Linking pollution and infectious disease

Thousands of seals died along the coasts of the heavily polluted Baltic Sea in the late 1980s. Scientists traced the deaths to a virus similar to the one that causes distemper in dogs. Last year, the same virus struck hundreds of seals in Maine. In both instances, researchers believe that persistent organic pollutants, such as polychlorinated biphenyls (PCBs), dioxins, and furans, played an indirect role in the seals’ demise.

The seals are one example of a phenomenon of increasing importance to toxicologists: the interplay between exposure to environmental contaminants and infectious disease. More than two decades ago, researchers reported that exposure to low levels of 2,3,7,8-tetrachlorodibenzo- p-dioxin (TCDD), the most toxic of the dioxins, decreases resistance to an influenza virus in mice (Fundam. Appl. Toxicol. 1996, DOI: 10.1006/faat.1996.0004).

Scientists have since shown that exposure to other chemicals, including perfluorooctanesulfonic acid (PFOS) and perfluorooctanoic acid (PFOA), mercury, and arsenic, can also alter the immune response and increase susceptibility to infectious diseases in multiple species of laboratory animals. Epidemiology studies in humans have linked exposure to various chemicals in the womb with reduced levels of antibodies triggered by childhood vaccines and increased risk of infectious diseases.

Chemicals also affect pathogens and in some cases can make them more dangerous. Researchers have shown a link between multidrug-resistant bacteria and exposure to zinc, lead, and disinfectants. Epidemiologists are investigating whether exposure to phthalates is also associated with multidrug-resistant bacteria.

“Environmental pollutants affect how we are infected,” Linda Birnbaum, director of the US National Institute of Environmental Health Sciences, said during opening remarks at a January workshop on the interactions between chemicals and pathogens sponsored by the US National Academies of Sciences, Engineering, and Medicine. Studies so far have revealed tantalizing clues about the scope and mechanisms of these interactions, but more work needs to be done to understand the full effects of chemical exposure on public health, said Birnbaum and other toxicologists, epidemiologists, and infectious disease experts who attended the workshop.

Fluorochemicals impair immune function

Birnbaum has been investigating the intersection between environmental pollution and infectious diseases for many years. She pioneered the work on exposure to TCDD and decreased resistance to an influenza virus in mice in the mid-1990s while working as a researcher for the US Environmental Protection Agency. Studies led by other scientists have linked perinatal exposure to PCBs (PLOS Med. 2006, DOI: 10.1371/journal.pmed.0030311) and to perfluoroalkyl substances (J. Immunotoxicol. 2017, DOI: 10.1080/1547691X.2017.1360968) with decreased immune responses to childhood vaccines in people living in the Faroe Islands in the North Atlantic Ocean. “Understanding environmental immunotoxicity will be critical to making successful vaccines,” Birnbaum emphasized at the workshop.

Researchers are now increasingly concerned about exposure to per- and polyfluoroalkyl substances (PFAS). These substances are widely used in consumer products because of their water- and oil-repellent properties. They are found in food packaging, nonstick products, stain-repellent clothing, cleaners, and many other household items. Firefighting foam is also a large source of PFAS groundwater contamination near military bases and airports.

Perfluoroalkyl carboxylates, particularly PFOA, have received a lot of attention because of their potential to suppress the immune system, says Berit Granum, a senior scientist at the Norwegian Institute of Public Health. Even though industry has phased out the use of PFOA, blood serum levels of PFOA in humans “have decreased at a slower rate than predicted,” she tells C&EN. Researchers attribute the slow decline to continuous exposure to perfluoroalkyl carboxylates from sources such as drinking water, milk and dairy products, meat, seafood, eggs, indoor air, house dust, and industrial emissions.

From these same sources, people are also exposed to other compounds that get metabolized into perfluoroalkyl carboxylates, Granum says. Such precursors include polyfluoroalkyl phosphate monoesters and diesters that can metabolize into PFOA, perfluorononanoate, perfluorohexanoate, and perfluoroheptanoate, she notes.

Philippe Grandjean and colleagues at the Harvard T.H. Chan School of Public Health were among the first to find an association between elevated PFAS concentrations in mothers’ blood and reduced response to vaccinations in their children, specifically how Faroese children responded to diphtheria vaccination (JAMA, J. Am. Med. Assoc. 2012, DOI: 10.1001/jama.2011.2034).

Granum and colleagues subsequently also linked exposure to perfluoroalkyl substances with decreased antibody levels triggered by vaccines and altered immune-related health effects in early childhood (J. Immunotoxicol. 2013, DOI: 10.3109/1547691X.2012.755580). The researchers measured concentrations of PFOA, perfluorononanoate, perfluorohexane sulfonate, and PFOS in blood collected from pregnant women in Norway when they gave birth in 2007 and 2008, and in blood of the children at three years of age. They found that higher maternal levels of the four PFAS correlated with fewer antibodies against rubella in the vaccinated children.

Granum and colleagues also observed a correlation between maternal levels of PFOA and perfluorononanoate and the number of common colds in the Norwegian children, and between maternal levels of PFOA and perfluorohexane sulfonate and gastroenteritis in the children. The scientists followed up with a study investigating gene expression associated with PFAS exposure in the Norwegian mother-child pairs (J. Immunotoxicol. 2015, DOI: 10.3109/1547691X.2015.1029147). They compared whole-genome transcriptomics data from umbilical cord blood with maternal blood concentrations of four PFAS. The team identified that PFAS exposure is associated with changes in expression of 52 genes involved in immunological and developmental functions. They discovered that expression of those same genes is also associated with decreased rubella antibodies and increased episodes of the common cold in children.

In other work, Granum and colleagues reported that exposure to several PFAS is associated with an increased number of respiratory tract infections in the first 10 years of life (Environ. Res. 2017, DOI: 10.1016/j.envres.2017.10.012). On the flip side, the researchers also found an inverse association between maternal concentrations of perfluoroundecanoic acid and atopic eczema in girls. However, they did not see an association with asthma or allergy-related health effects for other PFAS.

Most recently, they showed possible gender differences in respiratory infections and gastroenteritis associated with PFAS exposure, with the majority of correlations identified only in girls (Environ. Int. 2019, DOI: 10.1016/j.envint.2018.12.041).

The work is important for regulators around the world who are grappling with setting limits for PFAS in drinking water, food, and hazardous waste sites. Interestingly, risk assessors in the US and European Union came to similar conclusions for PFOA and PFOS using different approaches, Granum says. The European Food Safety Authority (EFSA) published its assessment of the risks of PFOA and PFOS in food in December 2018. EFSA used human epidemiological studies to recommend a maximum limit of 13 ng/kg of body weight per week for PFOS and 6 ng/kg of body weight per week for PFOA.

In the US, the Agency of Toxic Substances and Disease Registry (ATSDR) published a draft risk assessment of PFOS and PFOA in June 2018. ATSDR used animal studies to propose a minimum risk level of 14 ng/kg of body weight per week for PFOS and 21 ng/kg of body weight per week for PFOA. ATSDR is working with the US Centers for Disease Control and Prevention to conduct exposure assessments in communities near military bases that have PFAS in their drinking water.

The regulatory agencies have yet to set limits for PFAS other than PFOA and PFOS.

Arsenic alters vaccine response

Scientists have known for a long time that arsenic also affects the immune system. Case reports from the 1920s to 1940s show that arsenic given to patients at high doses to treat syphilis led to effects that sound like immunological responses, Molly Kile, an environmental epidemiologist at Oregon State University, said during the January workshop.

Kile presented a study demonstrating an association between total arsenic levels in urine and a lack of serum antibodies against varicella zoster, the virus that causes chicken pox and shingles (Environ. Health Perspect. 2015, DOI: 10.1289/ehp.1408731). She also showed that higher arsenic exposures were associated with higher odds of past hepatitis B infection (Environ. Res. 2018, DOI: 10.1016/j.envres.2018.06.023).

Kile also presented preliminary, unpublished data showing that the timing of arsenic exposure appears to be important. Working with colleagues in Bangladesh, her group examined the effects of in utero exposure to arsenic from drinking water on the risk of infectious disease in Bangladeshi children. One of the studies found a strong association between arsenic levels and a decrease in serum antibodies for diphtheria in vaccinated children. The researchers did not find an association between arsenic and tetanus.

Whenever there is an outbreak of infectious disease, people are quick to blame an unvaccinated child as the source of the infection. “But there are a lot of people who get caught up in these outbreaks who have been vaccinated,” Kile noted. People’s immunological responses to vaccines vary, and some vaccines are known to produce relatively short periods of protection—this is why adults should receive booster shots for tetanus and diphtheria every 10 years. But vaccine protection “may wane quicker if people are exposed to environmental pollutants,” Kile said.

Other studies looking at mother-child pairs in New Hampshire (Environ. Health Perspect. 2016, DOI: 10.1289/ehp.1409282) and Bangladesh (Environ. Health Perspect. 2011, DOI: 10.1289/ehp.1002265) found similar associations between developmental exposure to arsenic and increased risk of lower respiratory infections and diarrheal disease in children during the first year of life.

Mercury turns the body on itself

Scientists know that mercury has neurotoxic effects, but evidence is growing that it also interferes with the immune system. People can be exposed to many different forms of mercury: elemental mercury from broken thermometers and small-scale artisanal gold mining, inorganic mercury from fluorescent light bulbs and dental amalgams, methylmercury from eating contaminated fish, and ethylmercury from the preservative thimerosal used in some vaccines.

All forms of mercury produce some kind of effect on the immune response, says Jennifer Nyland, a professor of biology at Salisbury University. But they don’t all affect the immune system in the same ways, and they don’t have the same level of toxicity, she adds. They are also metabolized differently. Researchers have reported evidence of mercury altering immune function in animal studies, in vitro cell cultures, and human epidemiological studies.

As a postdoc at the Johns Hopkins Bloomberg School of Public Health, Nyland demonstrated that exposure to low doses of inorganic mercury exacerbate an autoimmune disease triggered by the Coxsackievirus in mice (Toxicol. Sci. 2011, DOI: 10.1093/toxsci/kfr264).

Coxsackievirus causes hand, foot, and mouth disease, and nearly everyone is infected with the virus at some point in their lives, Nyland says. Most people clear the infection without long-term adverse effects. But some susceptible individuals develop an autoimmune disease that affects their heart long after they have cleared the infection, she says. Somehow the immune system gets triggered to send cells called macrophages back into the heart. Those cells initiate an inflammatory response that eventually damages the heart tissue, leading to a flabby heart that is unable to function properly. Nyland and colleagues showed that exposure to mercury followed by Coxsackievirus infection makes this particular autoimmune disease worse in mice.

While at Hopkins, Nyland worked with others to investigate the effects of mercury on human immune cells in cell cultures (Toxicol. Lett. 2010, DOI: 10.1016/j.toxlet.2010.06.015). The researchers isolated the cells from blood samples and stimulated them with bacterial antigens to mimic an infection. They also dosed the cells with varying concentrations of mercuric chloride, methylmercury, and ethylmercury. All three forms of mercury affected the immune response by altering the release of signaling proteins called cytokines, but in different ways and to varying degrees. “The bottom line is that all of the different forms of mercury have the potential to be immunotoxic,” Nyland says.

The researchers also studied artisanal gold miners who use elemental mercury (Environ. Res. 2010, DOI: 10.1016/j.envres.2010.02.001) and people who eat contaminated fish downstream of gold mines in Brazil (Environ. Health Perspect. 2011, DOI: 10.1289/ehp.1103741).

They showed an association between mercury levels in urine from the gold miners and levels of an autoantibody in their blood. That autoantibody is one that physicians use to help diagnose the autoimmune disease lupus, Nyland says. The scientists also found a correlation between total levels of mercury and methylmercury in hair or blood of people who consumed contaminated fish and levels of the autoantibody in their blood, although the correlation was not as strong as in the gold miners.

At Salisbury, Nyland’s work recently moved toward looking at the mechanism of how mercury is interacting with the immune response. The work is just beginning, Nyland says. Her group is investigating the pathways involved in the inflammasome—a multiprotein complex that detects pathogens and activates the release of proinflammatory cytokines. The team is investigating how the interplay of various doses of mercury species, coexposure to other chemicals, and pathogen infections affects gene expression of inflammasome components.

“Mercury could be interacting with one of the steps that helps to make that inflammasome form,” Nyland says. “It could also be changing how that inflammasome does its job once it is formed.”

Chemicals boost pathogens’ virulence

In addition to changing the immune response in numerous species, some substances can also make pathogens more virulent or resistant to antibiotics.

Meghan Davis, a professor of environmental health and engineering at the Johns Hopkins Bloomberg School of Public Health, is investigating how various chemicals that people encounter every day affect methicillin-resistant Staphylococcus aureus (MRSA), leading to multidrug-resistant strains.

Unlike typical staph infections that respond well to antibiotics, MRSA infections are difficult to treat because the bacteria don’t make a particular protein that binds to penicillin, methicillin, and cephalaosporins. “None of the β-lactam antimicrobials bind” to MRSA, Davis says. These are the first-tier drugs used for treating skin and soft-tissue infections, which is where MRSA arises most commonly, Davis says.

Researchers like Davis get concerned when multidrug-resistant strains of MRSA pop up. These strains resist other antibiotics in addition to β-lactams, making it even more challenging to treat infections.

Scientists in the Netherlands discovered one of these superstrains of MRSA in pigs in 2005. This particular strain had the gene for tetracycline resistance and, in some cases, a gene for zinc resistance, Davis says. It spread throughout Europe, Canada, and the midwestern US. Danish scientists later reported a link between the bacteria in pigs and feed supplemented with tetracycline or zinc (Vet. Microbiol. 2011, DOI: 10.1016/j.vetmic.2011.05.025).

Both tetracycline and zinc in feed may have contributed to the strain’s surviving in animals and therefore being available to infect exposed people—who can then transmit it to other people, Davis says.

Zinc is not the only metal associated with more virulent strains of S. aureus. Researchers recently reported a correlation between exposure to lead and greater detection of MRSA (Environ. Health 2018, DOI: 10.1186/s12940-017-0349-7).

Davis and colleagues are investigating how disinfectants and other chemicals found in the home can affect MRSA. The work is starting to show that disinfectants might exert selective pressure on S. aureus, leading to multidrug-resistant strains.

Next on Davis’s list are phthalates, which are used as plasticizers in many plastics, inks, paints, and other consumer products and have been linked to respiratory diseases such as asthma. S. aureus, too, is a driver of inflammatory, noncommunicable disease in people; some strains of the bacteria produce superantigens that affect the immune system. Davis plans to analyze bacterial, fungal, and chemical elements in archived dust samples from several studies related to asthma. “I’m trying to explore the potential for phthalates to exert selective pressure on the microbial communities in the home and in the child” to promote disease, she says.

Complex interactions

The interplay between exposure to chemicals in the environment and disease is complicated. Chemicals can affect both pathogens and the infected person. And they can influence the person directly or indirectly through the microbiome—the bacteria, fungi, viruses, and other microorganisms living in and on an individual or other environments. Studies have shown alterations in the microbiomes of mice exposed to arsenic, lead, manganese, PCBs, or the pesticide diazinon, Birnbaum noted at the January workshop. Some of the studies suggest that those microbiome changes induce metabolic effects that could lead to obesity, atherosclerosis, and neurological diseases, she said.

Environmental contaminants may also play a role in emerging public health threats. The National Toxicology Program (NTP) is currently investigating why so many pregnant women in Brazil infected with the Zika virus in 2015 and 2016 gave birth to babies with microcephaly, a birth defect in which the baby’s head is smaller than expected.

“The severity of the brain effects of Zika in Brazil may have been related to coexposure to environmental pesticides in use there,” Birnbaum said. Birnbaum serves as director of the NTP as well as the National Institute of Environmental Health Sciences. NTP scientists are investigating the larvicide pyriproxyfen, which is used in drinking water in Brazil, Birnbaum noted. They did not see any effects in standard developmental, reproductive studies in rats and rabbits, but there might be an effect on the developing brain in studies in zebrafish, she said. The next step will be for the NTP to look at what happens when pyriproxyfen and the Zika virus are combined.

Evidence is growing that chemicals in the environment can alter how people and wildlife respond to pathogens. Wildlife, such as the seals in the Baltic and off the coast of Maine, serve as sentinels of exposure to toxic chemicals in the environment. “If we don’t learn from the critters around us, whether domestic or wildlife, even the plants, we are bound to ignore warning signs,” Birnbaum said.

Birnbaum and other participants at the workshop emphasized the need for more research on the effects of the environment combined with infectious diseases on human health. One particular challenge is finding funding for this type of research, which spans multiple disciplines. But such work is urgently needed, Birnbaum said, to inform public health policy.