Teachers' work- related non- literature- known building- related symptoms are also connected to indoor toxicity: A cross-sectional study

A previous study showed that classical building- related symptoms (BRS) were related to indoor dust and microbial toxicity via boar sperm motility assay, a sensitive method for measuring mitochondrial toxicity. In this cross- sectional study, we analyzed whether teachers’ most common work- related non- literature- known BRS (nBRS) were also associated with dust or microbial toxicity. Teachers from 15 schools in Finland completed a questionnaire evaluating 20 nBRS including general, eye, respiratory, hearing, sleep, and mental symptoms. Boar sperm motility assay was used to measure the toxicity of extracts from wiped dust and microbial fallout samples collected from teachers’ classrooms. 231 teachers answered a questionnaire and their classroom toxicity data were recorded. A negative binomial mixed model showed that teachers’ work- related nBRS were 2.9-


| INTRODUC TI ON
According to the World Health Organization, healthy indoor air is a human right. 1 However, poor indoor air quality is common, and building-related symptoms (BRS) have been reported worldwide for decades in different types of buildings. 2 Known BRS affect several organ systems, including the eyes, nose, skin, respiratory, and central nervous systems; poor indoor air quality can also cause general and mental health symptoms. [2][3][4] Microbial toxins indoors in moisture and mold-damaged buildings are suspected to cause some BRS. 5 In some reports of in vitro toxicity, differences have been found between moisture-damaged buildings and reference buildings. In schools with high levels of BRS, the inflammatory potential of deposited dust in human lung epithelial cell line A549 assay was statistically significantly higher than in control schools. 6 In offices, the inflammatory potential of the dusts above the floor in lung epithelial cell A549 assay was associated with central nervous system symptoms but not mucosal or skin symptoms. 7 In two moisture-damaged schools, inflammatory potential of indoor air particles and bioaerosols were examined before and after building renovation 8 ; in one school, a clear decrease in both TNFα and IL-6 levels was observed via the mouse RAW264.7 macrophage assay; in another school no differences in inflammatory potential were observed. In more recent work, the inflammatory potential of indoor air particles in mouse RAW264.7 macrophage assay was significantly lower as measured by TNFα and IL-6 after extensive renovation of a moisture-damaged school. 9 In the same study, inflammatory markers in teachers' nasal lavage samples were also lower after renovation, as were microbial concentrations in air samples. In a multicenter study in different climatic areas in Finland, the Netherlands and Spain, the inflammatory potential of school deposition dust was studied with mouse RAW264.7 macrophage cells. 10 Significant differences were found between countries, and in the pooled data, moisture-damaged schools had slightly higher, but not statistically significant, inflammatory potentials compared to reference schools. The toxicity of actively collected particle samples in the mouse RAW264.7 macrophage test did not differ significantly between moisture-damaged and control schools. 11 Boar sperm motility assay has been used to detect toxic microbes and dusts in moisture-and mold-damaged buildings associated with health complaints. [12][13][14][15][16][17][18][19][20][21] In a recent study, a clear temporal association was found between heavy occupational exposure to sperm-toxic dust, during renovation of a waterdamaged building, and a cluster of 21 new occupational asthma cases. 22 We previously showed that 20 literature-known BRS were associated with sperm-toxic dust and airborne microbes in teachers' working environments. 23 In the present study, we investigated whether teachers' other work-related nonliterature-known BRS (nBRS) are also linked to dust and microbial toxicity.

| Schools
A more detailed description of material and methods was previously published. 23 The Real Estate Department of Helsinki City chose 15 schools that represented different decades of construction, building technologies, and ventilation systems, some of which had been renovated. These schools are described more detailed earlier. 23 The schools were built between 1924 and 2004. Fourteen schools were between 2400 and 8300 m 2 and one school was 474 m 2 . Concrete was the main structural material for all the schools. Thirteen schools had a mechanical exhaust air system, while eight schools additionally had a mechanical supply air system. One building was renovated in 2009. Several moisture damage and indoor air studies had been conducted in eight schools, one study in one school, and no concerns about indoor air quality or moisture damage had been identified in six schools. These building-level data were not available to the research team during the research project.

| Teachers
Teacher eligibility for the study was based on working at least seven hours per week in the study school and for at least one year in the same principal classroom. Further, there had to be information available on his/her workplace and participants could not be pregnant at the time of survey. The symptoms of pupils in these schools were not studied.
We sent a questionnaire to all teachers in the 15 schools.
The questionnaire asked whether symptoms were present in the last 12 months, whether they were related to the workplace, and whether they were alleviated during holidays and aggravated during the school year. Demographic data included age, sex, smoking habits and the presence of atopy. The 20 most common work-related nBRS-symptoms which has not been published as building-relatedwere selected for further investigation. According to the Ethics Committee of Helsinki University Hospital, official approval was not required for this type of anonymous survey.

| Testing indoor samples with boar spermatozoa as biosensors
Two types of indoor samples (wiped dust and airborne microbes) were collected from each teacher's principal classroom. First team collected and analyzed health data, second team collected samples as previously described. 23 Dust samples were collected between Plates were incubated for 4-6 weeks, and all growing microbial biomasses were harvested. No colony-forming units were counted and no identification of species was performed. To process these two different types of samples, the wiped dust and microbial biomasses were extracted separately into ethanol and evaporated to dryness at 62°C. Then, the residues were re-dissolved in ethanol to a concentration of 10 mg dry weight per ml. 12 Next, boar spermatozoa were exposed to the extracts of wiped dust and microbial biomass samples for 3 days. 23 Then, we determined the lowest sample extract concentration at which ≥50% of the spermatozoa had lost motility in comparison with control vehicle (ethanol). 12,23 The half maximal effective concentration (EC 50 ) indicated the degree of toxicity (ie, the lower the EC 50 , the higher the toxicity).

| RE SULTS
The questionnaire was sent by email to all teachers at the 15 schools (n = 630); 464 recipients (74%) responded. 23 Of those, 231 teachers met the inclusion criteria with a complete questionnaire and at least one type of toxicity result (200 responders had microbial toxicity results and 169 responders had dust toxicity results). Table 1 shows the summary of the study population. The median age of responders was 43 years, 81.8% were women, 9.5% were current smokers, and 10.4% had atopy. Their median working time at the primary workplace was 22 hours per week. The median age of survey drop-outs was 44 years, 77.7% of them were women (missing data 3.9%), 9.4% were smokers, and 10.3% had atopy. Their median number of hours spent at the primary workplace was 20 h per week (missing data 41.2%). Table 2 shows the prevalence of the most common (prevalence over 5%) work-related symptoms in this dataset. In addition, the table indicates whether the symptom is reported in the literature as building-related (literature-known BRS) or not (non-literature-known BRS, nBRS). Table 2 does not include work-related literature-known BRS analyzed in previous work 23 with a prevalence of less than 5% in this dataset (fever, nose irritation, wheezing, exanthema, swollen eyelids, and difficulty concentrating). The 20 most common work-related nBRS included the following: three general symptoms (repeated or prolonged generalized feeling of sickness, decreased physical condition or performance, and indefinite feeling of thermoregulation failure); three throat symptoms (throat mucus and need to clear the throat, itching in the throat, and globus sensation); two other respiratory symptoms (sensation of pressure in the cheek or forehead, and getting out of breath easily); two eye symptoms (redness in the whites of the eyes, and eye discharge); three hearing symptoms (hearing impaired, difficulty distinguishing speech in noisy environments, and hypersensitivity to sound); three sleep symptoms (insomnia, difficulty falling asleep, and increased need for sleep); and four mental symptoms (depressiveness, irritability, anxiety, and decreased stress resistance). The median number of work-related nBRS per teacher was one symptom. Among the 231 teachers, 121 (52.4%) had at least one work-related nBRS and 32 (13.9%) had at least five work-related nBRS.
Dust toxicity was divided into three categories, and microbial toxicity was divided into two categories. The distribution of toxicities is shown in Table 3. There was no correlation between dust and microbial toxicities (Spearman's rho = 0.087, p = 0.31).
According to the negative binomial mixed model adjusted for age, gender, smoking status, and atopy (Model 1), the number of teachers' work-related nBRS was 2.9-fold higher (p = 0.024) when the dust sample EC 50 was 6 μg/ml, compared to symptoms in nontoxic classrooms (dust with EC 50 of 25 μg/ml or higher; Table 3).

TA B L E 4
Negative binomial mixed model 2 results show the relative ratios (RRs) of work-related non-literature-known BRS, in teachers exposed to materials with different toxicities, compared to those exposed to a reference material

| DISCUSS ION
Our results show that teachers' work-related non-literature-known BRS were linked to dust and microbial toxicity. This finding suggested that the spectrum of BRS is much broader than expected.
We evaluated the statistical effect of dust and microbial toxicity on teachers' 20 nBRS with a negative binomial mixed model adjusted for age, gender, smoking, and atopy (Model 1) or exposure time in the primary workplace (Model 2). We focused on collecting and analyzing individual and workstation-specific data instead of construction-and group-level data, and schools was used as a random effect in the mixed models. These symptoms, which we have called nBRS, have not been earlier linked to BRS. We found that the number of teachers' non-literature-known symptoms was significantly higher in the more toxic classrooms. When we earlier explored teachers' classroom toxicity and 20 literature-known BRS (like cough, wheezing, itchy eyes, stuffy nose, fatigue, headache) using Poisson regression, the RR values for work-related BRS were 2.8 for dust toxicity and 1.8 for microbial toxicity. 23 When the relationship between teachers' classroom toxicity and the presence of work-related nBRS were analyzed in this study using the negative binomial mixed model, the corresponding RR values were 2.9 for dust toxicity and 1.8 for microbial toxicity.
It was interesting to note that also teachers in toxic classrooms had significantly more nBRS that were alleviated during school holidays compared to teachers in less-toxic classrooms, when toxicity was analyzed using two different sampling methods. The same phenomenon was observed earlier in the relationship between toxic classrooms and literature-known BRS. 23 In this population, some of the teachers spent a relatively short Recently, severe occupational asthma patients have been reported who prior to asthma diagnosis had a high-level, prolonged exposure to dust that showed toxicity in the boar sperm assay. 22 During an eight-month renovation period, a majority of asthma patients' workplace dust samples were markedly toxic; 21% were highly sperm-toxic (EC 50 ≤6 μg/ml) and 53% were clearly toxic (EC 50 7-12 μg/ml). After one year, in most cases, severe or moderate asthma persisted, even though the individuals had moved to another workplace. Those authors concluded that heavy, long-lasting toxic exposure (months) probably produced permanent consequences. It is important to note that toxic exposure was different in our present study. First, the exposure time was relatively short (workdays were less than 4.5 h, on average); second, the classrooms were tidy, without visible dust; and third, based on the boar sperm assay, the dust toxicity was clearly milder in classrooms (35% with EC 50 ≤12 μg/ ml), compared to the environment tested in the asthma cases (74% with EC 50 ≤12 μg/ml). 22 These differences might explain why the teachers' symptoms were reversible (alleviated during vacations), while the work-related asthma symptoms persisted even two years later (20/21 patients [95%] requiring regular asthma medication).
Motility of boar spermatozoa is dependent on the energy produced by mitochondria, and boar sperm motility inhibition assay is a sensitive response-based method for detecting mitochondrial toxicants. 20,24,25 This method is suitable for large-scale field work and has been actively used for decades to detect toxigenic microbes and unknown toxins in moisture-damaged buildings. 14,25 Various toxins produced by microbes isolated from damaged buildings, including valinomycin, cereulide, amylosin, stephacidin A and B, fusaricidins, several peptaibols, ophiobolins, chaetoglobosins, and communesins, are known to inhibit boar sperm motility. [12][13][14][15][16][17][18][19]21 In addition to microbial toxins, environmental pollutants, consumer chemicals, tobacco smoke, and particulate matter are known to impair mitochondrial function. [26][27][28][29] Normal mitochondrial function and signaling are prerequisites for undisturbed physiological processes in humans, and mitochondrial dysfunction has been found to be associated with the pathophysiology of many common diseases, as well as aging. 30,31 In asthma and chronic obstructive pulmonary disease (COPD), dysfunction of the airway mucosa and smooth muscle mitochondria has been found to be central to pathophysiology. [32][33][34] Mitochondrial damage causes local and systemic inflammation through both innate and adaptive immunity. [35][36][37] The above findings have raised suspicion that environmental mitochondrial toxic exposure is one of the major causes of these lung diseases. 38,39 These findings also generate a hypothesis that in addition to asthma, 22 40 On the other hand, the microbial toxicity is originated from viable toxigenic microbes in indoor air. The toxicities of the dust and microbial samples were not correlated, which further supported the idea that they represented different phenomena.
Our study had some limitations. Of the 630 teachers in the original study population, 74% responded and 50% dropped out for different reasons. 23 In addition, data related to exposure in other areas visited by the study participants (eg, school, home, previous work places) were not available. Furthermore, an inherent limitation of our study design was that it was not possible to estimate the exact amount of teacher exposure using our sampling methods. We were also unable to assess the effect of ventilation differences on exposure and symptoms. However, we noted that, although we only tested the teachers' present principal classrooms, we found a significant association between the symptoms and the measured toxicity in settled dust and airborne microbial biomass.

| CON CLUS IONS
The spectrum of teachers' work-related BRS appeared to be broader than generally thought. The frequency of work-related non-literature-known BRS, which were alleviated during school holidays, was markedly higher among teachers exposed to classrooms containing toxic dusts or microbes compared to participants who worked in classrooms with non-toxic samples.
Non-literature-known BRS were almost as common compared to literature-known BRS.
Work-related symptoms were strongly associated with indoor toxicity measured as inhibition of boar sperm motility, a sensitive detector of mitochondrial toxicity. Thus, further studies are needed to clarify whether there is an association between BRS and mitochondrial function of people exposed to sources of indoor toxicity.

ACK N OWLED G EM ENTS
We acknowledge grants from the Finnish Work Environment Fund