Poor Communication of Essentially Good Research

It is no surprise to see a press release on scientific research that inflates the conclusions of the study or its potential impact. But usually the basic biology is correct, not so with the press release I read today. (I thank the twitterverse for bringing this to my attention.) Since I'll go through the press release paragraph by paragraph, you can thank the twitterverse too,

The release starts out well enough, but quickly goes off the rails. In fact, the writers couldn't get out of the first paragraph, which is also the first sentence, without stating half-truths.

Targeting serious and sometimes deadly fungal infections, a team of researchers at Worcester Polytechnic Institute (WPI) and the University of Massachusetts Medical School (UMMS) has discovered a chemical compound that prevents fungal cells from adhering to surfaces, (This would be a great place for a period.) which, typically, is the first step of the infection process used by the human pathogen Candida albicans (C. albicans). (Red note by AbC.)

The point being made is that 'adherence' is the first step of the infection process. Whatever the hell 'infection process' is escapes me, but it does sound pretty awesome and scary. However, the press release completely misses a key point, Candida albicans is a commensal in humans. C. albicans is easily culturable from 60% of all people and is almost certainly found in all humans. Let me be clear, C. albicans causes serious and sometimes deadly infections as noted in the first line of the press release. However, in the vast vast majority of people, C. albicans is not a problem. Herein lies the problem or obvious question ignored in order to write overimpactful press releases. Does C. albicans only adhere to surfaces during infection? Does C. albicans in most people simply float free in saliva, gastric juices, or vaginal fluids? (The answer to these questions is 'No'.)

After screening 30,000 chemical compounds in a series of tests with liveC. albicans, the team found one molecule that prevented the yeast from adhering to human cells or to polystyrene, a common plastic used in many medical devices. Named "filastatin" by the researchers, this molecule now emerges as a candidate for new anti-fungal drug development and as a potential protective material to embed on the surfaces of medical devices to prevent fungal infections.

This is all pretty good. It hits on the major finding and the potential impact of the research.  What is not explicit here, but comes up later, is that the major source of C. albicans in deadly systemic infections is contaminated medical devices, like catheters.

The team, led by co-principal investigators Paul Kaufman, PhD, professor of molecular medicine at UMMS, and Reeta Rao, PhD, associate professor of biology and biotechnology at WPI, reports its findings in the paper "Chemical screening identifies filastatin, a small molecule inhibitor of Candida albicansadhesion, morphogenesis, and pathogenesis," published online in advance of print by the journalProceedings of the National Academy of Sciences (PNAS).

Again nothing to complain about regarding the press release here. We are basically describing the leaders of the research teams and the source of the article. I do have an issue with part of the title of the paper 'a small molecule inhibitor of...morphogenesis...' The title suggests that filastatin inhibits the formation of cell structure or shape. In actuality, the drug inhibits the ability of yeast cells to form filamentous cells. This is a morphological transition, but it does not encompass the entirety of C. albicans structure and shape. A yeast cell making more yeast cells is an example of morphogenesis. The abstract makes this point clear, but the title is a bit colloquial. This is a problem in the field and not specific to the authors, I have probably used similarly short-hand. 

"In humans, the most widespread fungal pathogen is Candida albicans, which is also one of the most frequent causes of hospital-acquired infections," the authors write. "We conclude that filastatin is not toxic to the human cell line under our assay conditions, but is unique in that it can impair fungal adhesion both to inert surfaces and to cultured human epithelial cells."

I am not sure in what context filastatin is unique as a quick literature search revealed HIV-protease inhibitors, a 23 amino acid peptide, cranberry derived compounds, as well as antibodies have all been reported to inhibit C. albicans adhesion. (This is an issue of the authors inflating their work not of the press release writer(s).)

Infection by C. albicans causes common chronic illnesses like thrush and vaginitis, which affect millions of people globally each year and are not easily cleared by the handful of anti-fungal drugs now available. While most fungal infections do not cause serious harm, if one spreads to the bloodstream it can be deadly.

All good. Minor issue with referring to thrush and vaginitis as chronic illnesses as the vast majority of cases are not chronic.

Hospitalized patients with catheters or central intravenous lines are at risk as the fungi can grow on those devices and enter the body. Similarly, patients with implanted medical devices like pacemakers or prosthetic hips or knees are also at risk if the implant carries a fungus into the body. Also, people with compromised immune systems are at greater risk for serious fungal infections. Because of the lack of effective drugs against C. albicans and other pathogenic fungi, the mortality rate for systemic fungal infections is between 30 and 50 percent.

Again all good, but now we come to it. The paragraphs of FAIL.

Typically, a blood stream infection of C. albicans or a similar pathogen begins with fungal cells attaching to a surface—a catheter, for example, or epithelial cells lining the mouth—to form what is known as a biofilm. Next, the ovoid shaped yeast cells morph into an invasive filamentous form, extending pointed filaments that penetrate and damage surrounding tissues. In the current study, the team found that filastatin curtailed both steps: it largely prevented C. albicans from adhering to various surfaces, and it significantly reduced filamentation (inspiring the name filastatin).

First of all, we have established that C. albicans adheres to things independently of infection. So a basic part of C. albicans lifestyle is to adhere to things, things like epithelial cells which are the structures C. albicans normally finds itself on. This is a function of biology not infection.

Second, this press release builds on a dead mythos that yeast cells are commensals and filaments are infectious. This was the conventional wisdom 20+ years ago, but the field has (or has tried to) move on. It is true the filamentous cells do things related to disease that yeast cells do not. Filamentous cells secrete gobs of proteases which help degrade the host tissues. Filamentous cells are invasive. But you know what, yeast cells do things related to disease that filamentous cells do not. Yeast cells do not stimulate the immune response as much as filamentous cells do. Yeast cell populations grow much faster than filamentous cell populations. Yeast cells can readily disseminate throughout the body via the bloodstream, filamentous cells cannot. But ignoring these facts, there are basic predictions we can make based on the press release premise. For example, we should not observe filamentous cells in non-diseased individuals, but we do. We should not observe yeast cells during bloodstream infections, but we do. Other fungal pathogens should show similar morphological transitions, they don't. In fact, Candida glabrata, the second most common Candida species causing human disease only grows in the yeast form.

Third, filastatin inhibits adherence and filamentation. This becomes problematic because those who work on C. albicans are already aware that filamentous cells are extremely adherent, but yeast cells are not. Thus, if you inhibit filamentation you by definition inhibit adherence. However the press release tells us why this result is really the SHITZ (emphasis mine).

As a next step, the team tested filastatin's impact on C. albicans cells that had grown unfettered in test wells and had already adhered to the polystyrene walls. When the compound was added to the culture mix, it knocked off many of the fungal cells already stuck to the polystyrene. The inhibitory effects of filastatin were further tested on human lung cells, mouse vaginal cells, and live worms (C. elgans) exposed to the fungus to see if it would reduce adhesion and infection. In all cases, the novel small molecule had significant protective effects without showing toxicity to the host tissues.

Based on the bolded statement, you now think 'HOLYFUCKBALLS, that's interesting,' because the cells were already stuck to something. While the press release is technically correct, the implications are flat out wrong. First, in almost every experiment the drug was present throughout the experiment, in other words as soon as the cells were added, aka before adherence occurred. Thus, inhibition of filamentation is an obvious and unsurprising likelihood to explain these data. Second, in the one experiment referred to above, the effect of the drug was minor (see data from paper below). Even this minor effect could be an attribute of growth. 

From PNAS Figure 2D

In one case, the black bars, the cells were kept without (left) or with (right) drug for 8 hours and there is a profound difference between the two conditions. In the other case, the striped bars, the cells were grown for 4 hours and then drug left out (left) or added (right) and cells grown an additional 4 hours. Both samples lacking drug are similar (compare the leftmost of the black and striped bars). The authors suggest that the reduction observed 4 hours with drug (rightmost striped bar) represents a reversal of adherence, but this conclusion is not necessarily the best. How about growth after 4 hours is always ~8000 Fluorescence units, but when drug is added no further growth is observed. Knowing what the Fluorescence looked like at 4 hour in this experiment is lacking and important. Note the press release states the drug did something active, 'it knocked off many fungal cells already stuck to the polystyrene.' I say the drug inhibited growth or prevented additional adherence, like by preventing additional filamentous growth as observed in all the other experiments.

Research is now focused on teasing out the precise molecular mechanisms filastatin uses to prevent adhesion and filamentation. "We need to find the target of this molecule," Rao said. "We have some good leads, and the fact that we aren’t seeing toxicity with host cells is very encouraging, but there is more work to be done."

Why does the target of filastatin need to be found? It seems important, but I am not sure why and I wonder if a general audience reader knows why.

Additional studies on filastatin are underway at both WPI and UMMS. "The molecule affects multiple clinically relevant species, so we're pursuing the idea that it provides a powerful probe into what makes these organisms efficient pathogens," Dr. Kaufman said. "In this era of budget gridlock in Washington, our ability to get funding from the Center for Clinical and Translational Research at UMMS to support this work was essential for allowing us to pursue our ideas for novel ways to approach this important class of hospital-acquired infections."

What makes these organisms efficient pathogens? These organisms are shit pathogens. Virtually no one had issues with deadly fungal infections before the 1960s and even after 1960 that number was minute. It is only recently that these organisms have risen to the forefront of importance and it's not the organism that has changed. Rather it is the host, us, or more specifically modern medicine that is the culprit. Now we have a large and growing population of people who are immunocompromised because of cancer therapy, organ transplants, HIV, even extreme old age. But as noted 'in this era of budget gridlock in Washington,' we must inflate and oversell our studies 'to get funding from the Center of Clinical and Translational Research....' There fixed it.

Look, I agree filastatin is an interesting molecule worthy of additional research and development as a coating for catheters and other medical devices. These are some interesting studies and the fact the drug does not appear to be toxic to human cell lines and in a mouse vaginal model is extremely encouraging. It is important to know how this drug acts systemically in a mouse because the bloodstream infections are the critical ones to combat. But these studies are an important first step. 

The project was also funded by a grant from a WPI/UMMS pilot program established to promote collaborations between researchers at the universities to advance early stage translational research. "Joint research programs, such as the pilot program between our institutions, are central to WPI's work in the life sciences," said Michael Manning, PhD, associate provost for research ad interim, at WPI. "As this collaboration between Professors Rao and Kaufman demonstrates so well, both institutions can leverage their complementary expertise for the ultimate advancement of scientific discovery and public health."

Terence R. Flotte, MD, UMMS executive deputy chancellor, provost, and dean of the School of Medicine, agreed. "The faculty of UMass Medical School and WPI possess scientific knowledge and expertise in disciplines that complement each other," he said. "The creation of this type of multidisciplinary team collaboration between the two universities allows us to work synergistically to solve problems important for improving human health."

Synergistically, √. Novel, √. Translational, √. Paradigm shifting, oops missed that one.

I feel bad, slightly, picking on this press release and the author statements. We often give answers to question that upon reflection wish we had a do over. However, this is my field of interest and C. albicans is the organism I primarily study, so I am intimately familiar  with the issues in the field. I am sure the approach I take here could be applied to many press releases. I would like us, as scientists and communication officers, to work harder at getting the stories out, keeping it interesting and informative without jumping the shark. This likely requires some back and forth between the scientists and communications people with a touch of editorial oversight by an independent scientist.