This article was originally published on drugdeliveryleader.com.

A conversation between TAXIS Pharmaceuticals’ Chief Scientific Officer, Ajit Parhi; and Life Science Connect’s Michelle Raley

Antibiotic resistance is not only a matter of discovering new therapeutic agents. It also requires delivering those agents to the right pathogen, cellular target, and site of infection while limiting unnecessary exposure elsewhere. In this Q&A, Life Science Connect’s Michelle Raley caught up with Ajit Parhi, chief scientific officer of TAXIS Pharmaceuticals, to discuss strategies and to share his perspectives.

When we talk about “targeted” antibacterial therapy, how do you design a molecule so that it preferentially affects a pathogenic bacterium while minimizing exposure and effects on beneficial organisms? What PK/PD parameters are most important, and how do you actually prove that your drug is being selective in vivo — not just that it has selective antibacterial activity in vitro?

 

The most logical approach to targeted antibacterial therapy is target-based selectivity — identifying a target that is essential to pathogenic bacteria but absent from beneficial organisms.

Since the vast majority of beneficial bacteria reside in the gastrointestinal tract, particularly the colon, an important strategy is designing oral antibiotics that move through the digestive tract as quickly as possible and minimize exposure to the microbiome.

To target systemic infections, we design molecules with specific physicochemical properties to promote absorption. We generally target a sweet spot logP, approximately between 1 and 3 and aim for a relatively low polar surface area, preferably between 60 and 90 Ų. We also try to reduce the number of hydrogen-bond donors and acceptors and use prodrug strategies when appropriate to mask charged functional groups and facilitate absorption.

We incorporate counter-screening against beneficial organisms during lead optimization, rather than treating it as a late-stage checkpoint. We track the ratio of pathogen minimum inhibitory concentration (MIC) to commensal MIC as a live structure-activity relationship (SAR) parameter alongside potency and physicochemical properties.

When it comes to PK/PD, traditional blood and plasma measurements are not sufficient for targeted therapy. We need a dual-compartment PK/PD framework that considers both the site of infection and the gut lumen.

One important parameter is the compartmental free-exposure ratio — the ratio of free drug exposure at the site of infection to free drug exposure in the gut lumen.

Another important parameter is the clearance pathway ratio, comparing renal clearance with biliary clearance. For systemic therapies, we want to minimize biliary clearance and maximize renal clearance. If a drug is processed by the liver and excreted through bile, it can enter the intestinal tract. Renal elimination, by contrast, sends the drug through the urine and largely bypasses the digestive tract.

We also consider the fecal fraction unbound. A low fecal fraction unbound means that any drug reaching the colon is highly bound to fecal material, potentially limiting its biological activity and helping minimize exposure of beneficial bacteria.

In vitro MIC selectivity is important, but it does not by itself demonstrate in vivo selectivity. MIC testing is performed in relatively static monoculture conditions and does not capture the complex exposure conditions that occur in a living organism.

One of the studies we use is a resistance challenge study, which provides a more functional assessment of whether an antibiotic has preserved the protective capacity of the microbiome.

In this type of study, the animals are first treated with antibiotics. After treatment is completed, they are challenged with a surrogate pathogen, such as C. difficile. We then assess whether the microbiome that was exposed to the antibiotic can still provide protection against infection.

If the animals remain protected after antibiotic treatment, that provides evidence that the treatment did not significantly disrupt the protective function of the microbiome.

For infections such as C. difficile, localization is particularly interesting because the disease is occurring in the gastrointestinal tract. How does that influence your thinking about the route of administration, absorption, and formulation? Is systemic exposure desirable, undesirable, or dependent on the specific mechanism and infection you’re treating?

 

Antibiotic therapy for C. difficile is different from treatment of many other bacterial infections. For C. difficile, the oral route is generally the most logical because the infection occurs in the gastrointestinal tract.

C. difficile grows and produces toxins within the intestinal tract. An orally administered antibiotic can reach the site of infection directly. By contrast, an IV antibiotic must first enter systemic circulation, distribute throughout the body, and then reach the gastrointestinal tract. That is a more indirect and potentially less predictable route.

For our C. difficile program, the drug design objective is therefore very different from our programs targeting systemic infections. We aim for negligible systemic absorption.

The goal is to prevent the antibiotic from crossing the intestinal epithelial lining and entering the bloodstream. In this case, we intentionally design the molecule to have properties that limit passive membrane permeability.

This can mean moving beyond the traditional Lipinski Rule of Five parameters. We may intentionally make the molecule larger, increase its polar surface area, increase its molecular weight, and make it more hydrophilic. We can also introduce electrical charge, because charged molecules generally have greater difficulty crossing biological membranes through passive diffusion.

The formulation strategy depends on the molecule’s intrinsic physicochemical properties.

If a molecule is inherently non-absorbable and naturally remains in the gastrointestinal tract, a conventional oral capsule or tablet may be sufficient. If the molecule has meaningful passive permeability and tends to cross the intestinal membrane, additional approaches may be necessary, such as pH-dependent coatings or colon-targeted release systems.

For C. difficile, systemic exposure is undesirable because the target is localized within the intestinal lumen. The ideal PK profile is therefore minimal systemic exposure combined with high localized fecal concentrations.

Avoiding systemic circulation also helps reduce the potential for off-target toxicity.

TAXIS is developing efflux-pump inhibitors designed to keep antibiotics inside resistant bacterial cells rather than allowing the bacteria to expel them. Could you explain how you think about drug delivery inside the bacterium — essentially, getting sufficient intracellular exposure at the right bacterial target — and what makes that particularly challenging for Gram-negative organisms?

 

Intracellular delivery of antibiotics is particularly challenging in Gram-negative bacteria because it depends on net intracellular accumulation, the balance between how much drug enters the bacterial cell and how much is expelled.

A compound can have reasonable membrane permeability and enter a bacterial cell effectively but still achieve little or no net accumulation if efflux occurs faster than influx.

This is why efflux pump inhibitors, or EPIs, are particularly important for Gram-negative pathogens. These bacteria possess highly efficient efflux systems that can remove antibiotics from the cell before they reach sufficient intracellular concentrations to kill the bacteria.

An EPI can shift that balance back toward intracellular accumulation by reducing the rate at which the antibiotic is expelled.

Gram-negative bacteria present several additional barriers to antibiotic delivery. Unlike Gram-positive bacteria, they have an outer membrane that creates an additional permeability barrier. Porin-mediated uptake and self-promoted uptake can also limit drug entry.

At the same time, efflux systems can capture compounds from the periplasm and expel them directly into the extracellular environment.

The result is an unfavorable asymmetry: the drug has difficulty getting in and is efficiently pumped back out. This combination makes efflux a major contributor to antibiotic resistance in Gram-negative bacteria.

 

There is growing interest in antibiotic conjugates, prodrugs, siderophore-based delivery, and other strategies that essentially use bacterial biology to help deliver a drug to its target. Which of these approaches do you think has the greatest potential to change antibacterial drug development, and what scientific hurdles still need to be overcome?

 

Siderophore-based delivery has significant potential because it is a relatively mature and clinically validated approach. Cefiderocol provides an example of an approved antibiotic that uses this strategy.

Siderophore delivery essentially works as a molecular Trojan horse. Bacteria require iron for survival and replication and have active systems for acquiring it. An antibiotic can be chemically linked to a synthetic siderophore that binds iron, allowing the bacteria to recognize and actively transport the antibiotic into the cell.

This approach can bypass some of the traditional barriers to antibiotic entry, including limitations imposed by the outer membrane and porins.

There are also important challenges with siderophore-based approaches. Resistance to cefiderocol has already been documented. Bacteria can also express multiple alternative iron-acquisition systems, potentially allowing them to bypass a particular siderophore pathway.

In addition, siderophore receptor expression can vary among bacterial species. As a result, this approach may be particularly useful for developing narrow-spectrum antibiotics rather than broad-spectrum agents.

Prodrug strategies have their own applications, but they often focus on improving oral bioavailability or tissue penetration rather than exploiting a pathogen-specific uptake mechanism.

Antibiotic-antibody conjugates are less mature than comparable technologies in oncology, so their potential is still being established.

How should developers think about the relationship between delivery strategy and resistance development? Could selectively delivering an antimicrobial — or restricting its activity to a particular pathogen or compartment — actually reduce selective pressure for resistance, or are there circumstances in which highly targeted exposure could create new resistance risks?

 

There are two aspects to consider: pathogen selectivity and compartmental selectivity.

With pathogen selectivity, if an antimicrobial specifically targets one bacterial species or a relatively small group of pathogens, it can reduce unnecessary exposure of other bacteria that are not the intended targets. This could potentially reduce selective pressure on the normal bacterial flora.

However, selectivity does not eliminate the possibility of resistance developing in the targeted pathogen. Resistance can still arise through spontaneous mutations or horizontal gene transfer.

With compartmental selectivity, if an antibiotic reaches a sufficiently high concentration at the site of infection and effectively eliminates the pathogen, this can reduce unnecessary exposure elsewhere in the body.

The concern arises when drug concentrations are inadequate. If the drug kills susceptible bacteria while allowing less-susceptible bacteria to survive and multiply, that could increase the risk of resistance developing.

Another consideration is that pathogens may occupy multiple compartments within the body. For example, a respiratory infection may involve both the lungs and the upper respiratory tract. If a drug reaches the lungs but not another clinically relevant site of infection, surviving bacteria could persist and potentially develop resistance through mutation or horizontal gene transfer.

That does not mean compartment-specific delivery is inherently disadvantageous. Rather, developers need to ensure that the antibiotic reaches adequate concentrations at all clinically relevant sites of infection.

Overall, pathogen and compartmental selectivity can help reduce unnecessary antibiotic exposure and associated selective pressure. However, neither approach guarantees a lower risk of resistance in the target pathogen. That ultimately depends on the antibiotic’s mechanism of action, its intrinsic resistance potential and whether sufficient drug exposure is achieved at the relevant sites of infection.

Looking beyond TAXIS’ current programs, where do you see the biggest untapped opportunities for drug delivery technology in combating antimicrobial resistance? Are there particular routes, formulations, targeting mechanisms, or delivery platforms that you think the drug delivery community should be paying more attention to over the next five to 10 years?

 

We prioritize orally available antibiotics whenever possible. For severe infections, IV delivery may still be necessary.

Over the next five to 10 years, we expect nanoparticle-based delivery systems to receive significant attention. These platforms have the potential to direct drugs more precisely to sites of infection, potentially increasing treatment effectiveness while reducing unnecessary exposure. Different nanoparticle approaches, including liposomes and polymeric nanoparticles, could provide new delivery options.

I expect phage-based therapies to receive increased attention, particularly if these approaches can be developed for oral administration. Many of the phage-based treatments currently being explored are primarily used in hospital settings and require IV administration, so developing effective oral delivery could broaden their potential applications.

Antibiotic-antibody conjugates are another area that could mature significantly over the next decade. Antibody-drug conjugates are already well established in oncology, and applying similar principles to infectious diseases could allow antibiotics to be directed more specifically toward pathogens while minimizing exposure of the host microbiome.

Another promising area is smart trigger-based antimicrobial release from medical devices. Many antimicrobial devices currently rely on passive release mechanisms, in which antibiotics are released continuously. A smarter system could detect conditions associated with infection and release an antimicrobial only when it is needed. This could potentially reduce unnecessary antibiotic exposure and selective pressure for resistance.

Finally, pulmonary drug delivery represents an important opportunity. Today, when a patient has a lung infection, systemic IV therapy is often used with the expectation that sufficient drug will reach the lungs. In the future, more targeted pulmonary delivery systems could potentially deliver higher concentrations directly to the lungs.

This could be particularly important for patients with hospital-acquired pneumonia (HAP) and ventilator-associated pneumonia (VAP), where achieving adequate antibiotic concentrations at the site of infection can be challenging. Targeted pulmonary delivery could therefore become an important tool for improving treatment of these difficult infections.

About The Expert

Ajit Parhi, Ph.D., brings over 25 years of experience in the fields of organic chemistry, synthetic methodologies, natural product synthesis, medicinal chemistry, and drug discovery. His extensive expertise encompasses synthetic medicinal chemistry and microbiology, coupled with a deep understanding of pharmacology and toxicology.

Since the founding of TAXIS Pharmaceuticals, he has served as the chief medicinal chemist and has led the company’s discovery research efforts since 2018. Currently, as the chief scientific officer at TAXIS, he oversees all drug discovery programs.