRIPT A concern over formulating sCT in SmPill® was maintenance of sCT bioactivity after incorporation into emulsions at 65°C, followed by release. Two prototype sCT minispheres
T
were prepared with gelatin, sorbitol and the three oil phase components, but without any PE
IP
or CA. The influence of temperature was assessed by examining the effects of incubating
SC R
sCT in the emulsion at 65°C for either 5 min or 15 h. Negative control minispheres without sCT were prepared under parallel conditions. Dried minispheres were dissolved in PBS at 37°C using an orbital shaker prior adding to cells. T47D cell culture was carried out as
NU
described elsewhere (Ryan et al., 2009), with the density of seeded cells altered to 2.5x105
MA
cells/well. The theoretical loading of the two prototype formulations was approximately 0.4 mg sCT/100 mg minispheres (Table 1), and individual calculations were carried for
D
prototypes in order to allow the same concentration of minisphere-released sCT to be
TE
exposed to T47D cells. Dilutions were carried in PBS supplemented with 0.2mM IBMX to obtain a final concentration of 10 nM sCT to be incubated with cells. Native sCT solution
CE P
was prepared in PBS supplemented with 0.2 mM IBMX and added to T47D cells as a positive control. 0.2 mM IBMX in PBS was used as a control for basal levels of cAMP.
AC
Further experiments to assess the effect of heat exposure duration on free sCT bioactivity were carried by incubating a solution of sCT in water at 65°C for 1, 5, 20 and 900 min. In addition, selected PEs were added to solutions of native sCT in concentrations that mimicked similar ratios to those in minispheres (Table 1), and these combinations were incubated for 15 h at 65°C. The cyclic AMP ELISA on T47D cells was carried out according to previous descriptions (Fowler et al., 2005).
2.7 Rat intra-intestinal instillations
Animal experimental procedures were performed in compliance with the Irish Department of Health and Children licence number, B100/4193, in compliance with EC Directive 8
ACCEPTED MANUSCRIPT 86/609/EEC. Male Wistar rats weighing 280-340 g were housed under controlled environmental conditions with a 12:12 h light/dark cycle. Anaesthesia was induced with
T
isoflurane (Iso-Vet, 1000 mg/g isoflurane liquid for inhalation, Piramal Healthcare, UK)
IP
using a vaporising unit with suitable delivery mask (Blease Medical Equipment Ltd., UK), at
SC R
a rate of 4 L/min mixed with 1L/min O2. It was maintained by reducing the rate of isoflurane to 2 L/min. Rats were euthanized at the end of experiments by overdosing with 0.5ml EuthatalTM (Merial Animal Health Ltd., UK). Absorption studies were carried out using an in
NU
situ instillation method as previously described (Cheng et al., 2010), but with minor
MA
modifications. Rats were given water supplemented with 5% glucose two days prior to experiments to avoid hypoglycaemia during fasting period and surgery. Animals were fasted
D
for approximately 18 h before anaesthesia. The proximal jejunum or descending colon were
TE
exposed after a midline laparotomy. Approximately 5 cm jejunum or colon were isolated by tying the extremities up with size 4 braided silk suture (Mersilk®, Ethicon Ltd., UK), taking
CE P
care to avoid damage to blood vessels. Pre-minisphere emulsions with a concentration of 195 µg sCT/ml (1000 I.U./ml) were prepared maintaining component ratios (Table 1). Emulsions
AC
were warmed to 37°C before instilling 2000 I.U. sCT/kg (test) or PBS (control) using a 1ml syringe fitted with a 30G needle.
For instillation of sCT-SmPill® minispheres, the proximal extremity of the jejunal or colonic loop was tightened with silk and a small cut was made distally. A 2mm diameter cannula was inserted 10 mm into the hole, and sCT-SmPill® minispheres corresponding to ~2000 I.U. sCT/kg (~390 µg sCT/kg) were instilled with a forceps. The segment was then tightened as far as possible at the hole in order to avoid the absorption through damaged tissue; 500µl of water-for-injection was then instilled. Exposed segments were inspected to make sure there was no leakage of solutions or bleeding, and were then returned to the abdominal cavity and
9
ACCEPTED MANUSCRIPT the outer skin layer was sutured using Mersilk® W505 sutures with a cutting needle. Blood samples totalling 350 µl were taken from anaesthetised animals via retro orbital venepuncture
T
using a glass capillary tube of 1.35 mm internal diameter cut in two (75 mm total length) at
IP
pre-determined intervals. Samples were kept on ice before being centrifuged (6500 g, 5 min,
SC R
4ºC). The serum was transferred into fresh tubes and kept at -20ºC before analysis.
NU
2.8 Serum calcium analysis: pharmacodynamics (PD)
Serum samples were thawed, vortexed and centrifuged at 5,000 g at 4ºC for 5 min to remove
MA
excess lipid layers. 60-100 μl serum samples were transferre to micro-tubes fitted with 85 mm tubes. In some cases, 0.9% NaCl was used as a diluent to expand the sample sizes when
D
less than 60 μl serum was a ailable A fully automate clinical chemical analyser (Ran ox,
TE
UK) was used for total calcium measurement (Reagent CA 3871, Randox) by a colorimetric
CE P
method (Ryan et al., 2009). Decreases in calcium serum were calculated by setting the initial basal concentration of calcium at 100 %.
AC
2.9 Serum sCT analysis: pharmacokinetics (PK)
sCT was measured in rat serum using an extraction-free sCT ELISA (Cat. S-1155, Peninsula Laboratories, USA) with detection limit of 25 ng/ml (Ryan et al., 2009). The OD of each well was determined within 30 min, using a colorimeter microplate reader set to 450 nm. The standard curve had a sigmoid shape revealing an inverse relationship between sCT concentration and absorbance. Results were calculated from the standard curve using the software GraphPad® Prism 5.0. According to a validation carried, the assay is highly sensitive and reproducible regardless of source of sCT or rat serum (Aguirre, 2013). Absolute sCT bioavailability (F) was calculated according to Equation 1, which compares the 10
ACCEPTED MANUSCRIPT bioavailability (estimated as area under the curve (AUC)) obtained from extravascular (intestinal regional instillation) delivery of sCT-SmPill® with the bioavailability following
IP
T
intravenous (i.v.) dosing of sCT (Gabrielsson and Weiner, 2000).
SC R
Equation 1
NU
where AUCe.v. and AUCi.v. denote the area under the curve for extravascular and intravenous concentration-time profiles, respectively; Dosee.v. and Dosei.v. are the respective doses. The
MA
relative F was calculated according to Equation 2, and measures the bioavailability of sCTSmPill® formulation compared to the bioavailability obtained for native sCT delivered in the
TE
D
same intestinal segment (Gabrielsson and Weiner, 2000).
CE P
Equation 2
Where
denotes the absolute bioavailability calculated for the extravascular
AC
dosing of the selected sCT-SmPill® formulation and
denotes the absolute
bioavailability calculated for the extravascular dosing of native sCT.
2.10 Histology
Following intra-intestinal instillation, tissues were immersed in 10% (v/v) buffered formalin for at least 48 h an embe e in paraffin wax 5 μm sections were cut on a microtome (Leitz 1512; GMI, USA), mounted on adhesive coated slides, and stained with haematoxylin and eosin (H & E) or alcian blue. Slides were visualized under a light microscope (NanoZoomer 2.0-HT light microscopy, Hamamatsu) and images were taken with high-resolution camera 11
ACCEPTED MANUSCRIPT (Micropublisher 3.3 RTV; Q Imaging, Canada) and Image-Pro® Plus version 6.3 acquisition
T
software (Media Cybernetics Inc., USA).
IP
2.11 Statistics
SC R
Statistical analysis was carried out using Prism-5® software (GraphPad®, USA). Unpaired Stu ent’s t-tests or ANOVA with Dunnett's post-test were used as stated. Results were
NU
expressed as the mean ± SEM. A significant difference was considered to be present if P
