Removing the environmental stress with complete cooling (air-conditioning) is clearly the most effective method for alleviating the detrimental effects of occupational heat strain on physiological strain and work performance, as it effectively removes the environmental source of heat stress. Air-conditioning is, however, highly energy intensive, thereby incurring a significant economic cost , greenhouse gas emissions when energy supplies comes from non-renewable sources which further contributes to pollution related fatalities , and directly worsens global warming and the urban heat island effect due to the hot exhaust air production . Moreover, air-conditioning cannot be used outdoors, nor in large factory settings. It is possible, however, to improve the efficacy of air-conditioning by cooling small break rooms to provide workers reprieve from the heat at specific intervals. Further, as renewable sources of energy become more prevalent, this option will become increasingly attractive, however, the environmental impacts of the production and waste of this technology across its lifespan continues to have tremendous environmental impact regardless of energy source or efficiency .
Chemical carcinogenesis appears to provide a satisfactory explanation of tumors induced by many chemicals such as aflatoxin , benzo[a]pyrene , and others where high levels of DNA adduct formation are observed. In the case of benzene, and its role in leukemogenesis, DNA adducts have been described  but little DNA binding has been observed in vivo in animal test species . Alternatively, it has been suggested that the benzene metabolite 1,4-benzoquinone may inhibit topoisomerase II and, thereby, inhibit the annealing of strand breaks in DNA resulting in mutations [81,82,83]. The use of topoisomerase II inhibitors as cancer chemotherapeutic agents often results in leukemia following remission of the earlier tumor. Topoisomerase II inhibition may play a role in the induction of benzene-induced leukemia. Another relatively unexplored potential impact of benzene metabolites that may impact on leukemogenesis is extensive covalent binding to proteins . The many signaling pathways described above may be subject to attack by reactive metabolites of benzene leading to inactivation of critical signaling proteins or their receptors.
In 1939, the Chongqing asbestos factory began its operation by using water transportation on the Yangzi River to carry in raw materials and ship out manufactured products. This factory contained workshops for raw materials, textiles (carding, spinning, weaving), asbestos cement, rubber, and friction materials (Figs. 1 and 2). Its dust concentration, but not fiber concentration, was measured by Sichuan University (formerly West China University of Medical Sciences) every 5 years. In the 1980s, major innovations to improve the workplace environment included the installation of local ventilation systems throughout the factory, but, beginning in the 1990s, the systems were no longer operated to cut costs. Additionally, a small hut with a scrubber for the final treatment of exhaust air had been turned into a storage area for dumping sacks.
Although the manager of the factory claimed that the only type of asbestos used was chrysotile, he could not totally reject the possibility of contamination with amphibole. Using transmission electron microscopy, Dr. Kohyama, who is an asbestos mineralogy expert, examined samples from two Sichuan asbestos mines whose products were mainly used in the factory and found that the tremolite contamination was less than 0.001% . Later, airborne fiber samples were collected from the workplace and analyzed using scanning electron microscopy; only a small proportion of tremolite was found [2, 3].
At one point, the factory employed approximately 1200 individuals and, as with many Chinese companies, worker residences were on the same premises. Workers went back to their home residence during their lunch break, dined with their families in the evening, and family members often went to the factory to use the bathrooms commonly used by workers. Even after retirement, many workers stayed in the same residence and received a pension from the company. Incredibly, the present authors observed a child playing between factory buildings where suspended asbestos fibers were visible and shining in the sunlight. Thus, although it is beneficial to follow up with workers for epidemiological studies, their family members were also exposed to asbestos at work and in their personal lives throughout their life span and could have been affected as well.
The production of the factory severely decreased during the Cultural Revolution but increased in the late 1980s due to increased exports to Japan and Europe following restrictions that forbade the manufacturing of asbestos products in these countries. During the 1990s, production sharply decreased again, particularly in the cement and textile workshops, and shifted toward information technology-related areas such as electronic circuit cards. Traditionally, in this factory, the retirement age for women was 10 years earlier than that of men and workforce reductions were accomplished mainly by decreasing the number of female workers. The present author has published a number of epidemiological studies on the asbestos-related diseases of workers in this factory over the last two decades [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16]. Using these data, the present review will summarize important findings obtained from this factory with a focus on several issues that remain controversial in the field, namely, the amphibole hypothesis, the chrysotile/tremolite paradox, and the textile mystery.
Unlike lung cancer (bronchogenic carcinoma), the multiplicative interaction (synergistic effect) between asbestos exposure and cigarette smoking has not been observed for malignant mesothelioma . In the Chongqing factory, there were two confirmed cases of malignant mesothelioma, one was pleural and one was peritoneal, and one suspected case (unconfirmed by histopathology) of plural mesothelioma. The proportion of cases with mesothelioma compared to those with lung cancer in this factory was relatively small compared with other studies , which may suggest the important role of smoking in the causal mechanisms of lung cancer among asbestos-exposed populations.
In the Chongqing factory, the prevalence of smoking in asbestos workers was high. If smoking contributes to asbestos-related lung cancer, then it should be included as a consideration in the amphibole hypothesis. It is possible that asbestos causes cancer through multiple mechanisms and the absence of a synergistic effect between asbestos and cigarette smoking in the incidence of mesothelioma suggests that asbestos may cause mesothelioma without other carcinogen or with carcinogens other than cigarette smoking. On the other hand, in relation to lung cancer, asbestos may act with cigarette-derived chemical carcinogens. The production of reactive oxygen metabolites (ROMs), such as hydroxyl radicals, by the interaction between asbestos and phagocytic cells is well recognized, and thus, the mechanisms underlying ROM production induced by various types of asbestos in conjunction with phagocytic cells were investigated [21,22,23]. Although the potency and mechanism of each type of asbestos may not be the same, both amphibole and serpentine (chrysotile) asbestos types produce ROMs. This can relate to the carcinogenic mechanisms of asbestos, and thus, asbestos of any type can cause cancer.
Recently, our research group published the results of a fiber size analysis of airborne samples collected from each workshop in the Chongqing factory . Scanning electron microscopy was used to measure the lengths and widths of the fibers and identify the fiber types. This fiber type-specific size analysis revealed that tremolite was overrepresented among the total fibers even though it was estimated to be less than 1.5% at its highest concentration. Because tremolite is a natural contaminant of chrysotile and the proportion of tremolite in chrysotile may differ according to feedstock, it is not possible to absolutely avoid contamination with tremolite, i.e., despite industry claims, it is not possible to recognize natural chrysotile as non-carcinogenic due to its unavoidable contamination with amphibole. Taken together with the possible multiple mechanisms of asbestos carcinogenicity, these findings indicate that the amphibole hypothesis, which states that amphibole can cause cancer but chrysotile does not, cannot be accepted. Thus, as the International Agency for Research on Cancer (IARC) assessed, all forms of asbestos including chrysotile should be regarded carcinogenic to humans (Group 1) .
The second controversial issue involves the different asbestos fiber types present in the work environment versus those in the lungs of workers from workplaces exclusively using chrysotile. The vast majority of asbestos in the workplace was chrysotile whereas almost all of the fibers identified in lung tissues were tremolite with only exceptional instances of chrysotile . This odd contrast, which is known as the chrysotile/tremolite paradox , has been observed in several studies [25, 26]. For example, McConnochie et al.  conducted a fiber type analysis using the lung tissues of patients with mesothelioma and those of local sheep from around a chrysotile mine in Cyprus. A sizable proportion of tremolite that differed from the original composition found in the mine was identified in both the human and sheep lung tissues. A recent report from our research group proposed several possible mechanisms that may explain the paradoxically high levels of tremolite in these lung tissues despite the chrysotile-dominant work environment with only exceptional instances of tremolite . The first explanation relates to the curly shape of the chrysotile fiber that may be more likely to become entrapped in the narrow collider of the airway, which has many bifurcations in the bronchi and bronchioles. In contrast, the straight-shaped tremolite fibers can flow parallel with the air stream and penetrate to the deeper portion of the lung. Furthermore, the aerodynamic properties of the curly-shaped chrysotile fibers may differ from those of the straight-shaped tremolite fibers during the handling process of asbestos in a textile factory. During the raw material process in the factory, preparation of the material by mixing it with vibration and recycling may lead to the selective accumulation of curly chrysotile fibers in the air. Additionally, this may increase the proportion of chrysotile to tremolite in the ambient air from the original material. The accumulation of tremolite in lung tissue may also occur due to its biopersistence whereas chrysotile is rapidly cleared away . Compared with tremolite, chrysotile becomes easily bloated and dissolves in lung fluid, which may result in earlier leaching and clearance from the lung. 2b1af7f3a8