Dry Powder Aerosol Nanocomposite Microparticles for the Treatment of Pulmonary Diseases

Dry powder aerosols have attracted increasing attention for the treatment of pulmonary diseases due to their capability of providing direct and effiCient drug delivery to the lungs. Unfortunately, barriers exist in the implementation of dry powder pulmonary delivery systems, including: (1) dry powder aerosols with aerodynamic diameters smaller than 1 μm will often be exhaled; (2) particles with aerodynamic diameters above 5 μm tend to deposit in the mouth, throat or upper lung mucosa and will then be eliminated due to mucosal clearance mechanisms; and (3) particles larger than 1 μm that deposit in the deep lung area may be cleared from the alveoli via macrophages. Spray-dried nanocomposite microparticles (nCmP), combining the advantages of nanoscale and microscale carriers, can be employed to overcome these issues. Microscale powders facilitate effective deep lung deposition while the embedded nanoscale carriers can provide multiple functions such as the protection of active ingredients from degradation, enhancement of the solubility of drugs, controlled drug release, and reduction of systemic side effects. Several nCmP systems have been developed for various applications, but few comprehensive studies have been completed to illustrate how to effectively engineer optimal nCmP systems with desired properties including: targeted deposition in airways, controlled release of payloads, ability of overcoming physiological barriers, and favorable formulation stability. This dissertation focuses on the development, characterization, and optimization of various dry powder aerosol nanocomposite microparticle systems for the treatment of pulmonary diseases. In the production of nCmP, acetalated dextran (Ac-Dex) nanoparticles loaded with therapeutics are synthesized, suspended in an excipient solution or DI water, and transformed into microscale dry powder microparticles via spray drying. Upon pulmonary administration, the nCmP will deposit on the surface of the mucosal layer of the lungs and decompose into free NP, allowing the nanoparticles to reach the mucus and/or cell epithelium and then release drug(s) at sustained rate. Nanoparticles were prepared using Ac-Dex with different degradation rates to control the drug release profiles. In this work, various therapeutics were encapsulated in nanoparticles, and subsequently nCmP, for applications in the treatment of cystic fibrosis-related lung infections or pulmonary arterial hypertension. The physical properties of the nanoscale carriers in terms of size, size distribution, and surface properties were modified by manipulating the parameters of the nanoparticle preparation process. Dry powder aerosols embedded with nanoparticles were formulated using spray drying. The aerosol performance was optimized by manipulating the composition of feed solution and spray drying parameters. The optimal parameters to form nCmP with deep lung and whole lung deposition were studied. Furthermore, the favorable drying conditions necessary to achieve high therapeutic stability and formulation stability of the particles was also evaluated. Overall, dry powder aerosol nanocomposite microparticles have the ability to provide pulmonary drug delivery systems with tunable properties to fit a wide variety of desired applications. These general-purpose systems exhibit promising potential in the improved treatment of pulmonary diseases through particle engineering and design.

Pulmonary drug delivery has exhibited promising potential in the effective treatment of lung diseases, as it allows for the delivery of a wide range of therapeutics directly and efficiently to the lungs, thereby increasing local drug concentration, reducing systemic side effects, providing a rapid onset of pharmaceutical action, and avoiding the first-pass metabolism associated with the liver (1-4). As a result, various therapeutics such as antibiotics, proteins, peptides, anti-cancer drugs (5), plasmid DNA (6), siRNA (7), and anti-tuberculosis (TB) drugs have been employed in inhalation formulations for the treatment of pulmonary diseases such as asthma, chronic obstructive pulmonary disease (COPD), cystic fibrosis (CF)-related pulmonary infections, and lung cancer (5,8). In comparison to liquid aerosol formulations, aerosols based on dry powders offer additional benefits such as enhanced stability of the formulations, controllable particle size for targeting specific regions of the lung, and increased drug loading of hydrophobic or lipophilic payloads (9,10). Unfortunately, barriers exist in the implementation of dry powder pulmonary delivery systems, including: (1) dry powder aerosols with aerodynamic diameters smaller than 1 µm will often be exhaled; (2) particles with aerodynamic diameters above 5 µm tend to deposit in the mouth, throat or upper lung mucosa and will then be eliminated due to mucosal clearance mechanisms; and (3) particles larger than 1 µm deposited in the deep lung area may be cleared from the alveoli via macrophages (10)(11)(12). Spray-dried nanocomposite microparticles (nCmP) can be employed to overcome these issues, as the integration of nanoscale and microscale carriers combines the benefits of both types of systems. Microscale powders (1 -5 µm) facilitate effective deep lung deposition (12) while the embedded nanoscale carriers can provide multiple functions such as the protection of drugs from degradation, enhancement of the solubility of drugs, provision of controlled drug release, and reduction of systemic side effects (2,13,14). Although many nCmP systems have been developed, few comprehensive studies have been completed to illustrate the effective engineering of optimal nCmP systems for the pulmonary delivery of therapeutics.
To this end, the purpose of this work was to develop and optimize several dry

Significance of This Study
This is a comprehensive study on the development, characterization, and optimization of nCmP systems for the treatment of pulmonary diseases. The significance of the research discussed in this dissertation is summarized as the following: 1. The nCrnP system can provide a simply tailorable drug release profile of payloads by adjusting the synthesis of Ac-Dex polymer that is used for nanoparticle preparation.
2. The nCrnP system can be design to primarily deposit at a targeted location in the airways by manipulating the nCmP formulation process parameters.
3. The nCmP system can overcome various physiological barriers by tuning the physical properties of nanoparticles, thus providing effective delivery of therapeutics for the treatment of pulmonary diseases.

Objectives of This Research
The overall objective of this dissertation was to develop, characterize, and optimize multipurpose dry powder aerosol nanocomposite microparticle systems for the treatment of pulmonary diseases. This involved four projects and the specific objectives of these are outlined as follows:

Dissertation Organization
This dissertation will be organized as follows: Chapter 2 provides the background information on nanocomposite microparticle systems for pulmonary drug delivery applications including an introduction of pulmonary delivery of dry powder particle-based therapeutics, a review of nanocomposite microparticles for pulmonary delivery, and a discussion of nanocomposite microparticles preparation, optimization, and applications. Chapter 3 presents the project "Synthesis and characterization of nanocomposite microparticles (nCmP) for the treatment of cystic fibrosis-related infections," which involves the development of nanocomposite microparticle systems for the delivery of antibiotics through viscous lung mucus caused by cystic fibrosis . Chapter 4 includes the project "Nanocomposite microparticles (nCmP) for the delivery of tacrolimus in the treatment of pulmonary arterial hypertension (PAH)," which involves the development of nanocomposite microparticles for the deep lung delivery of tacrolimus, a novel treatment regime for P AH.
Chapter 5 includes the project "Optimization of nanocomposite microparticles (nCmP) for deep lung delivery of therapeutics via spray drying," which discusses the optimization strategy necessary in order to prepare nanocomposite microparticles with favorable properties including: deep lung deposition, desirable nanoparticle redispersion, high drug loading, and low water content.   (2), activators of the local immune system (3,4), and antibiotics, among others. Aerosol delivery of therapeutics to the lungs offers an attractive way to deliver high drug concentrations directly to the site of disease, reducing toxicity while improving the therapeutic potential. Pulmonary delivery is also an attractive route for systemic administration with a more rapid onset of action than is traditionally seen when given orally. The rapid onset is due to fast absorption of the therapeutic by the massive surface area of the alveolar region, abundant pulmonary vasculature, thin air-blood barrier, high solute permeability, and the avoidance of first pass metabolism, which degrades many proteins that have shown promise for use as biotherapeutics (5,6).
There is great promise for advances in pulmonary drug delivery using inhaled aerosols for the targeted treatment of respiratory diseases such as chronic obstructive pulmonary disease (COPD), asthma, cystic fibrosis, lung cancer, and infectious diseases (e.g. tuberculosis) (7). To date, several types of particle-based aerosol therapeutics have been developed including nanoparticulates suspended in aqueous form from inhalers and nebulizers as well as microparticle-based liquid and dry powder formulations. Both nanoparticle and microparticle-only formulations exhibit limitations that will be discussed in detail later, leading to a need in the development of advanced formulations using particle engineering techniques. In particular, it has been shown that solid drug-loaded nanoparticles can be effectively delivered to the lungs when they are encapsulated in a carrier system that dissolves after coming in contact with the aqueous environment of the lung epithelium. These so-called nanocomposite microparticles (nCmP) can enhance the delivery of therapeutics via the aerosol route as described in the following sections. In this chapter, a general overview of pulmonary delivery and the current state-of-the-art related to nCmP will be followed by an overview of current nCmP systems and their design considerations.

Dry powder aerosol delivery devices
The most common inhaler devices are nebulizers, pressurized metered-dose inhalers (MDI), and dry powder inhalers (DPI). Nebulizers deliver liquid medication in a steady steam of tiny droplets, do not require patient coordination, and can deliver larger doses compared to the other devices (5). MDis are the most popular devices used to treat local respiratory diseases and their mechanism utilizes a valve designed to deliver a precise aerosol amount each time the device is actuated (8,9). DPis are portable devices that deliver medication in the form of a dry powder directly to airways (10). Traditional dry powder blends are typically comprised of micronized drug particles with a mass median aerodynamic diameter (MMAD) less than 5 µm blended with inactive excipients of larger sizes (11)(12)(13). The drug is delivered when a patient inhales, pulling air through a punctured capsule, blister, or reservoir.
Advantages of DPis are that they are portable, compact, and have multi-dose functionality. DPis are breath-activated and unlike MDis do not require any outside energy source or propellant, eliminating the need to coordinate actuation and inhalation (14). Another practical advantage of using a dry powder system is the far slower degradation of drugs in the dry state as compared to their suspension in a liquid form, resulting in higher long-term stability and sterility (6). Dry formulations of nanoparticles also reduce inter-particle attractive forces as does blending with larger carrier molecules, which improves aerodynamic performance (15). The use of biodegradable particles in DPis may increase the bioavailability of therapeutic agents and change the pharmacokinetic plasma profile, making the agent more suitable for pulmonary delivery (16).
DPis have many advantages with respect to medication delivery; however, they are not without their disadvantages. These disadvantages include poor deposition in the lower bronchioles and alveoli as well as the limited availability of drugs from the device itself. In many cases, DPis are a relatively new technology still in the development stage, in comparison to the more established nebulizer and MDI technology. A more detailed overview of aerosol delivery devices can be found elsewhere (17)(18)(19)(20). Overall, the ease of use and powder storage advantages discussed further in this chapter provide strong reasoning for the application of DPis to deliver advanced dry powder formulations.

Overview of current dry powder aerosol therapeutics
As seen in formulation results in greater increases in forced expiratory volume over I-second (FEVI) at both trough and mean peak levels (22). In addition, a once-daily fluticasone-salmeterol DPI is FDA approved for COPD treatment (23).
While not yet FDA-approved, a once-daily DPI containing glycopyrronium and indacaterol is approved for use in Europe and Japan (24). Also, a twice-daily combination DPI comprised of aclidinium-formoterol is approved for use in Europe, the United Kingdom, and Canada (25).
Asthma maintenance therapy is often similar to that seen in COPD with the goals of bronchodilation and reduction in inflammation. Some examples of these DPis can be seen in Table 2 (36). Phagocytic activity is maximized for particles 1-2µm in diameter and less so for smaller or larger particles outside of this range (37).
The contribution of pulmonary endocytosis to the overall lung clearance is determined by the particle size and shape, solubility, particle burden, and chemical nature of the inhaled aerosols.
Macromolecular drugs are also susceptible to enzymatic degradation. The contribution of drug breakdown along with other pulmonary clearance mechanisms is minimal in comparison with the gastrointestinal tract; however, it requires consideration in enzyme-sensitive compounds. Therefore, to protect from rapid clearance or degradation and achieve sustained release within the lung, encapsulation into nanoparticles or microparticles has been pursued.
Systemic diseases can be treated via pulmonary delivery as formulations and their released drugs can translocate from lung tissue to the cardiovascular endothelium.
The mechanisms by which therapeutics are translocated from the lung occurs by either transcellular or paracellular transport. In transcellular transport, absorption of the therapeutic typically occurs through receptor/carrier-mediated endocytosis. This is a slow process (hours to days) and occurs for larger particles (>5-6 nm) having molecular weights of more than 40 kDa (36). Transcellular transport involves the internalization of caveolin-1, transcellular channels, and vesicular trafficking.
Paracellular transport involves diffusion of drug through alveoli, which is a fast process (5 to 90 minutes) and happens in the case of smaller molecules (<5-6 nm).
Several studies have evaluated the mechanisms of nanoparticle interaction with the lung and translocation of nanoparticles from nanocomposite microparticles through the epithelial barrier (38,39).
Nanocomposite microparticles offer the ability to overcome many of the barriers associated with the aerosol delivery of therapeutics. First, nCmP are typically an appropriate size for effective pulmonary delivery (e.g. micrometer) and thereby can be delivered to all locations in the lungs. While micro-sized particles are usually removed from the lungs via mucociliary clearance and/or phagocytosis, nCrnP can be designed to rapidly release nanoparticles from the nCmP bulk in a very quick fashion to avoid these clearance mechanisms. In addition to avoidance of pulmonary clearance, the nanoparticles released from nCmP offer their own advantages including the ability to translocate through mucus, penetrate into cells, and translocate across the pulmonary epithelium/cardiovascular endothelium barrier ( 40,41 ). At this point, the use of nCmP is the only effective way to delivery dry nanoparticle-based therapeutics.

Design for pulmonary deposition of particles
With respect to aerosol formulation and design, drug distribution and deposition along the respiratory tract depends on many characteristics of the inhaled formulation including the diameter, size distribution, shape, charge, density, and hygroscopicity of the particulate system. The mass median aerodynamic diameter (MMAD) and geometric standard deviation are what determine the site of deposition in the respiratory tract ( 42).
Dry powder particle formulations used in pulmonary applications must exhibit specific physical properties for successful implementation. Respirable particles with an MMAD between 0.5 and 5 µm undergo deposition in the alveolar region, which can facilitate systemic bioavailability (43,44). Particles larger than this may potentially be deposited in the inhaler, oropharynx, and/or larynx, or they may not reach the desired site within the lung due to size constraints. For example, it was shown that decreasing the particle size from 5.4 µm to 2.7 µm reduced the total throat deposition by half and increased the total lung deposition by over two-fold for mannitol particles ( 45).
The process of particle deposition in human airways includes: a) inertial impaction, which is dominant in the upper airways where velocities are at a maximum and where many particles impact and stick to the pulmonary surface, b) sedimentation, which is predominant throughout the central and distal tract, where the particles settle on the surface of the lung due to gravitational forces and air resistance, and c) diffusion, which is the most important for submicrometer-sized particles ( < 0.5 µm), which are in random motion and deposit on the lung walls via Brownian motion (33,42,46). Previous studies have shown that the size and distribution site of particles inhaled in the lungs is as follows: particles smaller than 1 µm tend to diffuse, remaining suspended in the airways, and are typically exhaled; particles 1 -3 µm in diameter deposit primarily in the alveolar region, 8 -10 µm-sized particles undergo tracheobronchial deposition; and particle larger than 10 µm often exhibit deposition in the mouth (36).

Long-term storage considerations
Another major concern in the development of powder aerosols is in the storage of particles, which tend to agglomerate (16). In general, the more uniformly the particles are distributed, the more stable their shelf life is (47). Spray-drying, freezedrying, and spray-freeze-drying often improve long-term particle stability during storage for dry formulations in comparison to aqueous formulations. For example, freeze-dried poly(methylidene malonate) nanoparticles were stored for twelve months at different temperatures and showed no significant variations in pH, size, turbidity and cytotoxicity ( 48). Suspensions of freeze-dried PLGA microspheres, either alone or loaded with cyclosporine, were stored at 8°C and room temperature (RT) for six months. The suspensions were substantially instable while the freeze-dried particles showed an alternative for long-term stability at low temperatures (49). In another study, a poorly soluble drug, celecoxib, was formulated into solid phospholipid nanoparticles and showed that the amorphous matrix generated via spray-drying and freeze-drying significantly enhanced the dissolution rates and apparent solubility (50).
Another technique applied to enhance stability and solubility of particles is spray-drying. TIP used for the treatment of cystic fibrosis was designed and formulated considering the process parameters related to critical temperature transitions. A room temperature stable product was obtained that reqmres no refrigeration (51). Gentamicin particles for cystic fibrosis developed with L-leucine showed no significant degradation for up to 6 months of storage (52). Spray-freezedrying is a relatively new method of producing biopharmaceutical powder preparations. It combines the advantages of freeze-and spray-drying techniques in order to get a stable product and increase the solubility of poorly water-soluble drugs.
This technique was used to obtain a fine stable probiotic powder of Lactobacillus casei in mannose and CaC0 3 (53). Finally, re-aggregation of palygorskite nanofibers containing ofloxacin was successfully overcome by freeze-drying. Comparing with the traditional oven-dried sample, the freeze-dried sample showed enhanced dispersion stability of palygorskite in deionized water (54).

Nanocomposite microparticles: a solution for overcoming the obstacles of particle-based pulmonary delivery
Nanocomposites are an emerging class of materials utilized in pulmonary drug delivery. Such systems are typically multiphase structures, comprised of a single continuous phase or matrix (e.g. polymers) with one or more discontinuous or dispersed phases (e.g. drugs). A formulation of nanocomposite microparticles (nCmP) utilizes a combination of phase or matrix components, active pharmaceutical ingredients (APls), and inert components (excipients) as seen in Figure 2.1. Although the overall size of the nanocomposite microparticles can vary between 0.1 -100 µm, a fraction of these particles having aerodynamic diameters 1-2 µm have been reported to be more effective in depositing into alveolar spaces (55). The properties of nanocomposite materials typically differ from conventional composite materials owing to the 'critical size' or 'nanometer range' of the dispersed drugs in the matrix, which enhances the properties of the nanocomposite particle.

Drug-Loaded Nano particles
Excipien.t  comprehensive overview of the current pulmonary nCmP systems that have been designed and evaluated and one significant conclusion is that the majority of the systems were fabricated via spray drying, which is a fairly straightforward, accessible, and high throughput synthesis method.

Preparation of nanocomposite microparticles (nCmP)
Particle engineering is a key factor in the development of nCmP systems with desirable properties (85). In this section, we review techniques used in the preparation of nanoparticles that are used in nCmP formulations as well as techniques used to prepare the nCmP systems themselves. (1) (2) (5)

Nanoprecipitation
Nanoprecipitation, also called solvent displacement, is used to prepare nanoparticles by precipitating polymer at the interface between the water and the organic solvent. The precipitation is driven by the diffusion of the organic solvent into the aqueous phase with or without surfactant. Therefore, water-miscible solvents like acetone, methanol, or ethanol are recommended. The nanoprecipitation method often results in high drug loading and encapsulation efficacy for hydrophobic drugs, however, it is not an efficient means to encapsulate water-soluble drugs because of the high diffusion rate of drug from organic solvent into water (88).

Emulsification/solvent diffusion
Emulsification/solvent diffusion 1s a modified vers10n of the solvent evaporation method. In this method, an oil phase is formed by dissolving drug and polymer in a partially water soluble solvent such as propylene carbonate. This mixture is saturated with water to ensure the initial thermodynamic equilibrium of both liquids and then emulsified in an aqueous solution of surfactant. The solvent will diffuse into the aqueous phase leading to nanoparticle or nanocapsule formation and will then be eliminated by evaporation or filtration, according to its boiling point. This method also allows a high drug loading and encapsulation efficacy for hydrophobic drugs as well as for hydrophilic ones (88).

Salting out
Salting out, regarded as the modified vers10n of emulsification/solvent diffusion, is based ori the separation of a water-miscible solvent from aqueous solution caused by a salting out effect. In this method, drug and polymer are dissolved in a water-miscible solvent such as acetone. The mixture is emulsified in an aqueous gel containing a salting-out agent such as magnesmm chloride, calcium chloride, magnesium acetate, or sucrose as well as surfactant and is then diluted with a sufficient volume of aqueous solution to enhance the diffusion of solvent into the aqueous phase. Nanoparticles are formed as the result of solvent diffusion. The salting out agent is the key factor in the method influencing the efficacy of encapsulation but should be eliminated with solvent in the end. This method exhibits an advantage over others by avoiding high stress and increased temperature, which is suitable for protein encapsulation. Salting out is ineffective for hydrophobic drug and requires extensive washing steps (88-90).

IJialysis
Dialysis can be used as a simple and effective method for the preparation of nanoparticles of small size and narrow distribution (88). However, the mechanism of this method is not completely understood right now and is regarded as similar to that of nanoprecipitation (91). In dialysis, polymer and drug are dissolved in an organic solvent inside a dialysis tube with a proper molecular weight cut off. As the solvent displacement progresses, aggregation of polymer happens due to loss of solubility, thus forming a nanoparticle suspension. The properties of resulting nanoparticles are influenced by the type of solvent used in salting out.

Supercriticalfluid (SCF) technology
A supercritical fluid is defined as any substance at a temperature and pressure above its critical point. Supercritical fluids are defined as compressed gases or liquids above their critical pressures and temperatures, which exhibit several fundamental advantages as solvent or non-solvent. For pharmaceutical manufacturing, supercritical C02 is widely used due to its low critical temperature (31.1 °C) and moderate pressure (73.8 bar), non-toxic inert nature, and low cost. Supercritical fluid technology offers an environmentally friendly way to make nanoparticles due to efficient extraction and separation of organic solvents in this process. The resulting nanoparticles are in pure dry form or as pure aqueous suspensions (92).

7. Preparation of nanoparticles from polymerization of monomers
Nanoparticles with desired properties can be prepared by polymerization of monomers. This method can be subdivided into emulsion polymerization, miniemulsion polymerization, interfacial polymerization, and controlled/living radical polymerization. For this method, toxic organic solvents, surfactants, monomers and initiators must be removed from formed nanoparticles, which limit the use of the method (88).

Techniques used to prepare nCmP
The aerosol performance of nCmP is directly related to their preparation techniques. In this section, both traditional and more advanced methods used in the preparation of nCmP are discussed (93). These methods are classified into two categories, including top-down as milling larger particles to reduce size and bottom-up precipitation of nanoparticles out of solution. Techniques such as the modification of spray drying also show promise to prepare nCmP (94).

Milling
Milling is the traditional method of drug powder micronization, which has been intensively studied. The milling method can involve the use of ball mills, colloid mills, hammer mills, and jet or fluid energy mills, in which the jet mill is applied to prepare most inhalation powders. Jet milling is defined as a process of micronization by inter-particle collision and attrition. In this method, coarse particles carried by a high velocity gas as milling medium pass through a nozzle into the jet mill. The particles are then suspended in the turbulent gas stream where they break into smaller particles due to inter-particle collisions. The fine particles are then taken up by the gas stream out through the exit, while the larger ones remain in the mill for further micronization by more inter-particle collision. The application of jet milling is limited by lack of control over resulting particle properties and the high energy in the process increasing the risk of degradation of therapeutics (94, 95).

Supercritical fluid technology (SCF)
Supercritical fluid technology can also be used for preparation of micro-size inhalation as discussed previously. More specific discussion of SCF could be found in several review articles (85, 94, 95).

Spray drying
Spray drying is the most commonly used method to prepare dry powders for inhalation (96). In spray drying, nanoparticle suspensions with or without excipients are atomized at the nozzle. The droplets of the suspension go through the drying chamber with heated gas, resulting in evaporation of solvent and generation of nCmP.
The resulting nCmP are separated in a cyclone separator and collected in the collector (97). The spray drying method provides easily tunable process conditions and high reproducible properties of resulting particles. However, challenges exist to determine optimal processing conditions to achieve desirable properties, due to the fact that the drying process depends on complex parameters including the solvent used, solute concentration, inlet temperature, outlet temperature, atomizing pressure, feed properties, pump rate, drying gas type, and gas flow rate (81,96,98,99).

Spray-freeze drying (SFD)
Spray freeze-drying is a modified version of spray-drying, in which the suspension or solution is atomized into a cryogenic liquid such as liquid nitrogen generating droplets (95). The droplets are then lyophilized, resulting in porous spherical particles suitable for inhalation. Application of SFD is limited by several disadvantages such as causing irreversible damage to the proteins by drying and freezing stress, time consumption, and high expense (100). and further nCmP preparation methods should be taken into consideration. In addition, the desired properties of nanoparticles are also an important concern when choosing polymers. Detailed discussion on the materials used for nanoparticle formation can be found in former reviews (6,84,88,89,(101)(102)(103)(104)(105)(106)(107)(108)(109).

Excipients used in nCmP synthesis
The nCmP carriers and excipients are the inactive ingredients used to improve the delivery efficacy and stability of nCmP formulations for pulmonary delivery. The excipients have no therapeutic effects, but still have a significant role to form a desired delivery system of therapeutics by enhancing the mechanical properties of dry powder, physical or chemical stabilization of drug, and improving redispersion and dissolution properties of encapsulated nanoparticles (94). The Food and Drug Administration (FDA) recommends the use of excipients that are either commercially established or "generally recognized as safe" (GRAS). Hence, the choice of excipients used for nCmP formation is limited since the current excipients approved for pulmonary drug delivery are very limited in number (110). Detailed discussion on the excipients used for inhalation application can be found in other review and research articles (43,94,111,112).

Optimization of nCmP manufacturing via spray drying
While there are obviously many methods available for the synthesis of nCmP, spray drying has been the primary method to date. Spray drying is a straightforward and high throughput method allowing for easy reproduction of nCmP and thus the remaining sections of this book chapter will focus on the optimization of nCmP via spray drying. In a spray drying process, several parameters can be optimized to achieve desirable properties for nCmP formulations. In this section, these parameters are classified into two major categories: 1) spray drying conditions and 2) feed solution composition. The spray drying conditions include the inlet temperature, aspirator, spray gas flow, feed speed (pump rate), which can be easily tuned through the control panel of spray dryer, and humidity of drying gas, which can be modified by incorporating a dehumidifier or using dry nitrogen or compressed air. The feed solution composition covers the ratio of nanoparticles to carriers ( excipients ), solid concentration (concentration of total dry materials), properties of carriers, properties of nanoparticles, and solvent composition (18,(113)(114)(115).
The relevant parameters for the spray process correlate with and depend on each other. Therefore, to design an optimal nCmP system with parameters suitable for desirable resulting particles, a comprehensive evaluation of all parameters should be taken. However, difficulties exist in the selection of the process conditions, lying in the facts that: 1) a single change in parameter may influence several other properties of spray drying products; 2) several parameters may influence one property of spray drying products and these parameters depend on each other; and 3) a parameter may have positive effects on products in one range and negative effects or no influence in another range. In summary, there is no best spray drying condition for any given system, but only an optimal condition that is most suited for a product intended for a certain therapeutic application. This section aims at providing a strategy in the design of optimal process conditions and to aid in the understanding of the internal relationship between the process parameters and corresponding properties of resulting particles.

How spray dryer settings influence dry powder formulations
There are many spray dryer settings that influence the characteristics of dry powder formulations like nCmP (116,117). These settings include the inlet temperature, aspiration rate, feed (pump) rate, spray-drying gas flow rate, composition of feed solution, and other external parameters discussed below.
The inlet temperature is directly proportional to outlet temperature, which is the actual temperature at which the spray-dried materials are exposed. A higher inlet temperature is able to reduce the relative humidity in the drying gas, resulting in dryer and less sticky powders, thus increasing the powder yield. Aspiration, which provides a gas flow depending on the pressure drop of the overall system, reflects the drying energy that the system exerts on the drying process. The aspiration rate leads to a positive effect on increasing outlet temperature, reducing the amount of residual moisture in the product, and thus improving the degree of product separation in the cyclone. The feed rate affects the amount of time that materials undergo the drying process. A higher feed rate results in larger droplet dispersion in the drying chamber, which decreases the outlet temperature. An increase in the feed rate also increases the moisture content in the gas, resulting in more humid or moist products that could adhere to the glassware (cyclone or collector), thereby decreasing the yield. The spraydrying gas flow rate is another parameter that affects the products. Higher gas leads to a reduction in outlet temperature and reduces the droplet size, thus decreasing the size of resulting particles.
While the previous parameters can be tuned through the control panel of the spray dryer, the feed composition of the spray-dried solution can also affect the final powder product. A direct result of increasing the feed concentration is a decrease in the solvent amount, which results in less liquid evaporation, leading to higher outlet temperature. An increase in the feed concentration also reduces the partial pressure of solvent in the gas, leading to less water in the final product. The feed concentration exerts a positive influence on the particle size due to an increase of solid in droplets and a higher yield contributes to the ease in the collection of larger particles. The use of an organic solvent for the feed solution will result in smaller particles due to lower surface tension of these solvents. The organic solvent also requires less energy to vaporize, leading to higher outlet temperatures, thus resulting in dry products. Finally, the humidity of the drying gas can be reduced through the use of a dehumidifier or using ultra dry nitrogen, and this has a negative influence on the vapor uptake capability in the gas stream, resulting in positive effects on outlet temperature. As a result, higher drying gas humidity leads to higher humidity of the final products with lower yield.

A theoretical framework of nanocomposite microparticle (nCmP) formation
After a greater understanding of how spray dryer parameter settings influence the spray-drying process and resulting materials, we can move on to the more complicated spray-drying process with nanoparticles. Nanoparticles with various properties add complexity into the drying process in the droplet of feed solution dispersed in the drying chamber, while the formation process of droplet is affected by parameters such as the feed speed, aspirator speed, and drying gas flow rate. The engineering of nCmP formation requires an overall understanding of microbiology, chemistry, formulation science, colloid and interface science, heat and mass transfer, solid state physics, aerosol and powder science, and nanotechnology (18). As a result, thermophoresis of the nanoparticles occurs, leading to their movement towards the surface of the droplets (18,118,119). A detailed mathematical model of this phenomenon can be found in Ferry Iskandar's research (118,120).
The other driving force is caused by the concentration gradients in the droplet.
The evaporation of solvent at the droplet surface increases the concentration of components in the same plane, causing diffusion of the solute and nanoparticles towards the droplet center (shown in Figure 2.3) (18,121,122). For the solute or nanoparticles that diffuse fast enough compared to the shrinkage of droplet caused by evaporation, their concentration gradient along the droplet radius will be insignificant, resulting in relatively uniform distribution of the dry materials in the nanoparticles.
However, if the diffusion is slower than the surface shrinkage, there will be a higher concentration of solute or nanoparticles at the droplet surface compared to the droplet center, resulting in a shell of particles. The ratio of evaporation rate to diffusion rate can be described by the dimensionless Peclet number, which is an important factor influencing the size, shape, surface morphology, and component distribution of resulting particles. The Peclet number is well discussed in other review and research articles (18,121,123) and can be used to explain how the inlet temperature influences the properties nCmP. In the final step during nCmP formation, solvent evaporation finalizes, and the particles undergo changes in morphology and thermal states.
Crystallization or re-crystallization may occur, resulting in crystals with different sizes located inside or at the surface of resulting particles. Whether nCmP exhibit spherical or wrinkled morphology is also determined in this final step (18,123,124).

Influence of spray drying parameters on nCmP formation
After covering the basics on how spray drying parameters influence the properties of nCmP and an overall theoretical framework of nCmP formation, we can continue with a detailed discussion of these parameters. As previously mentioned, the parameters relevant to the spray process correlate with and depend on each other. In this section we discuss the influences of these parameters in terms of primary parameters, secondary parameters, physicochemical properties, and application properties (desirable properties). It is assumed that while discussing a given parameter, that the others remain constant.

Influence of inlet temperature
The inlet temperature proportionally determines the outlet temperature, which is close to the actual temperatures that spray-dried materials are exposed to. At high spray drying temperatures evaporation is rapid, resulting in a higher Peclet number thus indicating a fast increase of the solute or nanoparticle concentration at the droplet surface. Once a critical concentration is reached at the surface, precipitation will take place. Early precipitation results in smaller differences between the solid particle shell and the initial droplet size, producing larger shells. In the preparation of nCmP, suspended nanoparticles will form a composite shell in early processes and insufficient solids remain to fill the internal space of the shell, thus forming large and hollow nCmP. On the contrary, lower temperatures lead to slower evaporation that may lag behind the solid diffusion rate, causing delayed shell formation and smaller nCmP. In general, higher inlet temperature leads to higher processing temperature, improving the formation of larger and hollow particles, which tend to exhibit favorable aerosol performance. However, exceptions have been reported indicating that temperature may not influence the particle size (125). In that case, a wider temperature range could be applied for study on influence of temperature (123). In another study, temperature firstly has a negative effect on size of nCmP, then presents a positive effect (72). In that case, an optimal inlet temperature should be chosen based on comprehensive consideration on other parameters.
The inlet temperature also influences the crystallization of carrier molecules during solvent evaporation at the droplet surface, thereby changing the surface roughness of the resulting particles. At very high temperatures, carriers may not have enough time to crystallize during the precipitation process, thus forming into an amorphous state. Given enough time for crystallization, lower temperatures ensure larger crystals due to low nucleation rates from the solution, resulting in more roughness at particle surfaces. On the contrary, spray drying at higher temperatures leads to higher nucleation rates with more nuclei, resulting in smaller crystals and smooth surfaces. However, converse results have also been reported, in which higher temperature leads to larger crystals, while at lower temperature smaller crystals are produced. This latter case usually happens in the spray drying of small droplets at lab scale, where high temperature results in higher possibility of supersaturation at the surface than in large droplets. Therefore, delayed crystallization from a highly supersaturated viscous liquid or even water-free melt may occur, leading to larger crystals and rougher surfaces. In summary, temperature is able to influence the crystalline states of carrier molecules, which may further affect the performance of pulmonary delivery (126,127). More detailed discussions on the influence of temperature on crystalline states of spray-dried particles can be found elsewhere (115 , 125, 128, 129).
Another consideration with process temperature is its influence on the bioactivity of the powder products. Although therapeutics are encapsulated and protected by nanoparticles and microcarriers, the process temperature should be run at a safe range to keep bioactivity of the payloads (68,73,113,114). In addition, temperature can impact the redispersion of the nanoparticles from the dissolved nCmP.
At high temperatures, the melting of dry materials can occur, which will lead to the fusion of nanoparticles when they diffuse towards the droplet center and form a nanocomposite shell. Poor redispersion can cause nanoparticles to lose their favorable properties as drug delivery systems and impair the efficacy of nCmP.

Influence of ratio and properties of nanoparticles
In the absence of other driving forces, the freely soluble carrier molecules driven by diffusion exhibit even distribution in the droplet during the evaporation. The initial saturation of carriers is small and the characteristic time for precipitation at the droplet surface is close to the droplet lifetime. Given that no nanoparticles are included, solid particles with a density close to the true density of the dry materials tend to form (115,116,123,125,126,(130)(131)(132)(133)(134). In the case of when a system with lower ratio of nanoparticles to carrier molecules mixture is spray-dried, nanoparticles in the droplet may not be able to form a shell (135). When the nanoparticles account for a larger part in the dry materials, a nanocomposite shell is able to form and this determines the shape and size of resulting nCmP. Although the formation of nCmP is initiated from the same initial droplet solidification at surface, the morphology varies significantly. For nanoparticles capable of building rigid shells quickly, solid hollow spherical nCmP are formed. Otherwise, dimpled or wrinkled particles are formed. This  (68,76,84,(136)(137)(138)(139)(140) or particles based on other colloids (114,135,(141)(142)(143)(144)(145). The ratio of nanoparticles to carriers also influences the redispersion of the nanoparticles from nCmP, as well as the particle size (72). Properties of nanoparticles including particle size and surface properties can influence the properties of resulting

Influence of excipients
Carriers (excipients) are inactive materials that are applied in spray drying to enhance the physical or chemical stability of the active pharmaceutical ingredient.
Sugars are widely used as excipients of nCmP, as they provide advantages such as rapid dissolution in aqueous environment, leading to the immediate release of encapsulated nanoparticles. These excipients have been well studied in aerosol manufacturing processes with highly reproducible products (94,148,149). Safety should be a primary consideration in choosing the carrier excipient for nCmP. Many excipients that can be used in drug formulations for delivery in non-pulmonary routes may not suitable for pulmonary delivery due to their potential to injure the lungs (149).
Excipients approved for pulmonary delivery or presented as interesting alternatives for DPI formulations should be considered in high priority. Carriers also have significant impact on properties of nCmP, including the size, surface morphology, water content, and redispersion potential of nanoparticles. Comparisons of the influence of different excipients on dry powder formulation that can provide insight into how the excipient properties will affect nCmP have been studied (43,94,111).

Summary
Several parameters m the spray drying process can be tuned to achieve desirable properties of nCmP. These parameters, which correlate with and depend on each other, should be considered comprehensively to design optimized particles. Some parameters may have variable effects on the nCmP systems, and not all will be regarded as favorable. In this case, some of the properties may be abandoned in an overall consideration.

Optimization of nCmP formation process
Pulmonary delivery of therapeutics using nCmP is increasingly recommended for the treatment of lung disease due to their high efficacy of delivery, more convenient administration, and more flexible storage conditions. These advantages are a result of effective particle engineering of the delivery system, which includes optimal aerodynamic size, stability, and recovery of primary nanoparticles. In this section, strategies to achieve these favorable properties are discussed.

2.4.4.J. Optimization of aerodynamic size and morphology
A schematic showing the parameters involved in optimization of aerodynamic size of nCmP is shown in Figure 2.4. The proper aerodynamic diameter will be dependent on the delivery strategy and final particle location for a given application.
Since higher temperatures may impair the redispersion of nanoparticles and bioactivity of encapsulated therapeutics, relatively lower temperature should be given priority if possible. A way to prepare nCmP with small aerodynamic sizing is to prepare particles with small geometric s1zmg, smce these sizes are positively proportional to one another. This goal can be achieved by reducing the feed speed and feed solution concentration during spray drying. The former way results in smaller droplet size with lower dry material content, while the latter one decreases the dry materials in droplet directly, thus leading to smaller particles. nCmP can be designed with rough surface morphology and low water content to allow for high aerosol dispersion and low cohesion of the resulting nCmP. Overall, the aerodynamic diameter can be influenced by the geometric diameter, structure, processing conditions, and materials used in the preparation of nCmP.  Minimum water content should be a target to enhance stability of the structure of nCmP so that the favorable amorphous state of the nCrnP is maintained.

Optimization of primary nanoparticle recovery
A schematic outlining the optimization of nanoparticle recovery from nCmP is shown in

Drug release behavior from nanocomposite microparticle (nCmP) systems
As with many other drug delivery systems, nCmP is comprised of mainly carriers and APL The carrier often serves as the most important component for successful and targeted drug delivery. Significant amounts of research has been underway to explore the potential of nanoparticles as drug delivery carriers as they offer several advantages including enhanced cellular uptake, increased drug stability, decreased side-effects and prolonged drug release (150). Drugs encapsulated in a carrier are released from the nanoparticles by various physicochemical processes such as diffusion, dissolution, osmosis, magnetically controlled and targeted delivery, and the mechanism of drug release depends on the selection of carrier system and the solubility of the drug. The general physicochemical process for drug release from NP in nCmP are discussed as follows: 1. Diffusion: As a result of hydration, polymeric NP swell and release drugs through diffusion. The rate of drug release from the inner core of the water insoluble polymeric material follows Fick's law of diffusion (151)(152)(153). The diffusion of drug from the nanoparticle depends on the type of polymer, thickness of polymer coating, solution-diffusion membrane, and rate of permeation.

Therapeutic applications of nanocomposite microparticles: Impact in pulmonary and non-pulmonary diseases
Over the past few years, research in the field of pulmonary drug delivery has gained momentum for treatment of pulmonary and non-pulmonary diseases.
Pulmonary route for drug delivery is an attractive target, yet it poses serious challenges for investigation. Particles are being engineered in such a way to overcome the lung clearance mechanism, targeting specific regions of the lungs and being retained in the lung for longer time period. The nCmP formulations form an attractive interdisciplinary area that brings together polymer/material science, nanotechnology and biology for a multitude of biomedical applications. These are 'intelligent' systems widely used to deliver small molecules, genes, protein/peptides or drug nanoparticles to various targeted locations in the body. nCmP help in protecting them from degradation during delivery, targeting them to specific sites and regulating their release properties with respect to time. There are various nCmP that have been used to deliver therapeutic agents to the lungs (pulmonary drug delivery) and other body parts (non-pulmonary drug delivery) with their key findings discussed in Table 2.1 and          ( 4), in which lung disease is the primary cause of mortality ( 4,5). The dysfunction of the respiratory tract results in frequent pulmonary infections, inflammation, bronchiectasis, and eventually respiratory failure, which causes over 90% of deaths in CF patients (6). Pulmonary infection is one of the primary complications among patients with CF and these patients tend to develop chronic infections within a year if no treatment is implemented, which will accelerate the decline in lung function, resulting in earlier mortality (7,8).
Pseudomonas aeruginosa (P. aeruginosa) is regarded as the most prevalent pathogen in CF patients' lungs (7,9). Azithromycin (AZI) is a macrolide antibiotic with a broad gram-negative antibacterial spectrum and is highly effective against planktonic, actively growing bacteria (10). AZI has been extensively studied for the treatment of CF-related infections due to its ability to decrease P. aeruginosa accumulation (10-15) as well as its pharmacokinetic advantages including high bioavailability, distribution, and extended half-life (13,14).
Burkholderia cenocepacia (B. cepacia) infection is also considered to be a lethal threat to CF patients because it causes severe and persistent lung inflammation and it is resistant to nearly all available antibiotics ( 16). Rapamycin (RAP), also known as sirolimus, is an immunosuppressive macrolide that is the most commonly used chemical to induce autophagy (17). It has been shown that RAP can markedly decrease B. cepacia infection in vitro by enhancing the clearance of this bacterium via induced autophagy. RAP has been shown to reduce bacterial burden and decrease inflammation in the lungs of CF infected mice (16).
Acetalated dextran (Ac-Dex) is an acid sensitive, biodegradable, biocompatible polymer that can be prepared in a one-step reaction by reversibly modifying dextran with acetal groups (18). This modification reverses the solubility properties of dextran from hydrophilic to hydrophobic, making it possible to form polymeric particles using standard emulsion or nanoprecipitation techniques. Drug loaded Ac-Dex nanoparticles exhibit sustained release profile, with the advantages of extended duration of action, decreased drug use, improved management of therapy, enhanced compliance and reduced side effects. (19,20) In comparison to other commonly used polymers in drug delivery such as poly(lactic-co-glycolic acid) (PLGA) and polyesters, Ac-Dex offers several advantages. Most notably, the degradation rate of Ac-Dex can be tuned from minutes to months by modifying the ratio of cyclic and acyclic acetal groups, which have different rates of hydrolysis. Also, Ac-Dex degrades into dextran, a biocompatible, biodegradable, FDA-approved by-product, and very low levels of methanol and acetone (20)(21)(22).
Mannitol, an FDA approved, non-toxic, readily degradable sugar alcohol commonly used in pharmaceutical products, was applied as the excipient of nCmP due to its beneficial properties.(23) First, mannitol can be rapidly dissolved into an aqueous environment, leading to a burst release of encapsulated nanoparticles. In addition, mannitol can improve the fluidity of mucus, thus enhancing the mucus penetration rate of nanoparticles (24). Mannitol has been extensively studied as a carrier in spray-dried powder aerosols for pulmonary drug administration and the resulting particles have been shown to exhibit desirable water content, size, and surface morphology for successful aerosol delivery (25,26).
Pulmonary antibiotic delivery is increasingly recommended as maintenance therapy to prolong the interval between pulmonary exacerbations and to slow the progression of lung disease of CF patients due to the capability of these systems to achieve high drug

Synthesis of Acetalated Dextran (Ac-Dex)
Ac-Dex was synthesized as described previously (22) with mmor modifications. Briefly, 1 g of lyophilized dextran and 25 mg of PPTS were dissolved in 10 mL anhydrous DMSO. The resulting solution was reacted with 5 mL of 2-MOP under nitrogen gas for 5 minutes and was quenched with 1 mL of TEA. The reaction mixture was then precipitated in basic water (water and TEA, pH 9), vacuum filtered, and lyophilized (-50 °C, 0.023 mbar) for 24 hours to yield a solid product.

Synthesis of Vitamin E Poly( ethylene glycol) (VPSk)
VP5k was prepared with some modifications to a previously described method

NMR Analysis of Ac-DEX and VPSk
The cyclic-to-acyclic (CAC) ratio of acetal coverage and degrees of total acetal coverage per 100 glucose molecules was confirmed by 1 H NMR spectroscopy (Bruker 300 MHz NMR, MA). 10 mg of Ac-Dex was added to 700 µL of D 2 0 and was hydrolyzed with 30 µL of DCl prior to analysis. The hydrolysis of one cyclic acetal group produces one acetone whereas one acyclic acetal produces one acetone and one methanol. Consequently, from the normalized integrations of peaks related to acetone, methanol, and the carbon ring of dextran, the CAC ratio of acetal coverage and degrees of total acetal coverage per 100 glucoses were determined.

Preparation of Drug-loaded Nanoparticles
Azithromycin (

Powder X-Ray Diffraction (PXRD)
Crystalline states of the nCmP were examined by PXRD using a Rigaku Multiflex X-ray diffractometer (The Woodlands, TX) with a Cu Ka radiation source (40 kV, 44 mA). The samples were placed on a horizontal quartz glass sample holder (3 mm). The scan range was 5 -65° in 28 with a step width of 0.02 and scan rate of 2°/min.

Differential Scanning Calorimetry (DSC)
The thermal phase transitions of the nCmP were determined by DSC using a TA Q200 DSC system (TA Instruments, New Castle, DE, USA) equipped with an automated computer-controlled RSC-90 cooling accessory. 1 -3 mg of sample was weighed into Tzero TM alodined aluminum pans that were hermetically sealed. The sealed pans were placed into the DSC furnace along with an empty sealed reference pan. The heating range was 0 -250 °C at a heating rate of 10 °C/min.

Scanning Electron Microscopy (SEM)
The shape and surface morphology of the NP and nCmP were evaluated by SEM using a Hitachi S-4300 microscope (Tokyo, Japan). nCmP samples were placed on aluminum SEM stubs (TedPella, Inc., Redding, CA, USA) with double-sided adhesive carbon tabs. Nanoparticles were dispersed in basic water (pH= 9, 10 mg/mL) and this suspension was dropped onto aluminum SEM stubs and then dried at room temperature. Both the NP and nCmP samples were coated with a thin film of a gold/palladium alloy using an Emscope SC400 sputter coating system at 20 µA for 75 seconds under argon gas. Images were captured at 5 kV.

Particle Size, Size Distribution and Zeta Potential Analysis
The size, size distribution, and zeta potential of the NP systems were measured

Karl Fischer (KF) Titration
The water content of the nCmP was quantified by Karl Fischer (KF) titration using a 737 KF coulometer (Metrohm, Riverview, FL). Approximately 10 mg of powder was dissolved in anhydrous methanol. The resulting solution was injected into the KF reaction cell filled with Hydranal® KF reagent and then the amount of water was analyzed. Pure solvent was also injected for use as a background sample.

Aerosol Dispersion Analysis
In vitro aerosol dispersion performance of the nCmP was evaluated using a

Analysis of Nanoparticle Drug Loading and Nanoparticle Loading in nCmP
Drug loading and encapsulation efficacy of AZI and RAP NP and nCmP were determined using high performance liquid chromatograph (HPLC) (Hitachi Elite LaChrom, Japan). Detection of AZI was performed using the following conditions:

Statistical analysis
All measurements were performed in at least triplicate. Values are given in the form of means ± SD. The statistical significance of the results was determined using t-Test.
A p-value of <0.05 was considered statistically significant.

NMR Analysis of Ac-Dex and VPSk
Successful synthesis of Ac-Dex and VP5k was confirmed by NMR ( Figure   A.1 in Supplemental Information). Ac-Dex exhibited 68.7% cyclic acetal coverage (CAC) and 79.1% total acetal coverage (conversion of -OH groups). A yield of approximately 95% was obtained for this Ac-Dex. An increase in CAC is known to slow drug release due to slower degradation. (21,22) A high total acetal coverage (higher than 75% according to our research) is required to stabilize the VP5k coating of nanoparticles, which ensures small particles size and narrow size distribution. The signals for ester groups were detected via NMR, indicating successful conjugation of mPEG to vitamin E succinate for the successful formation of VP5k ( 45). The yield of VP5k was approximately 30%.

Characterization of N anoparticles
The azithromycin-loaded nanoparticles (AZI-NP), shown in Figure 3.2A, appear as uniform spheres with smooth surface morphology. NP size, size distribution, and zeta potential are shown in Table 3.1. The resulting sizes of the NP analyzed via DLS (approximately 200 nm) were larger than those observed from SEM micrographs and ImageJ analysis (approximately 100 nm) due to shrinking of the particles during freeze-drying from the collapse of hydrated PEGylated chains (46). Both drug-loaded NP systems exhibited desirable size (less than 200 nm) with narrow size distribution to allow for potential mucus penetration. The relative surface charge of the NP systems were nearly neutral, confirming PEG coverage on their surfaces ( 45). Similar results were obtained for drug-loaded and blank nanoparticles with respect to their size, size distribution, and surface charge, indicating that drug encapsulation did not affect the formation of the NP.
Both AZI and RAP were successfully encapsulated into the described NP systems. 13.0 % of initial AZI and 25.8 % of initial RAP were effectively entrapped within the NP prepared using nanoprecipitation of Ac-Dex and drugs in VP5k solution.
The low encapsulation of the drugs may be due to the improved solubility of the drugs in the spinning solution by VP5K micelles. RAP-NP exhibited a higher encapsulation efficacy as a result of the lower solubility of RAP than AZI in the aqueous spinning solution.
Results of the in vitro release of AZI-NP and RAP-NP at physiological pH and temperature are reported in Figure 3.3 as the percentage of drug released over time.
Both NP systems displayed sustained release for approximately 12 hours, which matched the degradation profile of other Ac-Dex particle systems (43). Based on previous research, Ac-Dex made of lOkDa dextran and reacted for 5 minutes showed a maximum of degradation at 6 hours and negligible degradation after that ( 4 7). A possible explanation of the release profiles could be that the first release stage corresponds to Ac-Dex degradation as well as nanoparticle dissociation, whereas after 6 hours the rate of drug release is controlled by drugs passively diffuse out of the dissociated matrix of nanoparticles following the partial degradation of Ac-Dex.

Manufacturing of nCmP
With respect to nCmP manufacturing, the outlet temperatures of AZI-and RAP-nCmP were 30 -31 °C and 30 -33 °C, respectively, while the yields were 62.4% and 60.6%, respectively.

nCmP Morphology, Sizing, and Size Distribution
As seen in the SEM images, the AZI-nCmP were mostly spherical with a corrugated surface (Figure 3.2B) and encapsulated nanoparticles were visible on the surface of the nCmP as seen in

Karl Fisher (KF) Titration
The residual water contents of AZI-nCmP and RAP-nCmP were approximately 6% (Table 3.2). This is within the range of other nCmP in our group (results not published) and that of previously reported inhalable dry powder formulations prepared by other groups (25,(48)(49)(50)(51). Water in inhalable powders can significantly reduce their dispersion properties during aerosolization due to the interparticulate capillary forces acting at the solid-solid interface between particles (52) and also have a negative effect on the stability of the powders (50). Correspondingly, low water content in the powder is highly favorable for efficient dry powder aerosolization and effective particle delivery (52, 53). other phase transitions were present under 120 °C, which indicated that all the materials will be stable during manufacturing and storage.

7 Powder X-ray Diffraction (PXRD)
X-ray diffraction diffractograms of the raw materials and drug-loaded nCmP are shown in Figure 3.5. Strong peaks were present for raw AZI, RAP, and mannitol powders. These strong peaks indicate that the raw materials are in their crystalline forms prior to spray drying, which is in accordance with previous research (54)(55)(56). No strong peaks were present for raw Ac-Dex, indicating that it is non-crystalline. This is quite different from commercialized polymers such as PLGA, which exhibits strong XRD characterization peaks (54)(55)(56)(57). The absence of diffraction peaks in Ac-Dex is likely because the Ac-Dex is collected by rapid precipitation in water. XRD patterns of AZI-nCmP and RAP-nCmP showed the absence of any diffraction peaks, suggesting amorphization of raw AZI and RAP in the particle matrix. Also, the peaks characterizing mannitol were significantly reduced, indicating that mannitol is primarily in an amorphous state in the nCmP. The results obtained from the XRD diffractograms confirmed those from DSC thermograms, where raw AZI, RAP, and mannitol were converted into amorphous form in the nCmP manufacturing process.

Drug and Nanoparticles Loading in nCmP
HPLC was used to determine the amount of drug loading in nCmP, which can be used to calculate the resulting nanoparticles loading and nanoparticle encapsulation efficacy in nCmP. These results are shown in Table 3.2. Both AZI-and RAP-nCmP exhibited desirable drug loading, high nanoparticle loading, and nanoparticle encapsulation efficacy. In addition, standard deviations of these three values were very low, which indicated reproducible drug loading of the nCmP can be achieved.

Nanoparticle Redispersion from nCmP
The properties of NP redispersed from nCmP were evaluated usmg DLS (

In vitro Aerosol Performance of nCmP
In vitro aerosol dispersion performance properties (Figure 3.6

Conclusions
Both azithromycin and rapamycin were successfully encapsulated in Ac-Dex nanoparticles and can be released in a sustained rate. The drug-loaded nanoparticles were smooth spheres 200 nm in diameter with narrow size distribution and slightly negative surface charge, which is desirable for mucus penetration. Most nanoparticles maintained these properties during the nCrnP manufacturing process as shown in redispersion testing. The nCmP systems were corrugated spheres of 1 µm with observable nanoparticles present on their surfaces. The water content of the nCrnP systems was relatively low, which can enable efficient dry powder aerosolization and particle delivery. None of the raw materials underwent degradation during nCrnP manufacturing, indicating the stability of the therapeutics during formation. No crystalline structures of AZI and RAP were observed in the nCrnP, which confirmed that both drugs in the nCrnP were in their amorphous form. In vitro aerosol performance testing demonstrated desirable aerosol dispersion characteristics of nCmP, allowing them to deposit in deep lung regions for drug delivery.
This nCrnP system sheds a light on dry powder-based antibiotic delivery due to its novel features including targeted pulmonary delivery, rapid mucus penetration potential, and controlled drug release. It can be applied as a promising alternative of the traditional antibiotic treatment by providing effective delivery of therapeutics, more convenient administration, more flexible storage conditions, and lower risk of contamination in the device.

Acknowledgements
The authors gratefully acknowledge financial support from an Institutional

37.
A. Kuzmovand T. Minko. Nanotechnology approaches for inhalation treatment of lung diseases. J Control Release.

Introduction
Pulmonary arterial hypertension (P AH) is a progressive cardiovascular disease that leads to limited exercise capacity, right ventricular heart failure, and ultimately death (1). It is hemo-dynamically defined as an increased mean pulmonary arterial pressure higher than 25 mmHg with a pulmonary capillary wedge pressure lower than 15 mmHg at rest (2, 3). Patients often develop PAH because of genetic and/or environmental insult, resulting in endothelial cell apoptosis, loss of distal vessels, and occlusive vascular remodeling, which causes elevated pulmonary arterial pressures.
The primary treatment options currently available for patients with P AH are based on drugs with vasodilatory properties that improve cardiopulmonary function.
Unfortunately, these drugs show no effect on prohibiting the progress of obliterative vascular pathology. Because of this, the only option for many patients to improve their health and quality of life is heart-lung transplantation. Therefore, new approaches to utilize compounds that focus on activating cellular mechanisms to reverse vascular remodeling are under investigation ( 4). Improving the function of the bone morphogenetic protein receptor-2 (BMPR2) signaling pathway is a potential direction in the development of improved P AH therapies, as the mutations of BMPR2 have been identified as the main cause of inherited P AH, accounting for 60 to 80% of familial cases. BMPR2 mutations are also the main cause of 10 to 25% in sporadic PAH and 9% of PAH associated with fenfluramine use (4-8).
Since challenges exist in the clinical application of BMPR2 gene therapy, tacrolimus can be applied instead, as it increases signaling through the BMPR2 pathway. Tacrolimus (TAC) is an FDA-approved immunosuppressive drug with a known pharmacokinetic and toxicity profile. Previous research on the use of TAC to treat P AH has shown that: 1) low-dose TAC can reverse dysfunctional BMPR2 signaling in pulmonary artery endothelial cells (ECs) from patients with idiopathic P AH, 2) low-dose TAC can prevent exaggerated chronic hypoxic P AH in mice with conditional BMPR2 deletion in ECs, and 3) low-dose TAC can reverse severe PAH in rats with medial hypertrophy following monocrotaline exposure and in rats with neointima formation following vascular endothelial growth factor receptor blockade and chronic hypoxia ( 4 ).
Although TAC shows prom1smg potential in the treatment of PAH, its application is hindered by its poor solubility (water solubility of 1.3 µg/mL), instability, and limited bioavailability (9)(10)(11). In addition to these limiting pharmacokinetic properties, systemic side effects such as neurotoxicity and nephrotoxicity also complicate the clinical use of TAC (12). To overcome this problem, biodegradable polymeric acetalated dextran (Ac-Dex) nanoparticles (NP) can be applied to more safely and effectively deliver TAC. Ac-Dex NP can improve hydrophobic agent solubility, promote drug penetration into cell membranes to obtain higher plasma concentration, provide sustained drug release, and provide reduced systemic side effects.
Ac-Dex is an acid-sensitive, biodegradable, and biocompatible polymer that can be prepared in a one-step reaction by reversibly modifying dextran with acetal groups (13). This modification reverses the solubility properties of dextran from hydrophilic to hydrophobic, making it possible to form polymeric particles using standard emulsion or nanoprecipitation techniques. In contrast to poly(lactic-co-  (17)(18)(19)(20)(21)(22). While dry powder aerosolized nanoparticles (less than 500 nm) will be exhaled owing to their small size and mass, aerosolized particles with aerodynamic diameters of 1-5 µm can effectively deposit into the lungs. However, this larger microparticle size range is also the size at which particles are commonly phagocytosed by alveolar macrophages, which limits their efficacy as aerosol drug delivery vehicles (23). To overcome these aforementioned obstacles for pulmonary TAC delivery, we developed nanocomposite microparticles (nCmP) in the form of dry powder aerosols (Figure 4.1). This system is comprised of TAC-loaded Ac-Dex NP entrapped in microparticle carriers comprised using the excipient mannitol to allow for the delivery of TAC NP to the lungs.
Mannitol was used as the excipient of nCmP due to its beneficial properties in aerosol formulations. First, mannitol is highly water-soluble, which ensures rapid decomposition of nCmP in the aqueous environment of mucus and pulmonary surfactant in the lungs, leading to a burst release of NP from the nCmP system.
Furthermore, mannitol is able to enhance mucus penetration of nanoparticles by increasing fluidity of mucus in the lung (24). Finally, mannitol has been extensively studied as a carrier in spray-dried powder aerosols for pulmonary drug administration and the resulting particles have been shown to exhibit desirable water content, size, and surface morphology for successful aerosol delivery (25)(26)(27)(28)(29)(30).
In the present nCmP system, the drug-loaded nanoparticles contain TAC as a therapeutic and these NP are coated by polyvinyl alcohol (PV A). Upon pulmonary administration, the nCmP will deposit on the mucus in the respiratory tract, decompose into free NP and mannitol, and allow the nanoparticles to penetrate the mucus barrier and then release drug to the targeted site. The goal of the described research was the initial development and physicochemical characterization of the nCmP systems with ability of targeted delivery, improved drug solubility, and potential of efficient penetration of mucus barrier and of controlled release via particle engmeenng.

Materials
Dextran Resazurin was obtained from Acros Organics.

Synthesis and NMR Analysis of Acetalated Dextran (Ac-Dex)
Ac-Dex was synthesized as described previously (14)

Formation of TAC-Loaded Ac-Dex Nanoparticles (TAC NP)
Tacrolimus-loaded nanoparticles (TAC NP) were prepared via oil/water emulsion and solvent evaporation. 90 mg of Ac-Dex and 10 mg of TAC were dissolved in 1 mL of DCM over an ice bath, establishing the organic phase. 3% PVA in PBS, the aqueous phase, was added to the organic phase and the resulting solution was sonicated (Q500 Sonicator, Qsonica, Newtown, CT) for 30 seconds with a 1 second on/off pulse at 70% amplitude. The emulsion was transferred to a spinning solution of 0.3% PV A in PBS and stirred for 4 hours for evaporation of the organic solvent and particle hardening. The solution was then centrifuged at 12000 rpm for 20 minutes to collect the nanoparticles. Nanoparticles were washed once with basic water, redispersed in 0.1 % PVA, and lyophilized (-50 °C, 0.023 mbar) for 48 hours.

Particle Morphology and Shape Analysis via Scanning Electron Microscopy (SEM)
The shape and surface morphology of the NP and nCmP were evaluated by SEM using a Zeiss SIGMA VP Field Emission-Scanning Electron Microscope (FE-SEM) (Germany). nCmP samples were placed on aluminum SEM stubs with doublesided adhesive carbon tabs. Nanoparticles were dispersed in basic water (pH = 9, 10 mg/mL) and this suspension was dropped onto aluminum SEM stubs and then dried at room temperature. Both the NP and nCmP samples were coated with a thin film of a gold/palladium alloy using a BIO-RAD sputter coating system at 20 µA for 60 seconds under argon gas. Images were captured at 5 kV at various magnifications.

Particle Size, Size Distribution, and Zeta Potential Analysis
The size, size distribution, and zeta potential of the NP systems were measured  were analyzed for drug content via UPLC using the same methods described previously.

Karl Fischer Coulemetric Titration
The water content of the nCmP was quantified by Karl Fischer (KF) titration using a 737 KF coulometer (Metrohm, Riverview, FL). 10 mg of powder was dissolved in anhydrous methanol. The resulting solution was injected into the KF reaction cell filled with Hydranal® KF reagent and then the amount of water was analyzed.

Differential Scanning Calorimetry (DSC)
The thermal phase transitions of the nCmP, NP, and their raw components were determined via DSC using a TA QlO DSC system (TA Instruments, New Castle, DE, USA) equipped with an automated computer-controlled TA instruments DSC refrigerated cooling system. 1 -3 mg of sample was weighed into Tzero TM alodined aluminum pans that were hermetically sealed. The sealed pans were placed into the DSC furnace along with an empty sealed reference· pan. The heating range used was 0 -300 °C at a heating rate of 10 °C/min.

Powder X-Ray Diffraction (PXRD)
Crystalline states of the nCmP and its raw components were examined by PXRD using a Rigaku Multiflex X-ray diffractometer (The Woodlands, TX) with a Cu Ka radiation source (40 kV, 44 mA). The samples were placed on a horizontal quartz glass sample holder (3 mm). The scan range was 5 -80° in 28 with a step width of 0.02 and scan rate of2 °/min.

Impactor (NGI)
In vitro aerosol dispersion performance of the nCmP was evaluated using a Next Generation Impactor™ (

Tapped Density and Theoretical Aerodynamic Diameter Analysis
The density of the nCmP was evalulated via tapped density measurements as

In vitro cytotoxicity of TAC nCmP using resazurin assay
The in vitro cytotoxicity of TAC nCmP as well as free TAC were evaluated using the resazurin assay. Human adenocarcinomic alveolar basal epithelial cell line (A549,

Statistical Analysis
All measurements were performed in at least triplicate. Values are given in the form of means± SD. The statistical significance of the results was determined using ttests where a p-value of <0.05 was considered statistically significant.

Characterization of Ac-Dex
Successful synthesis of Ac-Dex was confirmed by NMR. Ac-Dex exhibited 71.5% cyclic acetal coverage and 79.2% total conversion of -OH groups. An increase in CAC is known to decrease the drug release rate due to slower degradation of Ac-Dex (13,14). A high total acetal coverage (higher than 75% according to our research) is required to stabilize the PV A coating of nanoparticles, which ensures small particles size and narrow size distribution  (34,35). The TAC NP system exhibited desirable size (less than 200 nm) with narrow size distribution to potentially allow for mucus penetration and pulmonary epithelial targeting. The surface charge of the NP system was slightly negative, which is suitable to reduce interactions with negatively charged mucin fibers, enhancing the particle permeation through the mucus barrier (35).   of TAC was released in the first 6 hours and 10% was released in the next 6 hours.
Previous research showed that Ac-Dex NP made of 1 OkDa dextran and reacted for 5 minutes exhibited a maximum of degradation at 6 hours and negligible degradation after that (15). Based on these results, the TAC release profiles in this study can likely be explained as follows: the first release stage (up to 6 hours) corresponds to Ac-Dex degradation as well as nanoparticle dissociation, whereas after 6 hours the rate of TAC release is controlled by drugs passively diffusing out of the dissociated matrix of nanoparticles following the partial degradation of Ac-Dex. This two-stage release profile of Ac-Dex based particles matches that of other drug-loaded Ac-Dex nanoparticle systems in our group and other previous studies (33).

nCmP Morphology, Size, and Size Distribution
TAC nCmP displayed a wrinkled and raisin-like surface with visibly encapsulated NP as seen in Figures 4.2B and 4.2C. The corrugated surface of the nCmP reduces their contact area, which prevents the particles from aggregating upon aerosolization, thus improving aerosol performance and the dispersion of the NP from the dry powder nCmP formulation. The cause of the raisin-like morphology of the nCmP may be attributed to the early formation of nanoparticle shells in droplets during spray drying. The formation of nanoparticle shells during spray drying determines the geometric size of nCmP. As the drying proceeds, the remaining water keeps evaporating from the droplet center, resulting in hollow particles that tend to shrink (36,37). The number average geometric diameter of the TAC nCmP system was 2.13 ± 0.52 µm, as determined by ImageJ analysis.

Karl Fischer Titration
The residual water content of TAC nCmP was approximately 8% (Table 4.

2),
which is acceptable in this dry powder formulation. Water in inhalable dry powders can significantly reduce their dispersion properties during aerosolization due to the interparticulate capillary forces acting at the solid-solid interface between particles (3 8) and the presence of water is also responsible for the instability of the powders during storage (39). Correspondingly, low water content in inhalable dry powders is highly favorable for efficient dry powder aerosolization and effective particle delivery (38,40).

Powder X-ray Diffraction (PXRD)
X-ray diffraction diffractograms of the raw materials and TAC nCmP are shown in Figure 4.5. Strong peaks were present for raw TAC and mannitol powders.
These strong peaks indicate that the raw materials are in their crystalline forms prior to spray drying, which is in accordance with previous studies (41,42). No strong peaks were present for raw Ac-Dex, indicating that it is non-crystalline. The non-crystallinity of Ac-Dex is quite different from commercialized polymers such as PLGA, which exhibits strong XRD characterization peaks (43)(44)(45). The absence of diffraction peaks in Ac-Dex is likely because the Ac-Dex was collected by rapid precipitation in water.
XRD patterns of TAC nCmP displayed the absence of any diffraction peaks of raw TAC, suggesting amorphization of raw TAC in the particle matrix, which is suitable for the improvement of tacrolimus solubility in physiological conditions. Also, the peaks characterizing mannitol were significantly reduced, indicating that mannitol is reduced to its amorphous state in the nCmP. The results obtained from the XRD diffractograms confirmed those from DSC thermograms, where raw TAC and mannitol were converted into amorphous forms in the nCmP manufacturing process.

Nanoparticle Redispersion from nCmP
The properties of NP redispersed from nCmP dissolved in water were evaluated using DLS (Table 4.1). The size and PDI of the redispersed NP exhibited slight increases after redispersion from nCmP, which is likely a result of agglomeration that occurred during the spray drying process. The NP surface charge remained slightly negative. These parameters were all within the desirable ranges for effective penetration of NP through the mucus barrier, demonstrating the favorable properties of nanoparticles remain intact after spray drying.

In Vitro Aerosol Dispersion Performance Using Next Generation Impactor (NGI)
In vitro aerosol dispersion performance properties of the nCmP were evaluated using a Next Generation Impactor™ coupled with a human DPI device (Figure 4.6 and Table 4 with TAC with a target blood level of 1.5 -5 ng/mL (48)(49)(50). The present nCmP system is expected to achieve a therapeutic effect using a low amount of TAC through its improved delivery efficacy over current systemic methods. 9.8% of TAC nCmP deposited on stages 5 -7 and it is predicted that these particles will deposit in the deep lung alveolar region due to diffusion mechanisms of deposition ( 51 ). Subsequently, 71.2% of nCmP deposited on stages 2 -4 and are thus predicted to deposit predominantly in the lungs by sedimentation due to gravitational settling (18,52,53).
Overall, the nCrnP exhibited desirable aerosol dispersion characteristics allowing them to deposit in deep lung regions for effective drug delivery of TAC.

Analysis of nCmP Density and Theoretical Aerodynamic Diameter
The density of the particles was determined via tapped density measurements.
The TAC nCmP exhibited a relative low density value of 0.138 ± 0.001 g/cm 3 compared with the raw materials (1.5 g/cm\ which can be attributed to the raisin-like surface morphology and hollow structure of the nCmP. The theoretical MMAD calculated with the geometric diameter and tapped density was 0. 79 ± 0.19 µm, which was lower than the experimental result. The lower MMADT in comparison to MMADE can be attributed to nCmP agglomeration during aerosolization, which increased the geometric size (and thus aerodynamic diameter) of the dry powder formulation.

In vitro cytotoxicity of TAC nCmP microparticles using resazurin assay
To evaluate the response of particles on cell viability, in vitro cytotoxicity of TAC nCmP as well as free TAC against A549 cells were conducted. TAC nCmP and free TAC were exposed to A549 cells at different concentrations for 48 h and 72 h, and the cell viability was analyzed using resazurin assay (Figure 4.7).

Conclusions
Tacrolimus was successfully encapsulated in Ac-Dex nanoparticles with a high encapsulation efficacy. The drug-loaded nanoparticles were 200 nm smooth spheres with narrow size distribution and slightly negative surface charge, which is desirable to overcome the mucus barrier in the lung. These favorable properties were maintained during the nCmP manufacturing process as shown in redispersion testing. The nCmP systems were 2 µm, raisin-like spheres with observable nanoparticles present on their surface. The water content of the nCmP systems was relatively low, which can enable efficient dry powder aerosolization and particle delivery. None of the raw materials    low water content. Acetalated dextran (Ac-Dex) was used as the polymer to form the nanoparticles, curcumin was used as a model drug, and mannitol was applied as an excipient in the nCmP formulation. Box-Behnken design was applied using Design-Expert software for nCmP parameter optimization. The results indicated that the nanoparticle ratio (NP%) and feed concentration (Fe) are significant parameters that affect the aerodynamic diameters of nCmP systems. NP% is also a significant parameter that affects the drug loading. Fe is the only parameter that influenced the water content of the particles significantly. All prepared nCmP systems could be completely redispersed into the parent nanoparticles, indicating that none of the factors have an influence on the redispersion of the nanoparticles within the design range. Overall, the optimal spray drying condition to prepare nCmP with a small aerodynamic diameter, complete redispersion of the nanoparticles, low water content, and high drug loading is that with an 80% nanoparticle ratio, 0.5% feed concentration, and an inlet temperature lower than 130 °C.

KEYWORDS:
Nanocomposite microparticles, pulmonary delivery, design of experiment, spray drying, particle engineering

Introduction
The deep lung (alveolar) region has attracted increasing attention as a target for drug delivery for several reasons (1). First, the alveolar surface area of the adult human ranges between 97 and 194 m 2 , providing a large surface for drug absorption. Second, the liquid layer over the alveoli is very thin (approximately 0.1 µm), ensuring rapid and unhindered drug absorption. Third, the plentiful capillary vessels situated beneath the thin alveolar epithelium provide efficient passive drug transport from the lungs to the bloodstream. Finally, the enzymatic activity in the lung is comparatively low, which enhances drug absorption (2, 3). As a result of the direct delivery of therapeutics to the lungs, aerosol drug delivery systems can provide several advantages including increased local drug concentration, reduced side effects, rapid onset of pharmaceutical action, and avoidance of the first-pass metabolism associated with the liver ( 4-6). These systems have been widely applied in the delivery of therapeutics such as antibiotics, proteins, peptides, and chemotherapeutics (7) for the treatment of diseases such as asthma, chronic obstructive pulmonary disease (COPD), cystic fibrosis (CF)-related pulmonary infections (8), pulmonary hypertension (9) and lung cancer (10,11).
In comparison to liquid aerosol formulations, inhalation aerosols based on dry powders offer additional benefits such as enhanced stability of formulations, controllable particle size for targeting specific regions of the lung, and increased drug loading of hydrophobic or lipophilic payloads (10,12). Unfortunately, barriers exist in the implementation of dry powder pulmonary delivery systems, including: (1) dry powder aerosols with aerodynamic diameters smaller than 1 µm will often be exhaled; (2) particles with aerodynamic diameters above 5 µm tend to deposit in the mouth, throat or upper lung mucosa and will then be eliminated due to mucus clearance; and (3) particles larger than 1 µm deposited in the deep lung area may be cleared from the alveoli via macrophages (10,13,14).
The application of traditional dry powder aerosols comprised of a mixture of drugs and excipients may be limited by the aforementioned barriers, but fortunately spray-dried nanocomposite microparticles (nCmP) can be employed to solve these issues.

44). Response surface methodologies (RSM) such as central composite design (CCD) or
Box-Behnken design (BBD) have been applied as optimization methods in the development and improvement of pharmaceutical formulations ( 45,46).
In the current study, Box-Behnken design was used to investigate the influence of the spray drying parameters: (1) weight ratio of NP (NP%), (2) total feed concentration of mannitol and NP (Fe), (3) and spray drying inlet temperature (Tin) in order to optimize the nCmP systems. These factors were selected as they have been proven to be significant in the successful formation of nCmP in previous research (22,38,39). Mannitol was used as excipient due to its ability to enhance the mucus penetration of nanoparticles by increasing the fluidity of mucus in the lungs (46). Acetalated dextran (Ac-Dex) was applied as the polymer to form nanoparticles as Ac-Dex is a novel, acid-sensitive, biodegradable, and biocompatible polymer that has attracted increasing attention in nanoparticle drug delivery applications. Compared to PLGA or other polyesters, Ac-Dex shows advantages including easily tunable degradation, minimal burst release at physiological conditions (pH 7.4), degradation into neutral by-products avoiding undesirable changes of micro-environmental pH, and acid-sensitivity, facilitating the targeted delivery of therapeutics (47)(48)(49). Curcumin (CUR) was used as model drug owing to its hydrophobic nature and fluorescent properties. The goal of this research was to identify the optimal spray drying condition(s) in order to prepare nCmP systems comprised of CUR-loaded Ac-Dex nanoparticles. The optimal nCmP should show favorable properties including: small aerodynamic diameter, effective nanoparticle redispersion, high drug loading, and low water content.

Synthesis and NMR Analysis of Acetalated Dextran (Ac-Dex)
Ac-Dex was synthesized as described previously (49) with minor modifications.
Briefly, 1 g of lyophilized dextran and 25 mg of PPTS were dissolved in 10 mL anhydrous DMSO. 5 mL of 2-MOP was added into the solution to initiate the reaction and 1 mL of TEA was added after 5 minutes to quench the reaction. The resulting mixture was then precipitated in basic water (water and TEA, pH 9), vacuum filtered, and lyophilized (-50 °C, 0.023 mbar) for 24 hours to yield a solid product.
The cyclic-to-acyclic (CAC) ratio of acetal coverage and the total acetal coverage was confirmed by lH NMR spectroscopy (Bruker 300 MHz NMR, MA). 10 mg of Ac-Dex was added to 700 µL of D 2 0 and was hydrolyzed with 30 µL of DCl prior to analysis. The hydrolysis of one cyclic acetal group produces one acetone whereas one acyclic acetal produces one acetone and one methanol. Consequently, from the normalized integrations of peaks related to acetone, methanol, and the carbon ring of dextran, the CAC ratio of acetal coverage and total acetal coverage per 100 glucoses were determined.

Formation of CUR-Loaded Ac-Dex Nanoparticles (CUR NP)
Curcumin-loaded nanoparticles (CUR NP) were prepared via oil-in-water emulsion solvent evaporation. 49 mg of Ac-Dex and 1 mg of CUR were dissolved in 1 mL of DCM over an ice bath, establishing the organic phase. The aqueous phase, comprised of 6 ml of 3% PVA in PBS, was added to the organic phase. The resulting mixture was sonicated (Q500 Sonicator, Qsonica, Newtown, CT) for 30 seconds with a 1 second on/off pulse at 70% amplitude. The emulsion was transferred to a spinning solution of 0.3% PVA in PBS and stirred for 3 hours to allow for evaporation of the organic solvent and particle hardening. The solution was then centrifuged at 19802 xg for 20 minutes to collect the nanoparticles. The nanoparticles were then washed once with DI water, redispersed in 0.1 % PV A, and lyophilized (-50 °C, 0.023 mbar) for 48 hours.  Table 1. The resulting nCmP were separated in a high-performance cyclone, dried for 15 minutes in the spray dryer, collected in a sample collector, and stored in amber glass vials in desiccators at -20°C.

Design of Experiment
The spray drying parameters evaluated by design of experiment (DOE) were: the weight ratio of NP (NP%), total feed concentration of excipient and NP (F c ), and spray dryer inlet temperature (Tin) ( Table 1). BBD was applied to determine the experimental conditions using Design Expert (Version 8 Stat-Ease, Inc.). The DOE responses including the aerodynamic diameter, properties of the nanoparticles following redispersion (characterized by particle size and PDI), drug loading, and water content were analyzed in Design Expert software using an analysis of variance (ANOVA) approach. The complete design and formulation parameters that were varied are shown in Table 2.

Particle Morphology and Shape Analysis via Scanning Electron Microscopy (SEM)
The shape and surface morphology of the NP and nCmP were evaluated by SEM using a Zeiss SIGMA VP Field Emission Scanning Electron Microscope (FE-SEM) (Germany). nCmP samples were placed on aluminum SEM stubs (Ted Pella, Inc., Redding, CA) with double-sided adhesive carbon tabs. Nanoparticles were dispersed in DI water (pH 7, 10 mg/mL) and the suspensions were dropped onto aluminum SEM stubs and then dried at room temperature. The samples were coated with a thin film of a gold/palladium alloy using a BIO-RAD sputter coating system at 20 µA for 60 seconds under argon gas. Images were captured at 5 kV at various magnifications.

Analysis of Nanoparticles Following Redispersion in Aqueous Solution
Following nCmP formation, the properties of the NP entrapped m and subsequently released from the nCmP were analyzed. The nCmP were dispersed in DI water (pH 7, 0.3 mg/mL) and the size, PDI, and zeta potential of the particles were In addition, the influence of the spray dryer inlet temperature on size and PDI at higher spray drying temperatures was evaluated by spray drying CUR NP without mannitol (100 nCmP) at 50,90,130,160,190, and 220 °C. The resulting particle samples were sonicated for 1, 2, and 15 minutes prior to DLS measurement.

Analysis of Drug Loading of CUR NP and CUR nCmP
Drug loading and encapsulation efficiency of CUR NP and CUR nCmP were initial mass of NP in nCmP formulation

Karl Fischer Coulometric Titration
The water content of the nCmP was quantified by Karl Fischer (KF) titration using a 737 KF coulometer (Metrohm, Riverview, FL). 3 mg of powder was dissolved in anhydrous methanol. The resulting solution was injected into the KF reaction cell filled with Hydranal® KF reagent and then the amount of water was analyzed. Pure solvent was also injected for use as a background sample.

Differential Scanning Calorimetry (DSC)
The thermal phase transitions of the raw particle components were determined via DSC using a TA QIO DSC system (TA Instruments, New Castle, DE, USA) equipped with an automated computer-controlled TA instruments DSC refrigerated cooling system.
1 -3 mg of sample was weighed into Tzero TM alodined aluminum pans that were hermetically sealed. The sealed pans were placed into the DSC furnace along with an empty sealed reference pan. The samples were analyzed at a heating range of 0 -250 °C and heating rate of 10 °C/min.

In Vitro Aerosol Dispersion Performance with the Next Generation Impactor (NGI)
The in vitro aerosol dispersion performance of nCmP was evaluated using a Next

Statistical analysis
All measurements were performed in at least triplicate. Values are given in the form of mean ± standard deviation. The statistical significance of the results was determined using analysis of variance (ANOV A). A p-value less than 0.05 was considered statistically significant.

Preparation and Characterization of Acetalated Dextran (Ac-Dex)
Successful synthesis of Ac-Dex was confirmed by lH NMR where the polymer exhibited 61.3% cyclic acetal coverage and 77.1% total conversion of -OH groups. A high CAC is known to decrease the degradation rate of Ac-Dex (50), and a high total acetal coverage is required to stabilize the PV A coating of nanoparticles, which ensures small particle size and narrow particle size distribution.

Preparation and Characterization of Original CUR NP
The average nanoparticle size, size distribution/polydispersion index (PDI), zeta potential, drug loading, and encapsulation efficiency of curcumin-loaded nanoparticles (CUR NP) are shown in Table 5.3. The original CUR NP exhibited an average diameter of approximately 200 nm, which is in the desirable range to avoid macrophage clearance and mucus entrapment (14). The low PDI value denotes a narrow size distribution of nanoparticles. The surface of nanoparticles was slightly negative, which is suitable to reduce the interaction of the NP with negatively charged mucin fibers present in airway mucus (52). Curcumin was successfully encapsulated into the CUR NP and the amount of the drug in CUR NP was sufficient for the quantification of CUR and NP in nCmP samples.

.2.1 Box-Behnken Design
The results of the Box-Behnken design are shown in Table 5 Table 5.5. The interaction of parameters was not included, as its effect on the results was negligible. The sign of the coefficient estimates signifies a positive or negative influence of the factor on the corresponding response. Since this study was designed to qualify the effects of the formulation parameters on particle properties and the final model will not be used for prediction purposes at this point, a lack of fit test was not a major consideration in model selection.

Drug Loading and Encapsulation Efficiency (EE)
The drug loading of CUR nCmP was only affected by the nanoparticle loading (NP%, p < 0.0001) and no other factors had a significant influence on this parameter.
Since the drug loading of the CUR NP was constant in this study, the NP% was the only parameter that should definitively affect the drug loading of nCmP. Therefore, either increasing the NP% in the nCmP formulation or using original NP with higher drug loading can be applied to achieve higher drug loading in the nCmP system. Unlike the formation of traditional spray-dried particles (53 , 54) composed of a mixture of drug and excipient, the drug loading capacity of nCmP can be optimized beforehand during nanoparticle preparation. By providing an optimal nanoparticle system, drug loading optimization during spray drying can be remarkably simplified.

.3 .2.4 Water Content of CUR nCmP
The CUR nCmP samples exhibited water content in the range of 2.6 -6.7% (Table 5.4). The water content of the nCmP was affected only by the feed solution concentration (p = 0.0009). Higher feed solution concentration resulted in lower water content in the particles, which corresponds to previous research on nCmP systems (22) and traditional spray-dried particles (56). However, unlike previous studies (22,56), Tin and NP% showed no effect on water content. The reason for this is likely the post drying process, which further removed water from the nCmP, thus reducing the difference in water content among the samples. Low water content is highly favorable in inhalable dry powders to allow for efficient aerosolization and particle delivery as it improves particle dispersion properties during aerosolization by reducing the interparticulate capillary forces acting at the solid-solid interface between particles (53,57). Furthermore, low water content enhances the stability of dry powders during storage (58). Since the water content of all of the samples were acceptable in this study, the post drying process can be applied to simplify optimization in future studies.

Morphology of CUR nCmP
As seen in the representative SEM images in The raisin-like morphology of nCmP may be attributed to the early formation of nanoparticle shells in droplets during the spray drying process, during which · the geometric size of nCmP is determined. As the drying process proceeds, the remaining water evaporates from the droplet center and the newly precipitated dry materials cannot fill the hollow interior of the nCmP, which results in shrinkage of the particles (59, 60).
The spherical particle systems with cracks in their surface are the typical morphology of particles with mannitol shells. The cracks evident on these types of particles were not caused by the spray drying process but instead were generated during SEM imaging due to the high energy of the electron beam ( Figure B.2).

Conclusions
The current study identified the optimal spray drying conditions to prepare Ac-Dex nanoparticles-based nCmP systems with favorable properties including: small aerodynamic diameter, effective nanoparticle redispersion, high drug loading, and low water content. Higher NP% resulted in nCmP systems with smaller MMAD, high FPF, higher RF, and higher drug loading. Lower Fe lead to smaller MMAD, higher FPF, and high RF, while higher Fe was beneficial for lower water content. Tin had a negligible influence on the nCmP properties in the design range, while in higher range it will hinder the redispersion of NP from the nCmP. The drug loading was only affected by the NP% of nCmP, which simplifies the optimization of spray drying nCmP given that high drug loading of NP can be achieved. The influence of each spray drying parameter on the properties of the resulting particles is summarized in Figure 5.6. The optimal condition to prepare nCmP with a small aerodynamic diameter, desirable nanoparticle redispersion, high drug loading, and low water content is via high NP%, low Fe, and a Tin under 130 °C.

Acknowledgements
The authors gratefully acknowledge financial support from an Institutional  of the Ac-Dex particles were tuned by modifying the Ac-Dex reaction time, particle size, and ratio of fast to slow degrading Ac-Dex. The pH of the environment was also a significant factor that influenced the drug release rate. All nCmP and MP systems exhibited desirable aerodynamic diameters that are suitable for deep lung delivery (e.g. below 5 µm). Overall, the engineered Ac-Dex aerosol particle systems have the potential to provide targeted and effective delivery of therapeutics into the deep lung.

Introduction
Pulmonary drug delivery has exhibited promising potential in the treatment of lung diseases, as it allows for the delivery of a wide range of therapeutics directly and efficiently to the lungs, thereby increasing local drug concentration, reducing systemic side effects, providing a rapid onset of pharmaceutical action, and avoiding the first-pass metabolism associated with the liver (1-4). The deep lung (alveolar) region can be utilized as a route for systematic drug delivery due to the enormous surface area available and nearby plentiful capillary vessels that facilitate drug absorption, the very thin (approximately 0.1 µm) liquid layer over the alveoli that ensures rapid and unhindered drug absorption, and low enzymatic activity, which enhances drug availability (1, 5, 6).
As a result, various therapeutics such as antibiotics, proteins, peptides, anti-cancer drugs (7), plasmid DNA (8), siRNA (9), and anti-tuberculosis (TB) drugs have been employed in the inhalation forms for the treatment of pulmonary diseases such as asthma, chronic obstructive pulmonary disease (COPD), cystic fibrosis (CF)-related pulmonary infections, and lung cancer (7,10).
Dry powders is a dosage form that delivers therapeutics to the lung, in the form of particles, using a dry powder inhaler (7). Compared with liquid aerosols, these formulations offer additional benefits such as enhanced stability of the formulation, controllable particle size for targeting different regions of the lung, and increased drug loading of hydrophobic payloads (10,11 ). Spray drying has proven to be a suitable technology in the preparation of dry powder therapeutics ( 10), as it is capable of producing respirable microparticles for deep lung delivery with acceptable aerosol dispersion properties (4). The properties of dry powder particles such as particle size, particle shape, and surface morphology can be modified by controlling the production process, thus providing desirable particle characteristics ( 4,7).
Biocompatible, biodegradable polymers such as poly( E-caprolactone) (PCL) and poly(lactic-co-glycolic acid) (PLGA) have been used as dry powder formulation excipients to carry drug molecules, protect drugs from degradation, and impart sustained release to aerosol formulations (3). However, the PLGA and PCL delivery systems show significant bolus release of their payloads due to bulk erosion of the polymers, and it is difficult to control the polymer degradation rate and modulate their release profiles (12) (Acetalated dextran (Ac-Dex) is an acid-sensitive, biodegradable, biocompatible polymer prepared via a one-step reaction by reversibly modifying dextran with acetal groups. This modification reverses the solubility properties of dextran from hydrophilic to hydrophobic, making it possible to form polymeric particles using standard emulsion or nanoprecipitation techniques. In contrast to other commonly used polymers such as PLGA, polylactic acid, and PCL, Ac-Dex exhibits attractive properties suitable for the controlled release of the payloads. By controlling the reaction time during the formation of Ac-Dex, the ratio of cyclic acetal groups with a slower degradation rate to acyclic acetal groups with a faster degradation rate can be adjusted. As a result, the degradation rate of the resulting Ac-Dex can be tuned from hours to months to suit various applications. Moreover, the acid-sensitivity of Ac-Dex enables it to degrade faster in lower pH environments, such as lysosomes in macrophage or tumor cells, allowing for controlled release of drug. Furthermore, Ac-Dex degrades into neutral by-products, which avoids undesirable changes in the micro-environmental pH in the body. Finally, Ac-Dex offers the potential of targeted delivery, due to the dextran chains present that can be further modified with a variety of functional moieties, which can enhance the efficacy of therapeutic delivery to a targeted site (13)(14)(15).
Owing to aforementioned advantages, Ac-Dex has been widely applied in the formation of polymeric carriers for drug delivery. Porous Ac-Dex microparticles loaded with the chemotherapeutic camptothecin for pulmonary delivery were developed using emulsion techniques, which exhibited in a respirable fraction of 37% and experimental mass mean aerodynamic diameters from 5.3 -11.9 µm (2). Ac-Dex nanoparticle systems have also been investigated in the application of protein delivery for imrnunotherapy (13), gene delivery to phagocytic and non-phagocytic cells (11), tandem delivery of peptide and chemotherapeutic for controlled combination chemotherapy (1 ), delivery of the host-mediated compound AR-12 (Arno Therapeutics; formerly known as OSU-03012) for the treatment of Leishmania donovani (6), and the control of Salmonella infection (6). Both Ac-Dex nanoparticles and microparticles loaded with horseradish peroxidase have been evaluated to improve vaccine stability outside cold chain conditions (16).
Despite this work, there is no comprehensive study on using Ac-Dex as an excipient for dry powder aerosol formulations. In this study, we aimed to develop and characterize pulmonary delivery systems based on spray-dried Ac-Dex particles with capabilities of (a) delivering therapeutics to the deep lung, (b) targeting the particles to a particular location in the lungs, and ( c) releasing therapeutics in controlled rate. Previous studies have shown that: (a) aerodynamic diameter (da) determines the region of the lungs where particles will deposit, where particles with an da of 1 -5 µm tend to deposit in the deep lung region (10); (b) geometric size plays an important role in the cellular uptake of particles, where nanoscale particles (approximately 150 nm) tend to escape phagocytic uptake ( 17), while particles larger than 1 µm will suffer from macrophage clearance in the alveoli (18); and (c) the drug release rate of Ac-Dex particles can be tuned by modifying the synthesis time of the Ac-Dex polymer (2,15).
To prepare the engineered particle systems, two types of Ac-Dex with rapid or slow degradation rates were synthesized. Nanocomposite microparticle (nCmP) and microparticle (MP) systems were successfully formulated using both kinds of Ac-Dex as the excipient and curcumin (CUR) as the model drug. The nCmP were prepared by spray drying an aqueous suspension of CUR-loaded Ac-Dex nanoparticles (200 nm) and the MP were formulated by spray drying a solution of Ac-Dex and CUR in a solution of tetrahydrofuran (THF) and acetone. We hypothesize that upon pulmonary administration, the nCmP will deposit in the deep lung, decompose into free NP, and will facilitate the sustained release of drug to the targeted site, while the MP will remain the original size after deposition in the deep lung region. A schematic of particle preparation and design is shown in Figure 6.1. Overall, the goal of the described research was the initial development and physicochemical characterization of dry powder Ac-Dex aerosol particle systems with the potential for effective delivery of therapeutics in to the deep lung, targeting at desired pulmonary location, and continuous controlled release of a drug.

Materials
Dextran

Synthesis and NMR Analysis of Acetalated Dextran (Ac-Dex)
Ac-Dex was synthesized as described previously (14)  minutes to prepare Ac-Dex with a rapid degradation rate (Ac-Dex-5min) or for 3 hours to prepare Ac-Dex with a slower degradation rate (Ac-Dex-3h). The reaction was quenched with 1 mL of TEA. The reaction mixture was then precipitated in basic water (water and TEA, pH 9), vacuum filtered, and lyophilized (-50 °C, 0.023 mbar) for 24 hours to yield a solid product.
The cyclic-to-acyclic (CAC) ratio of acetal coverage and degrees of total acetal coverage per 100 glucose molecules was confirmed by 1 H NMR spectroscopy (Bruker 300 MHz NMR, MA). 10 mg of Ac-Dex was added to 700 µL of D 2 0 and was hydrolyzed with 30 µL of DCl prior to analysis. The hydrolysis of one cyclic acetal group produces one acetone molecule whereas one acyclic acetal produces one acetone and one methanol molecule each. Consequently, from the normalized integration of peaks related to acetone, methanol, and the carbon ring of dextran, the CAC ratio of acetal coverage and degrees of total acetal coverage per 100 glucoses were determined.

Formation of CUR-Loaded Ac-Dex Nanoparticles (CUR NP)
Curcumin-lo.aded nanoparticles (CUR NP) were prepared via an oil/water emulsion solvent evaporation using Ac-Dex-5min, Ac-Dex-3h, or a mixture of both types of Ac-Dex (5U % w/w). 49 mg of Ac-Dex and 1 mg of CUR were dissolved in 1 mL of DCM over an ice bath, establishing the organic phase. were coated with a thin film of a gold/palladium alloy using a BIO-RAD sputter coating system at 20 µA for 60 seconds under argon gas. Images were captured at 8 kV at various magnifications. The geometric mean diameter and standard deviation of the MP were measured digitally from SEM images using ImageJ software (19). Representative micrographs (5000x magnification) for each sample were analyzed by measuring the diameter of at least 100 particles.

Tapped Density Evaluation
The tapped density of the particles was measured as described previously with minor modifications (20). where d is the geometric diameter determined by ImageJ, p is the tapped density of the particles, and p * = 1 g/cm 3 , which is the reference density of solid polymer.

Drug Loading Analysis of CUR nCmP and CUR MP
Drug loading and encapsulation efficiency of CUR nCmP and CUR MP were determined via fluorescence spectroscopy (Biotek Cytation 3, Winooski, VT). All particle samples were dissolved in DMSO and were evaluated at 420 nm (excitation) and 520 nm (emission). The CUR drug loading and encapsulation efficiency (EE) of the particles were determined by the following equations: volume of DMSO and analyzed for CUR content via fluorescence spectroscopy using the same method described for drug loading. The release data was fitted to several commonly utilized drug release models (S.6.1) to study the mechanism of drug release of Ac-Dex particles. The coefficient of determination (R 2 ) was applied to test the applicability of the release models.

Differential Scanning Calorimetry (DSC)
The thermal phase transitions of the nCmP, MP, and their raw components were determined via DSC using a TA QlO DSC system (TA Instruments, New Castle, DE, USA) equipped with an automated computer-controlled TA instruments DSC refrigerated cooling system. 1 -3 mg of sample was weighed into Tzero ™ alodine-coated aluminum pans that were hermetically sealed. The sealed pans were placed into the DSC furnace along with an empty sealed reference pan. The heating range was 0 -200 °C at a heating rate of 10 °C/min.

Powder X-Ray Diffraction (PXRD)
The crystalline states of the nCmP, MP, and its raw components were examined by PXRD using a Rigaku Multiflex X-ray diffractometer (The Woodlands, TX) with a Cu Ka radiation source (40 kV, 44 mA). The samples were placed on a horizontal quartz glass sample holder (3 mm) prior to analysis. The scan range was 5 -60° in 28 with a step width of 0.1 and scan rate of 1 ° /min.

Karl Fischer Coulometric Titration
The water content of the nCmP and MP was quantified by Karl Fischer (KF) titration using a 737 KF coulometer (Metrohm, Riverview, FL). 5 mg of powder was dissolved in anhydrous methanol. The resulting solution was injected into the KF reaction cell filled with Hydranal® KF reagent and then the amount of water was analyzed. Pure solvent was also injected for use as a background sample.

In Vitro Aerosol Dispersion Performance with the Next Generation Impactor (NGI)
In vitro aerosol dispersion performance of the nCmP and MP was evaluated using

Statistical analysis
All measurements were performed in at least triplicate. Values are given in the form of mean ± standard deviation. The statistical significance of the results was determined using analysis of variance (ANOV A) and student's t-test. A p-value of <0.05 was considered statistically significant.

NMR Analysis of Ac-Dex
Successful synthesis of Ac-Dex was confirmed by 1 H NMR. Ac-Dex-5min exhibited 61.2% cyclic acetal coverage (CAC) and 71.6% total conversion of -OH groups, while Ac-Dex-3h exhibited 82.5% CAC and 80.0% total conversion of -OH groups. The Ac-Dex with longer synthesis time (Ac-Dex-3h) exhibited a higher CAC, which was in accordance with previous studies. An increase in CAC is known to decrease polymer degradation and ultimately, the drug release rate, due to the slower degradation of the cyclic acetal groups on the Ac-Dex backbone (13,14). Ac-Dex-3h also showed a higher total conversion of -OH groups, which could be a result of elongated reaction time. This higher total acetal coverage is favorable in the enhancement of the stability of the PV A coating of nanoparticles (data not shown), thus ensuring small particle size and narrow size distribution.

.1.2 Dynamic Light Scattering (DLS) Analysis of Original and Redispersed CUR NP
Average nanoparticle (NP) size, size distribution/polydispersion index (PDI), and zeta potential are shown in Table 6.1. No significant changes in NP size, PDI, or zeta potential was found between the original and redispersed NP (p < 0.05), indicating that the CUR NP maintained their desirable properties after redispersion. The original and redispersed NP exhibited an average diameter of approximately 200 nm, which is in the desirable range to avoid macrophage clearance and mucus entrapment (18). The low PDI value denotes a narrow size distribution of the NP, and the slightly negatively charged surface of nanoparticles, as measured by zeta potential, is desirable in order to reduce the interactions with negatively charged mucin fibers in the airway mucus (22). According to our preliminary experiments (data not shown), a low total conversion of -OH groups on the Ac-Dex results in NP with larger sizes and PDI due to NP agglomeration. This phenomenon could be a result of the lessened hydrophobicity of Ac-Dex with fewer -OH groups converted to acetal groups, which leads to insufficient absorption of PV A on the NP surface. However, the Ac-Dex in this study was prepared to produce NP with small sizes and low PDI, as the total conversion of -OH groups was kept in a higher range to prevent NP agglomeration.  which determines the geometric size of nCmP during spray drying. As the drying process proceeds, the remaining solvent evaporates from the droplet center, resulting in hollow particles that tend to shrink (23 , 24).
CUR MP were collapsed spheres with smooth surfaces as seen in Figures 6.2D-F.
Differing the Ac-Dex composition of particles had no impact on particle morphology.
The geometric diameters ( dg) of the CUR nCmP and MP system are shown in Table 6.2.
All of the MP dg were approximately 1 µm in size, which is reported to make the particles vulnerable to macrophage uptake (25). On the contrary, the nCmP system can escape macrophage clearance as they dissociate into NP upon reaching the deep lung. relatively low compared with the raw materials ( ~ 1 g/cm 3 ), which can be attributed to the wrinkled morphology and hollow structures of the particle systems. It has been reported that particles > 1 µm in diameter and having greater density will get deposited in the lung by sedimentation (26). Therefore, the increased density of CUR nCmP system as compared to MP will enhance their rate of deposition into the deep lung.

CUR Loading and in Vitro Release of CUR
CUR was successfully encapsulated into both the nCmP and MP systems as seen in  Figure 6.3 as the percentage of cumulative drug released over time. As shown in Table C.1, all particle systems exhibited shorter release duration and increased release of CUR (p < 0.05) at acidic pH (except nCmP-5min), which is in accordance with previous studies (2,27). The pH-sensitive release profiles suggest that the drug will be released at significantly higher rates once the carrier particles reaches an acidic environment. This can allow the Ac-Dex particles the ability to provide the controlled Time (hours) 20 40 Time (hours) -<>nCmP-3h OnCmP-5min """"nCmP-h In addition, the nCmP systems exhibited faster release than MP systems, which is likely due to the larger surface area available in the nano-sized delivery systems. Upon reaching an aqueous environment, the nCmP dissociate into nanoparticles with large surface area and a PVA coating that facilitates particle dispersity, while the MP may agglomerate due to their highly hydrophobic surfaces. As a result, the nCmP systems undergo faster polymer degradation and drug diffusion, resulting in a faster release of payloads than MP at both acidic and physiologic pH. Particles made of Ac-Dex-3h exhibited slower release rates than those comprised of Ac-Dex-5min, indicating that the drug release rate can be controlled by the polymer reaction time. At pH 7.4, particles made of Ac-Dex-h exhibited a drug release rate between Ac-Dex-5 min and Ac-Dex-3h, suggesting that the ratio of different types of Ac-Dex can also act as an important factor in adjusting the release profiles of particle systems. However, the drug release rates of the Ac-Dex particles at pH 5 did not follow this trend, which could be explained by the following two mechanisms: (1) The release profile of Ac-Dex particles is polymer degradation/erosion controlled, and the decomposition of the Ac-Dex matrix is greatly impacted by buffer pH. A study based on design of experiment will be conducted in our further research to illustrate the influence of potential factors (reaction time, polymer ratio, particle size, and pH) on the drug release profile of particle systems. (2) The release profile of Ac-Dex particles is both polymer degradation/erosion and drug diffusion controlled. In previous studies, the drug release of Ac-Dex particles is believed to be associated with Ac-Dex degradation (2,14,15). However, Ac-Dex degradation may cause surface erosion of particles, formation of large pores in the particles that facilitate drug diffusion, or both at the same time. As a result, the drug release profile could be controlled by drug diffusion through water-filled pores (diffusion controlled), polymer erosion on the particle surface (erosion controlled), or both drug diffusion and surface erosion (diffusion and erosion controlled), respectively (13).
In order to further illustrate the mechanism of drug release of Ac-Dex particles, we fit the release data to several commonly utilized drug release models, including: (1) a first order model and (2) Table C.2, and the modified Higuchi and Baker-Lonsdale models exhibited higher R 2 compared with other models, indicating that the drug release profile of Ac-Dex particles in both pH buffers was due primarily to drug diffusion. For Baker's model (28) that describes a degradation and diffusion process, the optimal coefficient k was 0, making the Baker's model the same with Higuchi model.
Considering that the degradation of Ac-Dex was observable during the release test, the release profile of Ac-Dex particles can be explained by the mechanism of drug diffusion through water-filled pores (32). In this process, water was absorbed by Ac-Dex particles and filled in the pores of the polymer matrix, through which the drug diffused into the buffer. As polymer degraded, both pore size and number increased, leading enhanced drug release. Therefore, the reaction time of Ac-Dex affected the drug release rate significantly by controlling the formation of pores of particle matrix but not polymer degradation on the surface. Meanwhile, the water absorption into the particles may also influence the drug release rate, which can be supported by the fact that Ac-Dex-3h had a higher ratio of total hydrophobic acetalated group conversion than Ac-Dex-5min. The fitted release curves using modified Higuchi model are shown in Figure S6.1 along with the original data points. The model was modified to fit the burst release at time 0 of some of the particle systems, which is attributed to the CUR initially available on the surface of particles. The nCmP systems exhibited a higher release at time 0 because that the nanoparticle suspension was sonicated before spray drying to form a uniform dispersion, which may lead to CUR release in to the suspension.

Karl Fischer Titration
The residual water content of CUR nCmP and MP is shown in Table 6.2. The water content of nCmP system was approximately 8%, while that of MP system was approximately 6%. The lower water content of MP samples is likely due to the absence of water during the closed mode spray drying process. All the particle systems showed acceptable water content. In general, reducing the water content in inhalable dry powders can significantly improve their dispersion properties and enhance the stability of the powders during storage (33,34). Correspondingly, low water content in inhalable dry powders is highly favorable for efficient dry powder aerosolization and effective particle delivery (34, 35). Figure 6.4 shows DSC thermograms of the raw materials used in particle preparation and the final CUR nCmP and CUR MP systems. Both the raw Ac-Dex-5min
The peaks were broad because of the wide size distribution of Ac-Dex polymer crystallites. The raw CUR powder displayed a sharp peak at 180 °C corresponding to the melting point of crystalline CUR, which is consistent with earlier studies (36). None of the CUR nCmP systems exhibited a peak corresponding to Ac-Dex melting, which indicates that the Ac-Dex was turned amorphous by rapid precipitation during NP formation. All of the CUR MP systems exhibited a broad peak near 160 °C, which corresponds to the melting of Ac-Dex. This phase transition shifted to the lower temperature range, indicating an increase in the amorphous state of Ac-Dex after the spray drying process. None of the CUR nCmP or CUR MP displayed melting peaks that would correspond to raw CUR, indicating that the crystalline CUR was converted to an amorphous phase during particle preparation. Temperature ( 0 C) Figure 6.4. Representative differential scanning calorimetry (DSC) thermograms of raw curcumin (CUR), raw acetalated dextran-5min (Ac-Dex-5min), raw acetalated dextran-3h (Ac-Dex-3h), CUR nCmP-5min, CUR nCmP-h, CUR nCmP-3h, CUR MP-5min, CUR MP-h, and CUR MP-3h.

Powder X-ray Diffraction (PXRD)
X-ray diffraction diffractograms of the raw materials, CUR nCmP, and CUR MP are shown in Figure 6.5. Strong peaks were present for raw CUR indicating that it is in a crystalline form prior to spray drying. No peaks were present for either raw Ac-Dex samples, suggesting an irregular distribution of Ac-Dex crystallites. The absence of diffraction peaks from Ac-Dex is quite different from commercialized polymers such as PLGA, which exhibits strong XRD characterization peaks (34). This phenomenon is likely because the Ac-Dex was collected by rapid precipitation in water, which prevents the formation of large polymer crystallites. XRD diffractograms of the CUR nCmP and CUR MP were absent in any diffraction peaks corresponding to raw CUR, indicating that the crystalline CUR was converted to an amorphous phase in the spray drying or emulsion evaporation process. The amorphization of raw payloads into the particle matrix is suitable for the improvement in solubility of hydrophobic drugs in physiological conditions. The results obtained from the XRD diffractograms confirmed those from DSC thermograms, which proves that raw CUR was converted to amorphous forms during the particle manufacturing process.

In Vitro Aerosol Dispersion Performance Using Next Generation Impactor (NGI)
In vitro aerosol dispersion performance properties of the nCmP were evaluated using a Next Generation Impactor™ coupled with a human DPI device (Figure 6.6 and Figure 6. 7). The results indicated that the formulated nCmP and MP are favorable for efficient dry powder aerosolization and effective targeted pulmonary delivery. The mass mean aerodynamic diameter (MMAD) value of all particle systems were approximately 2 µm , while the geometric standard deviation (GSD) value was 2 -3 µm. The MMAD value is within the range of 1 -5 µm that is required for predominant deposition of particles into the deep lung region (3 7), which would be desirable to deliver therapeutics for the treatment of both local pulmonary and systematic diseases through the lung. The theoretical MMAD (MMADr, shown in Table 2), calculated using the geometric diameter and tapped density, was lower than the experimental results (MMAD 1 ). This could be due to particle agglomeration, which increased the geometric size of the dry powder. The GSD values were within the range of those previously reported and the respirable fraction (RF), fine particle fraction (FPF), and emitted dose (ED) values were all higher than reports from similar systems (10,37,38). The formulated Ac-Dex particle systems are expected to achieve an improved therapeutic effect with a reduced amount of payloads by effectively delivering the drug into deep lung region.

Conclusions
Two types of pulmonary delivery systems were successfully formulated using Ac-Dex with two different degradation rates. MP were wrinkled spheres (approximately 1 µm) with smooth surfaces, while nCmP were 1.5 µm with wrinkled surfaces that dissociated into 200 nm nanoparticles upon being reconstituted in water. The variation in the drug release rates from the Ac-Dex particles were determined by the Ac-Dex reaction time, ratio of two types of Ac-Dex, and the particle size, which could be easily tuned during the manufacturing process. The pH value of environment also had a significant influence on the release profiles, allowing the Ac-Dex particles to release the payload in a controlled fashion. All nCmP and MP systems exhibited desirable properties as dry powder inhalation therapeutics, including small aerodynamic diameters, which is suitable for deep lung delivery, low water content, which is favorable for particle storage, and amorphization of a crystalline payload, which improves the efficiency of drug dissolution. Overall, the engineered Ac-Dex aerosol particle systems have the potential for targeted delivery of therapeutics in to the deep lung. Meanwhile, these dry powder formulations also provide more convenient administration, more flexible storage conditions, and lower risk of contamination in the device.

Acknowledgements
The authors gratefully acknowledge financial support from an Institutional

CHAPTER 7
Conclusions and Future Work

Conclusions
This dissertation focuses on the development, characterization, and optimization of dry powder aerosol nanocomposite microparticle systems for the treatment of pulmonary diseases. Azithromycin and rapamycin were encapsulated in Ac-Dex nanoparticles for potential application in the treatment of cystic fibrosis-related lung infections. These nanoparticles were coated with VP5k to ensure effective dispersion in and facilitate penetrating through viscous mucus caused by cystic fibrosis. Nanoparticles coated with PV A were able to maintain a favorable dispersion and can be used to deliver tacrolimus for the treatment of pulmonary arterial hypertension.
In the production of nanoscale carriers, Ac-Dex with different degradation rates were applied to form polymer matrices with tunable release profiles. The drug release rates of the Ac-Dex particles were determined by the Ac-Dex reaction time, ratio of two types of Ac-Dex, and particle size, which could be easily tuned during the manufacturing process. The pH value of environment also had a significant influence on the release profiles, allowing the Ac-Dex particles to release the payload in a controlled fashion.
Nanocomposite microparticles were formulated using spray drying a suspension of nanoparticles. The optimal spray drying conditions were identified to prepare nCmP systems with favorable properties including: small aerodynamic diameter, effective nanoparticle redispersion, high drug loading, and low water content. Meanwhile, the safe range of spray drying parameters were discussed to ensure the stability of therapeutics and formulations.
This is the first time that the dry powder aerosol nanocomposite microparticle systems have been comprehensively studied for the treatment of pulmonary diseases. The dissertation presents a strategy to develop nanoscale carriers with capability of protecting of drugs from degradation, enhancing of the solubility of drugs, releasing payloads at controlled rate, and overcoming physiological barriers. In addition, the optimal conditions to formulate microscale carriers with advantages including targeted deposition in the airways, high drug loading, enhanced stability, and maintenance of the favorable properties of embedded nanoparticles were elucidated.

Future Work
This dissertation provides the first comprehensive study on developing and optimizing dry powder aerosol nanocomposite microparticle systems for the treatment of pulmonary diseases. Many opportunities exist to conduct further research based on this delivery system, which may include, but are not limited to: • Investigation of the safety and antibacterial efficacy of azithromycin-loaded Acdex nanoparticle with VP5k coating for the treatment of cystic fibrosis-related infections.
• Development of Ac-dex nanoparticles with vanous coating materials and investigation of the safety of these nanoparticles, which could be used to targeted delivery of therapeutics.
• Optimization of the spray drying conditions to formulate nCmP encapsulated with Ac-dex nanoparticles with various coating materials for pulmonary delivery or systemic delivery of therapeutics through the lungs.
• Investigation of the safety and efficacy of tacrolimus-loaded nCmP for the treatment of pulmonary arterial hypertension by in vivo study using an animal model.
• Development of nCmP comprised of hydrophilic encapsulated nanoparticles and a hydrophobic excipient of microscale carriers, which could be applied to provide sustained release of hydrophilic drugs with short half-lives.

D.1 Drug Release Model Description
The drug release data of the particle systems was fitted to several models (equations shown below) to aid in the determination of the type of release the systems experienced. For the models that can be linearized (all models except for Baker's model), the coefficient of determination (R 2 ) was utilized to determine the applicability of the release models. For Baker's model, Microsoft Excel with Solver add-in was applied to determine the parameters that minimize the sum of squares of the residues of the model.
The models, equations, and their corresponding parameters for are as follows: First order model: K log Mi= logMo+--t 2.303 where Mt is the amount of drug released at time t, Mo is the initial amount of drug in the solution, and K is the first order release constant.
Weibull model: where m is the accumulated fraction of the drug released at time t, a is the scale parameter, which defines the time scale of the process, Ti is the location parameter, which represents the lag time before the onset of the dissolution or release process, and b is the shape parameter, which characterizes the shape of release curve.
Higuchi model: where Mt is the amount of drug released at time t, K is the Higuchi dissolution constant, and b is the amount of drug released at time 0.
Hixson-Crowell model: where K is a constant equal to k 0 /C 0 ao, where k 0 is the erosion rate constant, Co is the initial concentration of drug in the matrix, and ao is the initial radius for particles. m is the accumulated fraction of the drug released at time t.
where M 1 is the amount of drug released in time t, Po is the drug permeability, A is the total area of the particle, Co is the drug concentration at the initial time, and k is the firstorder rate constant of bond cleavage of the polymer carrier.