Infectious diseases caused by enveloped viruses, such as influenza, severe acute respiratory syndrome (SARS), and Middle East respiratory syndrome (MERS), cause thousands of deaths and billions of dollars of economic losses per year. Studies have found a relationship among temperature, humidity, and influenza virus incidence, transmission, or survival; however, there are contradictory claims about whether absolute humidity (AH) or relative humidity (RH) is most important in mediating virus infectivity. Using the enveloped bacteriophage Phi6, which has been suggested as a surrogate for influenza viruses and coronaviruses, we designed a study to discern whether AH, RH, or temperature is a better predictor of virus survival in droplets. Our results show that Phi6 survived best at high (>85%) and low (<60%) RHs, with a significant decrease in infectivity at mid-range RHs (∼60 to 85%). At an AH of less than 22 g · m−3, the loss in infectivity was less than 2 orders of magnitude; however, when the AH was greater than 22 g · m−3, the loss in infectivity was typically greater than 6 orders of magnitude. At a fixed RH of 75%, infectivity was very sensitive to temperature, decreasing two orders of magnitude between 19°C and 25°C. We used random forest modeling to identify the best environmental predictors for modulating virus infectivity. The model explained 83% of variation in Phi6 infectivity and suggested that RH is the most important factor in controlling virus infectivity in droplets. This research provides novel information about the complex interplay between temperature, humidity, and the survival of viruses in droplets.IMPORTANCE Enveloped viruses are responsible for a number of infectious diseases resulting in thousands of deaths and billions of dollars of economic losses per year in the United States. There has been a lively debate in the literature over whether absolute humidity (AH) or relative humidity (RH) modulates virus infectivity. We designed a controlled study and used advanced statistical modeling techniques specifically to address this question. By providing an improved understanding of the relationship between environmental conditions and virus infectivity, our work will ultimately lead to improved strategies for predicting and controlling disease transmission.

Legionella is a genus of pathogenic Gram-negative bacteria responsible for a serious disease known as legionellosis, which is transmitted via inhalation of this pathogen in aerosol form. There are two forms of legionellosis: Legionnaires' disease, which causes pneumonia-like symptoms, and Pontiac fever, which causes influenza-like symptoms. Legionella can be aerosolized from various water sources in the built environment including showers, faucets, hot tubs/swimming pools, cooling towers, and fountains. Incidence of the disease is higher in the summertime, possibly because of increased use of cooling towers for air conditioning systems and differences in water chemistry when outdoor temperatures are higher. Although there have been decades of research related to Legionella transmission, many knowledge gaps remain. While conventional wisdom suggests that showering is an important source of exposure in buildings, existing measurements do not provide strong support for this idea. There has been limited research on the potential for Legionella transmission through heating, ventilation, and air conditioning (HVAC) systems. Epidemiological data suggest a large proportion of legionellosis cases go unreported, as most people who are infected do not seek medical attention. Additionally, controlled laboratory studies examining water-to-air transfer and source tracking are still needed. Herein, we discuss ten questions that spotlight current knowledge about Legionella transmission in the built environment, engineering controls that might prevent future disease outbreaks, and future research that is needed to advance understanding of transmission and control of legionellosis.

We hypothesized that the interrupted corrosion control and associated release of iron, nutrients, and depleted chlorine residual in the distribution system would lead to high levels of Legionella. A tap water survey conducted throughout Flint in August and October 2015 confirmed Legionella pneumophila in two hospitals (mean of 1890 ± 2220 gene copy numbers/mL, 48% positivity), but not small single-story buildings. The hospitals frequently had optimal Legionella growth temperatures and were located in high-water age zones of the distribution system (3 to >6 days). Relatively high concentrations of iron were present (mean of 51.0 ± 37.2 ppb), and Cl 2 residual was sporadic (mean of 0.700 ± 0.775 mg/L) throughout the Flint distribution system. This study addresses knowledge gaps linking legionellosis outbreaks to changes in municipal water quality and distribution system operation.

We hypothesize that the increase in reported Legionnaires' disease from June 2014 to November 2015 in Genesee County, MI (where Flint is located) was directly linked to the switch to corrosive Flint River water from noncorrosive Detroit water from April 2014 to October 2015. To address the lack of epidemiological data linking the drinking water supplies to disease incidence, we gathered physiochemical and biological water quality data from 2010 to 2016 to evaluate characteristics of the Flint River water that were potentially conducive to Legionella growth. The treated Flint River water was 8.6 times more corrosive than Detroit water in short-term testing, releasing more iron, which is a key Legionella nutrient, while also directly causing disinfectant to decay more rapidly. The Flint River water source was also 0.8-6.7 °C warmer in summer months than Detroit water and exceeded the minimum Legionella growth temperature of 20 °C more frequently (average number of days per year for Detroit was 63 versus that for the Flint River, which was 157). The corrosive water also led to 1.3-2.2 times more water main breaks in 2014-2015 compared to 2010-2013; such disruptions have been associated with outbreaks in other locales. Importantly, Legionella spp. and Legionella pneumophila decreased after switching back to Detroit water, in terms of both gene markers and culturability, when August and October 2015 were compared to November 2016.

To examine the impact of environmental factors on Legionella in drinking water distribution systems, the growth and survival of Legionella under various conditions was studied. When incubated in tap water at 4 °C, 25 °C, and 32 °C, L. pneumophila survival trends varied amongst the temperatures, with the stable populations maintained for months at 25 °C and 32 °C demonstrating that survival is possible at these temperatures for extended periods in oligotrophic conditions. After inoculating coupons of PVC, copper, brass, and cast iron, L. pneumophila colonized biofilms formed on each within days to a similar extent, with the exception of cast iron, which contained 1-log less Legionella after 90 days. L. pneumophila spiked in a model drinking water distribution system colonized the system within days. Chlorination of the system had a greater effect on biofilm-associated Legionella concentrations, with populations returning to pre-chlorination levels within six weeks. Biofilms sampled from drinking water meters collected from two areas within central Arizona were analyzed via PCR for the presence of Legionella. Occurrence in only one area indicates that environmental differences in water distribution systems may have an impact on the survival of Legionella. These results document the impact of different environmental conditions on the survival of Legionella in water.