The inhalation unit described herein can generate, sample for characterization, and uniformly deposit a drug aerosol in the lungs of rodents. This enables the pre-clinical determination of the efficacy and safety of drug doses deposited in the lungs; key data enabling clinical inhaled drug development.
Obstructive respiratory diseases like asthma and chronic obstructive pulmonary disease (COPD) are currently treated by inhaled anti-inflammatory and bronchodilator drugs. Despite the availability of multiple treatments, both diseases are growing public health concerns. The majority of asthma patients are well controlled on current inhaled therapies but a substantial number of patients with severe asthma are not. Asthma affects an estimated 300 million people worldwide and approximately 20 percent have a severe form of the disease. In contrast to asthma, there are few effective therapies for COPD. An estimated 10% of the population has COPD and the trend in death rates is increasing for COPD while decreasing for other major diseases. Although developing drugs for inhaled delivery is challenging, the nose-only inhalation unit enables direct delivery of novel drugs to the lung of rodents for pre-clinical efficacy and safety/toxicology studies. Inhaled drug delivery has multiple advantages for respiratory diseases, where high concentration in the lung improves efficacy and minimizes systemic side effects. Inhaled corticosteroids and bronchodilators benefit from these advantages and inhaled delivery may also hold potential for future biologic therapies. The inhalation unit described herein can generate, sample for characterization, and uniformly deposit a drug aerosol in the lungs of rodents. This enables the pre-clinical determination of the efficacy and safety of drug doses deposited in the lungs of rodents, key data required before initiating clinical development.
There are many advantages to inhaled administration of drugs for the treatment of respiratory disease. Inhaled delivery applies the therapeutic agent directly to the site of action, the lungs. A high local concentration of the drug in the lungs offers a significant advantage that minimizes dose and systemic exposure, and maximizes efficacy. This is an important advantage that can greatly increase the therapeutic index (TI, ratio of dose of drug that causes a side-effect over the dose of drug that provides efficacy) of a drug. Inhaled β2-adrenergic agonists, corticosteroids, and anti-cholinergic drugs have proven effective at improving lung function in patients with asthma and COPD, while minimizing the systemic side effects (tachycardia, immunosuppression, and constipation) observed when these drugs are taken orally. New drug classes (for example, PDE4 inhibitors1 and BTK inhibitors2) have recently proven effective at improving lung function in pre-clinical animal disease models but, similar to β2-agonists, corticosteroids, and anti-cholinergic drugs, suffer from systemic side effects that may be minimized by inhaled delivery. Due to the added cost of development of inhaled vs. oral drugs, an inhaled formulation should only be considered for respiratory indications after successful oral/systemic administration reveals dose limiting mechanism-based systemic side effects.
Pre-clinically, inhaled compounds are optimized to increase the TI, which requires in vivo efficacy and side effect measurements. Initially these measurements can be made in separate assays, usually a topically delivered efficacy measurement and a systemically delivered side-effect measurement, but to truly compare compounds, efficacy and side effects should be measured in the same animals after inhaled administration. This requires dose/response studies that achieve enough compound administered to the lungs to induce a measureable side effect. The only way currently to uniformly distribute large doses of drug into the lungs of multiple small animals simultaneously is nose-only inhalation3,4,5. The strengths and weaknesses of different inhalation exposure techniques have been recently reviewed6,7,8. Specialized equipment and a large amount of test compound (gram quantities) are required for nose-only inhaled drug delivery, but proof-of-concept studies may be possible by other means.
When the amount of drug is limited (mg quantities), direct administration methods are an option but all suffer from non-homogeneous deposition, with more drug concentrated along the central airways and less well represented in the parenchymal/alveolar regions3,4,5. The effective dose delivered by direct instillation is always higher and can never be directly compared to inhaled doses4. Direct instillation methods including intranasal9, intratracheal liquid10,11 and spray instillation12, or dry-powder insufflation13,14 can be used as a screening tool to determine the approximate dose range for later nose-only inhalation studies, or to determine the ranking of efficacy/toxicity for a series of structurally similar drugs15. Due to the central airway deposition pattern, direct administration methods may be more useful to determine the effects of compounds that act on the central airways (bronchodilators or mast cell inhibitors) than in the peripheral lung (anti-inflammatories).
Unlike humans, who can inhale a substantial dose of concentrated aerosol from an inhaler in a single deep breath, continuous generation of a respirable (0.5-5 μm mass median aerodynamic diameter, MMAD) aerosol, for up to an hour, is required to deposit an efficacious drug dose into a spontaneously breathing rodent's lungs in a nose-only inhalation system16. Aerosol generators (jet nebulizer or the Wright dust feed17) that can continuously produce the required aerosol particle size and concentration for nose-only inhalation studies are not very efficient at generating high quality (respirable) aerosols. The drug feed rates to these aerosol generators for potent (IC50 pM to nM in functional cell based assays) compounds are commonly in the 1 to 10 mg/min range and usually less than 1% of that drug aerosol makes it to the breathing zone of the animals (Figure 1). Many of the particles generated are too big (>5 μm) to enter the lungs and are removed by an aerosol classifier (a pre-separator or cyclone with a 5 μm cut point) to avoid a large dose of drug in the nose. Adding to the inefficiency of nose-only inhalation systems is the small particle size range (0.5 to 5 μm MMAD) for respirable particles. Many of the aerosol particles less than 0.5 µm are exhaled (like cigarette smoke) and not deposited in the lungs18. Also, many of the "larger" aerosol particles (~ 5 µm) deposit in the nose are absorbed or transported by mucociliary clearance towards the back of the throat where they are swallowed into the stomach19. When using nose-only inhalation, the dose deposited in the nose is always larger than the dose deposited in the lungs and the nasal dose can contribute to systemic exposure and side-effects20. Inherently, inhaled drug doses are small (in the microgram range) minimizing any systemic side effect potential of drug absorbed by the nasal, lung, or gastrointestinal tissues. Even when the particle size of the aerosol supplied to the animals is in the respirable range, an average of only 4% of the aerosol particles that make it to the breathing zone of the animals deposit in the lungs. More efficient aerosol generators are available, but the jet nebulizer and Wright dust feed are unmatched for their ability to produce continuous aerosols from diverse liquid and dry powder formulations, respectively.
Respirable aerosol from the pre-separator passes into the nose-only exposure inhalation unit21 that is based on a flow-past the nose design22. The inhalation unit has 3 tiers (only two tiers are shown in Figure 1) and each tier contains 10 exposure ports for animals and aerosol sampling. The ports are located peripherally around the central aerosol plenum. Conscious mice are placed in glass restraining holders (6 inches long by 1.2 inch diameter) and breathe the aerosol contents of the inhalation unit. The mice are not acclimated to the restraint devices23. Previous experience has shown that mice tolerate tube restraint of less than an hour duration similarly, with or without adaptation2.
The inhalation unit is designed to deliver drug aerosols directly to the lungs of the animals while avoiding exposure to the operators. The potency/toxicity of these drugs is usually unknown and multiple engineering safety controls are used to avoid exposure to the operators. The operators must always wear personal protective equipment (gloves, laboratory coats, respirators, and safety glasses). The outer plenum of the inhalation unit is under negative pressure at all times during operation, allowing for the safe removal of single or groups of animals without shutting off the aerosol generator. The inhalation unit is also contained in a secondary enclosure maintained at a negative pressure by an exhaust port in the ceiling to prevent any escape of aerosol into the room in case of malfunction. All effluent air from the inhalation system is filtered by a HEPA filter prior to release into the environment. The nose-only exposure system used in this manuscript was purchased from a single vendor (see supplemental Table of Materials).
The mice used in these studies were cared for in accordance with the Guide for the Care and Use of Laboratory Animals, eighth edition24. Mice were group housed at an Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) international-accredited facility in sterile ventilated microisolator housing on corn cob bedding. When measuring bronchoconstriction, mice were anesthetized with 100 mg/kg i.p. pentobarbital and depth of anesthesia was monitored by lack of toe pinch reflex and maintained by i.p. anesthesia as necessary. At the end of the experiments, mice were euthanized by cervical dislocation after barbiturate overdose. Verification of euthanasia was confirmed by lack of respiration. No survival surgeries were performed on the mice. All research protocols were approved by the Institutional Animal Care and Use Committee (IACUC).
1. Formulation and Device Selection for Generation of Pharmaceutical Aerosols
NOTE: Formulation and device selection are dependent on the physicochemical properties of the individual drug to be aerosolized, therefore general protocols are presented below and the reader is referred to the reviews by Zeng25 and O'Riordan26.
2. Aerosol Exposure Experiment Setup
3. Aerosol Delivery
4. Calculation of the Deposited Dose
The bronchodilator ipratropium bromide was dissolved in 0.9% normal saline at a concentration of 1 mg/mL. A 100 mL syringe was filled with the ipratropium formulation solution and the syringe was inserted into a syringe pump set to feed the jet nebulizer (Figure 1) at a flow rate of 1 mL/min. Activating the control system for the inhalation unit initiates a 10 L/min air flow to the jet nebulizer, 5 L/min dilution air flow to the dilution chamber, and 15.5 L/min of vacuum air flow to draw the drug aerosol out of the inhalation unit once it passes the nose of the mice. The syringe pump and real time aerosol monitor sample pump were activated. Twenty-four mice were inserted into the inhalation unit after the real time aerosol monitor confirms the aerosol concentration in the inhalation unit had come to equilibrium (15-30 min, Figure 2). The sample pumps for the absolute filter (1 L/min) and cascade impactor (0.5 L/min) were turned on once all mice were connected to the inhalation unit. The first group of 8 mice was removed from the inhalation unit after 5 min, the second group after 15 min, and the third group after 45 min. Bronchoconstriction32 was measured as the increase in respiratory system resistance (Rrs) to a single nebulized dose of methacholine (30 mg/mL) delivered 2 h after dosing ipratropium using the rodent respirator 33. The percent increase in Rrs for each animal, calculated as the maximum Rrs over the 3-min interval after nebulized methacholine (Rmax) minus the Rrs value of the baseline measurement before methacholine (Rbase) divided by Rbase (% increase in Rrs = (Rmax-Rbase)/Rbase), was used to quantitate the bronchoconstriction.
The aerosol deposited on the cascade impactor filters and absolute filter during the 45 min of aerosol exposure was dissolved in 50% acetonitrile and the mass of ipratropium quantitated by HPLC (Table 1). The MMAD x/ GSD of the ipratropium aerosol was calculated to be 1.7 x/ 1.5 μm (Figure 3) and a deposition fraction (DF) of 0.037 for mice was used for an aerosol with an MMAD of 1.7 μm (Figure 4). The inhalable fraction (IF) was 1 as the aerosol was passed through a pre-separator to remove non-respirable particles. The average concentration of ipratropium in the aerosol (0.5 μg/L) was calculated from mass of ipratropium on the absolute filter (22 μg) divided by the flow rate of air pulled through the filter (1 L/min) multiplied by the sampling time (45 min). The average body weight of the mice was 0.019 kg and their predicted minute ventilation was calculated as 0.021 L/min. Using Equation 1, the deposited doses for the 5, 15, and 45 min exposure groups were 0.1, 0.3, and 0.9 μg/kg; respectively.
Ipratropium inhibited nebulized methacholine induced bronchoconstriction in 8 week old C57Bl/6 mice. Methacholine increased Rrs within seconds of nebulized administration (Figure 5, top panel). The Rrs of the control group (exposed to room air instead of ipratropium aerosol) increased from a baseline value of 0.62 ± 0.03 cmH2O*s/mL to a maximum value of 1.66 ± 0.12 cmH2O*s/mL, representing a 168 ± 9% increase in Rrs, 70 s after methacholine aerosol administration. The percent increase in respiratory system resistance was inhibited in a dose dependent manner by inhaled doses of ipratropium. The percent increase in Rrs was inhibited by 51 ± 9%, 79 ± 14%, and 89 ± 2% (Figure 5, bottom panel) at inhaled deposited doses of ipratropium of 0.1, 0.3, and 0.9 µg/kg, respectively.
Figure 1: Inhalation unit with dry powder aerosol generator attached. The nebulizer used to generate aerosol from liquid formulations is shown on the right. 1) Wright dust feed, 2) cyclone, 3) syringe pump, 4) jet nebulizer inserted into pre-separator, 5) dilution air mixer. When delivering nebulized aerosols, the dust feed and cyclone are replaced by the syringe pump, jet nebulizer/pre-separator, and dilution air mixer. The enlarged view (modified from Oldham21) shows the aerosol flow path at the breathing zone around animal's nose that is created by inserting the restraining tube into the inhalation unit. The aerosol fills the central aerosol (gray) plenum of the inhalation unit, flows out to the animal's nose where it is pulled back by a slight vacuum in the outer (white) plenum and into the waste collection filters (not shown). Please click here to view a larger version of this figure.
Figure 2: Real time aerosol monitor measurements confirm the aerosol concentration in the inhalation unit reaches equilibrium ~30 min after turning on the aerosol generator. Please click here to view a larger version of this figure.
Figure 3: Ipratropium mass collected on each stage of the cascade impactor (blue bar graph) overlaid with curve fit (black curve) used to calculate the MMAD and GSD of the aerosol particle size distribution. Please click here to view a larger version of this figure.
Figure 4: Deposition fraction for mouse lungs as a function of MMAD for aerosols delivered by nose-only inhalation (as modified from Hsieh31). For particle sizes less than 0.5 µm most of the aerosol is exhaled and does not deposit in the lungs (like cigarette smoke). For particle sizes greater than 5 µm, most of the aerosol is filtered out by the nose. Please click here to view a larger version of this figure.
Figure 5: The rapid increase in respiratory system resistance induced by nebulized methacholine administered to mice is blocked by ipratropium delivered by nose-only inhalation to mice (n = 8 per group, top panel). Ipratropium inhibited the methacholine induced increase in respiratory system resistance with an ED50 of 0.1 µg/kg deposited dose (*p <0.05 vs. control, bottom panel). Please click here to view a larger version of this figure.
Cascade impactor data used to calculate MMAD and GSD | ||
Stage Filter | Ipratropium on filter by HPLC (mg) |
Stage Cut Diameter (mm) |
1 | 0.002 | 6 |
2 | 0.19 | 4.5 |
3 | 1.43 | 2.5 |
4 | 2.55 | 1.5 |
5 | 2.95 | 1.2 |
6 | 1.03 | 0.7 |
7 | 0.37 | 0.5 |
Table 1: Cascade impactor data used to calculate MMAD and GSD
A nose-only inhalation system and its operation to deliver pharmaceutical aerosols to the lungs of rodents has been described. Restraining animals in nose-only holders is a commonly used method for exposing animals to airborne material. An experiment was conducted demonstrating the anticholinergic bronchodilator ipratropium bromide34 can potently reverse methacholine induced bronchoconstriction when delivered by nose-only inhalation to mice with an ED50 of 0.1 μg/kg deposited dose. A greater than 10-fold increase in effective dose of ipratropium was required for bronchodilation following intratracheal delivery (ED50intratracheal = 1.3 μg/kg, data not shown) delivery. This is due to the non-homogeneous deposition pattern of drug in the lungs produced by intratracheal dosing3,5,10. 10-fold and greater effective dose differentials between inhaled and intratracheal dosing techniques for other drugs have been observed previously by others4.
The non-homogeneous deposition pattern of drug in the lungs produced by intratracheal dosing also slows drug absorption by the lung35, decreasing the rate at which the drug enters the systemic circulation and decreasing the chances of seeing systemic side effects. Therefore, to optimize the safety/efficacy (therapeutic index)36 of novel inhaled drugs, nose-only inhalation delivery should be used. Intratracheal deposition will give an inaccurate estimate of the TI by generating erroneously high effective doses in the lung and low doses to the systemic circulation.
As new drugs delivered by the inhaled route are developed, it is essential to appropriately translate the efficacious drug dose from preclinical efficacy studies to predict an efficacious human dose for clinical trials. The infamous death of Tusko the elephant37 is commonly cited in the literature to remind us to use allometric scaling to predict interspecies drug doses and that interspecies drug doses should not be linearly extrapolated on the basis of a simple comparison of body masses. It is common to use an allometric approach with an allometric exponent b of 0.67 to predict human drug doses from preclinical efficacy studies38. Using the mouse inhaled ED90 of 0.9 μg/kg for ipratropium, an allometric exponent of 0.67, a mouse body mass of 0.03 kg, and a human body mass of 60 kg; an estimated human ED90 of 0.07 μg/kg ((0.9*(0.03/60)(1-0.67) = 0.07) can be calculated as the predicted human deposited dose from our mouse data. This predicted value is comparable to the actual human efficacious deposited dose of 0.26 μg/kg, which can be calculated from the delivered dose of 40 μg39 divided by the human body mass of 60 kg multiplied by a human oral inhalation pulmonary deposition fraction16 of 0.4 ((40/60)*0.4 = 0.26). An estimate of the effective human inhaled deposited dose also helps to plan the delivered doses used in inhalation toxicology studies required for clinical trials40.
The specialized equipment and large amount of test compound (gram quantities) required for nose-only inhaled drug delivery can be significant limitations when developing the nose-only inhalation technique. A calibration run without animals present is necessary if a specific dose is required for a (usually toxicology and not an efficacy dose/response) study, and this calibration run will require more drug to be available. This calibration run is necessary because the physicochemical properties of each drug/formulation can vary enough to greatly affect the deposited dose of drug in the lungs. Optimization of the air and drug feed rates to the aerosol generator during the calibration run is required to achieve an aerosol particle size and concentration for the specific dose required. While the aerosol generator air and drug feed rates suggested in the methods are a reasonable starting point, there is potential that for a specific drug formulation a non-respirable (particle size >5 mm) aerosol will be generated. At the preclinical stage, there is often not enough drug available to do a calibration run and it is impossible to know a priori what doses (if any) are getting into the lungs. Also, static electricity is produced during the aerosol generation process and can influence the time required for the aerosol concentration to equilibrate. It is important to properly ground the equipment to minimize static charge. Another option to minimize static electricity is to add humidity to the air supplied to the inhalation unit, to increase conductivity and dissipate electrostatic charges on particles25. Humidifying the air supplied to the aerosol generator is not required for animal comfort during short (< 1 h) dosing sessions but should be considered if longer dosing times are used.
Drugs can be delivered to the lungs of animals by passive nose-only inhalation or direct intratracheal administration methods that bypass nasopharyngeal deposition. Nose-only inhalation delivery is commonly used in the field of inhalation toxicology41 but is sparingly used early in the drug discovery process. A multidisciplinary research team is needed for conducting nose-only inhalation studies due to the: need for large quantities of drug, specialized knowledge required to formulate, generate, and characterize aerosols; operate complex equipment, and measure drug efficacy in animal models of respiratory disease. The delivery techniques described herein are used to develop small molecule inhaled drugs but in the future can be applied to develop inhaled biologics42,43. Hopefully the procedures and tips documented in this manuscript will facilitate new pre-clinical drug discovery and toxicology researchers to acquire the skills necessary to deliver drugs to rodents by aerosol inhalation.
The authors have nothing to disclose.
We acknowledge: Dr. Thomas Budiman at TSE Systems GmbH for his technical expertise and equipment customization. John Fry (Battelle Inc.) and Dr. Rudy Jaeger (CH Technologies Inc.) for their helpful discussions. Tian Wu, Sam Mboggo, April Miller, and Sean Davis (Amgen) for help with the experiments.
Nose-only exposure inhalation unit | TSE systems | 700100-KNES-040-ss | Custom configurations available |
DACO data acquisition system | TSE systems | 700400-PRO-C-D/1 | |
MC One Jet Mill | Jetpharma | DEC MicroJet 10 | |
Turbula Mixer | GlenMills Inc | T2F | |
Micronized Lactose | DFE Pharma | Lactohale 200 | |
Hydraulic press | Specac | GS15011 | |
Cascade impactor filters | Pall Life Sciences | 7219 | Emfab filter |
Absolute filters | Whatman | 10370302 | 5 cm diameter |
Real time aerosol monitor Microdust Pro Monitor |
Casella | CEL-712 | |
Ipratropium bromide | Spectrum Chemical | I1178 | pre-micronized |
flexiVent FX1 system | scireq | FV-FXCS |