- Email: [email protected]
PEGylation of polylysine dendrimers improves absorption and lymphatic targeting following SC administration in rats
Journal of Controlled Release 140 (2009) 108–116
Contents lists available at ScienceDirect
Journal of Controlled Release j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j c o n r e l
PEGylation of polylysine dendrimers improves absorption and lymphatic targeting following SC administration in rats Lisa M. Kaminskas a, Jagannath Kota a, Victoria M. McLeod a, Brian D. Kelly b, Peter Karellas b, Christopher JH. Porter a,⁎ a b
Drug Delivery, Disposition and Dynamics, Monash Institute of Pharmaceutical Sciences, Monash University (Parkville Campus). 381 Royal Pde, Parkville, VIC, 3052, Australia Starpharma Holdings Ltd. 75 Commercial Rd, Melbourne, VIC, 3004, Australia
a r t i c l e
i n f o
Article history: Received 10 July 2009 Accepted 6 August 2009 Available online 15 August 2009
a b s t r a c t Polylysine dendrimers have potential as highly flexible, biodegradable nanoparticular carriers that may also promote lymphatic transport. The current study was undertaken to determine the impact of PEGylation on the absorption and lymphatic transport of polylysine dendrimers modified by surface derivatisation with PEG (200, 570 or 2000 Da) or 4-benzene sulphonate following SC or IV dosing. PEGylation led to the PEG200 derived dendrimer being rapidly and completely absorbed into the blood after SC administration, however only 3% of the administered dose was recovered in pooled thoracic lymph over 30 h. Increasing the PEG chain length led to a systematic decrease in absorption into the blood and an enhancement of the proportion recovered in the lymphatics (up to 29% over 30 h). For the PEG570 and PEG2000 derived dendrimers, indirect access to the lymph via equilibration across the capillary beds also appeared to play a role in lymphatic targeting after both IV and SC dosing. In contrast, the anionic benzene sulphonate-capped dendrimer was not well absorbed from the SC injection site (26% bioavailability) into either the blood or the lymph. The data suggest that PEGylated poly-L-lysine dendrimers are well absorbed from SC injection sites and that the extent of lymphatic transport may be enhanced by increasing the size of the PEGylated dendrimer complex. © 2009 Elsevier B.V. All rights reserved.
1. Introduction The lymphatic system plays a central role in the trafficking and transport of white blood cells, in the pathogenesis of diseases including HIV and metastitial tuberculosis and in the metastatic spread of many cancers [1–4]. For lymph resident diseases therefore, lymphatic targeting of therapeutic drugs (e.g. antivirals, cytotoxics or immunomodulators) or imaging agents is expected to provide advantage over conventional approaches that focus on drug delivery via the blood [5]. Previous studies have shown that after interstitial injection preferential access to the lymphatics may be facilitated by the use of high molecular weight molecules or colloidal materials [6–8]. In general, increasing the size of polymers or carrier particles appears to increase the proportion of the dose absorbed by the lymphatics and decreases vascular uptake [9,10]. In the colloidal size range, however, increasing size eventually reduces absorption from the injection site [11]. The charge associated with nanoparticular or macromoleculate drug carriers also appears to influence the extent of lymphatic transport. For example, increasing the size of liposomes has been
⁎ Corresponding author. Tel.: +61 399039649; fax: +61 399039583. E-mail address: [email protected] (C.J.H. Porter). 0168-3659/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2009.08.005
shown to increase retention within lymph nodes, (retention being achieved by both physical filtration and removal by fixed phagocytic macrophages [9]) and cationic liposomes appear to be retained to a greater extent than uncharged or anionic liposomes [12]. The potential for preferential lymphatic uptake of particulate systems has therefore been described; however few examples of clinical applications of such systems are evident. In part, this reflects issues associated with colloidal stability, polydispersity and the potential antigenicity of colloidal structures that are retained at the injection site [9,13]. Many of these problems may be overcome by the use of dendronised polymers as drug delivery vectors. Dendrimers are high molecular weight repeating polymers which have physical sizes intermediate to that of typical proteins and liposomal or nanoparticular systems [14–16]. As such, dendrimers have the potential to be sufficiently large that specific lymphatic access is promoted, but sufficiently small that good drainage from the interstitial injection site occurs. Furthermore, dendrimers can be produced with narrow polydispersity and structural flexibility and can be designed with a variety of structural architectures [17]. Drugs may be associated with dendrimers via encapsulation into the core scaffold (via hydrophobic, electrostatic or H-bonding interactions) or via covalent conjugation to the surface via cleavable linkages [18]. The capacity of dendrimer based delivery systems to manipulate pharmacokinetic and pharmacodynamic profiles in vivo has also been described following a range of administration routes
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
including intravenous, intramuscular, oral and transdermal delivery (see [14,18–21]). Dendrimers therefore show significant potential as vehicles to improve the plasma pharmacokinetics, biodistribution and pharmacodynamics of associated drug molecules [19–23]. To this point, however, their utility as delivery systems to specifically target agents to the lymphatics has not been explored in detail. Kobayashi and colleagues [24] have examined the lymphatic transport of a series of Gd+ 3 labelled polyamidoamine (PAMAM) dendrimers administered subcutaneously with a view to improved lymphatic imaging and have shown a size (Generation) dependence in lymphatic uptake and lymph node retention. In these studies, optimal lymphatic retention was achieved approximately 30 min after injection of a Generation 6 PAMAM dendrimer. In contrast, smaller dendrimers diffused into the surrounding tissues rather than the lymphatics and the uptake of larger dendrimers into the lymph was extremely slow. PAMAM dendrimers, however, are not biodegradable and are typically retained in the liver and spleen after access to the systemic circulation [25], raising the possibility of toxicity on chronic dosing. In contrast, poly-L-lysine dendrimers are biodegradable after intravenous administration and the rate of biodegradation and circulation time in the plasma may be manipulated by surface derivatisation with polyethylene glycol (PEG) [26,27]. In this study we have therefore explored the lymphatic uptake and lymph node retention of several Generation 4 (16 surface lysine groups) PEGylated and 4-benzene sulphonate-capped poly-L-lysine dendrimers after subcutaneous administration in rats. The data suggest that PEGylation may be an effective mechanism to promote dendrimer drainage from SC administration sites and to enhance subsequent access to the lymphatics.
109
2. Methods 2.1. Materials Tritiated dendrimers (Lys16(BS)16, Lys16(PEG200)32, Lys16(PEG570)32 and Lys16(PEG2000)32) (Fig. 1) were synthesised and characterised as previously described [26]. The nomenclature used for the dendrimers has also been reported elsewhere [26,28]. Dendrimers were prepared at 10 mg/ml in sterile saline and were stored at −20°C prior to use. For the current studies, the hydrodynamic radius and polydispersity of the dendrimers (Table 1) was also determined by photon correlation spectroscopy using a Nano ZS zeta sizer (Malvern Instruments, Worestershire, UK). Dendrimers were dissolved to 10 mg/ml in isotonic saline and filtered through a 20 nm membrane (Anotop, Whatman, Singapore) prior to analysis with the refractive index set at 1.450. Medical grade polyvinyl, polyethylene and silastic tubing (0.58 mm internal diameter, 0.96 mm external diameter) was purchased from Microtube Extrusions (NSW, Australia). Starscint and Soluene were from Packard Biosciences (Meriden, CT). 2.2. Experimental design Experiments were conducted in parallel groups of animals (n = 3–5 rats per group) and four separate sets of data are reported for each dendrimer: 1) an IV control group, 2) an SC control group, 3) an SC lymph cannulated group and 4) an IV lymph cannulated group (for the PEG570 and PEG2000 dendrimers only). Data from the IV control group are reproduced from previous publications [26,28] to facilitate estimation of absolute bioavailability in animals where dendrimer
Fig. 1. Structure of Generation 4 poly-L-lysine dendrimers, where X represents benzene sulphonate (BS) or PEG groups. For PEG, n repeating ethylene oxide groups of 3, 10 or 42 produce PEG chains with respective molecular weights of 200, 570 and 2000 Da.
110
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
Table 1 Physical properties of dendrimers. Hydrodynamic radius and polydispersity index (PDI) were determined by dynamic light scattering in isotonic saline [29]. Detailed chemical characterisation of the dendrimers is reported in [26,28].
Lys16(BS)32 Lys16(PEG200)32 Lys16(PEG570)32 Lys16(PEG2000)32
MW (kDa)
Radius (nm)
PDI
10 11 22 68
2.8 3.0 5.7 6.7
0.243 0.190 0.179 0.092
subcutaneously into the right hind leg approximately 1 cm above the ankle. 2.4.3. Sampling Blood (0.15 ml) was sampled via the carotid artery and lymph collected continuously via the thoracic lymph duct cannula as described in detail in the supplementary information (available at http://www.sciencedirect.com). 2.5. Sample analysis
was administered SC. Whilst care must be exercised when drawing conclusions from cross-study comparisons, previous studies have shown that the IV pharmacokinetics of dendrimers such as those examined here are reproducible. Ethical considerations therefore precluded re-running the IV studies simply to confirm absolute bioavailability. Rats from the SC control group (group 2) were cannulated via the carotid artery to allow for blood collection. Comparison of data from these rats (group 2) with the IV control data (group 1) allowed calculation of the absolute bioavailability of each dendrimer following SC administration. Animals in the SC lymph cannulated group (group 3) and IV lymph cannulated group (group 4) were cannulated via the carotid artery (to allow blood collection), the jugular vein (to replace fluid lost via collection of thoracic lymph and to enable IV dosing in group 4) and the thoracic lymph duct (to allow collection of lymph). These groups therefore gave estimations of the extent of dendrimer absorption into both the blood and the lymph. Sampling continued for 30 h for Lys16(BS)32, Lys16(PEG200)32 and Lys16 (PEG570)32 and for 7 days for Lys16(PEG2000)32 due to the extended plasma circulation of the largest PEG derived dendrimer. However, animal ethics requirements dictated that thoracic lymph could only be drained and collected for a maximum of 30 h post dose. 2.3. Surgical procedures Male Sprague Dawley rats (270–350 g) were fasted overnight prior to dosing and for 8 h post dose. Food was supplied at all other times and water was available ad libitum. Following surgery and during the experimental period rats were housed individually in metabolic cages. All animal experimentation was approved by the local Animal Ethics Experimentation Committee. In the IV control group, animals had polyethylene cannulae implanted under isoflurane anaesthesia in the right carotid artery and jugular vein as described previously [27]. In the SC control group rats had polyethylene cannulae (as above) implanted only in the carotid artery to allow blood collection. In the lymph cannulated groups, the carotid artery and jugular vein cannulae were implanted as described above. The thoracic lymph duct was subsequently cannulated as previously reported [30] and described in detail in the supplementary information (available at http://www.sciencedirect.com). All cannulae were exteriorised to the back of the neck to enable sample collection via a swivel-tether apparatus that allowed free movement of the rats in metabolic cages. Rats were allowed to recover overnight prior to dosing the following morning. 2.4. Dendrimer administration and sampling 2.4.1. IV groups Rats in the IV control and IV lymph cannulated groups were administered dendrimer intravenously (via the jugular vein cannula) at a dose of 5 mg/kg as a 1 ml bolus in saline over a 2 min infusion period as previously reported [27]. 2.4.2. SC groups Rats in the SC control and SC lymph cannulated groups were administered dendrimer (5 mg/kg) in a volume of 5 ml/kg (in saline)
Whole blood (collected into heparinised (100 U) Eppendorf tubes) was centrifuged at 3500 g for 5 min to separate plasma (50 to 100 µl) which was mixed with 1 ml Starscint in 6 ml polyethylene scintillation vials and scintillation counted (Packard Tri-carb LS) as described previously [27]. Lymph was mixed with Starscint (1:2 to 1:10 v/v depending on lymph volume) to give a final volume of 10 to 20 ml, vortexed and scintillation counted for tritium radiolabel. Blank plasma and lymph were used to correct samples for background counts. Size exclusion chromatography was performed on samples of plasma and lymph as previously described to confirm that the radiolabel measured was associated with intact dendrimer (or the products of dendrimer opsonisation in the case of Lys16(BS)32) [26,28]. At the end of the study (30 h for rats administered Lys16(BS)32, Lys16(PEG200)32 and Lys16(PEG570)32 and 7 days for Lys16(PEG2000)32) rats were sacrificed by intravenous infusion of 1 ml Lethabarb (60 mg/ml pentobarbitone sodium) and the lymph nodes (popliteal and iliac nodes) surrounding the SC injection site and major organs collected into scintillation vials. In addition, lymph nodes were collected after 48 h in a separate set of rats administered Lys16(PEG2000)32 to determine lymph node localisation of injected radiolabel at a time that was more consistent with rats administered the smaller dendrimers. Major organs, injection site and whole lymph node samples were processed and analysed for retained radiolabel as described previously [27]. 2.6. Pharmacokinetic calculations Plasma concentrations were calculated from the measured 3H activity in plasma samples and the specific activity of the dendrimers (reported previously) [27,28]. Plasma concentrations were expressed as ng equivalents/ml and were calculated based on the assumption that all radioactivity was derived from intact dendrimer. Cmax and Tmax were obtained directly from the data obtained. The terminal elimination rate constants (k in Table 1) were calculated by regression analysis of the individual post-distributive plasma concentration vs. time profiles. The area under the plasma-concentration time curves (AUC0−∞) were calculated using the trapezoid rule to the last measured time point (Clast) and extrapolated to infinity by dividing Clast by k. The fraction of the dose absorbed into the blood following SC dosing in control (nonlymph-cannulated) animals (F) was calculated by dividing AUCsc by AUCiv. For the lymph cannulated group, the fraction of the dose absorbed directly into the blood (Fblood) was calculated as the ratio of the AUC0−∞ of dendrimer in plasma after SC administration to lymph cannulated animals and the AUC0−∞ after IV administration. In the case of the two larger dendrimers where lymphatic transport after IV administration was significant and had an impact on IV plasma clearance, the IV AUC0−∝ was taken from the plasma data obtained in lymph-cannulated animals. For the smaller dendrimers, where lymphatic transport was insignificant, IV data from non-lymph cannulated animals was employed. The plasma AUC0−∞ in the SC group after administration of Lys16(PEG570)32 was calculated by extrapolating from the last measured data point (30 h) to infinity using the terminal rate constant estimated from the 3 data points available in the elimination phase (data were only available to 30 h in this group).
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
The fraction of the dose transported into the lymph over the collection period (Flymph) was determined by the quantity of 3H recovered in the combined fractions of lymph divided by the quantity of administered 3H. The total fraction of the dose that was absorbed from the SC injection site (Fabs) in lymph-cannulated animals was calculated from the sum of Flymph and Fblood. Preliminary attempts were made to apply a compartmental modelling approach to the data analysis as previously described for a number of therapeutic proteins [6]. However, the complexity of the required model (including a requirement for lymph and blood transfer processes from the injection site, and blood to lymph to blood transfer processes within the systemic circulation), in combination with relatively sparse data sets that could be obtained in lymph cannulated animals (i.e. data was available only up to 30 h), precluded confident analysis. 2.6. Statistical analyses Statistical analysis of the pharmacokinetic data was performed using GraphPad Prism V4.0. The terminal elimination rate constant (k) was compared across administration routes and k, Cmax, Tmax, AUC, F, Flymph, Fabs and Fblood compared across dendrimers by one-way ANOVA with subsequent Tukey's test for significant differences between groups. Tmax, Cmax and F vs Fabs in SC control vs SC lymph cannulated groups were compared by two-tailed, unpaired T-tests. Plasma concentration curves in control vs lymph cannulated groups were compared by two-way ANOVA with Bonferroni test for significant differences at each time point. Biodistribution into lymph nodes was compared by one-way ANOVA with subsequent Tukey's test. Significant differences are reported where p < 0.05. 3. Results 3.1. Pharmacokinetics and lymphatic uptake After IV dosing of Lys16(BS)32 to non-lymph cannulated animals plasma concentrations declined monoexponentially with an elimination half life of 1 h (reproduced from [26]) (Fig. 2, Table 2). After SC administration, absorption was relatively rapid and plasma concentrations peaked approximately 3 h after dosing (Fig. 2). However, plasma levels were low and bioavailability (F) in the SC control group was only 26% (Table 2). The plasma profile and plasma AUC in the SC lymph cannulated group was not significantly different to that in the SC control group, suggesting a limited role for absorption via the lymphatics. Consistent with the plasma data, only 2% of the administered SC dose of Lys16(BS)32 was recovered in 30 h pooled thoracic lymph (Table 2, Fig. 3). Following IV administration of Lys16(PEG200)32, plasma concentrations declined rapidly and biexponentially with a terminal half life of approximately 40 min (reproduced from [28]) (Fig. 4, Table 2). In the SC control group, absorption of 3H from the injection site was again rapid, reaching peak plasma concentrations approximately 2 h after dosing (Fig. 4). Unlike the anionic Lys16(BS)32 dendrimer, however, Lys16(PEG200)32 was well absorbed after SC administration and bioavailability was approximately 100% (Table 2). In thoracic lymph-cannulated rats, Lys16(PEG200)32 transport into the systemic circulation from the SC injection site occurred predominantly via the blood, and plasma levels were not significantly different to those in the non-cannulated group (Fig. 4,Table 2). Approximately 3% of the administered dose was recovered in 30 h pooled thoracic lymph (Fig. 3,Table 2). Following IV administration of Lys16(PEG570)32 to non-lymphcannulated rats plasma concentrations declined biexponentially with a terminal elimination half life of 9.5 h (reproduced from [26]) (Fig. 5A, Table 2). After SC administration to control (non-lymph cannulated) rats, exposure of animals to Lys16(PEG570)32 was more prolonged than
111
Fig. 2. Plasma concentration — time profile of Lys16(BS)32 following 5 mg/kg dosing via IV administration in non-lymph cannulated control rats (▲), SC administration in nonlymph cannulated rats (●) and SC administration in lymph cannulated rats (○). Data represent mean ± s.d. (n = 3).
for the smaller PEGylated dendrimers as evidenced by the significantly lower k and significantly higher Tmax values (Fig. 5B. Table 2). Similarly to the other PEGylated dendrimers, the bioavailability of Lys16 (PEG570)32 after SC administration was approximately 100% (Table 2). In the lymph cannulated group, the fraction of the dose recovered in the lymph (Flymph) was 29%. Consistent with the contribution of lymphatic transport to drug transport, the plasma AUC in lymph cannulated animals was significantly lower than that in the SC control group. Given the apparent involvement of the lymphatic system in the systemic disposition of Lys16(PEG570)32 after SC injection, the involvement of the lymphatic system in plasma exposure after IV dosing was also examined in lymph cannulated animals. Redirection and collection of thoracic lymph after IV administration resulted in a small but statistically insignificant increase in the elimination rate constant when compared to IV dosed non-lymph cannulated animals (Fig. 5A,Table 2). Approximately 12% of the administered dose, however, was recovered in pooled thoracic lymph and this appeared to reach a plateau 24 h after dosing (Fig. 5C). Elimination of the largest Lys16(PEG2000)32 dendrimer was slow after IV administration and the terminal half life was approximately 3 days (reproduced from [28]) (Fig. 5D,Table 2). After subcutaneous administration to the control (non-lymph-cannulated) group, plasma concentrations of Lys16(PEG2000)32 peaked after 26 h and plasma concentrations remained high and relatively constant for several days. The terminal elimination half life in the SC control group was approximately 7 days, indicating the possibility of flip-flop pharmacokinetics and an extended period of absorption from the SC injection site. Bioavailability in the SC control group was 94 ± 31% suggesting good drainage from the injection site. In the SC lymph-cannulated group, Fblood was calculated by comparison of the plasma AUC0−∞ and the IV AUC0−∞ obtained in lymph cannulated animals and was 32 ± 5% (Table 2). The fraction of the dose recovered directly in the lymphatics over 30 h was 29% and was not significantly different to that observed after administration of Lys16(PEG570)32 (Fig. 5F). As was the case with Lys16(PEG570)32 the recovery of large quantities of dendrimer in the lymph after SC administration of Lys16(PEG2000)32 stimulated an examination of the extent of lymphatic transport after IV administration. Plasma concentrations after IV administration of Lys16(PEG2000)32 to lymph-cannulated rats declined monoexponentially and significantly more rapidly than was observed in non-lymph-cannulated control animals (Fig. 5D, Table 2). The elimination rate constant in the IV lymph-cannulated group was also similar to that in SC lymph cannulated animals. In the IV and SC administered lymph-cannulated groups, k was 7 and 10 fold higher (respectively) than in the non-lymph cannulated groups. The differences in the IV plasma profiles in lymph-cannulated and non-
112
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
Table 2 Pharmacokinetics of dendrimers following IV or SC administration.
Data was collected to 30 h except for non-lymph cannulated rats in the Lys16(PEG2000)32 dosed group. Data represent mean ± s.d. (n = 3–5). e Data extracted from previous publications. ND — not determined. Pharmacokinetic parameters were compared for statistical significance as reported in the Methods section. ⁎Indicates p < 0.05 for SC control vs SC lymph cannulated groups, †indicates p < 0.05 for SC control or SC lymph cannulated groups vs. IV control animals, †† indicates p < 0.05 for groups b and d. Statistical differences between dendrimer groups is reported in the right hand column where bars connecting sets of data indicate data sets that do not differ significantly (p > 0.05).
cannulated animals were consistent with dendrimer transfer from plasma into lymph, where subsequent lymph collection resulted in an effective increase in the rate of elimination from plasma in the lymph cannulated groups. The data also support the use of the IV data in lymph cannulated animals (rather than control animals) to estimate Fblood in SC lymph-cannulated animals. Consistent with the plasma data, 29% of the administered dose of Lys16(PEG2000)32 was recovered directly in thoracic lymph after IV administration. The similarity in lymphatic recovery of Lys16(PEG2000)32 after IV and SC administration (Fig. 5F) suggests that a significant proportion of the lymphatic
Fig. 3. Cummulative recovery of Lys16(BS)16 (●) and Lys16(PEG200)32 (○) in thoracic lymph after SC dosing via the right hind-leg in rats. Data represent mean ± s.d. (n = 3).
recovery of dendrimer after SC administration may have resulted from transfer from the systemic circulation in addition to direct absorption into the lymph from the injection site. 3.2. Lymph node retention and tissue biodistribution after SC administration At the end of the study regional lymph nodes (popliteal and iliac) were collected from rats in the SC control group. For all PEGylated
Fig. 4. Plasma concentration — time profile of Lys16(PEG200)32 following 5 mg/kg dosing via IV administration in non-lymph cannulated rats (▲), SC administration in nonlymph cannulated rats (●) and SC administration in lymph cannulated rats (○). Data represent mean ± s.d. (n = 3 rats).
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
113
Fig. 5. Plasma concentration — time profile of Lys16(PEG570)32 (Panel A and B) and Lys16(PEG2000)32 (Panel D and E) following 5 mg/kg dosing via IV administration (Panels A and D) and SC administration (Panels B and E) in non-lymph cannulated rats (●) and lymph cannulated rats (○). Cummulative recovery of Lys16(PEG570)32 (Panel C) and Lys16(PEG2000)32 (Panel F) are shown in the lower panels in lymph cannulated rats dosed SC (●) and IV (○). Data represent mean ± s.d. (n = 3 rats). *Indicates significant difference between lymph cannulated and control groups.
dendrimers, retention in regional lymph nodes was relatively low (0.03 to 0.4% of the administered dose, Fig. 6). This represented approximately 0.6 to 8% of the injected dose/g of wet lymph nodes. Dendrimers with larger PEG chains showed greater retention in lymph nodes than dendrimers containing smaller molecular weight PEG. The proportion of the dose retained in the lymph nodes 7 days after SC administration of Lys16(PEG2000)32 appeared to decrease when compared to uptake after 2 days, however this difference was not statistically significant. For Lys16(BS)32, the data were variable but suggested that although the anionic dendrimer was similar in size to the smallest PEG200 dendrimer, greater lymph node retention was attained. Retention of the largest PEGylated dendrimers in popliteal and iliac nodes after IV administration was only 10% of the levels obtained after SC administration (data not shown), suggesting improved access to the local lymph nodes after SC administration rather than IV administration. Biodistribution of injected radiolabel to major organs after SC administration was similar to that reported previously [26,28] for the same dendrimers after IV administration (see supplementary information).
Fig. 6. Recovery of 3H-dendrimers in regional lymph nodes (popliteal and iliac) 30 h after 5 mg/kg SC administration in control (non-lymph cannulated) rats. Lymph nodes from rats administered Lys16(PEG2000)32 were collected 2 or 7 days after dosing. Data represent mean ± s.d. (n = 3–4).
114
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
4. Discussion The non-PEGylated anionic (Lys16(BS)32) dendrimer was poorly absorbed from the injection site and the absolute bioavailability was only 26%. The low bioavailability may have been due to hindered absorption from the injection site due to electrostatic interactions with components of the interstitium as has been suggested previously for liposomes [9,31]. Indirect support for this suggestion is evident in the size exclusion chromatography data that shows that Lys16(BS)32 exists mainly as high molecular weight opsonised species in plasma [26], consistent with interaction of the highly charged dendrimer with endogenous components. Interestingly, whilst absorption of Lys16 (BS)32 into the lymph was relatively limited, a significant proportion of the absorbed dose was recovered in lymph nodes neighbouring the injection site (approximately 17% of the quantity collected in thoracic lymph was recovered in lymph nodes). The mechanism of retention of the dendrimer in the lymph nodes was not studied here, but likely resulted from both physical entrapment and phagocytosis of the opsonised dendrimer by lymph node resident macrophages [32]. PEGylation of the dendrimer with relatively low molecular weight (200 Da) PEG (producing a dendrimer that was similar in size to the anionic species, Table 1) led to a significant increase in absorption from the SC injection site and essentially 100% bioavailability. PEGylation therefore appears to reduce potential electrostatic interactions of the dendrimer with the interstitium and facilitate drainage from the injection site. This is consistent with previous reports that have shown that conjugation of anionic PLGA nanospheres with uncharged PLA:PEG increases drainage from SC injection sites [33]. In addition, PEGylation of the dendrimer surface reduced the proportion of the injected dose that was retained in regional lymph nodes. Again, this seems likely to have resulted from reduced electrostatic interaction of the dendrimer with tissue and recognition sites within the node. Surprisingly, the extent of lymphatic transport of the 11 kDa PEGylated dendrimer (3% of the dose) was lower than that reported in previous studies for a similarly sized, albeit somewhat smaller, protein (insulin, 6 kDa, 17% of dose) after SC administration in sheep [7]. Differences in the data reported may reflect both animal model variation and/or differences in the behaviour of dendrimers vs proteins. Consistent with the data obtained for the smaller dendrimer, PEGylation with larger PEG570 or PEG2000 also resulted in essentially complete (>90%) absorption from the SC injection site. Complete absorption in combination with the reduced systemic clearance inherent in these highly PEGylated systems resulted in extremely prolonged plasma exposure and in the case of the Lys16(PEG2000)32 system an apparent terminal half life of approximately 7 days. PEGylation therefore appears to provide for extended systemic exposure after both intravenous and subcutaneous administration, the latter providing additional benefit in terms of potential downstream patient compliance. In contrast to the smaller PEG200 modified system, for both Lys16 (PEG570)32 and Lys16(PEG2000)32, a significant quantity of the administered dose was recovered in the lymph up to 30 h post dose. Whilst the lymphatic recovery of dendrimer was not significantly different for both Lys16(PEG570)32 and Lys16(PEG2000)32 (suggesting that hydrodynamic radius (Table 1) rather than molecular weight may have determined the rate of lymphatic uptake), it seems likely that over longer time periods a more divergent pattern may be evident. In the current studies, animal ethics requirements dictated that data could only be collected for up to 30 h in the lymph cannulated animals. In the case of the Lys16(PEG570)32 dendrimer this appeared to allow collection of much of the required data, since plasma profiles in both cannulated and non-cannulated animals (Fig. 4B) and recovery of dendrimer in the lymph (Fig. 4C) appeared largely complete over this timescale. Fabs in lymph cannulated animals was also not significantly different to F in non-cannulated animals. In
contrast, for the larger dendrimer (Lys16(PEG2000)32) recovery in the lymph appeared to continue beyond the 30 h collection period (Fig. 4F), suggesting that an increase in hydrodynamic radius of approx. 1 nm was sufficient to further hinder absorption into the blood and to promote more extended transport into the lymph. The extended terminal elimination half life in SC administered control animals when compared with IV administered animals is also consistent with extended absorption periods. It seems likely therefore that the 29% recovery of (Lys16(PEG2000)32) in the lymph after SC administration was a conservative estimate of total lymphatic recovery. Linear increases in lymphatic recovery with time were evident for the larger dendrimers over extended time periods (and periods likely to be in excess of the time scale of absorption from the injection site). This suggested the potential for lymphatic access of dendrimer from the systemic circulation via equilibration across capillary beds in addition to direct access from the injection site [34–36]. To examine the potential for direct transfer of dendrimer from the systemic circulation to the lymphatics, thoracic lymph was collected from animals administered the two largest PEGylated dendrimers intravenously. The small PEGylated dendrimers were not examined since the extent of lymphatic transport after subcutaneous administration was extremely low. The area under the plasma concentration-time profile for Lys16(PEG570)32 after IV administration decreased by 14% when compared to non-lymph cannulated animals. The decrease in systemic exposure was essentially matched by the recovery of 12% of the administered dose in the thoracic lymph suggesting that lymphatic redistribution had a small, but significant role in the access of Lys16(PEG570)32 to the lymphatics after SC administration. In contrast, the area under the plasma concentration-time profile of Lys16(PEG2000)32 after IV administration to lymph cannulated animals decreased by approximately 70% when compared to non-lymph cannulated animals. Approximately 30% of the administered dose was also recovered in 30 h lymph, although it appeared that transfer of Lys16(PEG2000)32 into lymph was continuing after this time. In lymph cannulated animals therefore, collection of thoracic lymph after IV administration effectively provides an alternate route of dendrimer elimination which is significant for Lys16(PEG2000)32. Collectively, the data suggests that after SC administration the extended plasma exposure of Lys16(PEG2000)32 is due to reduced systemic clearance, relatively slow absorption from the injection site and lymphatic recycling from the systemic circulation into the lymphatics via the capillary beds and return to the systemic circulation via the thoracic duct as shown in Fig. 7. Although a trend towards increased lymph node retention for the larger PEGylated dendrimers was observed, the proportion of the injected dose recovered in lymph nodes neighbouring the injection site was relatively low (less than 1% of the injected dose in regional
Fig. 7. Schematic representation of the effects of PEGylation on the absorption of polylysine dendrimers from a subcutaneous injection site.
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
nodes). This corresponds to approximately 5–6% of the injected dose per gram of lymph nodes for the larger PEG dendrimers. Others have reported the recovery of 10 to >100% of the injected dose/g of colloidal material in lymph nodes following administration of liposomes [5,31,37] or nanospheres [38], suggesting an effect of both particle size and composition [39]. These previous studies, however, did not report the extent of uptake into lymph and quantified only the amount retained in lymph nodes via deposition studies. The fully PEGylated surface of the dendrimers used here likely hindered the effective retention of the particles within lymph nodes, but promoted passage through the nodes into the lymph. This is consistent with previous reports that have shown that increasing PEGylation decreases lymph node retention due to reduced recognition and sequestration by fixed macrophages [31,33]. The current data also suggest that higher lymph node concentrations are achieved with localised (SC) administration rather than systemic (IV) administration, since after IV dosing of the largest PEGylated dendrimers, retention of the dose in popliteal and iliac nodes near the SC injection site was reduced by up to 90% (data not shown). These data are consistent in principle with the data obtained in mice by Kobayashi et al [24] who examined the lymphatic uptake and retention of PAMAM dendrimers, and suggested that after SC administration, smaller dendrimers diffused throughout the interstitium, whereas larger dendrimers were well absorbed into the lymph and retained by regional lymph nodes. In conclusion, this study has described the lymphatic uptake and lymph node retention of a series of PEGylated poly-L-lysine dendrimers and compared these data with a similar non-PEGylated dendrimer. PEGylation enhanced drainage from the injection site and resulted in sustained plasma exposure for periods in excess of 6 days after a single SC dose. PEGylation also increased lymphatic recovery, and lymphatic access of the dendrimers appeared to occur both via direct access from the SC injection site, and via redistribution of the complex into the lymph from the systemic circulation. The results suggest that PEGylated dendrimers may offer an improved mechanism of lymphatic targeting and may also provide a means for extended exposure to both the systemic circulation and lymphatics after SC and IV administration. Acknowledgements LMK was supported by a National Health and Medical Research Council Australian Biomedical Training Fellowship. This work was funded by an Australian Research Council Linkage grant. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at 10.1016/j.jconrel.2009.08.005. References [1] G. Pantaleo, C. Graziosi, J.F. Demarest, L. Butini, M. Montrone, C.H. Fox, J.M. Orenstein, D.P. Kotler, A.S. Fauci, HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease, Nature 362 (1993) 335–358. [2] K. Dowlatshahi, M. Fan, H.C. Snider, F.A. Habib, Lymph node micrometastases from breast carcinoma: reviewing the dilemma, Cancer 80 (1997) 1188–1197. [3] P. Benedetti-Pancini, F. Maneschi, G. Scambia, S. Greggi, G. Cutillo, G. D'Andrea, C. Rabitti, F. Coronetta, A. Capelli, S. Mancuso, Lymphatic spread of cervical cancer: an anatomical and pathological study based on 225 radical hysterectomies with systemic pelvic and aortic lymphadenectomy, Gynecol. Oncol. 62 (1996) 19–24. [4] Y. Kodama, K. Inokuchi, T. Okamura, Tumor cell aggregation and mode of cancer spread in linitis plastics type of gastric carcinoma, Gann 70 (1979) 721–729. [5] Y. Nishioka, H. Yoshino, Lymphatic targeting with nanoparticulate system, Adv. Drug Deliv. Rev. 47 (2001) 55–64. [6] J. Kota, K.K. Machavaram, D.N. McLennan, G.A. Edwards, C.J. Porter, S.A. Charman, Lymphatic absorption of subcutaneously administered proteins: influence of different injection sites on the absorption of darbepoetin alfa using a sheep model, Drug Metab. Dispos. 35 (2007) 2211–2217.
115
[7] D.N. McLennan, C.J.H. Porter, S.A. Charman, Subcutaneous drug delivery and the role of the lymphatics, Drug Disc. Today: Technol. 2 (2005) 89–96. [8] S.M. Moghimi, A.E. Hawley, N.M. Christy, T. Gray, L. Illum, S.S. Davis, Surface engineered nanospheres with enhanced drainage into lymphatics and uptake by macrophages of the regional lymph-nodes, FEBS Lett. 344 (1994) 25–30. [9] C. Oussoren, G. Storm, Liposomes to target the lymphatics by subcutaneous administration, Adv. Drug Deliv. Rev. 50 (2001) 143–156. [10] C.J.H. Porter, S.A. Charman, Lymphatic transport of proteins after subcutaneous administration, J. Pharm. Sci. 89 (2000) 297–310. [11] V. Manolova, A. Flace, M. Bauer, K. Schwarz, P. Saudan, M.F. Bachmann, Nanoparticles target distinct dendritic cell populations according to their size, Eur. J. Immunol. 38 (2008) 1404–1413. [12] C.K. Kim, J.H. Han, Lymphatic delivery and pharmacokinetics of methotrexate after intramuscular injection of differently charged liposome-entrapped methotrexate to rats, J. Microencaps. 12 (1995) 437–446. [13] R.S. Spangler, Insulin administration via liposomes, Diabetes Care 13 (1990) 911–922. [14] N.K. Jain, A. Asthana, Dendritic systems in drug delivery applications, Expert Opin. Drug Deliv. 4 (2007) 495–512. [15] E. Gillies, J. Frechet, Dendrimers and dendritic polymers in drug delivery, Drug Disc. Today 10 (2005) 35–43. [16] C.C. Lee, J.A. MacKay, J.M. Frechet, F.C. Szoka, Designing dendrimers for biological applications, Nature Biotechnol. 23 (2005) 1517–1526. [17] E.R. Gillies, E. Dy, J.M.J. Frechet, F.C. Szoka Jr., Biological evaluation of polyester dendrimer: Poly(ethylene oxide) “bow-tie” hybrids with tunable molecular weight and architecture, Mol. Pharm. 2 (2005) 129–138. [18] Y. Cheng, Z. Xu, M. Ma, T. Xu, Dendrimers as drug carriers: applications in different routes of drug administration, J Pharm. Sci. 97 (2008) 123–143. [19] S. Svenson, Dendrimers as versatile platform in drug delivery applications, Eur. J. Pharm. Biopharm. 71 (2009) 445–462. [20] Y. Cheng, T. Xu, The effect of dendrimers on the pharmacodynamic and pharmacokinetic behaviors of non-covalently or covalently attached drugs, Eur. J. Med. Chem. 43 (2008) 2291–2297. [21] Y. Cheng, Y. Gao, T. Rao, Y. Li, T. Xu, Dendrimer-based prodrugs: design, synthesis, screening and biological evaluation, Comb. Chem. High Throughput Screen. 10 (2007) 336–349. [22] C. Lee, E. Gillies, M. Fox, S. Guillaudeu, L. Frechet, E. Dy, F. Szoka, A single dose of doxorubicin-functionalized bow-tie dendrimer cures mice bearing C-26 colon carcinomas, Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 16679-16654. [23] G. Wu, R. Barth, W. Yang, S. Kawabata, L. Zhang, K. Green-Church, Targeted delivery of methotrexate to epidermal growth factor receptor-positive brain tumors by means of cetuximab (IMC-C225) dendrimer bioconjugates, Mol. Cancer Ther. 5 (2006) 52–59. [24] H. Kobayashi, S. Kawamoto, M. Bernado, M.W. Brechbiel, M.V. Knopp, P.L. Choyke, Delivery of gadolinium-labeled nanoparticles to the sentinel lymph node: Comparison of the sentinel node visualization and estimations of intra-nodal gadolinium concentration by the magnetic resonance imaging, J. Control. Release 111 (2006) 343–351. [25] N. Malik, R. Wiwattanapatapee, R. Klopsch, K. Lorenz, H. Frey, J.W. Weener, E.W. Meijer, W. Paulus, R. Duncan, Dendrimers: relationship between structure and biocompatibility in vitro, and preliminary studies on the biodistribution of I-125labelled polyamidoamine dendrimers in vivo, J. Control. Release 65 (2000) 133–148. [26] L.M. Kaminskas, B.J. Boyd, P. Karellas, S.A. Henderson, M.P. Giannis, G. Krippner, C.J.H. Porter, Impact of surface derivatisation of poly-L-lysine dendrimers with anionic arylsulphonate or succinate groups on intravenous pharmacokinetics and disposition, Mol. Pharm. 4 (2007) 949–961. [27] B.J. Boyd, L.M. Kaminskas, P. Karellas, G. Krippner, R. Lessene, C.J.H. Porter, Cationic poly-L-lysine dendrimers: pharmacokinetics, biodistribution and evidence for metabolism and bioresorption after intravenous administration in rats, Mol. Pharm. 3 (2006) 614–627. [28] L.M. Kaminskas, B.J. Boyd, P. Karellas, G.Y. Krippner, R. Lessene, B. Kelly, C.J.H. Porter, The impact of molecular weight and PEG chain length on the systemic pharmacokinetics of PEGylated poly-L-lysine dendrimers, Mol. Pharm. 5 (2008) 449–463. [29] T. Jossang, J. Feder, E. Rosenqvist, Photon correlation spectroscopy of human IgG, J. Protein Chem. 7 (1988) 165–171. [30] M. Boyd, V. Risovic, P. Jull, E. Choo, K.M. Wasan, A stepwise surgical procedure to investigate the lymphatic transport of lipid-based oral drug formulations: cannulation of the mesenteric and thoracic lymph ducts within the rat, J. Pharmacol. Toxicol. Methods 49 (2004) 115–120. [31] C. Oussoren, G. Storm, Lymphatic uptake and biodistributions of liposomes after subcutaneous injection: III. influence of surface modification with poly(ethyleneglycol), Pharm. Res. 14 (1997) 1479–1484. [32] C. Oussoren, M. Velinova, G. Scherphof, J.J. van der Want, N. van Rooijen, G. Storm, Lymphatic uptake and biodistribution of liposomes after subcutaneous injection IV. fate of liposomes in regional lymph nodes, Biochim. Biophys. Acta. 1370 (1998) 259–272. [33] A.E. Hawley, L. Illum, S.S. Davis, Preparation of biodegradable, surface engineered PLGA nanospheres with enhanced lymphatic drainage and lymph node uptake, Pharm. Res. 14 (1997) 657–661. [34] S.A. Chen, R.J. Sawchuk, R.C. Brundage, C. Horvath, H.V. Mendenhall, R.A. Gunther, R.A. Braeckman, Plasma and lymph pharmacokinetics of recombinant human interleukin-2 and polyethylene glycol-modified interleukin-2 in pigs, J. Pharm. Exp. Ther. 293 (2000) 248–259. [35] L. Kagan, P. Gershkovich, A. Mendelman, S. Amsili, N. Ezov, A. Hoffman, The role of the lymphatic system in subcutaneous absorption of macromolecules in the rat, Eur. J. Pharm. Biopharm. 67 (2007) 759–765.
116
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
[36] K. Kojima, T. Takahashi, Y. Nakanishi, Lymphatic transport of recombinant human tumor necrosis factor in rats, J. Pharmacobiodyn. 11 (1988) 700–706. [37] S.M. Moghimi, The effect of methoxy-PEG chain length and molecular architecture on lymph node targeting of immuno-PEG liposomes, Biomaterials 27 (2006) 136–144. [38] L. Illum, A.E. Church, M.D. Butterworth, A. Arien, J. Whetstone, S.S. Davis, development of systems for targeting the regional lymph nodes for diagnostic
imaging: in vivo behaviour of colloidal PEG-coated magnetite nanospheres in the rat following interstitial administration, Pharm. Res. 18 (2001) 640–645. [39] S.M. Moghimi, A.R. Rajabi-Siahboomi, Advanced colloid based systems for efficient delivery of drugs and diagnostic agents to the lymphatic tissue, Prog. Biophys. Molec. Biol. 65 (1996) 221–249.
Contents lists available at ScienceDirect
Journal of Controlled Release j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j c o n r e l
PEGylation of polylysine dendrimers improves absorption and lymphatic targeting following SC administration in rats Lisa M. Kaminskas a, Jagannath Kota a, Victoria M. McLeod a, Brian D. Kelly b, Peter Karellas b, Christopher JH. Porter a,⁎ a b
Drug Delivery, Disposition and Dynamics, Monash Institute of Pharmaceutical Sciences, Monash University (Parkville Campus). 381 Royal Pde, Parkville, VIC, 3052, Australia Starpharma Holdings Ltd. 75 Commercial Rd, Melbourne, VIC, 3004, Australia
a r t i c l e
i n f o
Article history: Received 10 July 2009 Accepted 6 August 2009 Available online 15 August 2009
a b s t r a c t Polylysine dendrimers have potential as highly flexible, biodegradable nanoparticular carriers that may also promote lymphatic transport. The current study was undertaken to determine the impact of PEGylation on the absorption and lymphatic transport of polylysine dendrimers modified by surface derivatisation with PEG (200, 570 or 2000 Da) or 4-benzene sulphonate following SC or IV dosing. PEGylation led to the PEG200 derived dendrimer being rapidly and completely absorbed into the blood after SC administration, however only 3% of the administered dose was recovered in pooled thoracic lymph over 30 h. Increasing the PEG chain length led to a systematic decrease in absorption into the blood and an enhancement of the proportion recovered in the lymphatics (up to 29% over 30 h). For the PEG570 and PEG2000 derived dendrimers, indirect access to the lymph via equilibration across the capillary beds also appeared to play a role in lymphatic targeting after both IV and SC dosing. In contrast, the anionic benzene sulphonate-capped dendrimer was not well absorbed from the SC injection site (26% bioavailability) into either the blood or the lymph. The data suggest that PEGylated poly-L-lysine dendrimers are well absorbed from SC injection sites and that the extent of lymphatic transport may be enhanced by increasing the size of the PEGylated dendrimer complex. © 2009 Elsevier B.V. All rights reserved.
1. Introduction The lymphatic system plays a central role in the trafficking and transport of white blood cells, in the pathogenesis of diseases including HIV and metastitial tuberculosis and in the metastatic spread of many cancers [1–4]. For lymph resident diseases therefore, lymphatic targeting of therapeutic drugs (e.g. antivirals, cytotoxics or immunomodulators) or imaging agents is expected to provide advantage over conventional approaches that focus on drug delivery via the blood [5]. Previous studies have shown that after interstitial injection preferential access to the lymphatics may be facilitated by the use of high molecular weight molecules or colloidal materials [6–8]. In general, increasing the size of polymers or carrier particles appears to increase the proportion of the dose absorbed by the lymphatics and decreases vascular uptake [9,10]. In the colloidal size range, however, increasing size eventually reduces absorption from the injection site [11]. The charge associated with nanoparticular or macromoleculate drug carriers also appears to influence the extent of lymphatic transport. For example, increasing the size of liposomes has been
⁎ Corresponding author. Tel.: +61 399039649; fax: +61 399039583. E-mail address: [email protected] (C.J.H. Porter). 0168-3659/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jconrel.2009.08.005
shown to increase retention within lymph nodes, (retention being achieved by both physical filtration and removal by fixed phagocytic macrophages [9]) and cationic liposomes appear to be retained to a greater extent than uncharged or anionic liposomes [12]. The potential for preferential lymphatic uptake of particulate systems has therefore been described; however few examples of clinical applications of such systems are evident. In part, this reflects issues associated with colloidal stability, polydispersity and the potential antigenicity of colloidal structures that are retained at the injection site [9,13]. Many of these problems may be overcome by the use of dendronised polymers as drug delivery vectors. Dendrimers are high molecular weight repeating polymers which have physical sizes intermediate to that of typical proteins and liposomal or nanoparticular systems [14–16]. As such, dendrimers have the potential to be sufficiently large that specific lymphatic access is promoted, but sufficiently small that good drainage from the interstitial injection site occurs. Furthermore, dendrimers can be produced with narrow polydispersity and structural flexibility and can be designed with a variety of structural architectures [17]. Drugs may be associated with dendrimers via encapsulation into the core scaffold (via hydrophobic, electrostatic or H-bonding interactions) or via covalent conjugation to the surface via cleavable linkages [18]. The capacity of dendrimer based delivery systems to manipulate pharmacokinetic and pharmacodynamic profiles in vivo has also been described following a range of administration routes
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
including intravenous, intramuscular, oral and transdermal delivery (see [14,18–21]). Dendrimers therefore show significant potential as vehicles to improve the plasma pharmacokinetics, biodistribution and pharmacodynamics of associated drug molecules [19–23]. To this point, however, their utility as delivery systems to specifically target agents to the lymphatics has not been explored in detail. Kobayashi and colleagues [24] have examined the lymphatic transport of a series of Gd+ 3 labelled polyamidoamine (PAMAM) dendrimers administered subcutaneously with a view to improved lymphatic imaging and have shown a size (Generation) dependence in lymphatic uptake and lymph node retention. In these studies, optimal lymphatic retention was achieved approximately 30 min after injection of a Generation 6 PAMAM dendrimer. In contrast, smaller dendrimers diffused into the surrounding tissues rather than the lymphatics and the uptake of larger dendrimers into the lymph was extremely slow. PAMAM dendrimers, however, are not biodegradable and are typically retained in the liver and spleen after access to the systemic circulation [25], raising the possibility of toxicity on chronic dosing. In contrast, poly-L-lysine dendrimers are biodegradable after intravenous administration and the rate of biodegradation and circulation time in the plasma may be manipulated by surface derivatisation with polyethylene glycol (PEG) [26,27]. In this study we have therefore explored the lymphatic uptake and lymph node retention of several Generation 4 (16 surface lysine groups) PEGylated and 4-benzene sulphonate-capped poly-L-lysine dendrimers after subcutaneous administration in rats. The data suggest that PEGylation may be an effective mechanism to promote dendrimer drainage from SC administration sites and to enhance subsequent access to the lymphatics.
109
2. Methods 2.1. Materials Tritiated dendrimers (Lys16(BS)16, Lys16(PEG200)32, Lys16(PEG570)32 and Lys16(PEG2000)32) (Fig. 1) were synthesised and characterised as previously described [26]. The nomenclature used for the dendrimers has also been reported elsewhere [26,28]. Dendrimers were prepared at 10 mg/ml in sterile saline and were stored at −20°C prior to use. For the current studies, the hydrodynamic radius and polydispersity of the dendrimers (Table 1) was also determined by photon correlation spectroscopy using a Nano ZS zeta sizer (Malvern Instruments, Worestershire, UK). Dendrimers were dissolved to 10 mg/ml in isotonic saline and filtered through a 20 nm membrane (Anotop, Whatman, Singapore) prior to analysis with the refractive index set at 1.450. Medical grade polyvinyl, polyethylene and silastic tubing (0.58 mm internal diameter, 0.96 mm external diameter) was purchased from Microtube Extrusions (NSW, Australia). Starscint and Soluene were from Packard Biosciences (Meriden, CT). 2.2. Experimental design Experiments were conducted in parallel groups of animals (n = 3–5 rats per group) and four separate sets of data are reported for each dendrimer: 1) an IV control group, 2) an SC control group, 3) an SC lymph cannulated group and 4) an IV lymph cannulated group (for the PEG570 and PEG2000 dendrimers only). Data from the IV control group are reproduced from previous publications [26,28] to facilitate estimation of absolute bioavailability in animals where dendrimer
Fig. 1. Structure of Generation 4 poly-L-lysine dendrimers, where X represents benzene sulphonate (BS) or PEG groups. For PEG, n repeating ethylene oxide groups of 3, 10 or 42 produce PEG chains with respective molecular weights of 200, 570 and 2000 Da.
110
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
Table 1 Physical properties of dendrimers. Hydrodynamic radius and polydispersity index (PDI) were determined by dynamic light scattering in isotonic saline [29]. Detailed chemical characterisation of the dendrimers is reported in [26,28].
Lys16(BS)32 Lys16(PEG200)32 Lys16(PEG570)32 Lys16(PEG2000)32
MW (kDa)
Radius (nm)
PDI
10 11 22 68
2.8 3.0 5.7 6.7
0.243 0.190 0.179 0.092
subcutaneously into the right hind leg approximately 1 cm above the ankle. 2.4.3. Sampling Blood (0.15 ml) was sampled via the carotid artery and lymph collected continuously via the thoracic lymph duct cannula as described in detail in the supplementary information (available at http://www.sciencedirect.com). 2.5. Sample analysis
was administered SC. Whilst care must be exercised when drawing conclusions from cross-study comparisons, previous studies have shown that the IV pharmacokinetics of dendrimers such as those examined here are reproducible. Ethical considerations therefore precluded re-running the IV studies simply to confirm absolute bioavailability. Rats from the SC control group (group 2) were cannulated via the carotid artery to allow for blood collection. Comparison of data from these rats (group 2) with the IV control data (group 1) allowed calculation of the absolute bioavailability of each dendrimer following SC administration. Animals in the SC lymph cannulated group (group 3) and IV lymph cannulated group (group 4) were cannulated via the carotid artery (to allow blood collection), the jugular vein (to replace fluid lost via collection of thoracic lymph and to enable IV dosing in group 4) and the thoracic lymph duct (to allow collection of lymph). These groups therefore gave estimations of the extent of dendrimer absorption into both the blood and the lymph. Sampling continued for 30 h for Lys16(BS)32, Lys16(PEG200)32 and Lys16 (PEG570)32 and for 7 days for Lys16(PEG2000)32 due to the extended plasma circulation of the largest PEG derived dendrimer. However, animal ethics requirements dictated that thoracic lymph could only be drained and collected for a maximum of 30 h post dose. 2.3. Surgical procedures Male Sprague Dawley rats (270–350 g) were fasted overnight prior to dosing and for 8 h post dose. Food was supplied at all other times and water was available ad libitum. Following surgery and during the experimental period rats were housed individually in metabolic cages. All animal experimentation was approved by the local Animal Ethics Experimentation Committee. In the IV control group, animals had polyethylene cannulae implanted under isoflurane anaesthesia in the right carotid artery and jugular vein as described previously [27]. In the SC control group rats had polyethylene cannulae (as above) implanted only in the carotid artery to allow blood collection. In the lymph cannulated groups, the carotid artery and jugular vein cannulae were implanted as described above. The thoracic lymph duct was subsequently cannulated as previously reported [30] and described in detail in the supplementary information (available at http://www.sciencedirect.com). All cannulae were exteriorised to the back of the neck to enable sample collection via a swivel-tether apparatus that allowed free movement of the rats in metabolic cages. Rats were allowed to recover overnight prior to dosing the following morning. 2.4. Dendrimer administration and sampling 2.4.1. IV groups Rats in the IV control and IV lymph cannulated groups were administered dendrimer intravenously (via the jugular vein cannula) at a dose of 5 mg/kg as a 1 ml bolus in saline over a 2 min infusion period as previously reported [27]. 2.4.2. SC groups Rats in the SC control and SC lymph cannulated groups were administered dendrimer (5 mg/kg) in a volume of 5 ml/kg (in saline)
Whole blood (collected into heparinised (100 U) Eppendorf tubes) was centrifuged at 3500 g for 5 min to separate plasma (50 to 100 µl) which was mixed with 1 ml Starscint in 6 ml polyethylene scintillation vials and scintillation counted (Packard Tri-carb LS) as described previously [27]. Lymph was mixed with Starscint (1:2 to 1:10 v/v depending on lymph volume) to give a final volume of 10 to 20 ml, vortexed and scintillation counted for tritium radiolabel. Blank plasma and lymph were used to correct samples for background counts. Size exclusion chromatography was performed on samples of plasma and lymph as previously described to confirm that the radiolabel measured was associated with intact dendrimer (or the products of dendrimer opsonisation in the case of Lys16(BS)32) [26,28]. At the end of the study (30 h for rats administered Lys16(BS)32, Lys16(PEG200)32 and Lys16(PEG570)32 and 7 days for Lys16(PEG2000)32) rats were sacrificed by intravenous infusion of 1 ml Lethabarb (60 mg/ml pentobarbitone sodium) and the lymph nodes (popliteal and iliac nodes) surrounding the SC injection site and major organs collected into scintillation vials. In addition, lymph nodes were collected after 48 h in a separate set of rats administered Lys16(PEG2000)32 to determine lymph node localisation of injected radiolabel at a time that was more consistent with rats administered the smaller dendrimers. Major organs, injection site and whole lymph node samples were processed and analysed for retained radiolabel as described previously [27]. 2.6. Pharmacokinetic calculations Plasma concentrations were calculated from the measured 3H activity in plasma samples and the specific activity of the dendrimers (reported previously) [27,28]. Plasma concentrations were expressed as ng equivalents/ml and were calculated based on the assumption that all radioactivity was derived from intact dendrimer. Cmax and Tmax were obtained directly from the data obtained. The terminal elimination rate constants (k in Table 1) were calculated by regression analysis of the individual post-distributive plasma concentration vs. time profiles. The area under the plasma-concentration time curves (AUC0−∞) were calculated using the trapezoid rule to the last measured time point (Clast) and extrapolated to infinity by dividing Clast by k. The fraction of the dose absorbed into the blood following SC dosing in control (nonlymph-cannulated) animals (F) was calculated by dividing AUCsc by AUCiv. For the lymph cannulated group, the fraction of the dose absorbed directly into the blood (Fblood) was calculated as the ratio of the AUC0−∞ of dendrimer in plasma after SC administration to lymph cannulated animals and the AUC0−∞ after IV administration. In the case of the two larger dendrimers where lymphatic transport after IV administration was significant and had an impact on IV plasma clearance, the IV AUC0−∝ was taken from the plasma data obtained in lymph-cannulated animals. For the smaller dendrimers, where lymphatic transport was insignificant, IV data from non-lymph cannulated animals was employed. The plasma AUC0−∞ in the SC group after administration of Lys16(PEG570)32 was calculated by extrapolating from the last measured data point (30 h) to infinity using the terminal rate constant estimated from the 3 data points available in the elimination phase (data were only available to 30 h in this group).
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
The fraction of the dose transported into the lymph over the collection period (Flymph) was determined by the quantity of 3H recovered in the combined fractions of lymph divided by the quantity of administered 3H. The total fraction of the dose that was absorbed from the SC injection site (Fabs) in lymph-cannulated animals was calculated from the sum of Flymph and Fblood. Preliminary attempts were made to apply a compartmental modelling approach to the data analysis as previously described for a number of therapeutic proteins [6]. However, the complexity of the required model (including a requirement for lymph and blood transfer processes from the injection site, and blood to lymph to blood transfer processes within the systemic circulation), in combination with relatively sparse data sets that could be obtained in lymph cannulated animals (i.e. data was available only up to 30 h), precluded confident analysis. 2.6. Statistical analyses Statistical analysis of the pharmacokinetic data was performed using GraphPad Prism V4.0. The terminal elimination rate constant (k) was compared across administration routes and k, Cmax, Tmax, AUC, F, Flymph, Fabs and Fblood compared across dendrimers by one-way ANOVA with subsequent Tukey's test for significant differences between groups. Tmax, Cmax and F vs Fabs in SC control vs SC lymph cannulated groups were compared by two-tailed, unpaired T-tests. Plasma concentration curves in control vs lymph cannulated groups were compared by two-way ANOVA with Bonferroni test for significant differences at each time point. Biodistribution into lymph nodes was compared by one-way ANOVA with subsequent Tukey's test. Significant differences are reported where p < 0.05. 3. Results 3.1. Pharmacokinetics and lymphatic uptake After IV dosing of Lys16(BS)32 to non-lymph cannulated animals plasma concentrations declined monoexponentially with an elimination half life of 1 h (reproduced from [26]) (Fig. 2, Table 2). After SC administration, absorption was relatively rapid and plasma concentrations peaked approximately 3 h after dosing (Fig. 2). However, plasma levels were low and bioavailability (F) in the SC control group was only 26% (Table 2). The plasma profile and plasma AUC in the SC lymph cannulated group was not significantly different to that in the SC control group, suggesting a limited role for absorption via the lymphatics. Consistent with the plasma data, only 2% of the administered SC dose of Lys16(BS)32 was recovered in 30 h pooled thoracic lymph (Table 2, Fig. 3). Following IV administration of Lys16(PEG200)32, plasma concentrations declined rapidly and biexponentially with a terminal half life of approximately 40 min (reproduced from [28]) (Fig. 4, Table 2). In the SC control group, absorption of 3H from the injection site was again rapid, reaching peak plasma concentrations approximately 2 h after dosing (Fig. 4). Unlike the anionic Lys16(BS)32 dendrimer, however, Lys16(PEG200)32 was well absorbed after SC administration and bioavailability was approximately 100% (Table 2). In thoracic lymph-cannulated rats, Lys16(PEG200)32 transport into the systemic circulation from the SC injection site occurred predominantly via the blood, and plasma levels were not significantly different to those in the non-cannulated group (Fig. 4,Table 2). Approximately 3% of the administered dose was recovered in 30 h pooled thoracic lymph (Fig. 3,Table 2). Following IV administration of Lys16(PEG570)32 to non-lymphcannulated rats plasma concentrations declined biexponentially with a terminal elimination half life of 9.5 h (reproduced from [26]) (Fig. 5A, Table 2). After SC administration to control (non-lymph cannulated) rats, exposure of animals to Lys16(PEG570)32 was more prolonged than
111
Fig. 2. Plasma concentration — time profile of Lys16(BS)32 following 5 mg/kg dosing via IV administration in non-lymph cannulated control rats (▲), SC administration in nonlymph cannulated rats (●) and SC administration in lymph cannulated rats (○). Data represent mean ± s.d. (n = 3).
for the smaller PEGylated dendrimers as evidenced by the significantly lower k and significantly higher Tmax values (Fig. 5B. Table 2). Similarly to the other PEGylated dendrimers, the bioavailability of Lys16 (PEG570)32 after SC administration was approximately 100% (Table 2). In the lymph cannulated group, the fraction of the dose recovered in the lymph (Flymph) was 29%. Consistent with the contribution of lymphatic transport to drug transport, the plasma AUC in lymph cannulated animals was significantly lower than that in the SC control group. Given the apparent involvement of the lymphatic system in the systemic disposition of Lys16(PEG570)32 after SC injection, the involvement of the lymphatic system in plasma exposure after IV dosing was also examined in lymph cannulated animals. Redirection and collection of thoracic lymph after IV administration resulted in a small but statistically insignificant increase in the elimination rate constant when compared to IV dosed non-lymph cannulated animals (Fig. 5A,Table 2). Approximately 12% of the administered dose, however, was recovered in pooled thoracic lymph and this appeared to reach a plateau 24 h after dosing (Fig. 5C). Elimination of the largest Lys16(PEG2000)32 dendrimer was slow after IV administration and the terminal half life was approximately 3 days (reproduced from [28]) (Fig. 5D,Table 2). After subcutaneous administration to the control (non-lymph-cannulated) group, plasma concentrations of Lys16(PEG2000)32 peaked after 26 h and plasma concentrations remained high and relatively constant for several days. The terminal elimination half life in the SC control group was approximately 7 days, indicating the possibility of flip-flop pharmacokinetics and an extended period of absorption from the SC injection site. Bioavailability in the SC control group was 94 ± 31% suggesting good drainage from the injection site. In the SC lymph-cannulated group, Fblood was calculated by comparison of the plasma AUC0−∞ and the IV AUC0−∞ obtained in lymph cannulated animals and was 32 ± 5% (Table 2). The fraction of the dose recovered directly in the lymphatics over 30 h was 29% and was not significantly different to that observed after administration of Lys16(PEG570)32 (Fig. 5F). As was the case with Lys16(PEG570)32 the recovery of large quantities of dendrimer in the lymph after SC administration of Lys16(PEG2000)32 stimulated an examination of the extent of lymphatic transport after IV administration. Plasma concentrations after IV administration of Lys16(PEG2000)32 to lymph-cannulated rats declined monoexponentially and significantly more rapidly than was observed in non-lymph-cannulated control animals (Fig. 5D, Table 2). The elimination rate constant in the IV lymph-cannulated group was also similar to that in SC lymph cannulated animals. In the IV and SC administered lymph-cannulated groups, k was 7 and 10 fold higher (respectively) than in the non-lymph cannulated groups. The differences in the IV plasma profiles in lymph-cannulated and non-
112
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
Table 2 Pharmacokinetics of dendrimers following IV or SC administration.
Data was collected to 30 h except for non-lymph cannulated rats in the Lys16(PEG2000)32 dosed group. Data represent mean ± s.d. (n = 3–5). e Data extracted from previous publications. ND — not determined. Pharmacokinetic parameters were compared for statistical significance as reported in the Methods section. ⁎Indicates p < 0.05 for SC control vs SC lymph cannulated groups, †indicates p < 0.05 for SC control or SC lymph cannulated groups vs. IV control animals, †† indicates p < 0.05 for groups b and d. Statistical differences between dendrimer groups is reported in the right hand column where bars connecting sets of data indicate data sets that do not differ significantly (p > 0.05).
cannulated animals were consistent with dendrimer transfer from plasma into lymph, where subsequent lymph collection resulted in an effective increase in the rate of elimination from plasma in the lymph cannulated groups. The data also support the use of the IV data in lymph cannulated animals (rather than control animals) to estimate Fblood in SC lymph-cannulated animals. Consistent with the plasma data, 29% of the administered dose of Lys16(PEG2000)32 was recovered directly in thoracic lymph after IV administration. The similarity in lymphatic recovery of Lys16(PEG2000)32 after IV and SC administration (Fig. 5F) suggests that a significant proportion of the lymphatic
Fig. 3. Cummulative recovery of Lys16(BS)16 (●) and Lys16(PEG200)32 (○) in thoracic lymph after SC dosing via the right hind-leg in rats. Data represent mean ± s.d. (n = 3).
recovery of dendrimer after SC administration may have resulted from transfer from the systemic circulation in addition to direct absorption into the lymph from the injection site. 3.2. Lymph node retention and tissue biodistribution after SC administration At the end of the study regional lymph nodes (popliteal and iliac) were collected from rats in the SC control group. For all PEGylated
Fig. 4. Plasma concentration — time profile of Lys16(PEG200)32 following 5 mg/kg dosing via IV administration in non-lymph cannulated rats (▲), SC administration in nonlymph cannulated rats (●) and SC administration in lymph cannulated rats (○). Data represent mean ± s.d. (n = 3 rats).
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
113
Fig. 5. Plasma concentration — time profile of Lys16(PEG570)32 (Panel A and B) and Lys16(PEG2000)32 (Panel D and E) following 5 mg/kg dosing via IV administration (Panels A and D) and SC administration (Panels B and E) in non-lymph cannulated rats (●) and lymph cannulated rats (○). Cummulative recovery of Lys16(PEG570)32 (Panel C) and Lys16(PEG2000)32 (Panel F) are shown in the lower panels in lymph cannulated rats dosed SC (●) and IV (○). Data represent mean ± s.d. (n = 3 rats). *Indicates significant difference between lymph cannulated and control groups.
dendrimers, retention in regional lymph nodes was relatively low (0.03 to 0.4% of the administered dose, Fig. 6). This represented approximately 0.6 to 8% of the injected dose/g of wet lymph nodes. Dendrimers with larger PEG chains showed greater retention in lymph nodes than dendrimers containing smaller molecular weight PEG. The proportion of the dose retained in the lymph nodes 7 days after SC administration of Lys16(PEG2000)32 appeared to decrease when compared to uptake after 2 days, however this difference was not statistically significant. For Lys16(BS)32, the data were variable but suggested that although the anionic dendrimer was similar in size to the smallest PEG200 dendrimer, greater lymph node retention was attained. Retention of the largest PEGylated dendrimers in popliteal and iliac nodes after IV administration was only 10% of the levels obtained after SC administration (data not shown), suggesting improved access to the local lymph nodes after SC administration rather than IV administration. Biodistribution of injected radiolabel to major organs after SC administration was similar to that reported previously [26,28] for the same dendrimers after IV administration (see supplementary information).
Fig. 6. Recovery of 3H-dendrimers in regional lymph nodes (popliteal and iliac) 30 h after 5 mg/kg SC administration in control (non-lymph cannulated) rats. Lymph nodes from rats administered Lys16(PEG2000)32 were collected 2 or 7 days after dosing. Data represent mean ± s.d. (n = 3–4).
114
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
4. Discussion The non-PEGylated anionic (Lys16(BS)32) dendrimer was poorly absorbed from the injection site and the absolute bioavailability was only 26%. The low bioavailability may have been due to hindered absorption from the injection site due to electrostatic interactions with components of the interstitium as has been suggested previously for liposomes [9,31]. Indirect support for this suggestion is evident in the size exclusion chromatography data that shows that Lys16(BS)32 exists mainly as high molecular weight opsonised species in plasma [26], consistent with interaction of the highly charged dendrimer with endogenous components. Interestingly, whilst absorption of Lys16 (BS)32 into the lymph was relatively limited, a significant proportion of the absorbed dose was recovered in lymph nodes neighbouring the injection site (approximately 17% of the quantity collected in thoracic lymph was recovered in lymph nodes). The mechanism of retention of the dendrimer in the lymph nodes was not studied here, but likely resulted from both physical entrapment and phagocytosis of the opsonised dendrimer by lymph node resident macrophages [32]. PEGylation of the dendrimer with relatively low molecular weight (200 Da) PEG (producing a dendrimer that was similar in size to the anionic species, Table 1) led to a significant increase in absorption from the SC injection site and essentially 100% bioavailability. PEGylation therefore appears to reduce potential electrostatic interactions of the dendrimer with the interstitium and facilitate drainage from the injection site. This is consistent with previous reports that have shown that conjugation of anionic PLGA nanospheres with uncharged PLA:PEG increases drainage from SC injection sites [33]. In addition, PEGylation of the dendrimer surface reduced the proportion of the injected dose that was retained in regional lymph nodes. Again, this seems likely to have resulted from reduced electrostatic interaction of the dendrimer with tissue and recognition sites within the node. Surprisingly, the extent of lymphatic transport of the 11 kDa PEGylated dendrimer (3% of the dose) was lower than that reported in previous studies for a similarly sized, albeit somewhat smaller, protein (insulin, 6 kDa, 17% of dose) after SC administration in sheep [7]. Differences in the data reported may reflect both animal model variation and/or differences in the behaviour of dendrimers vs proteins. Consistent with the data obtained for the smaller dendrimer, PEGylation with larger PEG570 or PEG2000 also resulted in essentially complete (>90%) absorption from the SC injection site. Complete absorption in combination with the reduced systemic clearance inherent in these highly PEGylated systems resulted in extremely prolonged plasma exposure and in the case of the Lys16(PEG2000)32 system an apparent terminal half life of approximately 7 days. PEGylation therefore appears to provide for extended systemic exposure after both intravenous and subcutaneous administration, the latter providing additional benefit in terms of potential downstream patient compliance. In contrast to the smaller PEG200 modified system, for both Lys16 (PEG570)32 and Lys16(PEG2000)32, a significant quantity of the administered dose was recovered in the lymph up to 30 h post dose. Whilst the lymphatic recovery of dendrimer was not significantly different for both Lys16(PEG570)32 and Lys16(PEG2000)32 (suggesting that hydrodynamic radius (Table 1) rather than molecular weight may have determined the rate of lymphatic uptake), it seems likely that over longer time periods a more divergent pattern may be evident. In the current studies, animal ethics requirements dictated that data could only be collected for up to 30 h in the lymph cannulated animals. In the case of the Lys16(PEG570)32 dendrimer this appeared to allow collection of much of the required data, since plasma profiles in both cannulated and non-cannulated animals (Fig. 4B) and recovery of dendrimer in the lymph (Fig. 4C) appeared largely complete over this timescale. Fabs in lymph cannulated animals was also not significantly different to F in non-cannulated animals. In
contrast, for the larger dendrimer (Lys16(PEG2000)32) recovery in the lymph appeared to continue beyond the 30 h collection period (Fig. 4F), suggesting that an increase in hydrodynamic radius of approx. 1 nm was sufficient to further hinder absorption into the blood and to promote more extended transport into the lymph. The extended terminal elimination half life in SC administered control animals when compared with IV administered animals is also consistent with extended absorption periods. It seems likely therefore that the 29% recovery of (Lys16(PEG2000)32) in the lymph after SC administration was a conservative estimate of total lymphatic recovery. Linear increases in lymphatic recovery with time were evident for the larger dendrimers over extended time periods (and periods likely to be in excess of the time scale of absorption from the injection site). This suggested the potential for lymphatic access of dendrimer from the systemic circulation via equilibration across capillary beds in addition to direct access from the injection site [34–36]. To examine the potential for direct transfer of dendrimer from the systemic circulation to the lymphatics, thoracic lymph was collected from animals administered the two largest PEGylated dendrimers intravenously. The small PEGylated dendrimers were not examined since the extent of lymphatic transport after subcutaneous administration was extremely low. The area under the plasma concentration-time profile for Lys16(PEG570)32 after IV administration decreased by 14% when compared to non-lymph cannulated animals. The decrease in systemic exposure was essentially matched by the recovery of 12% of the administered dose in the thoracic lymph suggesting that lymphatic redistribution had a small, but significant role in the access of Lys16(PEG570)32 to the lymphatics after SC administration. In contrast, the area under the plasma concentration-time profile of Lys16(PEG2000)32 after IV administration to lymph cannulated animals decreased by approximately 70% when compared to non-lymph cannulated animals. Approximately 30% of the administered dose was also recovered in 30 h lymph, although it appeared that transfer of Lys16(PEG2000)32 into lymph was continuing after this time. In lymph cannulated animals therefore, collection of thoracic lymph after IV administration effectively provides an alternate route of dendrimer elimination which is significant for Lys16(PEG2000)32. Collectively, the data suggests that after SC administration the extended plasma exposure of Lys16(PEG2000)32 is due to reduced systemic clearance, relatively slow absorption from the injection site and lymphatic recycling from the systemic circulation into the lymphatics via the capillary beds and return to the systemic circulation via the thoracic duct as shown in Fig. 7. Although a trend towards increased lymph node retention for the larger PEGylated dendrimers was observed, the proportion of the injected dose recovered in lymph nodes neighbouring the injection site was relatively low (less than 1% of the injected dose in regional
Fig. 7. Schematic representation of the effects of PEGylation on the absorption of polylysine dendrimers from a subcutaneous injection site.
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
nodes). This corresponds to approximately 5–6% of the injected dose per gram of lymph nodes for the larger PEG dendrimers. Others have reported the recovery of 10 to >100% of the injected dose/g of colloidal material in lymph nodes following administration of liposomes [5,31,37] or nanospheres [38], suggesting an effect of both particle size and composition [39]. These previous studies, however, did not report the extent of uptake into lymph and quantified only the amount retained in lymph nodes via deposition studies. The fully PEGylated surface of the dendrimers used here likely hindered the effective retention of the particles within lymph nodes, but promoted passage through the nodes into the lymph. This is consistent with previous reports that have shown that increasing PEGylation decreases lymph node retention due to reduced recognition and sequestration by fixed macrophages [31,33]. The current data also suggest that higher lymph node concentrations are achieved with localised (SC) administration rather than systemic (IV) administration, since after IV dosing of the largest PEGylated dendrimers, retention of the dose in popliteal and iliac nodes near the SC injection site was reduced by up to 90% (data not shown). These data are consistent in principle with the data obtained in mice by Kobayashi et al [24] who examined the lymphatic uptake and retention of PAMAM dendrimers, and suggested that after SC administration, smaller dendrimers diffused throughout the interstitium, whereas larger dendrimers were well absorbed into the lymph and retained by regional lymph nodes. In conclusion, this study has described the lymphatic uptake and lymph node retention of a series of PEGylated poly-L-lysine dendrimers and compared these data with a similar non-PEGylated dendrimer. PEGylation enhanced drainage from the injection site and resulted in sustained plasma exposure for periods in excess of 6 days after a single SC dose. PEGylation also increased lymphatic recovery, and lymphatic access of the dendrimers appeared to occur both via direct access from the SC injection site, and via redistribution of the complex into the lymph from the systemic circulation. The results suggest that PEGylated dendrimers may offer an improved mechanism of lymphatic targeting and may also provide a means for extended exposure to both the systemic circulation and lymphatics after SC and IV administration. Acknowledgements LMK was supported by a National Health and Medical Research Council Australian Biomedical Training Fellowship. This work was funded by an Australian Research Council Linkage grant. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at 10.1016/j.jconrel.2009.08.005. References [1] G. Pantaleo, C. Graziosi, J.F. Demarest, L. Butini, M. Montrone, C.H. Fox, J.M. Orenstein, D.P. Kotler, A.S. Fauci, HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease, Nature 362 (1993) 335–358. [2] K. Dowlatshahi, M. Fan, H.C. Snider, F.A. Habib, Lymph node micrometastases from breast carcinoma: reviewing the dilemma, Cancer 80 (1997) 1188–1197. [3] P. Benedetti-Pancini, F. Maneschi, G. Scambia, S. Greggi, G. Cutillo, G. D'Andrea, C. Rabitti, F. Coronetta, A. Capelli, S. Mancuso, Lymphatic spread of cervical cancer: an anatomical and pathological study based on 225 radical hysterectomies with systemic pelvic and aortic lymphadenectomy, Gynecol. Oncol. 62 (1996) 19–24. [4] Y. Kodama, K. Inokuchi, T. Okamura, Tumor cell aggregation and mode of cancer spread in linitis plastics type of gastric carcinoma, Gann 70 (1979) 721–729. [5] Y. Nishioka, H. Yoshino, Lymphatic targeting with nanoparticulate system, Adv. Drug Deliv. Rev. 47 (2001) 55–64. [6] J. Kota, K.K. Machavaram, D.N. McLennan, G.A. Edwards, C.J. Porter, S.A. Charman, Lymphatic absorption of subcutaneously administered proteins: influence of different injection sites on the absorption of darbepoetin alfa using a sheep model, Drug Metab. Dispos. 35 (2007) 2211–2217.
115
[7] D.N. McLennan, C.J.H. Porter, S.A. Charman, Subcutaneous drug delivery and the role of the lymphatics, Drug Disc. Today: Technol. 2 (2005) 89–96. [8] S.M. Moghimi, A.E. Hawley, N.M. Christy, T. Gray, L. Illum, S.S. Davis, Surface engineered nanospheres with enhanced drainage into lymphatics and uptake by macrophages of the regional lymph-nodes, FEBS Lett. 344 (1994) 25–30. [9] C. Oussoren, G. Storm, Liposomes to target the lymphatics by subcutaneous administration, Adv. Drug Deliv. Rev. 50 (2001) 143–156. [10] C.J.H. Porter, S.A. Charman, Lymphatic transport of proteins after subcutaneous administration, J. Pharm. Sci. 89 (2000) 297–310. [11] V. Manolova, A. Flace, M. Bauer, K. Schwarz, P. Saudan, M.F. Bachmann, Nanoparticles target distinct dendritic cell populations according to their size, Eur. J. Immunol. 38 (2008) 1404–1413. [12] C.K. Kim, J.H. Han, Lymphatic delivery and pharmacokinetics of methotrexate after intramuscular injection of differently charged liposome-entrapped methotrexate to rats, J. Microencaps. 12 (1995) 437–446. [13] R.S. Spangler, Insulin administration via liposomes, Diabetes Care 13 (1990) 911–922. [14] N.K. Jain, A. Asthana, Dendritic systems in drug delivery applications, Expert Opin. Drug Deliv. 4 (2007) 495–512. [15] E. Gillies, J. Frechet, Dendrimers and dendritic polymers in drug delivery, Drug Disc. Today 10 (2005) 35–43. [16] C.C. Lee, J.A. MacKay, J.M. Frechet, F.C. Szoka, Designing dendrimers for biological applications, Nature Biotechnol. 23 (2005) 1517–1526. [17] E.R. Gillies, E. Dy, J.M.J. Frechet, F.C. Szoka Jr., Biological evaluation of polyester dendrimer: Poly(ethylene oxide) “bow-tie” hybrids with tunable molecular weight and architecture, Mol. Pharm. 2 (2005) 129–138. [18] Y. Cheng, Z. Xu, M. Ma, T. Xu, Dendrimers as drug carriers: applications in different routes of drug administration, J Pharm. Sci. 97 (2008) 123–143. [19] S. Svenson, Dendrimers as versatile platform in drug delivery applications, Eur. J. Pharm. Biopharm. 71 (2009) 445–462. [20] Y. Cheng, T. Xu, The effect of dendrimers on the pharmacodynamic and pharmacokinetic behaviors of non-covalently or covalently attached drugs, Eur. J. Med. Chem. 43 (2008) 2291–2297. [21] Y. Cheng, Y. Gao, T. Rao, Y. Li, T. Xu, Dendrimer-based prodrugs: design, synthesis, screening and biological evaluation, Comb. Chem. High Throughput Screen. 10 (2007) 336–349. [22] C. Lee, E. Gillies, M. Fox, S. Guillaudeu, L. Frechet, E. Dy, F. Szoka, A single dose of doxorubicin-functionalized bow-tie dendrimer cures mice bearing C-26 colon carcinomas, Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 16679-16654. [23] G. Wu, R. Barth, W. Yang, S. Kawabata, L. Zhang, K. Green-Church, Targeted delivery of methotrexate to epidermal growth factor receptor-positive brain tumors by means of cetuximab (IMC-C225) dendrimer bioconjugates, Mol. Cancer Ther. 5 (2006) 52–59. [24] H. Kobayashi, S. Kawamoto, M. Bernado, M.W. Brechbiel, M.V. Knopp, P.L. Choyke, Delivery of gadolinium-labeled nanoparticles to the sentinel lymph node: Comparison of the sentinel node visualization and estimations of intra-nodal gadolinium concentration by the magnetic resonance imaging, J. Control. Release 111 (2006) 343–351. [25] N. Malik, R. Wiwattanapatapee, R. Klopsch, K. Lorenz, H. Frey, J.W. Weener, E.W. Meijer, W. Paulus, R. Duncan, Dendrimers: relationship between structure and biocompatibility in vitro, and preliminary studies on the biodistribution of I-125labelled polyamidoamine dendrimers in vivo, J. Control. Release 65 (2000) 133–148. [26] L.M. Kaminskas, B.J. Boyd, P. Karellas, S.A. Henderson, M.P. Giannis, G. Krippner, C.J.H. Porter, Impact of surface derivatisation of poly-L-lysine dendrimers with anionic arylsulphonate or succinate groups on intravenous pharmacokinetics and disposition, Mol. Pharm. 4 (2007) 949–961. [27] B.J. Boyd, L.M. Kaminskas, P. Karellas, G. Krippner, R. Lessene, C.J.H. Porter, Cationic poly-L-lysine dendrimers: pharmacokinetics, biodistribution and evidence for metabolism and bioresorption after intravenous administration in rats, Mol. Pharm. 3 (2006) 614–627. [28] L.M. Kaminskas, B.J. Boyd, P. Karellas, G.Y. Krippner, R. Lessene, B. Kelly, C.J.H. Porter, The impact of molecular weight and PEG chain length on the systemic pharmacokinetics of PEGylated poly-L-lysine dendrimers, Mol. Pharm. 5 (2008) 449–463. [29] T. Jossang, J. Feder, E. Rosenqvist, Photon correlation spectroscopy of human IgG, J. Protein Chem. 7 (1988) 165–171. [30] M. Boyd, V. Risovic, P. Jull, E. Choo, K.M. Wasan, A stepwise surgical procedure to investigate the lymphatic transport of lipid-based oral drug formulations: cannulation of the mesenteric and thoracic lymph ducts within the rat, J. Pharmacol. Toxicol. Methods 49 (2004) 115–120. [31] C. Oussoren, G. Storm, Lymphatic uptake and biodistributions of liposomes after subcutaneous injection: III. influence of surface modification with poly(ethyleneglycol), Pharm. Res. 14 (1997) 1479–1484. [32] C. Oussoren, M. Velinova, G. Scherphof, J.J. van der Want, N. van Rooijen, G. Storm, Lymphatic uptake and biodistribution of liposomes after subcutaneous injection IV. fate of liposomes in regional lymph nodes, Biochim. Biophys. Acta. 1370 (1998) 259–272. [33] A.E. Hawley, L. Illum, S.S. Davis, Preparation of biodegradable, surface engineered PLGA nanospheres with enhanced lymphatic drainage and lymph node uptake, Pharm. Res. 14 (1997) 657–661. [34] S.A. Chen, R.J. Sawchuk, R.C. Brundage, C. Horvath, H.V. Mendenhall, R.A. Gunther, R.A. Braeckman, Plasma and lymph pharmacokinetics of recombinant human interleukin-2 and polyethylene glycol-modified interleukin-2 in pigs, J. Pharm. Exp. Ther. 293 (2000) 248–259. [35] L. Kagan, P. Gershkovich, A. Mendelman, S. Amsili, N. Ezov, A. Hoffman, The role of the lymphatic system in subcutaneous absorption of macromolecules in the rat, Eur. J. Pharm. Biopharm. 67 (2007) 759–765.
116
L.M. Kaminskas et al. / Journal of Controlled Release 140 (2009) 108–116
[36] K. Kojima, T. Takahashi, Y. Nakanishi, Lymphatic transport of recombinant human tumor necrosis factor in rats, J. Pharmacobiodyn. 11 (1988) 700–706. [37] S.M. Moghimi, The effect of methoxy-PEG chain length and molecular architecture on lymph node targeting of immuno-PEG liposomes, Biomaterials 27 (2006) 136–144. [38] L. Illum, A.E. Church, M.D. Butterworth, A. Arien, J. Whetstone, S.S. Davis, development of systems for targeting the regional lymph nodes for diagnostic
imaging: in vivo behaviour of colloidal PEG-coated magnetite nanospheres in the rat following interstitial administration, Pharm. Res. 18 (2001) 640–645. [39] S.M. Moghimi, A.R. Rajabi-Siahboomi, Advanced colloid based systems for efficient delivery of drugs and diagnostic agents to the lymphatic tissue, Prog. Biophys. Molec. Biol. 65 (1996) 221–249.