|Home | About | Journals | Submit | Contact Us | Français|
Plectasin is a 4.4-kDa antimicrobial peptide with the potential to be a treatment of infections caused by gram-positive bacteria. Since plectasin is a large molecule compared to conventional antibiotics, the development of antidrug antibodies (ADAs) could be anticipated. The immunogenic properties of plectasin were assessed through immunization studies. In mice treated for 5 days with one to two daily subcutaneous doses of plectasin, no antibody response was observed. If the animals were immunized again, after a rest period, low levels of antibodies developed in approximately half the animals. Additionally, mice were immunized with plectasin in Freund's incomplete adjuvant (FIA). Ninety-two percent of these mice developed ADAs after repeated immunizations, with two-thirds having high levels of antibodies. An agar diffusion bioassay showed that sera from animals immunized with plectasin did not inhibit the efficacy of the drug, while hyperimmune sera from animals in which an immune response was provoked by immunization with plectasin in FIA reduced the efficacy of plectasin at the lowest concentration tested. Studies in the murine peritonitis model showed an excellent efficacy of plectasin for the treatment of Streptococcus pneumoniae infections both in naïve animals and in animals with ADAs. No difference in bacterial counts was seen when the animals were treated with plectasin at 2.5 mg/kg of body weight, a dose below the expected therapeutic level. When animals were treated with plectasin at 0.625 mg/kg, the effect was reduced but not neutralized in animals with high levels of ADAs. No animals showed signs of hypersensitivity or injection site reactions toward plectasin, and the half-life of the compound did not vary between animals with and without antibodies.
Plectasin is a defensin-type antimicrobial peptide (4.4 kDa) derived from the saprophytic ascomycete Pseudoplectania nigrella. Plectasin has shown potent activity comparable to that of vancomycin both in vitro and in vivo against various strains of Streptococcus pneumoniae, including strains resistant to conventional antibiotics (20). Plectasin is, at present (September 2009), in the preclinical stage of development, with the first dose expected to be administered to humans in 2010.
Increasing worldwide problems with antimicrobial resistance in both S. pneumoniae and Staphylococcus aureus threaten the future use of many conventional antibiotics, and the need for new compounds to fight these infections is increasing (13, 14).
Peptides, including plectasin, and other protein-based pharmaceutical candidates are relatively large molecules (with molecular masses of 2 to 4 kDa or larger) compared to conventional antibiotics, which are small molecules. Protein therapeutics have the potential to generate an immune response in animals or humans through the development of antidrug antibodies (ADAs). ADAs are seldom associated with direct adverse effects such as acute hypersensitivity or infusion reactions, and the major concern is generally the development of antibodies capable of neutralizing the drug by binding to the active region. Neutralizing antibodies may inhibit drug efficacy or cross-react with endogenous proteins. Furthermore, antibodies can alter the pharmacokinetics of a drug, thereby changing the desired biological effect and toxicity profile. Therefore, it is important to investigate if new compounds have the potential to induce ADAs, and such data have become a vital part of regulatory considerations for the registration of biologics.
Clinical sequelae such as the induction of antibodies have been described for many types of therapeutic proteins, including monoclonal antibodies, cytokines, hormones, and clotting factors (9, 16, 24). The incidence of neutralizing antibodies varied from <1% to >70%, and all of the compounds tested elicited some level of immunogenic response (16). Many antimicrobial agents, both protein based and others, have been described to provoke immune responses and to give rise to allergic reactions, such as urticaria, erythema, and pruritus; but no studies have investigated whether ADAs developed against the compounds and if these could have neutralizing properties (28, 30). In various areas of therapy (e.g., treatment for multiple sclerosis), neutralizing antibodies against therapeutic proteins are a well-described problem (29). However, to our knowledge no earlier investigations of the potential neutralizing effect of ADAs on an antimicrobial peptide have been described.
Plectasin is a promising candidate for the future treatment of infections caused by resistant bacteria. We therefore decided to investigate whether plectasin can induce an antibody response and whether such antibodies, as well as antibodies elicited by the use of plectasin and an adjuvant, would have an impact on drug efficacy, the toxicity profile, and/or the pharmacokinetics of plectasin in animals.
(Parts of this study were presented at the 47th Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, IL, 2007.)
Plectasin wild type (amino acid sequence, GFGCNGPWDEDDMQCHNHCKSIKGYKGGYCAKGGFVCKCY) was provided by Novozymes A/S. The plectasin was diluted in 50 mM acetate and 500 mM NaCl, pH 4. The pH was adjusted to 6.0 with 10 mM phosphate and NaCl at 9 g/liter. The molecular mass was 4,402 Da, and the purity was determined to 96.9% by high-pressure liquid chromatography. The cloning, purification, and structural features have previously been described in detail (20). Vancomycin was used for comparisons in the agarose diffusion assays and was from Sigma (V-2002, lot no. 093K0937).
All study plans involving the use of animals were approved by Novozymes' Laboratory Animal Review Committee and complied with Danish legislation on the use of laboratory animals. Female outbred NMRI mice (starting weight, 26 to 30 g) from Harlan Europe were used in all studies. Animals were kept in standard Macrolon type III cages, were fed a standard pellet diet (2016 Harlan) ad libitum, and had free access to bottled drinking water. Their bedding was aspen wood, and animals had nesting material and polycarbonate mouse houses as environmental enrichment. The animals were acclimatized for 5 to 6 days prior to the initiation of the study. All immunized animals were observed daily for any adverse effects related to treatment.
Six mice were injected subcutaneously (s.c.) with 10 mg/kg of body weight plectasin (0.3 ml of a 1-mg/ml dilution) on day 1. Blood samples were collected on days 0, 14, 28, 42, and 56. Serum was separated by centrifugation at 2,000 × g for 10 min (4°C). The serum samples were stored at −20°C until analysis. The immunization dose of 10 mg/kg was determined from a pilot immunization study, in which that dose elicited the most pronounced antibody response in the mice.
Six mice were injected s.c. with 10 mg/kg plectasin (0.3 ml of a 1-mg/ml dilution) twice a day for 5 consecutive days (days 1 to 5). On days 15 to 19, the injection protocol was repeated. Blood was collected on days 0, 9, 15, 23, and 30. Serum was separated as described above. Mice treated with 0.3 ml vehicle (phosphate-buffered saline [PBS] and acetic acid, pH 6.0) served as controls.
Agarose diffusion assays were performed in Omnitray single-well plates containing 10% Mueller-Hinton agarose (1%) mixed with Staphylococcus carnosus (ATCC 51365) at 5 × 105 CFU/ml. The mixture was poured onto Omnitray single-well plates, and a transferable solid phase plate (Nunc) was placed inside each tray. The plates were cooled and allowed to solidify at 4°C for at least 30 min. Antimicrobial agents (plectasin or vancomycin) were diluted in 0.1% bovine serum albumin-0.01% acetic acid, mixed 1:1 with sera, and added in a volume of 10 μl/well. Twofold dilutions of antimicrobial agents (1.25 to 320 μg/ml) were employed. The serum samples used were pools of samples from immunized mice from the repeated-dose immunization study. Sera from naïve mice served as controls. The plates were incubated at 37°C for 18 to 22 h before measurement of the inhibition zones. The concentration of plectasin in the wells was estimated by extrapolation on a standard curve obtained with the twofold dilutions of plectasin. The diameters of the inhibition zones around the wells with immune serum were used to calculate the percent inhibition. The results from the agarose diffusion assay were validated by repetition of the assay both on the same day and on the following day. The results of double determinations were compared to the results for the controls, and coefficients were calculated for day-to-day and plate-to-plate variations.
Fifty-two mice were immunized s.c. on three occasions with 2-week intervals between immunizations. Group 1 was immunized with vehicle (PBS with acetic acid, pH 6.0) at 0.3 ml/mouse; group 2 was immunized with 150 μl Freund's incomplete adjuvant (FIA) per mouse; group 3 was immunized with 10 mg/kg plectasin, the dose chosen as a result of the findings from a pilot immunization study (data not shown); and group 4 was immunized with 2.5 mg/kg plectasin in FIA. Before each immunization and 9 days after the last immunization, the animals were bled and the serum was separated as described above. The peritonitis study was performed 2 weeks after the last immunization.
The doses used for treatment in the peritonitis study were determined from a dose-response study in which plectasin doses ranging from 0.16 to 30 mg/kg were applied (5). Dose-response curves from the peritonitis data were calculated with GraphPad Prism software by use of the Hills equation (variable slope). The log difference values were used to calculate the 50% effective dose. The 50% effective dose was used as the low dose in the peritonitis study (after immunization), and the high dose was chosen by determining the point at which the curve flattened, thereby obtaining two values within the steep part of the curve.
The mouse peritonitis study was a modified version of a previously described model (7, 15). The inoculum consisted of colonies of S. pneumoniae D39 (plectasin MIC, 1 mg/liter), which were suspended in saline to 3.25 × 107 CFU/ml. At the time of −1 h, all animals were inoculated intraperitoneally with 0.5 ml inoculum. At 0 h, the animals were treated with plectasin. Six animals in each group were treated s.c. with 2.5 mg/kg or with 0.625 mg/kg. The doses were determined from the dose-response study described above. One animal in each group (except group 4, in which one animal died during blood sampling) served as an untreated control. At the time of +4 h, the animals were euthanized by cervical dislocation, and a peritoneal wash was performed. Samples were spotted on 5% horse blood agar and incubated at 35°C in 5% CO2 overnight before CFU counts were performed.
CFU counts were not assumed to follow a Gaussian distribution. Therefore, to determine statistical significance, the median colony counts between groups were compared by analysis of variance (ANOVA; the Kruskal-Wallis test followed by Dunn's posttest for multiple comparison). A P value of <0.05 was considered significant. Since the antibody levels varied greatly between the immunized animals, comparisons were made both between immunized groups and between mice with high and low levels of antibodies (from both group 3 [immunized with plectasin] and group 4 [immunized with plectasin in FIA]). Animals in groups 1 and 2, which were immunized with vehicle or FIA alone, still served as negative controls (all animals in those groups had low levels of antibodies). Animals with high ADA levels were defined as animals with optical density (OD) values more than five times above the cutoff value (see the section on the enzyme-linked immunosorbent assay [ELISA] for the definition of the cutoff value).
MaxiSorp plates (Nunc) were coated with 0.33 mg/liter plectasin diluted in PBS (pH 7.4), and the plates were incubated at 4°C overnight. On the next day, the plates were washed in washing buffer (PBS, 0.05% Tween) and blocked with 1% casein (casein nach Hammarsten in 1 M NaOH). The plates were incubated for 1 h at room temperature and washed. The mouse sera were diluted 100-fold (duplicates) in 0.1% casein and incubated for 1 h. Hyperimmune sera and sera from naïve animals served as positive and negative controls, respectively. The plates were washed and incubated for 1 h with polyclonal goat anti-mouse immunoglobulin G (IgG)-horseradish peroxidase (Serotec, Oxford, United Kingdom) diluted 1:5,000. The plates were washed and incubated with ready-to-use TMB (tetramethylbenzidine) Plus (KemEnTec) for 10 min, and the reaction was stopped with 0.2 M H2SO4.
Mouse ELISA data were compared at 100-fold dilution by determination of the cutoff value for blood samples from naïve mice (samples obtained on day 0). Samples were considered positive when the OD values were above 1.625 times the mean value for the samples from naïve mice (background level), as suggested by Mire-Sluis (18). However, the value should not be less than 0.2 OD units since the precision deteriorates at absorbance levels below that value (8).
The mice were immunized on four occasions, and there were 2-week intervals between immunizations. Six mice served as controls and received either no treatment (group 1) or were immunized with 150 μl adjuvant (FIA) (group 2). Six mice were immunized with 2.5 mg/kg plectasin in FIA (group 3). Blood samples for determination of antibody responses were taken on days 0 (before the first immunization), 28, and 49. Serum was separated as described above.
One week after the last immunization, all animals received a single, weight-adjusted dose of plectasin. The dose of 10 mg/kg was administered s.c. Blood samples (100 μl) were obtained from each animal after 40, 80, 120, and 180 min. Serum was separated as described above, and the plectasin in serum was quantified by chromatography/reverse mass spectrometry (Micromass Quattro Micro). The elimination rate constant (kel) was estimated by using the Excel program, and the serum half-life (t1/2) was calculated as t1/2 = ln(0.5)/kel. The t1/2s for the vehicle control and the treated groups were compared by applying the one-way ANOVA, followed by Dunnett's multiple-comparison test (P < 0.05). Serum samples were analyzed for plectasin-specific antibodies by use of a direct ELISA format, as described above, where the cutoff values were calculated and the levels of antibodies were determined on the basis of the OD values at a 100× dilution of serum.
To investigate the immunogenic effects of different dosing regimens, mice were immunized s.c. with plectasin at different time intervals. The s.c. route was chosen to provoke an immune reaction in the animals, and some animals were immunized with plectasin alone and some were immunized with plectasin in FIA to further enhance the immune response. No mice treated with plectasin showed any signs of adverse effects, such as anaphylactic or injection site reactions, related to the treatment.
Mice were injected s.c. with a single dose of plectasin, and the antibody response was monitored over time to investigate if a single treatment could elicit an immune response in the animals. Analyses of serum samples for IgG 2 to 8 weeks after immunization (days 14, 28, 42, and 56) showed no signs of antibody development in six of seven animals (86%) (Fig. (Fig.1A).1A). The cutoff value for blood samples from naïve mice was calculated to be an OD value of 0.20. One animal developed antibody levels slightly above the cutoff value after 4 weeks, and these levels remained increased throughout the study period. At day 30, all animals had OD values slightly above the cutoff value.
Mice were immunized s.c. twice daily for five consecutive days, followed by a rest period of 10 days before the immunization protocol was repeated. The results of the IgG ELISA are shown in Fig. Fig.1B.1B. The cutoff value for blood samples from naïve mice was calculated to be an OD of 0.31. No animals developed antibodies after the first 5 days of treatment. After the second treatment, the animals began to develop antibodies with OD values slightly above the cutoff value. One animal had a significant increase in OD values after the second immunization.
An in vitro assay was performed with pooled serum from naïve animals, immune animals (animals immunized with plectasin alone), and hyperimmune animals (animals immunized with plectasin in FIA) to investigate the impact of ADAs on drug efficacy in vitro.
The percent inhibition from serum by the agarose diffusion assay is shown in Table Table1.1. Sera from naïve animals and animals immunized with plectasin alone did not have any marked effect on the inhibition zone sizes. Hyperimmune sera inhibited the efficacy of plectasin at the lowest drug concentrations, while other differences in zone diameters were not significantly different. Day-to-day and plate-to-plate variations were calculated, and the coefficient of variation varied from 4.3 to 11.2%, with no marked plate-to-plate and day-to-day differences being noted. The activity of vancomycin was not affected by sera containing plectasin-specific antibodies (data not shown).
Before assessment of drug efficacy in the peritonitis model, mice were immunized with either vehicle, adjuvant (FIA) alone, plectasin (10 mg/kg) alone, or plectasin in FIA on three occasions with a 2-week interval between immunizations. These dosing regimens were chosen to generate the most pronounced antibody response toward plectasin and compare the potency of the drug in animals with ADAs compared to that in naïve animals.
ELISAs for plectasin-specific IgG antibodies showed that 92% and 58% of animals immunized with plectasin in FIA (Fig. (Fig.2A)2A) and plectasin alone (Fig. (Fig.2B),2B), respectively, had OD values above the cutoff value (OD = 0.28) after three immunizations. The development of antibodies typically began after the second immunization, but a large variation in immune response was observed, with most (two-thirds) animals immunized with plectasin alone having relatively low antibody levels, while mice immunized with plectasin in FIA had higher levels. In control animals immunized with vehicle or adjuvant, two animals had OD values marginally above the cutoff value at the last blood sampling occasion (day 37).
The peritonitis efficacy study was performed with mice immunized with plectasin as described above. The efficacy of plectasin in mice with antibodies against plectasin were compared to those in naïve animals to investigate if ADAs affected the potency of plectasin. The animals in groups 1 and 2 had been immunized with vehicle or adjuvant and therefore had no antibodies against plectasin, whereas the animals in groups 3 and 4 had been immunized with plectasin or plectasin in FIA, the latter of which was done in order to generate the largest antibody response to plectasin possible. The bacterial counts from peritoneal fluid after the treatment of infection with S. pneumoniae with either 2.5 mg/kg or 0.625 mg/kg plectasin, both of which were subtherapeutic doses, between the immunized groups were compared. As shown in Fig. Fig.3,3, no significant (P > 0.05) changes in median CFU counts were observed when the antimicrobial killing in immunized animals (groups 3 and 4) were compared with that in naïve animals (groups 1 and 2) for both doses tested (0.625 and 2.5 mg/kg).
Since not all animals immunized with plectasin or plectasin in FIA (groups 3 and 4) had high antibody levels, the CFU results were compared after correlation to the OD values from the ADA ELISA. Animals with high antibody levels were defined as those with OD values greater than five times the cutoff value.
The results (Fig. (Fig.4)4) showed that CFU counts were significantly higher (P < 0.01) in animals with high levels of ADAs than in those with low levels of ADAs after treatment with plectasin at very low doses (0.625 mg/kg). The CFU counts were still significantly lower than those for the untreated controls. When the animals were treated with 2.5 mg/kg, which is still a subtherapeutic dose, no differences in the CFU counts between any groups were seen.
The impact of antibodies on the kinetics of the drug was studied in a pharmacokinetic study performed with immunized mice. After four immunizations with plectasin in FIA, five of six animals had values above the cutoff value (OD = 0.24). The elimination curves are illustrated in Fig. Fig.55.
The pharmacokinetic study compared the t1/2 in control animals (treated with vehicle or adjuvant) to that in animals immunized with plectasin in FIA. The immunized mice had a t1/2 of 38 to 51 min, while the t1/2 in naïve animals and FIA-immunized animals was 42 to 47 min. The antibody levels varied within the immunized groups, and three animals had ADA levels greater than five times the cutoff value. Therefore, the terminal t1/2 was also compared between animals with high and low levels of antibodies. Mice with high levels of ADA had a mean t1/2 of 51 min (standard deviation, ±11.8 min), while mice with low ADA levels had a mean t1/2 of 38 mice (standard deviation, ±4.1 min). No significant differences in t1/2s between naïve and immunized mice or between mice with high and low levels of ADA were observed when a one-way ANOVA was performed to compare the group means to the mean for the vehicle control (Table (Table22).
Plectasin has potent activity both in vitro and in animal models of infection against various strains of S. pneumoniae, including drug-resistant strains, and a new variant of plectasin (NZ2114) is even effective against methicillin-resistant S. aureus (21, 31). We conducted immunizations with plectasin followed by efficacy studies in order to assess whether the potency of the drug was affected by ADAs in a study setup chosen to provoke an antibody response in the animals. Our studies showed that an antibody response could be induced in mice after repeated s.c. immunizations with plectasin. The antibody response varied between individuals, and high antibody levels developed mainly in the group receiving plectasin in combination with FIA (hyperimmune animals).
Drug efficacy was not affected by antibodies generated after immunization with plectasin alone when it was tested in in vitro assays. A detectable neutralizing effect was seen at low drug levels with serum from hyperimmune animals, in which an immune response had been provoked by immunization with plectasin in FIA. The results of the in vivo studies with the murine peritonitis model supported our in vitro findings: plectasin had a strong effect against S. pneumoniae, even in hyperimmune animals with high levels of antibodies, at 2.5 mg/kg, a dose below the expected therapeutic level. With a dose of 0.625 mg/kg, some level of inhibitory effect of the ADAs on drug efficacy was seen in animals with high levels of antibodies against plectasin.
Neither the pharmacokinetic profile nor the toxicity profile of plectasin was affected by ADAs, and no animals developed signs of anaphylactic reactions. No injection site reactions were observed, even though this is considered the most common toxicity associated with the s.c. administration of biological agents (10). It should be noted that our studies were conducted by the use of s.c. immunization, which is considered the most immunogenic route of administration (25), and was often conducted with adjuvant (FIA) to further enhance the immune response.
The results from the peritonitis study indicated that the model was sufficiently sensitive for the detection of small differences in CFU counts and therefore a useful tool for investigations of the impact of ADAs on the in vivo efficacy of an antimicrobial agent.
Not many previous studies of ADAs and drug efficacy of antimicrobial agents have been reported. It has been shown that phage-encoded lysins protect mice against bacteremia with S. pneumoniae even in animals which had developed ADAs (12). However, the mice were immunized only once before rechallenge, and therefore, the potential for the development of neutralizing ADAs is likely to be markedly higher since therapy often requires a multiple-dosing regimen.
Two studies of lysostaphin (3, 4) investigated the long-term effect of treatment and showed that antibodies developed in rabbits injected weekly (intravenously or s.c.) for nine consecutive weeks. The antibodies neutralized the drug in vitro (there was an eightfold increase in the MIC), but studies with animals showed no impact on the in vivo antibacterial effect. Conversely, the studies with animals were performed only with therapeutic doses, and therefore, the neutralizing effect was not quantified.
Treatment with vancomycin (1.4 kDa) can result in the development of IgG antibodies associated with thrombocytopenia and refractoriness to platelet transfusion in patients with leukemia, but the impact of the antibodies on the efficacy of the drug was not investigated (2, 32).
Overall, knowledge on the impact of ADAs on treatment with antimicrobial agents is still limited. Since many new antibiotics are compounds of a size likely to generate an immune response in the host, it would be of great interest to investigate more thoroughly the potential of other antimicrobial agents to provoke the formation of drug-specific antibodies. The problem can also occur during therapy with conventional antibiotics, since these compounds can also induce an immune response if they are bound to albumin (haptens).
During therapy with other classes of pharmaceuticals based on peptides/proteins, antibody formation is a well-known problem, especially in the treatment of chronic diseases such as multiple sclerosis (6, 22, 23). Since many antimicrobial agents are used for the treatment of long-term infections, the neutralizing effect of ADAs could be of great importance for treatment protocols. Therefore, further studies of ADAs and their potential to neutralize both conventional and novel anti-infective compounds should be conducted in the future.
Animal models of immunogenicity are not necessarily predictive of immunogenicity in humans; but they may be useful in assessments of ADAs and their influence on activity, toxicity, and pharmacokinetics (26, 27, 33). The compounds are often found to be more immunogenic in animals than in humans because many therapeutic proteins are endogenous human proteins. Plectasin is a protein foreign to both animals and humans. Therefore, it is likely that antigenicity does not differ as much as is seen, i.e., when human proteins are tested in animal models (1). However, the present studies cannot predict if the immunogenicity profile in humans will be similar to the profile found in mice.
Antibody levels above the cutoff value were found in a few animals naïve to plectasin or after a single immunization. These findings could be false-positive results since a level of false-positive results of approximately 5% must be expected. Furthermore, it is important to remember that some responders can always be found even among naïve animals (19).
Drug activity and elimination in vivo, toxicity, the development of resistance, and production costs are all potential impediments when an attempt is made to develop peptides into pharmaceuticals (11, 17). Plectasin has potent in vivo activity and exhibits low systemic toxicity and a novel mode of action without cross-resistance to other antibiotics (5, 20, 21; T. Schneider et al., unpublished data). The pharmacokinetic profile is favorable, and the drug can be produced in a cost-efficient manner. Furthermore, the findings from the present study prove that plectasin retains its bactericidal effect even after the generation of ADAs and show that ADAs do not induce any undesirable changes in drug kinetics or cause hypersensitivity reactions. All in all, these properties make plectasin a promising candidate for clinical development.
We thank S. Buskov for the high-pressure liquid chromatography analysis of sera from the pharmacokinetic study and P. H. Mygind, N. K. Soni, and D. Sandvang for advice and fruitful discussions on analysis setups. Furthermore, we thank C. Rasmussen, J. Theil, and M. R. R. Markvardsen from Novozymes A/S and J. M. Andersen from Statens Serum Institute for expert technical assistance.
This work was supported by a grant from the Danish Ministry of Science, Technology and Innovation.
K. S. Brinch and H. H. Kristensen are employed by Novozymes A/S. N. Høiby and N. Frimodt-Møller are members of the Novozymes advisory associated with the plectasin project.
Published ahead of print on 17 August 2009.