TO PREVENT TUBERCULOSIS, BREATHE IN
They were all dead. For Yun-Ling Wong, that finding was not new. Luckily, "they" were only mycobacteria. Finding them all dead, though, meant that Wong just made another failed batch of vaccine for tuberculosis. She seemed stuck in an endless cycle of trial-and-error as she tried to prepare a live vaccine in a new way - one that would increase potency and portability.
As a graduate student in David Edwards' lab at the Harvard School of Engineering and Applied Science, Wong is part of a community of researchers around the world who seek a better vaccine for TB. The existing one, Bacillus Calmette-Guérin (BCG), was developed at the Pasteur Institute in France in the 1920s, and its age is showing. Made from cultures of an attenuated strain of Mycobacterium bovis rather than its disease-causing relative, M. tuberculosis, BCG is an efficacy lottery, with studies reporting that anywhere between zero and 80% of those inoculated are protected; in other words, vaccination can be entirely useless or it can protect four out of five people, depending on a range of factors. These factors include batch-to-batch variation, the Tuberculin status of subjects, differences in infective strains of M. tuberculosis and the intensity of infecting dose, genetic and nutritional differences among inoculated populations, and of course methodological variations in the studies. Other drawbacks of BCG are the need for refrigeration and for the vaccine to be injected.
Despite the shortcomings, Jerald Sadoff, president and CEO of the Aeras Global TB Vaccine Foundation in Rockville, MD, notes that BCG saves the lives of 40,000-100,000 children each year who otherwise would have died from disseminated TB. But he also notes that worldwide, 1.5 billion people - that's one in three - are infected with TB, and two million die of the disease each year. "There is an enormous need for an effective TB vaccine" says Carol Dukes Hamilton of Duke University Medical Center, The ideal new vaccine would be easy to deliver, not require refrigeration, have a long shelf life, and provide a high degree of protection.
AVOIDING INJECTIONS
"What's needed is a vaccine more potent and as safe as the current vaccine" says TB researcher Marcus Horwitz of the University of California, Los Angles; adding, "A handful of vaccines have demonstrated superior efficacy to BCG in animal models." These include several recombinant forms of BCG being studied in Horwitz's lab. Likewise, James Triccas of the University of Sydney has tested an attenuated form of M. tuberculosis - the actual infecting bacteria - and found promising results in mice, which indicate that this could make a more efficacious vaccine than BCG. Several other examples also exist.
While these and other researchers are focused on developing new vaccine candidates, new delivery mechanisms are also of interest. Inhalation is a front-runner for TB. "Pulmonary tuberculosis in adults is the most prevalent form of this disease," says Stefan Kaufmann, director of the Max Planck Institute for Infection Biology in Berlin. "An inhalable TB vaccine would make a lot of sense." Annie De Groot, a developer of TB vaccines at Brown Medical School says, "The mucosal route is one of the most important avenues of engaging protective immune response in the lung. It is also the most immunogenic."
With a variety of researchers agreeing on the potential value of a nasal approach, do they also agree on applying that to BCG? Some do. McShane says, "Making a stable aerosol of BCG is a good place to start." Others in the field, however, care less about how the vaccine gets administered and more about the vaccine itself. "The inhalation versus other-delivery-route issue doesn't matter too much to those of us who work on TB," says Hamilton, "as long as it is safe and effective." Still, she is open to new ideas. She adds, "Given the difficulty we've had finding inoculations that are particularly immunogenic, using the respiratory mucosa may be the novel approach that works."
Similar thinking led Barry Bloom, dean of Harvard's School of Public Health, to encourage Edwards to try to make an inhaled version of BCG. Edwards has the right background. He already knew how to make inhaled drugs. As a postdoc in Robert Langer's lab at the Massachusetts Institute of Technology in the early 1990s, Edwards worked on a project that led to inhaled insulin. In 1997, he was one of the founders of Advanced Inhalation Research, a company sold to Alkermes in 1999. And in 2003, he founded Medicine in Need (MEND), a not-for-profit providing innovative drug-delivery technologies, especially oral delivery, that counter diseases associated with those in poverty. Edwards hopes that an inhalable TB vaccine will turn into one of MEND's technologies.
SOLUTIONS USING SOLUTIONS
Although Edwards knew how to develop an aerosolized drug and put it into a delivery device for inhalation, doing so with BCG proved particularly challenging. The first step was to develop a powder fine enough for aerosolization. Edwards envisioned a powder that could withstand environmental heat and humidity - typical conditions in the tropical and subtropical regions of Africa and south Asia where TB is a major scourge. And he knew that this approach would require a delivery device that could be used successfully with a squirming, crying infant.
"Frankly," says Edwards, "most vaccines are prepared by freeze drying - lyophilization; you create a powder that is not inhalable." So, he explains, "we needed to make the organisms in a dry-powder form that is inhalable, and the question was, how to do that." Edwards decided that his best bet was spray drying, which is used to make many food products, including powdered milk. Simply put, this technique sprays a liquid as droplets through hot gas. The result is a dry, powder-form of the starting liquid. With an objective and an idea of how to go about it, Edwards just needed someone to try it. That's where Wong came into the project. She used M. smegmatis in early studies, which is generally not pathogenic. By comparison, the M. bovis used in BCG can cause TB in some cases.
When spray drying food, processes typically include cryoprotectants and salt. So, it made sense to do that with bacteria, too. Edwards says, "It seemed to me a priori to introduce cryoprotectants within the organism or outside the organism to eliminate osmotic stress." But suspending the bacteria across a wide range of concentrations of salt and cryoprotectant didn't work. Says Wong, "There wasn't very much guidance from what had been done in the past."
Then, Edwards and Wong wondered if the salt and cryoprotectants killed the bacteria instead of preserving them. "Cryoprotectants are useful to the degree that the change in hydration in the organism happens on a timescale that is long relative to the time it takes for water to diffuse across the membrane," explains Edwards. They modeled the process for bacteria - reported as part of this entire study in the February 20, 2007, Proceedings of the National Academy of Sciences - and it indicated very fast drying: A water droplet would dry completely in less than 400 milliseconds, which is much less than the amount of time it takes water to diffuse across the cell membrane. In fact, the speed of drying was at least an order of magnitude faster than the diffusion time.
So as the water around a bacterium dried, the concentration of the salts and cryoprotectants kept increasing. Moreover, water moving from inside the bacterium to the outside could not flow fast enough to balance the osmotic pressure. As the drying continued and the solutes outside the bacteria grew thicker and thicker, the salts and cryopreservants turned into crystals, killing the bacteria. At least, that was the theory.
To test it, Wong put bacteria in solutions of increasing salt concentration, essentially mimicking the environment around a bacterium as its droplet got spray dried. With the concentration of salt being equal inside and outside of the bacteria, Wong found that the cells lived for a day. When she doubled the external salt concentration, most of the bacteria died in half an hour. Doubling the salt once more, the bacteria lasted less than five minutes. At that point, Edwards and Wong started to believe the results of their model.
"The answer was not manipulating the concentration of cryoprotectant or salt," says Edwards, "but just eliminating them." Wong prepared bacteria in three solutions: one that contained no salt or cryoprotectant; one with cryoprotectant but no salt; and one with salt but no cryoprotectant. She sprayed dried samples of each solution and measured the colony forming units, essentially a measurement of how many bacteria are alive. Far more bacteria survived in the solution without salt or cryoprotectant.
As a general rule in biology, cells only survive with some salt around. Here, the cells did best without it. As Edwards says, "The membrane is so solid in these bacteria, actually, they did totally fine." In addition to merely keeping the bacteria alive, Edwards and Wong found other benefits from this new finding. As one crucial element for something inhalable, the bacteria must dry to an appropriate flowable consistency, and they did. Moreover, the spray-dried bacteria showed good shelf stability, even under high temperatures and humidity.
At that point, though, Edwards and Wong knew that everything looked promising with M. smegmatis. But what about with the real thing? When they applied the same technique to BCG, it worked just about the same. That left one hanging question: How well did spray drying preserve BCG in comparison with lyophilizing? Edwards and Wong made a batch of BCG and sprayed dry half and lyophilized the rest and tested the two solutions for bacterial viability. Even one day after the process, the spray-dried bacteria survived better - 10-fold better, in fact. The spray-dried form also survived better over time and somewhat better at high temperatures. Whether or not that means that spray-dried BCG would make a more effective vaccine remains to be tested.
DELIVERY PROS AND CONS
Delivery of the vaccine was another issue. Aerosolized "delivery to newborns is not something that's been done," Edwards explains. "There are many challenges to doing that, just because their breathing rate is so, so small." In the 1990s, Edwards had worked on a project with mechanical engineer Rich Miller of Manta Product Development in Cambridge, Massachusetts, and they collaborated again to investigate a way to deliver BCG. Making an inhalable vaccine for newborns poses unique challenges. "One, they're very small," says Miller, "so their anatomy is quite different from a large child or an adult, [and] two, they can't cooperate. You can't ask them to inhale on demand." Edwards points out that another limiting factor in designing an inhalation mechanism is that "newborns don't like facemasks." Still, Edwards says that Miller and his colleagues at Manta have "come up with a very innovative solution. It involves a pacifier-like interface." As Miller explains, "This is the first time anyone has attempted to use this particular technology on a newborn infant." Manta has several prototypes undergoing testing.
Some scientists worry about safety. As Kaufmann of Max Planck says, "The lung is a highly sensitive organ, and therefore a safe vaccine is needed and this issue needs further investigation." UCLA's Horwitz agrees. "As with any biological, a novel delivery route raises new safety issues," he says. "BCG is typically administered into the skin and this route of administration has been shown to be very safe. BCG also has been safely administered orally." Although those delivery routes prove quite safe, the same is not necessarily true of inhalation. "Delivering this live, replicating vaccine by inhalation, especially if it is inhaled through the nose," says Horwitz, "raises concerns regarding rhinitis, sinusitis, otitis media; and, given the proximity of the nasal passages to the brain, brain inflammation." He adds, "In this regard, a vaccine that is non-replicating or replication-limited may be a safer alternative to conventional ones.
THE ROAD AHEAD
For now, that's just one of many challenges to Edwards and Wong. At the top of the list: They still need to show that their BCG blocks TB. As with any new vaccine - even a reformulation of an existing one - they also need to address issues of cost. In the industry, it is generally accepted that spray drying is less expensive than lyophilizing, but it is too early to make supportable claims about the economics of an inhalable BCG.
Regardless of work by Edwards or anyone else, BCG will continue to be used, at least into the near future and maybe much longer. Although it might not protect against TB as well as it has in the past, it does protect against infection with another pathogenic mycobacterium, M. leprae, which causes leprosy. It also could play a role as the prime element in a neonatal prime-boost vaccine regimen - and not just against TB. Sadoff of Aeras explains that any vaccine response that involves cellular immunity can be primed with BCG, or even recombinant BCG that has been manipulated to make it more immunogenic. "Since it's universally given to children at birth, it's very convenient for the [vaccination] schedule because they're getting it anyway," says Sadoff. He foresees BCG as serving as a prime for vaccines against malaria, HIV, and TB - "something that would protect against three diseases that haven't been conquered yet."
How the Edwards and Wong approach plays into this new world of vaccination remains to be seen. "David Edwards is on the right track," says Kaufmann, "but as always in science, it is hard to make predictions." Moreover, a disease as dangerous as TB should be attacked from as many directions as possible. As Kaufmann says, "It would be good to exploit additional approaches."
