Novel delivery systems for controlled release of bacterial therapeutics

Abstract

As more is learned about the benefits of microbes, their potential to prevent and treat disease is expanding. Microbial therapeutics are less burdensome and costly to produce than traditional molecular drugs, often with superior efficacy. Yet, as with most medicines, controlled dosing and delivery to the area of need remain key challenges for microbes. Enabling higher precision in living biotherapeutic delivery would relieve the burden of strict dosing regimens and adverse effects for patients, and such delivery and release are well-established for traditional drugs. Advances in materials to control small-molecule delivery are expected to translate to microbes, enabling similar control with equivalent benefits. In this perspective, recent advances in living biotherapeutics are discussed within the context of new methods for their controlled release. The integration of these advances provides a roadmap for the design, synthesis, and analysis of controlled microbial therapeutic delivery systems.

Bacteria are promising next-generation therapeutics

Despite the often-negative connotations of bacteria as disease agents,[1] microbes are essential for our survival. In fact, there are more microbes in and on us than mammalian cells.[2] Microbial communities in the gut,[3] female reproductive system,[4] and skin [5] are necessary for our health but must be in homeostasis to provide benefits. Dysregulation can lead to disease,[6] which is often then treated with small-molecule therapeutics that either intentionally (e.g., antibiotics) or unintentionally further disrupt microbiomes. As we learn more about the importance of microbes for our health, there is growing interest in their use to prevent and treat disease as living biotherapeutics (LBPs).

Importantly, the role of microbes in disease prevention has been known for centuries.[7] Smallpox inoculation was mandated for soldiers during the Revolutionary War, and the bacterial vaccine Mycobacterium Bacillus Calmette-Guerin (BCG) has been used for tuberculosis prevention for over 100 years.[8] Microbes like BCG also have immunogenic activity, enabling their use as inexpensive immuno-oncogenics.[9] Despite the long history of LBP use, their formulation has remained rudimentary. As technologies to study and engineer bacteria improve, more microbes are anticipated to be developed as LBPs, and recent work in these areas has been well-reviewed.[10,11] Thus, as the complexity of these therapeutics increases, we expect that control over their delivery and release will be increasingly important, as has been shown for small-molecule therapeutics. Controlled delivery of molecular medicines has been shown to improve compliance and efficacy, while decreasing uncomfortable side effects. Here, we discuss improvements in Recent advances in small-molecule drug delivery are expected to translate to LBPs, providing them with similar advantages (Figure 1).

Figure 1.

(Key Figure): Microbial delivery roadmap. By following the roadmap developed for the controlled delivery of small-molecule therapeutics, including formulation for maintaining efficacy (through maintaining viability for microbes), selection of the proper materials for controlled release, and evaluation of the release, similar advantages to small-molecule delivery are expected for bacteria.

Emerging Bacterial Therapeutics

As our understanding of microbe-human interactions and non-model organism engineering methods improve, the diversity of diseases addressed with LBPs is rapidly growing (Figure 2). Recent advances in microbial vaccine development have mainly centered on strain engineering strains to produce antigens; Lactobacillus casei have been engineered to produce human papillomavirus 16 (HPV16) E7, a cervical intraepithelial neoplasia and cancer antigen [12], as well as a pancreatic necrosis virus antigen [13]. Additionally, Lactobacillus gasseri can express antigens against Streptococcus group A infections [14].

Figure 2:

Recent developments in microbial therapeutic development. Microbial therapeutics are increasingly used to prevent and treat disease, including as whole-cell vaccines, as probiotics to support health, and as native or engineered living biotherapeutic microbes.

Microbes have also long been used as probiotic supplements to support health and wellness in non-regulated products. Novel strains are continually introduced, including the recent Akkermansia muciniphilia. This strain has anti-inflammatory properties in mice with chronic colitis, showing reduction in interleukin-8 and other pro-inflammatory cytokines following treatment [15]. Similarly, a multi-strain probiotic was recently developed to treat asymptomatic inflammatory bowel disease (IBD)[16]. Though patient experience was unchanged, inflammatory markers (such as fecal calprotectin levels) indicate potential prevention of clinical relapse. However, such products remain challenging to evaluate because of their variability and lack of regulation, leading to efforts focused on clinically-approved LBPs.

Food and Drug Administration (FDA)-approved microbial therapeutics remain in their nascency, though they can target therapeutic molecule delivery to disease microenvironments, minimizing off-target effects. One of the best-known microbial therapeutics is the fecal transplant, which involves the direct transplantation of fecal matter or microbes into a patient, mainly for the treatment of Clostridium difficile infections. Despite promising results, fecal transplants have had limited use because of their inconsistency and the potential to infect the patient with illness-causing microbes. Efforts to develop LBPs to treat C. difficile have therefore focused on individual strains rather than human-derived consortia. Recent examples include clinically-validated MRx1234, Blautia hydrogenotrophica to treat IBS; the multi-strain RBx2660 for recurrent C. difficile infection; and LACTIN-V, a Lactobacillus crispatus-based bacterial vaginosis treatment [17–19]. The critical limitation with these examples, though, remains their limited application to treat microbiome dysbiosis or infection.

Beyond dysbiosis, recent advances in LBPs include their use as cancer therapeutics and inflammation treatments. A 2019 study showed that an 11-strain consortium induced CD8 T cells in the intestine, propagating anti-cancer immunity [20]. Salmonella typhimurium have also been engineered to express and deliver L-asparaginase, a tumoricidal protein, directly to a tumor, as Salmonella naturally aggregate in tumor cells [21]. In addition, Lactococcus lactis can express interleukin-10 to treat Crohn’s disease [14]. These recent examples demonstrate the importance of leveraging the natural diversity and engineerability of bacteria. Despite these advances, new formulations are needed to maintain cell viability, standardize production, and consistently deliver bacteria [14,22,23].

Critical challenges must be overcome to enable the universal application of LBPs. For example, BCG (developed to protect against tuberculosis) can also prevent leprosy and pertussis (whooping cough) [24,25]. Yet, its efficacy varies based on geography, prior exposures, and non-tuberculosis mycobacterium (NTM) infections [26,27]. Importantly, simple formulation changes impact its efficacy; recent studies suggest that intranasal, oral, or mucosal routes improve antibody production as compared to conventional transdermal delivery [28,29]. Thus, even established LBPs can benefit from reformulation. BCG is also a critical immuno-oncogenic in the treatment of bladder cancer. In fact, upon regimen completion, it has a 95.4% cure rate [30,31], but patience compliance remains low due to burdensome and painful administration with fewer than 20% of patients completing treatment. The patient discomfort and side effects experienced render BCG ripe for innovation in formulation and delivery. As LBPs, bacteria are uniquely advantageous but will not enter mainstream medical treatment without innovation in their formulation for controlled delivery and release. These strategies are well-established for small-molecule therapeutics, and guidance from these technologies is expected to enable equivalent control over LBPs.

Roadmap to Controlled Bacterial Delivery

To-date, LBPs have generally incorporated minimal formulation; as an example, BCG is simply freeze-dried for storage and transport, followed by reconstitution in saline for delivery. Additionally, platforms developed for microbial encapsulation are mainly for oral delivery to the gastrointestinal (GI) tract and are often too specific for use beyond GI delivery. One material for GI delivery with applications beyond is alginate-chitosan particles, which provide pH-responsive release based on chitosan protonation [32]. Such microparticles have also been used for BCG encapsulation to enable intranasal lung delivery [33].

Small-molecule drug delivery platforms have not had the equivalent limitations; one of many delivery systems can be selected based on the mode of delivery, the desired release profile, and the material-therapeutic interactions, which have been thoroughly reviewed (Figure 3).[34–38] For microbial delivery beyond niche applications, similar design considerations are needed to enable the selection of a delivery system. These include material biocompatibility, stability under environmental stressors, chemistry for controlled release, and macroscale delivery. Further, materials must both protect microbes and maintain their viability. Polymers best fits these criteria (Table 1), especially biopolymers (e.g., chitosan, alginate, and hyaluronic acid) and polyethylene glycol (PEG)-containing block copolymers [39]. As an example, pullulan-poly(lactic-co-glycolic acid) (PLGA) combinations have been shown to protect L. rhamnosus [40].

Figure 3:

Design criteria for materials to deliver microbe-based therapeutics. Criteria range from the chemistry used to release therapeutics to the target organ system. At the molecular level, both the chemistry used to control the release of the active agent and the selection of polymer matrix are critical. These materials can then be formulated into hydrogels, fibers, or spheres for delivery with input based on the target organ for delivery. Together, these factors enable tunable delivery of therapeutics to different areas of the body.

Table 1.

Summary of recent microbial encapsulation and delivery studies.

Bacteria	Target site/purpose	Material platform	Release mechanism	Citation
Bacillus subtilis natto	Intestines	Carboxymethyl cellulose + chitosan + alginate	pH-responsive swelling and ionic crosslinks	[44]
Ligilactobacillus salivarius	Gut	Chitosan-alginate layer-by-layer encapsulation	pH-responsive ionic crosslinks	[47]
BCG	Lungs	Chitosan and chitosan-alginate beads	pH responsive ionic crosslinks	[32]
E. coli		Alginate + PEO + surfactant electrospun fibers	n/a (storage capabilities only)	[41]
L. rhamnosus	Intestines	Electrospun pullulan and PLGA fibers	Hydrolysis-degradable PLGA outer layers	[40]
L. rhamnosus	Wound healing	Hyaluronic acid with polysaccharides PF127 and FD hydrogel	Degradable schiff-base crosslinks	[46]
L. rhamnosus		Hyaluronic acid hydrogel	Redox-responsive degradable disulfide crosslinks	[45]
L. plantarum	Intestines	Chitosan-coated agar-gelatin particles	pH-responsive release in intestines	[66]
L. casei	Gut	PVA air-dried films, with additives (MRS broth, glycerol, NaCMC)	Burst release upon rehydration (in conjunction with enteric or other delivery capsule)	[43]
B. adolescentis (anaerobic)	Gut	PVA + skim milk air-dried films	n/a (encapsulation evaluated only)	[67]
L. casei	Gut	Alginate microspheres with calcium chloride crosslinks	Simulated intestinal fluid breaks ion crosslinks	[42]

Following selection of a polymer, the material must be assembled into macroscopic structures for delivery. Most often, hydrogels are used because their properties can be tuned based on polymer density and pH-responsive pendant groups, but electrospun fibers and polymer films have also been shown to protect microbes. Electrospun fibers generated from alginate, polyethylene oxide (PEO), and surfactant have been shown to protect microbes for delivery [41], and polyvinyl alcohol-based films protected encapsulated anaerobic L. casei during storage [42,43]. Further, these macroscopic assemblies enable control over release kinetics; for the L. casei-based films, carboxymethyl cellulose could be added to control the film thickness and therefore the rate of release.

Controlled release for small-molecule therapeutics is generally enabled by matrix degradation, but for microbials, diffusion within the matrix can also play a significant role. Most work on bacterial release has focused on tuning the timing of a burst of microbes, which can be controlled by incorporating polysaccharides such as sodium carboxymethylcellulose (NaCMC) to impede diffusion and delay degradation [43,44]. Similarly, degradation can be controlled by tuning the crosslinking mechanism in hydrogels using redox-responsive thiol crosslinks [45], dynamic Schiff-base crosslinks [46], or modifying the ionic crosslinking density of chitosan and/or alginate layers [42,47]. Of the limited number of studies on controlled microbial release, most focus on the timescale of hours or days, though longer release timelines are needed for many applications. We anticipate that polymers used for the extended release of small molecules, such as tunable PLGA-PEG triblock copolymers, will enable sustained microbial delivery.

Though there are few non-polymer materials that have successfully enabled LBP delivery, one such material, metal-phenolic networks (MPNs), has recently been applied to protect microbes from processing and delivery stresses with significant success.[48,49] These materials were shown to protect anaerobic B. thetaiotaomicron from oxygen exposure and low-pH conditions as well as B. subtilis from processing stressors. Combining emerging materials like MPNs with polymeric materials are anticipated to enable both protection and controlled release of microbes. As more microbes are developed for therapeutic applications, materials that enable general protection and controlled delivery will become increasingly important, as will a better understanding of their mechanisms of protection.

4. Characterization and release assays

Though analytical methods for small-molecule drug characterization are well-established, evaluating the protective efficacy and release profiles of formulated microbes presents unique challenges. Bacteria are living organisms; their viability and activity are tied to their efficacy, making both determining the total number of cells released and the fraction that are viable following release critical. Thus, assays to evaluate LBPs must extend beyond conventional drug analysis to include growth, gene expression, and metabolic activity (Figure 4). The selection of assays to evaluate LBPs is dependent on the information they provide, their relative cost, and their ease.

Figure 4:

Methods to measure bacteria release and viability of released cells. Complementary strategies can be used based on the budget for analysis balanced against the need for quantitative readout. Methods such as spot assays for colony forming unit (CFU) counting and growth curves are relatively inexpensive but can also be less precise than other methods. Colorimetric assays provide a balance between quantitative results with inexpensive measurement, while flow cytometry, fluorescence microscopy, and transcriptomics can be costly but also provide additional information about the microbes.

Growth assays including serial dilutions and optical density at 600 nm (OD₆₀₀) measurements enable determination of the viability of released microbes and are straightforward and inexpensive. Unfortunately, while these approaches are experimentally simple and inexpensive, the (comparatively) long incubation times make rapid characterization impossible, especially for slower-growing bacteria such as BCG [50] [51]. Characterization of nucleic acids is similarly popular to analyze microbial release. Quantitative Polymerase Chain Reaction (qPCR) [52] measures the amplification rate of specific nucleic acid sequences to quantify the original amount of that sequence in a sample. This technique has been used to quantify microbes [53,54] but must be coupled with a method to determine viability. Thus, both of these well-established methods provide critical information but must be used in combination to obtain necessary information.

Microscopy, in contrast to growth or nucleic acid-based assays, enables immediate visualization of released bacteria with morphological information, which can inform materials design to better protect the cells, but it is comparatively low throughput. However, microscopy can be coupled with fluorescent stains such as live/dead or gram staining[55,56], cell-expressed fluorescent proteins[57], or enzymatically-activated fluorophores[58], to quantify the total number of microbes released and the fraction that are viable. Recent innovations in imaging can further provide accurate real-time bacterial quantification; as an example, Fluid-Screen, which is based on real-time dielectrophoresis, captures bacteria from dilute samples, fluorescently stains them, and performs direct on-chip quantification [59]. A method that combines the speed of imaging with the throughput of viability assays is flow cytometry, which can be paired with viability staining to quantify released cells and determine the fraction that are viable [60]. However, flow cytometry and fluorescence microscopy often require trained personnel and costly equipment, which can limit access. Thus, recent innovations often focus on autonomous or lower-cost data collection methods. One recent example to utilize less costly equipment is a fiber-based, portable spectroscopic device (optrode) was developed to measure live/dead bacterial staining [61] [62] to enable rapid, on-site enumeration of live and dead bacteria.

Colorimetric assays are similarly growing in popularity, as they can provide critical information on microbial viability and metabolic activity as well as on the pathway-specific activity of the microbes, which can be important to establish therapeutic efficacy. Colorimetric assays can also be used to monitor the internal redox balance of microbes. As an example, tetrazolium can be used as a redox indicator, as it can be reduced by NADH in the microbe to generate formazan [63]. Similarly, adenosine triphosphate (ATP) assays can provide quantitative insight into bacterial metabolism by using glycerol phosphorylation to generate a purple dye correlated to the ATP content [63]. Importantly, both NADH and ATP assays have both been adapted to study in situ viability of bacteria in hydrogels, expanding our analytical capabilities beyond released bacteria [64]. Many of these assays, though, lack reproducibility. Thus, recent efforts have focused on improving workflows to enable quantitative measurements from previously-qualitative assays. We recently improved an assay for monooxygenase activity, enabling substrate regioselectivity determination and quantification for the first time [65]. We anticipate that the ease of colorimetric assays will lead to their development to monitor more sophisticated cellular processes.

Conclusion and Future Outlook

Small-molecule therapeutics have enabled us to lead long, healthy lives, but off-target effects diminish patient quality of life, and overuse of drugs such as chemotherapeutics and antibiotics had led to resistances against them. As we reach the limits of what can be accomplished with small-molecule medicines, emergent therapeutics including microbes have become increasingly important for disease prevention and treatment. Fortunately, years of development to enable the controlled delivery of small-molecule drugs have yielded materials advances that are expected to translate to emerging classes of therapeutics, namely microbes.

To-date, microbial products for human health have focused on maintaining homeostasis (probiotics), preventing disease (vaccines), or treating infection (e.g., for C. difficile infections). However, the scope of microbial therapeutics is expected to expand as we learn more about their benefits and develop new technologies for their engineering. Along with these advances come additional knowledge gaps regarding potential interactions between small-molecule drugs and microbes, regulatory barriers to the implementation of LBPs, and, critically, how to best deliver them (see outstanding questions). As can be seen from recent work in microbial formulation, the development of materials for their controlled delivery and release has lagged behind the development of the microbes themselves. Though we have focused here on the potential of existing materials for small-molecule drug delivery to translate to microbial delivery, new polymeric materials are expected to be critical for the implementation of LBPs. Further, the rules for materials design and analysis outlined here are expected to translate to the development of novel materials for maintaining microbial viability while controlling their delivery.

Glossary

LBP

A living biotherapeutic product is a medicinal product where the active ingredient is a live microbe

BCG

Bacillus Calmette-Guerin is an early example of a microbial therapeutic and has found use as a vaccine as well as an immuno-oncogenic to treat certain forms of cancer

PEG

Polyethylene glycol is a critical bio-compatible polymer used in many drug formulations because of its demonstrated safety