Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Comparison of stainless and mild steel welding fumes in generation of reactive oxygen species

Comparison of stainless and mild steel welding fumes in generation of reactive oxygen species Background: Welding fumes consist of a wide range of complex metal oxide particles which can be deposited in all regions of the respiratory tract. The welding aerosol is not homogeneous and is generated mostly from the electrode/wire. Over 390,000 welders were reported in the U.S. in 2008 while over 1 million full-time welders were working worldwide. Many health effects are presently under investigation from exposure to welding fumes. Welding fume pulmonary effects have been associated with bronchitis, metal fume fever, cancer and functional changes in the lung. Our investigation focused on the generation of free radicals and reactive oxygen species from stainless and mild steel welding fumes generated by a gas metal arc robotic welder. An inhalation exposure chamber located at NIOSH was used to collect the welding fume particles. Results: Our results show that hydroxyl radicals ( OH) were generated from reactions with H O and after exposure 2 2 to cells. Catalase reduced the generation of OH from exposed cells indicating the involvement of H O .The 2 2 welding fume suspension also showed the ability to cause lipid peroxidation, effect O consumption, induce H O 2 2 2 generation in cells, and cause DNA damage. Conclusion: Increase in oxidative damage observed in the cellular exposures correlated well with OH generation in size and type of welding fumes, indicating the influence of metal type and transition state on radical production as well as associated damage. Our results demonstrate that both types of welding fumes are able to generate ROS and ROS-related damage over a range of particle sizes; however, the stainless steel fumes consistently showed a significantly higher reactivity and radical generation capacity. The chemical composition of the steel had a significant impact on the ROS generation capacity with the stainless steel containing Cr and Ni causing more damage than the mild steel. Our results suggest that welding fumes may cause acute lung injury. Since type of fume generated, particle size, and elapsed time after generation of the welding exposure are significant factors in radical generation and particle deposition these factors should be considered when developing protective strategies. Background Although full-time welders may be easier to track, the There are approximately 390,000 full-time welders in profession has a large number of part-time, small shop the United States [1] and an estimated 5 million persons welders who may also be exposed as well as others occupationally exposed to welding fumes worldwide. working in the vicinity of the welding activities. These The welding process which joins materials by causing part-time and cross workplace exposures make the coalescence using a filler material, usually wire, to form effects difficult to monitor. Welding is frequently carried a molten pool which then cools bonding the surfaces out in areas with poor ventilation such as ship hulls, together. During this process an occupational exposure metal tanks, or pipe and crawl spaces leading to a can occur through inhalation of the fume and particles. greater potential for exposure. Welding fumes have been demonstrated to cause toxicity among exposed workers [2,3] Welding fumes consist of a wide range of metal * Correspondence: [email protected] Pathology and Physiology Research Branch, Health Effects Laboratory oxide particles, including iron, manganese, chromium, Division, National Institute for Occupational Safety and Health, Morgantown, and nickel, which are generated mostly from the WV, USA © 2010 Leonard et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 2 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 electrode/wire feed [4]. Two of the major feed wire continuous weld (figure 1). We used this automated sys- types which are used in the welding process are mild tem to collect welding fume from both SS wire and MS steel (MS) and stainless steel (SS). wire while using an arc welding process similar to that Inhalation of the fume has been related to bronchitis reported previously [22]. Differences in toxicity between [5], metal fume fever, occupational asthma [6] cancer mild steel and stainless steel fume have been previously and possible increases in lung tumorigenicity [7,8], observed [31,32]. Although, the reasons for these differ- suppression of lung defenses [9,10], and functional ences are not fully understood, some studies have indi- changes in the lung [11-14]. Investigations have also cated metal content in the weld wire as being the cause shownanincreaseinROS production afterwelding [33-35]. fume generation [15-17] and initiation of downstream The goal of our study was to access differences in mediators, such as HO-1, VEGF, and MAP kinases reactive oxygen species (ROS) generation during the [18-20]. The generated fume ranges in size and can be welding process. This study will investigate the effects of deposited throughout the respiratory tract [4,21-23]. weld wire type, particle size, surface area and time after Previous studies have demonstrated the impact of par- generation of fume on the potential generation of ROS ticle size and surface area on pulmonary effects of and its downstream effects. The observations reported inhaled toxicants [24]. here hope to elucidate the mechanism behind some of The effects seen from exposure to welding fume are the biological effects observed in occupational exposures under investigation but not well understood nor are the to welding fume. mechanisms behind potential toxicity from fume expo- sure. However, the act of welding causes the generation Results of unstable metal oxides due to the energy at the point Particle morphology and characteristics of the weld leading to an uncommon form of first expo- Electron micrographs of welding fume sample collected sure to newly formed unstable and potentially more on MOUDI filters (Figure 2) showed large agglomerated reactive particles, similar to that seen in sandblasting chains of particles linked together which not only [16,25-27]. Biological effects of welding fume exposure formed larger aggregate particles but greatly increased have been investigated employing cellular [28,29] and surface area. Particle sizes are not related to one single animal models [30] using a welding exposure system large particle but many smaller spherical particles linked located at NIOSH Health Effects Laboratory Division. as a chain which by its form creates a much greater sur- This system can simulate real workplace exposures and face area than would ordinarily be associated with a sin- allows the collection of fresh welding fume from a gle spherical particle of this size. This greater surface Figure 1 Schematic of the welding fume generation system. Welder operated from screened computer control room as welding occurs on a table. Generated fume collected immediately off the surface and passed through a heat exchanger then collected in the MOUDI particle collector or midget impinger. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 3 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Figure 2 Electron micrographs of two of the 15 filters used in the MOUDI and nano-MOUDI.A)shows stage4(3.2 μmcut-off)larger particles and the formation of agglomerated chains of particles. B) shows stage 12 (0.032 μm cut-off) smaller individual particles and less chain formation. area allows much more reaction surface with the gener- contained Fe, Mn and Cu. However, stainless steel also ated fumes. contained the transition metals Cr (20.2 ± 1.52% bw) Most particles, calculated by mass, generated in our and Ni (8.76 ± 0.18% bw) while mild steel showed a welding fume system were between 0.56 and 0.1 μmin small amount of Si (2.75 ± 0.28% bw) mean diameter as demonstrated in figure 3. This size range deposits mostly in the alveolar and bronchiolar Free radical generation and ROS regions in the lung; however, particles were found at all Free racial generation was examined using three sepa- size ranges measured using the MOUDI. Note that even rate systems; sized fume reaction with hydroxyl radical though grease was not used on the MOUDI stages in precursor H O , direct bubbling of whole fume through 2 2 this study, the profile of the mass collected on the stages solution containing H O and the spin trap DMPO, and 2 2 was similar to that obtained when the grease was used sized fume effects on exposed cells. [22], demonstrating that the phenomenon of potential Figure 4 shows individual filter sized welding fume particle bounce, if any, was minimal. reacting with hydrogen peroxide to measure generation Table 1 shows the elemental analysis of both stainless of hydroxyl radical ( OH). Strong radical generation is and mild steel wire which indicated that each wire observed at the 1 hour post collection time in both the Figure 3 Mass concentration trapped on each filter of the MOUDI. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 4 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Table 1 Stainless steel versus mild steel elemental sizes. When the OH radical signal was adjusted for par- analysis ticle weight/filter, mass normalized radical activity, it Stainless Steel was observed that the smaller particles had increased radical generation potential per unit weight, figure 5. Metal μg/sample Weight % of metal Stainless steel showed higher generation of OH radicals Fe 1207 ± 161 57 ± 1.28 at every time point (except 5.6 μm) with significantly Cr 427.5 ± 69.1 20.2 ± 1.52 higher generation;41%,57% and59% atsizes0.1 μm, Mn 295.0 ± 48.4 13.8 ± 0.45 0.056 μm, and 0.032 μm respectively. Addition of defer- Ni 185 ± 24.0 8.76 ± 0.18 oxamine, a metal chelator showed a decrease in radical Cu 3.30 ± 0.492 0.155 ± 0.004 signal strength in a concentration dependent manner Mild Steel with a complete abolishment of the DMPO/ OH signal Fe 776 ± 9.8 80.6 ± 0.17 at 2 mM deferoxamine. Additon of catalase, an H O 2 2 Mn 142 ± 2.0 14.7 ± 0.11 catalyst, to the cellular reactions decreased the radical signal as well (data not shown). Si 26.6 ± 2.9 2.75 ± 0.28 Figure 6 shows the results of total fumes drawn Cu 17.2 ± 0.20 1.79 ± 0.02 directly from the welding surface and bubbled through PBS in the presence of H O and DMPO, to act as a 2 2 stainless steel and mild steel samples. The number of spin trap. Results demonstrate the total sample potential radicals generated is reduced over time as shown at the to generate OH radicals when reacted with H O . Stain- 2 2 24 hour and 1 week post generation. Using the 0.32 μm less steel fumes showed a significantly higher generation mean aerodynamic diameter as a reference; stainless of OH radicals than was an equal mass of fume gener- steel showed a 68% reduction in OH radical signal ated by mild steel. between1 hour to 24 hours post generation which The results of RAW 264.7 cellular exposure to weld- became a 77% reduction at the 1 week time point. Mild ingfumebysizegroup andtypeoffumeare shownin steel showed similar reduction over time with a 64% figure 7. Stainless steel fumes incubated with cells drop in OH signal from 1 hour to 24 hours and a 67% showed a significant increase at all three size groups drop at the 1 week time point. The strongest OH radi- when compared to mild steel fumes. There was also an cal signals were measured at the 0.32 μmto0.056 μm increase in radical production within each steel type as size ranges; however, it should be noted this closely cor- particle size decreased. It should also be noted that the responds to the observed higher mass collected at those Cr (V) electron spin resonance signal (figure 7 inset) Figure 4 The generation of short-lived OH radicals upon reaction of H O with individual filter sizes using different wire type and 2 2 time periods after generation of fume. Open symbols represent stainless steel while filled symbols are mild steel wire. ESR spectrum recorded 3 min after reaction was initiated in PBS (pH 7.4), 1 mM H O and 100 mM DMPO. ESR settings were; center field, 3385 G; scan width, 100 G; 2 2 time constant, 40 msec; scans, 5; modulation amplitude, 1 G; receiver gain, 2.5 × 10 ; frequency, 9.793 GHz; and power, 63 mW. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 5 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 steel also showed a significant increase over mild steel when comparing fine and ultrafine size groups. H O production 2 2 Figure 9 shows the effects of welding fume on H O 2 2 generation after RAW 264.7 cells were exposure to welding fume. All welding fume sizes and steel types at equal mass showed a significant rise in H O produc- 2 2 tion in exposed RAW 264.7 cells when compared to control. A significant difference was also observed between stainless steel and mild steel fumes in the fine and ultrafine size groups. A size dependent increase in H O production was also observed in both types of 2 2 Figure 5 Mass normalized radical activity, OH radical peak metals. heights of individual filters sizes adjusted for radicals/mg on 1 hour post generation samples. ESR settings were; center field, O consumption 3385 G; scan width, 100 G; time constant, 40 msec; scans, 5; A significant rise in oxygen consumption was observed modulation amplitude, 1 G; receiver gain, 2.5 × 10 ; frequency, 9.793 GHz; and power, 63 mW. Asterisks indicate a significant increase in in RAW 264.7 cells in all stainless steel size exposures stainless steel fume compared to mild steel fume (P < 0.05). and in the mild steel ultrafine size as shown in figure 10. A significant difference was observed between stain- less steel fumes and mild steel fumes at equal mass in was observed only in spectra from stainless steel fume the ultrafine size group. A size dependent increase was exposure to cells. also observed within steel types. Lipid peroxidation DNA damage Stainless steel fumes showed a significant increase in Comet assay results for DNA damage are shown in fig- lipid peroxidation in exposed RAW 264.7 cells com- ure11. Thedatademonstrates total groupedsizefilters pared to control at all three sizes groups at equal mass for both stainless and mild steel at equal mass showed a as seen in figure 8. Mild steel fumes caused a significant significant increase in DNA damage in exposed RAW increase at the ultrafine particle size group. Stainless 264.7 cells. Furthermore, the stainless steel fumes caused Figure 6 OH radical peak heights of midget impinger bubbled samples. Flow rate 3.80 L/min, 20 min collection time. Welding fume was bubbled directly through PBS (pH 7.4), 1 mM H O and 100 mM DMPO. ESR settings were; center field, 3385 G; scan width, 100 G; time 2 2 constant, 40 msec; scans, 1; modulation amplitude, 1 G; receiver gain, 2.5 × 10 ; frequency, 9.793 GHz; and power, 63 mW. Asterisks indicate a significant increase in stainless steel fume compared to mild steel fume (P < 0.05). Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 6 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Figure 7 Radicals generated from RAW 264.7 cells (1 × 10 /ml) exposed to grouped filter sample welding fume (250 μg/ml) for 10 min at 37°C in a shaking water bath. ESR settings were; center field, 3385 G; scan width, 100 G; time constant, 40 msec; scans, 5; modulation amplitude, 1 G; receiver gain, 2.5 × 10 ; frequency, 9.793 GHz; and power, 63 mW. Asterisks indicate a significant increase in radicals compared between steel types; (+) indicate significant difference between fume sizes. significantly more DNA damage when compared to mild potential cellular damage from ROS generation and to steel fumes. determine if the biological effects were size, time and surface area dependent. Our results determined that Discussion both stainless steel and mild steel welding fumes are Thepresent studywas undertaken to investigatepossi- able to generate OH radicals in a Fenton-like system. ble radical generation of fume generated from two of These OH radials are precursors and initiators of many the most common types of welding processes to access forms of ROS that can produce damage to cellular membranes, proteins and DNA as well as initiate further downstream damage and signaling associated with respiratory burst and inflammation. Our results also determined that freshly generated fume samples, 1 hour post, were significantly more reactive in generating OH radials than samples aged for 24 hours and 1 week. This reactivity is attributed to the change in transition states of the freshly generated metals that are produced during the welding process due to the high energy involved. Transition metals have the ability to accept and donate single electrons. Metals used in welding may temporarily attain a different and possibly more reactive, transition or valence state and in this changed transition state they are able to overcome the spin restriction on direct reac- Figure 8 Welding fume induced lipid peroxidation in tion of O with non-radicals. These reactions can lead incubation mixture containing 250 μg/ml welding fume 2 sample and 5 × 10 RAW 264.7 cells. Data presented are means to the generation of ROS and further damage to biologi- of ± S.D. for 4 sets of experiments. (*) indicate a significant increase cal systems. Addition of metal clelators and catalase in lipid peroxidation compared to control. (+) indicate significant confirmed the involvement of metals and H O in the 2 2 difference between metal types at the same sizes. (P < 0.05) radical generation observed. A consistent result Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 7 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Figure 9 H O production in incubation mixtures containing 250 μg/ml welding fume sample and 1 × 10 RAW 264.7 cells.Data 2 2 presented are means of ± S.D. for 4 sets of experiments. (*) indicate a significant increase in lipid peroxidation compared to control. (+) indicate significant difference between metal types at the same sizes. (P < 0.05) throughout our various ESR measurements (H O , cellu- their ability to cycle through transition states once in a 2 2 lar and direct whole fume) was the ability of stainless cellular system can lead to extensive and ongoing ROS steel fumes to generate higher amounts of free radicals production. This metal cycling ability was demonstrated than mild steel. Animal studies indicate that stainless in our investigation by the ESR spectra results showing steel welding fumes induce more lung inflammation and the production of Cr(V) in stainless steel fume exposed injury compared to mild steel fumes [17] Elemental ana- cells. This Cr(V) radical signal came from the reduction lysisofthe fumesshowedthatstainless steelcontained of Cr(VI) in the welding fume in a cellular system. This two transition metals not found in mild steel, chromium Cr(VI) ® Cr(V) system has been researched and shown and nickel. Both chromium and nickel have had exten- to be toxic and to responsible for the generation ROS sive research performed on their reactivity and abilities associated damage [37]. to produce ROS in biological systems [36]. These metals The relationship between free radical generation and have been shown to be highly toxic and result in gen- different sizes at equal mass was also investigated to eration of ROS and associated damage. Furthermore, determine the effect on free radical generation after exposure to H O . The greatest free radical generation 2 2 was observed when analyzing the ultrafine particles demonstrating that reactivity of the particles is depen- dent on available surface area with equal mass. Gener- ated welding fumes are long chains of agglomerated particles which are trapped on filters in size groups. However, the large chains are actually made up of many smaller primary particles linked together, creating a much higher surface area at equal mass than spherical particles that are more commonly observed in the envir- onment and workplace. The welding fume chains formed under the high heat conditions of welding acts synergistically to create fresh particles of a high surface area which generate reactive species. Freshly generated Figure 10 Oxygen consumptioninwelding fume stimulated welding fume has been previously reported to be more RAW 264.7 cells. Incubation mixtures contain 3 × 10 RAW 264.7 cells + 500 μg/ml welding fume sample. Data presented are means biologically active in an animal model [16]. of ± S.D. for 4 sets of experiments. (*) indicate a significant increase Welding fumes were also found to cause lipid perox- in oxygen consumption compared to control. (+) indicate significant idation, an indicator of cell membrane damage and difference between metal types at the same sizes. (P < 0.05) precursor for other radical generation systems. Lipid Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 8 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Figure 11 Comet assay results after incubation of RAW 264.7 cells for 1 h exposed to 250 μg/ml welding fume.Datapresented are means of ± S.D. for 4 sets of experiments. (*) indicate a significant increase in DNA damage compared to control. (+) indicate significant difference between metal types at the same sizes. (P < 0.05) peroxidation results in the release of lipid-derived radi- containing Cr causing significantly greater DNA . . . cals (R ,RO and ROO ) [38], which can lead to a cas- damage. In addition, H O and OH generated by cells 2 2 cade effect and release reactive iron [39] causing the exposed to welding fume may cause other cellular generation of more ROS. Lipid peroxidation and its damage via mechanisms associated with reactions effects have been found to cause DNA damage [40] initiated by ROS, for example, dG hydroxylation and and may function as tumor initiators [41]. Once again protein-DNA cross-links. Furthermore these ROS may the stainless steel fumes cause significantly greater also cause activation of nuclear transcription factors, damage to the cells than mild steel. This same trend, such as NF-B, over-expression of certain oncogenes as well as a trend in smaller size i.e. greater surface and induction of p53 mutation [44,45]. area was also demonstrated in H O generation, O 2 2 2 consumption and DNA damage in exposed RAW 264.7 Conclusions cells. H O generation and O consumption are indica- The significant increase in oxidative damage observed in 2 2 2 tors of a respiratory burst in the exposed cells creating the cellular exposures correlates well with the results as downstream effects and activation of inflammatory determined by ESR of OH generation in size and type pathways and signaling. The significantly higher DNA of welding fumes, indicating the influence of metal type damage observed in the stainless steel fumes may be and transition state on radical production as well as due to the different metals present. Hydroxyl radicals associated damage. Our results demonstrate that both generated with certain metals, such as nickel and cop- types of welding fumes are able to generate ROS and per, exhibit less reactivity with DNA. Possible reasons ROS-related damage over a range of particle sizes; how- for this non-reactivity include; that OH radicals are ever, the stainless steel fumes consistently showed a sig- generated within the domain of certain macromole- nificantly higher reactivity and radical generation cules and, therefore, are not able to exhibit significant capacity. The chemical composition of the steel had a reactivity and the structural contribution of metal significant impact on the ROS generation capacity with toward DNA-binding and metal interactions with the the stainless steel, but not mild steel, containing Cr and DNA [42,43]. The results from our study show that Ni causing more damage than the mild steel. Both mate- OH radicals generated from both welding fumes have rials contained Fe, a known Fenton radical producer, the potential to cause DNA damage with stainless steel Mn and Cu. Our results suggest that the marked Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 9 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 difference between the radicals produced in the two Electric), a water cooled arc welding torch (WC 650 types of fumes is due to the presence of Cr found in the amp, Lincoln Electric), a wire feeder that supplied the stainless steel. Cr has been shown in many investiga- wire to the torch at a programmed rate up to 300 tions to a cause a number of toxicities. It is also noted inches/min, and an automatic welding torch cleaner that that the smaller particle size of the fume the greater the kept thewelding nozzlefreeof debrisand spatter. Gas ROS potential at equal mass with the 0.180 - 0.032 μm metal arc welding was performed using a mild steel sizes showing the most reactivity in both types of weld- electrode (carbon steel ER70S-6, Lincoln Electric) or a ing fume which can penetrate deepest into the lung of stainless steel electrode (Blue Max E308LSi wire, Lin- an exposed welder. Our results further demonstrated coln Electric). Welding took place on A36 carbon steel that the freshly generated fumes were more reactive and plates for fume collection times of 20 min at 25 V and caused more oxidative damage than the aged particles, 200 amps. During welding, a shielding gas combination indicating that metal transition state also plays an of 95% Ar and 5% CO (Airgas Co., Morgantown, WV) important role in welding fume reactivity. Therefore, was continually delivered to the welding nozzle at an air our results suggest that welding fumes may cause acute flow of 20 L/min. lung injury. Since type of fume generated, particle size, Figure 1 shows an outline of the generation and col- and elapsed time after generation of the welding expo- lection system. The system is contained in three rooms; sure are significant factors in radical generation and par- the control room, the robotic arm welding fume genera- ticle deposition these factors should be considered when tor and the fume collection chamber. All three rooms developing protective strategies. were separated from each other. Aerodynamically size-selected aerosol samples were Methods collected with the Micro-Orifice Uniform Deposit Impac- Reagents tor (MOUDI), model # 110 and Nano-MOUDI MSP Chelex 100 resin was purchased from Bio-Rad Labora- Model #115, with rotator (MSP, Inc., Minneapolis, MN, tories (Richmond, CA, USA). Phosphate-buffered saline USA). This MOUDI/Nano-MOUDI arraignment pro- (PBS), (KH PO (1.06 mM), Na HPO (5.6 mM), NaCl vided a 15-stage research-grade cascade impactor. Each 2 4 2 4 (154 mM), pH 7.4), was purchased from Biowhittaker filter stage has a cutoff size with particles collected at Inc. (Walkersville, MD, USA). The PBS was treated with each stage being aerodynamically size-selected between Chelex 100 to remove transition metal ion contaminants. stage sizes. Cutoff sizes were; 18, 10, 5.6,3.2, 1.8, 1.0, 0.56, Dulbecco’s modified eagles medium (DMEM), 5,5- 0.32, 0.18, 0.10, 0.006, 0.03, 0.02, 0.01 μmand thefinal dimethyl-1-pyroline-oxide (DMPO), fetal bovine serum filter. Filters used for ESR and cellular exposures were (FBS), FeSO H O and penicillin/streptomycin were from Millipore Corp. (Billerica, MA, USA) 47 mm, 0.8 4, 2 2, purchasedfromSigma Chemical Company(St.Louis, μm, PVC model PVC0847600. PVC was selected because MO, USA). The spin trap, DMPO, was purified by char- it has previously been demonstrated to have no effect in coal decolorization and vacuum distillation and was free the free radical analysis. The MOUDI substrates are nor- of ESR detectable impurities. Quartz sample tubes were mally coated with grease to ensure adherence of depos- purchased from Wilmad Glass (Buena, NJ, USA). ited particles and to avoid bounce of large particles to lower stages of the impactor. However, grease can alter Cell culture the surface of collected aerosol particles and is not suita- RAW 264.7 mouse peritoneal monocytes were pur- ble for use in collecting samples for free radical analysis. chased from American Type Culture Collection (Rock- Therefore, the cascade impactor was operated without ville, MD). RAW 264.7 cells are commonly used and grease substrates to collect and fractionate the welding have been found to respond to particle exposure in a fume. Filters were collected for 20 min. Filters were then manner similar to primary alveolar macrophages splitinto3groups foreither 1 hour,24houror1week [46-49]. RAW 264.7 cells were cultured in DMEM with post generation analysis. 10% FBS, 2 mM L-glutamine, and 50 mg/ml pen/strep Filter suspensions, which were used for H O , lipid 2 2 at 37°C in a 5% CO incubator. Cells were split after peroxidation, oxygen consumption and DNA damage confluence approximately every 3 days. analysis, were prepared by splitting the filters into three size groups. For our study the size groups were defined Generation and collection of welding fumes as ultrafine (0.01 - 0.056 μm), fine (0.1 - 1.0 μm) and The welding fume generation system was similar to that coarse (1.8 - 18 μm) particles. The groups consisted of previously outlined [22]. Briefly, the generation system five filters each of which were placed in PBS. Pre- and consisted of a welding power source (Power Wave 455, post filter weights were used to prepare a final concen- Lincoln Electric, Cleveland, OH), an automated, pro- tration of 1 mg/ml welding fume suspension. The slurry grammable six-axis robotic arm (Model 100 Bi, Lincoln was centrifuged, and the fume suspension was decanted Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 10 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 from the filter pellet. A clean control suspension was due to its specificity and sensitivity. All ESR measure- also prepared at a ratio of 4 filters/1 ml PBS. ments were conducted using a Bruker EMX spectro- meter (Bruker Instruments Inc. Billerica, MA 01821, Welding particle morphology, transmission electron USA) and a flat cell assembly. Hyperfine couplings microscopy (TEM) were measured (to 0.1 G) directly from magnetic field Welding fume samples, collected separately from the separation using potassium tetraperoxochromate fumes used in free radial analysis, were collected at 30- (K CrO ) and 1,1-diphenyl-2-picrylhydrazyl (DPPH) as 3 8 minute intervals during 3 h of welding directly onto for- reference standards [51,52]. The relative radical con- mvar-coated TEM grids and viewed using a JEOL 1220 centration was estimated by multiplying half of the transmission electron microscope (JOEL, Inc.). peak height by (ΔH ) ,where ΔH represents peak- pp pp to peak width. The Acquisit program was used for Welding particle size distribution data acquisitions and analyses (Bruker Instruments Inc. Particle size distribution was determined by using a Billerica, MA 01821, USA). Micro-Orifice Uniform Deposit Impactor (MOUDI) model # 110 which is intended for the general purpose Radicals trapped by bubbler aerosol sampling, and a Nano-MOUDI (MSP Model Freshly generated total welding particle reactivity was 115) that is specifically designed for sampling aerosols measured using a midget bubbler (Ace Glass, Vineland in the size range down to 0.010 μm. NJ) containing a reaction mixture made up of H O (10 2 2 mM) DMPO (100 mM), and PBS to instantly trap gen- Welding particle metal content analysis erated radicals for ESR measurement. The midget bub- Mild and stainless steel welding particles were collected bler was attached to a collection tube with a flow rate of onto 5.0 mm polyvinyl chloride membrane filters in 37 1 L/min positioned approximately 18 inches from the mm cassettes during 30 min of welding. Metal analysis welding surface in order to directly collect the freshly samples were collected separately from the samples used generated fume. Fume was collected for 20 min. The for free radical analysis. The particle samples were reaction mixture was then immediately measured using digested and the metals analyzed by inductively coupled ESR in order to determine OH generation from freshly plasma atomic emission spectroscopy (ICP-AES) by generated whole fume. Clayton Group Services (A Bureau Vertis Company, Novi, MI) as coordinated with the Division of Applied Grouped filters Research and Technology (DART, Cincinnati, OH) Grouped filter suspensions, were used for H O , lipid 2 2 according to NIOSH method 7300 modified for micro- peroxidation, O consumption and DNA damage ana- wave digestion [50]. Metal content of blank filters also lysis. Suspensions were prepared by splitting the filters was analyzed for control purposes. into three size groups. For this study the size groups were defined as ultrafine (0.01 - 0.056 μm), fine (0.10 - Free radical measurements 1.0 μm) and coarse (1.8 - 18 μm) particles. The groups ESR spin trapping was used to detect short-lived free consisted of five filters each of which were placed in radical intermediates. Hydroxyl radicals were measured PBS and blended on ice into a fine slurry using a Tis- using the addition-type reaction of a short-lived radical sueTearor(BiospecProductsInc.Racine,WI).Pre- with a compound (spin trap) to form a relatively long- andpostfilter weightswereusedtoprepareafinal lived paramagnetic free radical product (spin adduct), suspension of 1 mg/ml welding fume. The slurry was which can then be studied using conventional ESR. For centrifuged, and the welding fume suspension was dec- hydroxyl radical measurements on individual filter anted from the filter pellet. A clean control suspension sizes, reactants were mixed in test tubes at a final was also prepared at a ratio of 5 filters/1 ml PBS. Sus- volume of 1.0mlofPBS in thepresenceof1 mM pension not used immediately was aliquoted and fro- H O . The reaction mixture was then transferred to a zen at -70°C. 2 2 quartz flat cell for ESR measurement. Experiments were performed at room temperature and under ambi- Lipid peroxidation ent air. The concentrations given in the figure legends Lipid peroxidation of RAW 264.7 mouse peritoneal are final concentrations. The intensity of the signal is monocytes was measured by using a colormetric assay used to measure the relative amount of short-lived for malondialdehyde (LPO-586 Oxis International Inc. radicals trapped, and the hyperfine couplings of the Portland, OR, USA). A reaction mixture contained var- spin adduct are characteristic of the original trapped ious size groups (refer to filter methods) of welding par- radicals. Spin trapping is the method of choice for ticle samples [250 μg/ml], H O (1 mM) and 1 × 10 2 2 detection and identification of free radical generation cells in a total volume of 1.0 ml PBS (pH 7.4). A Fenton Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 11 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 reaction, FeSO (1 mM), H O (1 mM) and 1 × 10 Cranberry Township, PA). A minimum of 50 cells were 4 2 2 cells, was also carried out as a positive control. The scored for each sample at 400× magnification. The dis- mixtures were exposed for 1 h in a shaking water bath tance between the edge of the head and the end of the at 37°C. The measurement of lipid peroxidation is based tail was measured using an automated image analysis on the reaction of a chromogenic reagent with malonal- system (Optimas 6.51, Media Cybernetics Inc., Silver dehyde at 45°C [53,54]. The absorbance of the supernate Spring, Md) [55,56]. was measured at 586 nm. Statistics H O production Data were expressed as mean ± standard error of the 2 2 H O production was monitored using a Bioxytech mean (SEM) (n = 4) for each group. One-way ANOVA 2 2 quantitative hydrogen peroxide assay kit (H O -560, test was performed using SigmaStat statistical software 2 2 Oxis International Inc. Portland, OR, USA). Measure- (Jandel Scientific, San Rafael, CA, USA) to compare the ments were made on a system containing 5 × 10 RAW responses between treatments. Statistical significance 264.7 mouse peritoneal monocytes/ml in pH 7.4 PBS was set at p < 0.05. and exposing them to the size groupings (refer to filter methods) of welding particle solution. Cells were Abbreviations exposed to welding particle solution [250 μg/ml], for 30 Defined in text. minutes in a 37°C incubator. Absorbance was monitored Acknowledgements at a wavelength of 560 nm using a Spectra Max 250 Disclaimer: The findings and conclusions of this paper have not been formally multi-well plate reader (Molecular Devices, Sunnyvale, dissemination by NIOSH and should not be construed to represent any agency CA, USA). determination or policy. Authors’ contributions O consumption SSL, BTC, DF and JMA contributed to the idea and design of the study. SSL, Oxygen consumption measurements were carried out BTC and SGS carried out welding generation and particle collection. SSL and AJK carried out all electron spin resonance and toxicity measurements. DSB using a Gilson oxygraph (Gilson Medical Electronics, carried out EM analysis of particles. All authors read and approved the final Middleton, WI). Measurements were made on a system manuscript. containing 3 × 10 RAW 264.7 mouse peritoneal mono- Competing interests cytes/mL and welding particle solution [500 μg/ml], in The authors declare that they have no competing interests. pH 7.4 phosphate buffer. The oxygraph was calibrated with media and equilibrated with known concentrations Received: 25 June 2010 Accepted: 3 November 2010 Published: 3 November 2010 of oxygen. References DNA damage, Comet assay 1. Bureau of Labor Statistics. [http://www.bls.gov/oes/current/oes514121. The Comet assay was performed using methods htm]. 2. Moulin JJ, Wild P, Haguenoer JM, Faucon D, De Gaudemaris R, Mur JM, described in the Trevigen (Gaithersburg, Md., U.S.A.) Mereau M, Gary Y, Toamain JP, Birembaut Y: A mortality study among assay kit. A typical reaction mixture contained welding mild steel and stainless steel welders. Br J Ind Med 1993, 50(3):234-43. particle solution [250 μg/ml], H O (1 mM), and 1 × 3. Simonato L, Fletcher AC, Andersen A, Anderson K, Becker N, Chang- 2 2 Claude J, Ferro G, Gérin M, Gray CN, Hansen KS: A historical prospective 10 RAW 264.7 mouse peritoneal monocytes cells study of European stainless steel, mild steel, and shipyard welders. Br J brought to a total volume of 1.0 mL in PBS (pH 7.4). Ind Med 1991, 48(3):145-54. Briefly, (all steps performed in the dark or low light con- 4. Zimmer AT, Biswas P: Characterization of the aerosols resulting from arc welding processes. J Aerosol Sci 2001, 32:993-1008. ditions) RAW264.7 mouse cells were exposed to the var- 5. Contreras GR, Chan-Yeung M: Bronchial reactions to exposure to welding ious sized welding fume and incubated in media at 37 ° fumes. Occup Environ Med 1997, 54(11):836-9. C for 1 h. The cells were washed (2×) with PBS and 6. Hannu T, Piipari R, Kasurinen H, Keskinen H, Tuppurainen M, Tuomi T: Occupational asthma due to manual metal-arc welding of special combined with LMAgarose, then pipetted onto a Comet stainless steels. Eur Respir J 2005, 26(4):736-9. slide. The slide was placed in a refrigerator for 30 min, 7. d’Errico A, Pasian S, Baratti A, Zanelli R, Alfonzo S, Gilardi L, Beatrice F, immersed in lysis solution, chilled for 60 min, and then Bena A, Costa G: A case-control study on occupational risk factors for sino-nasal cancer. Occup Environ Med 2009, 66(7):448-55. immersed in alkaline solution for 55 min. Slides were 8. Langård S: Nickel-related cancer in welders. Sci Total Environ 1994, 148(2- placed in a horizontal electrophoresis chamber for 40 3):303-9, Review. min at 300 mA. Slides were washed and SYBR green 9. Antonini JM, Taylor MD, Millecchia L, Bebout AR, Roberts JR: Suppression in lung defense responses after bacterial infection in rats pretreated with stain was added to each. Slides were visualized using different welding fumes. Toxicol Appl Pharmacol 2004, 200(3):206-18. fluorescence microscopy, with an image capturing sys- 10. Anderson SE, Meade BJ, Butterworth LF, Munson AE: The humoral immune tem (Olympus AX70 and sample PCI, Compix, response of mice exposed to manual metal arc stainless steel-welding fumes. J Immunotoxicol 2007, 4(1):15-23. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 12 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 11. Antonini JM, Lewis AB, Roberts JR, Whaley DA: Pulmonary effects of free radical production, DNA damage, and apoptosis induction. Mol Cell welding fumes: review of worker and experimental animal studies. Am J Biochem 2005, 279(1-2):17-23. Ind Med 2003, 43(4):350-60. 32. Edmé JL, Shirali P, Mereau M, Sobaszek A, Boulenguez C, Diebold F, 12. Sferlazza SJ, Beckett WS: The respiratory health of welders. Am Rev Respir Haguenoer JM: Assessment of biological chromium among stainless steel Dis 1991, 143(5 Pt 1):1134-48. and mild steel welders in relation to welding processes. Int Arch Occup 13. Solano-Lopez C, Zeidler-Erdely PC, Hubbs AF, Reynolds SH, Roberts JR, Environ Health 1997, 70(4):237-42. Taylor MD, Young SH, Castranova V, Antonini JM: Welding fume exposure 33. Antonini JM, Roberts JR: Chromium in stainless steel welding fume and associated inflammatory and hyperplastic changes in the lungs of suppresses lung defense responses against bacterial infection in rats. J tumor susceptible a/j mice. Toxicol Pathol 2006, 34(4):364-72. Immunotoxicol 2007, 4(2):117-27. 14. Sobaszek A, Boulenguez C, Frimat P, Robin H, Haguenoer JM, Edme JL: 34. Alexopoulos EC, Cominos X, Trougakos IP, Lourda M, Gonos ES, Acute respiratory effects of exposure to stainless steel and mild steel Makropoulos V: Biological monitoring of hexavalent chromium and welding fumes. J Occup Environ Med 2000, 42(9):923-31. serum levels of the senescence biomarker apolipoprotein j/clusterin in 15. Liu HH, Wu YC, Chen HL: Production of ozone and reactive oxygen welders. Bioinorg Chem Appl 2008, 420578. species after welding. Arch Environ Contam Toxicol 2007, 53(4):513-8. 35. Serageldin M, Reeves DW: Development of welding emission factors for 16. Antonini JM, Clarke RW, Krishna Murthy GG, Sreekanthan P, Jenkins N, Cr and Cr(VI) with a confidence level. J Air Waste Manag Assoc 2009, Eagar TW, Brain JD: Freshly generated stainless steel welding fume 59(5):619-26. induces greater lung inflammation in rats as compared to aged fume. 36. Leonard SS, Harris GK, Shi X: Metal-induced oxidative stress and signal Toxicol Lett 1998, 98(1-2):77-86. transduction. Free Radic Biol Med 2004, 37(12):1921-1942. 17. Taylor MD, Roberts JR, Leonard SS, Shi X, Antonini JM: Effects of welding 37. Leonard SS, Wang S, Castranova V, Vallyathan V, Shi X: Role of Molecular fumes of differing composition and solubility on free radical production Oxygen in the Generation of hyrdroxyl and superoxide anion radicals and acute lung injury and inflammation in rats. Toxicol Sci 2003, during enzymatic Cr (VI) reducation and its implication to (CrVI)-induced 75(1):181-91. carcinogenisis. J Environ Path Toxicol Oncol 2000, 19:49-60. 18. Stark M, Zubareb J, Jacovovitz R, Schwartz Y, Lerman Y, Grinberg N, 38. Shi X, Dalal NS, Vallyathan V: ESR evidence for the hydroxyl radical Fireman E: HO-1 and VEGF gene expressions are time dependant during formation in aqueous suspension of quartz particles and its possible exposure to welding fumes. Cytokine 2009, 46(2):290-5. significance to lipid peroxidation in silicosis. J Toxicol Environ Health 1988, 19. Seel EA, Zaebst DD, Hein MJ, Liu J, Nowlin SJ, Chen P: Inter-rater 25:237-245. agreement for a retrospective exposure assessment of asbestos, 39. Vladimirov YA: In Free Radicals, Aging and Degenerative Diseases. Edited by: chromium, nickel and welding fumes in a study of lung cancer and Johnson JE Jr, Walford R, Harman D, Miquel J. New York: Liss; 1986:141-195. ionizing radiation. Ann Occup Hyg 2007, 51(7):601-10. 40. Vaca CE, Wilhelm J, Harms-Ringdahl M: Interaction of lipid peroxidation 20. Tessier DM, Pascal LE: Activation of MAP kinases by hexavalent products with DNA. A review. Mutat Res 1988, 195:137-149. chromium, manganese and nickel in human lung epithelial cells. Toxicol 41. Comporti M: Lipid peroxidation and cellular damage in toxic liver injury. Lett 2006, 167(2):114-21. Lab Invest 1985, 53(6):599-623. 21. Hewett P: Estimation of regional pulmonary deposition and exposure for 42. Keyhani E, Abdi-Oskouei F, Attar F, Keyhani J: DNA strand breaks by metal- fumes from SMAW and GMAW mild and stainless steel consumables. induced oxygen radicals in purified Salmonella typhimurium DNA. Ann N Am Ind Hyg Assoc J 1995, 56(2):136-42. Y Acad Sci 2006, 1091:52-64. 22. Antonini JM, Afshari AA, Stone S, Chen B, Schwegler-Berry D, Fletcher WG, 43. Valko M, Morris H, Cronin MT: Metals, toxicity and oxidative stress. Curr Goldsmith WT, Vandestouwe KH, McKinney W, Castranova V, Frazer DG: Med Chem 2005, 12(10):1161-208. Design, construction, and characterization of a novel robotic welding 44. Chen F, Ding M, Lu Y, Leonard SS, Vallyathan V, Castranova V, Shi X: fume generator and inhalation exposure system for laboratory animals. J Participation of MAP kinase p38 and IkB kinase in chromium(VI)-induced Occup Environ Hyg 2006, 3(4):194-203. NF-kB and AP-1 activation. J Environ Pathol Toxicol Oncol 2000, 19:231-238. 23. Jenkins NT, Pierce WM-G, Eagar TW: Particle Size Distribution of Gas Metal 45. Shi X, Chiu A, Chen CT, Halliwell B, Castranova V, Vallyathan V: Reduction of and Flux Cored Arc Welding Fumes. Welding J 2005, 84:156s-163s. chromium(VI) and its relationship to carcinogenesis. J Toxicol Environ 24. Leonard SS, Castranova V, Chen BT, Schwegler-Berry D, Hoover M, Health 1998, 2:101-118. Piacitelli C, Gaughan D: Particle size dependent radical generation from 46. Hiura TS, Kaszubowski MP, Li N, Nel AE: Chemicals in diesel exhaust Wildland fire smoke. Toxicology 2007, 236:103-113. particles generate reactive oxygen radicals and induce apoptosis in 25. Antonini JM, Taylor MD, Zimmer AT, Roberts JR: Pulmonary responses to macrophages. J Immunol 1999, 163(10):5582-91. welding fumes: role of metal constituents. J Toxicol Environ Health A 2004, 47. Jalava PI, Salonen RO, Hälinen AI, Penttinen P, Pennanen AS, Sillanpää M, 67(3):233-49, Review. Sandell E, Hillamo R, Hirvonen MR: In vitro inflammatory and cytotoxic 26. du Plessis L, Laubscher P, Jooste J, du Plessis J, Franken A, van Aarde N, effects of size-segregated particulate samples collected during long- Eloff F: Flow cytometric analysis of the oxidative status in human range transport of wildfire smoke to Helsinki. Toxicol Appl Pharmacol peripheral blood mononuclear cells of workers exposed to welding 2006, 215(3):341-53. fumes. J Occup Environ Hyg 2010, 7(6):367-74. 48. Leonard SS, Roberts JR, Antonini JM, Castranova V, Shi X: PbCrO mediates 27. Vallyathan V, Castranova V, Pack D, Leonard S, Shumaker J, Hubbs AF, cellular responses via reactive oxygen species. Mol Cell Biochem 2004, Shoemaker DA, Ramsey DM, Pretty JR, McLaurin JL, Khan A, Teass A: Freshly 255(1&2):171-9. fractured quartz inhalation leads to enhanced lung injury and 49. Zhang Y, Fong CC, Wong MS, Tzang CH, Lai WP, Fong WF, Sui SF, Yang M: inflammation: potential role of free radicals. Am J Respir Crit Care Med Molecular mechanisms of survival and apoptosis in RAW 264.7 1995, 152:1003-1009. macrophages under oxidative stress. Apoptosis 2005, 10(3):545-56. 28. Antonini JM, Lawryk NJ, Murthy GG, Brain JD: Effect of welding fume 50. NIOSH: Elements (ICP): Method 7300. Manual of Analytical Methods. 4 solubility on lung macrophage viability and function in vitro. J Toxicol edition. US Dept of Health and Human Services publication No. 98-119. Environ Health A 1999, 58(6):343-63. Washington, D.C.: NIOSH; 1994, 2. 29. McNeilly JD, Heal MR, Beverland IJ, Howe A, Gibson MD, Hibbs LR, 51. Janzen EG, Blackburn BJ: Detection and identification of short-lived free MacNee W, Donaldson K: Soluble transition metals cause the pro- radicals by an electron spin resonance trapping technique. J Am Chem inflammatory effects of welding fumes in vitro. Toxicol Appl Pharmacol Soc 1968, 90:5909-10. 2004, 196(1):95-107. 52. Buettner GR: Spin trapping: ESR parameters of spin adducts. Free Rad Biol 30. Zeidler-Erdely PC, Kashon ML, Battelli LA, Young SH, Erdely A, Roberts JR, Med 1987, 3:259-303. Reynolds SH, Antonini JM: Pulmonary inflammation and tumor induction 53. Brambilla G, Martelli A, Marinari UM: Is lipid peroxidation associated with in lung tumor susceptible A/J and resistant C57BL/6J mice exposed to DNA damage? Mutat Res 1989, 214(1):123-7. welding fume. Part Fibre Toxicol 2008, 5:12. 54. Imamoglu N, Yerer MB, Donmez-Altuntas H, Saraymen R: Erythrocyte 31. Antonini JM, Leonard SS, Roberts JR, Solano-Lopez C, Young SH, Shi X, antioxidant enzyme activities and lipid peroxidation in the erythrocyte Taylor MD: Effect of stainless steel manual metal arc welding fume on membrane of stainless-steel welders exposed to welding fumes and gases. Int J Hyg Environ Health 2008, 211(1-2):63-8. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 13 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 55. Collins AR: The comet assay for DNA damage and repair: principles, applications, and limitations. Mol Biotechnol 2004, 26(3):249-61. 56. Botta C, Iarmarcovai G, Chaspoul F, Sari-Minodier I, Pompili J, Orsière T, Bergé-Lefranc JL, Botta A, Gallice P, De Méo M: Assessment of occupational exposure to welding fumes by inductively coupled plasma- mass spectroscopy and by the alkaline Comet assay. Environ Mol Mutagen 2006, 47(4):284-95. doi:10.1186/1743-8977-7-32 Cite this article as: Leonard et al.: Comparison of stainless and mild steel welding fumes in generation of reactive oxygen species. Particle and Fibre Toxicology 2010 7:32. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Particle and Fibre Toxicology Springer Journals

Comparison of stainless and mild steel welding fumes in generation of reactive oxygen species

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/comparison-of-stainless-and-mild-steel-welding-fumes-in-generation-of-bb0oNdAGHx

References (62)

Publisher
Springer Journals
Copyright
Copyright © 2010 by Leonard et al; licensee BioMed Central Ltd.
Subject
Biomedicine; Pharmacology/Toxicology; Pneumology/Respiratory System; Nanotechnology and Microengineering
eISSN
1743-8977
DOI
10.1186/1743-8977-7-32
pmid
21047424
Publisher site
See Article on Publisher Site

Abstract

Background: Welding fumes consist of a wide range of complex metal oxide particles which can be deposited in all regions of the respiratory tract. The welding aerosol is not homogeneous and is generated mostly from the electrode/wire. Over 390,000 welders were reported in the U.S. in 2008 while over 1 million full-time welders were working worldwide. Many health effects are presently under investigation from exposure to welding fumes. Welding fume pulmonary effects have been associated with bronchitis, metal fume fever, cancer and functional changes in the lung. Our investigation focused on the generation of free radicals and reactive oxygen species from stainless and mild steel welding fumes generated by a gas metal arc robotic welder. An inhalation exposure chamber located at NIOSH was used to collect the welding fume particles. Results: Our results show that hydroxyl radicals ( OH) were generated from reactions with H O and after exposure 2 2 to cells. Catalase reduced the generation of OH from exposed cells indicating the involvement of H O .The 2 2 welding fume suspension also showed the ability to cause lipid peroxidation, effect O consumption, induce H O 2 2 2 generation in cells, and cause DNA damage. Conclusion: Increase in oxidative damage observed in the cellular exposures correlated well with OH generation in size and type of welding fumes, indicating the influence of metal type and transition state on radical production as well as associated damage. Our results demonstrate that both types of welding fumes are able to generate ROS and ROS-related damage over a range of particle sizes; however, the stainless steel fumes consistently showed a significantly higher reactivity and radical generation capacity. The chemical composition of the steel had a significant impact on the ROS generation capacity with the stainless steel containing Cr and Ni causing more damage than the mild steel. Our results suggest that welding fumes may cause acute lung injury. Since type of fume generated, particle size, and elapsed time after generation of the welding exposure are significant factors in radical generation and particle deposition these factors should be considered when developing protective strategies. Background Although full-time welders may be easier to track, the There are approximately 390,000 full-time welders in profession has a large number of part-time, small shop the United States [1] and an estimated 5 million persons welders who may also be exposed as well as others occupationally exposed to welding fumes worldwide. working in the vicinity of the welding activities. These The welding process which joins materials by causing part-time and cross workplace exposures make the coalescence using a filler material, usually wire, to form effects difficult to monitor. Welding is frequently carried a molten pool which then cools bonding the surfaces out in areas with poor ventilation such as ship hulls, together. During this process an occupational exposure metal tanks, or pipe and crawl spaces leading to a can occur through inhalation of the fume and particles. greater potential for exposure. Welding fumes have been demonstrated to cause toxicity among exposed workers [2,3] Welding fumes consist of a wide range of metal * Correspondence: [email protected] Pathology and Physiology Research Branch, Health Effects Laboratory oxide particles, including iron, manganese, chromium, Division, National Institute for Occupational Safety and Health, Morgantown, and nickel, which are generated mostly from the WV, USA © 2010 Leonard et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 2 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 electrode/wire feed [4]. Two of the major feed wire continuous weld (figure 1). We used this automated sys- types which are used in the welding process are mild tem to collect welding fume from both SS wire and MS steel (MS) and stainless steel (SS). wire while using an arc welding process similar to that Inhalation of the fume has been related to bronchitis reported previously [22]. Differences in toxicity between [5], metal fume fever, occupational asthma [6] cancer mild steel and stainless steel fume have been previously and possible increases in lung tumorigenicity [7,8], observed [31,32]. Although, the reasons for these differ- suppression of lung defenses [9,10], and functional ences are not fully understood, some studies have indi- changes in the lung [11-14]. Investigations have also cated metal content in the weld wire as being the cause shownanincreaseinROS production afterwelding [33-35]. fume generation [15-17] and initiation of downstream The goal of our study was to access differences in mediators, such as HO-1, VEGF, and MAP kinases reactive oxygen species (ROS) generation during the [18-20]. The generated fume ranges in size and can be welding process. This study will investigate the effects of deposited throughout the respiratory tract [4,21-23]. weld wire type, particle size, surface area and time after Previous studies have demonstrated the impact of par- generation of fume on the potential generation of ROS ticle size and surface area on pulmonary effects of and its downstream effects. The observations reported inhaled toxicants [24]. here hope to elucidate the mechanism behind some of The effects seen from exposure to welding fume are the biological effects observed in occupational exposures under investigation but not well understood nor are the to welding fume. mechanisms behind potential toxicity from fume expo- sure. However, the act of welding causes the generation Results of unstable metal oxides due to the energy at the point Particle morphology and characteristics of the weld leading to an uncommon form of first expo- Electron micrographs of welding fume sample collected sure to newly formed unstable and potentially more on MOUDI filters (Figure 2) showed large agglomerated reactive particles, similar to that seen in sandblasting chains of particles linked together which not only [16,25-27]. Biological effects of welding fume exposure formed larger aggregate particles but greatly increased have been investigated employing cellular [28,29] and surface area. Particle sizes are not related to one single animal models [30] using a welding exposure system large particle but many smaller spherical particles linked located at NIOSH Health Effects Laboratory Division. as a chain which by its form creates a much greater sur- This system can simulate real workplace exposures and face area than would ordinarily be associated with a sin- allows the collection of fresh welding fume from a gle spherical particle of this size. This greater surface Figure 1 Schematic of the welding fume generation system. Welder operated from screened computer control room as welding occurs on a table. Generated fume collected immediately off the surface and passed through a heat exchanger then collected in the MOUDI particle collector or midget impinger. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 3 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Figure 2 Electron micrographs of two of the 15 filters used in the MOUDI and nano-MOUDI.A)shows stage4(3.2 μmcut-off)larger particles and the formation of agglomerated chains of particles. B) shows stage 12 (0.032 μm cut-off) smaller individual particles and less chain formation. area allows much more reaction surface with the gener- contained Fe, Mn and Cu. However, stainless steel also ated fumes. contained the transition metals Cr (20.2 ± 1.52% bw) Most particles, calculated by mass, generated in our and Ni (8.76 ± 0.18% bw) while mild steel showed a welding fume system were between 0.56 and 0.1 μmin small amount of Si (2.75 ± 0.28% bw) mean diameter as demonstrated in figure 3. This size range deposits mostly in the alveolar and bronchiolar Free radical generation and ROS regions in the lung; however, particles were found at all Free racial generation was examined using three sepa- size ranges measured using the MOUDI. Note that even rate systems; sized fume reaction with hydroxyl radical though grease was not used on the MOUDI stages in precursor H O , direct bubbling of whole fume through 2 2 this study, the profile of the mass collected on the stages solution containing H O and the spin trap DMPO, and 2 2 was similar to that obtained when the grease was used sized fume effects on exposed cells. [22], demonstrating that the phenomenon of potential Figure 4 shows individual filter sized welding fume particle bounce, if any, was minimal. reacting with hydrogen peroxide to measure generation Table 1 shows the elemental analysis of both stainless of hydroxyl radical ( OH). Strong radical generation is and mild steel wire which indicated that each wire observed at the 1 hour post collection time in both the Figure 3 Mass concentration trapped on each filter of the MOUDI. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 4 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Table 1 Stainless steel versus mild steel elemental sizes. When the OH radical signal was adjusted for par- analysis ticle weight/filter, mass normalized radical activity, it Stainless Steel was observed that the smaller particles had increased radical generation potential per unit weight, figure 5. Metal μg/sample Weight % of metal Stainless steel showed higher generation of OH radicals Fe 1207 ± 161 57 ± 1.28 at every time point (except 5.6 μm) with significantly Cr 427.5 ± 69.1 20.2 ± 1.52 higher generation;41%,57% and59% atsizes0.1 μm, Mn 295.0 ± 48.4 13.8 ± 0.45 0.056 μm, and 0.032 μm respectively. Addition of defer- Ni 185 ± 24.0 8.76 ± 0.18 oxamine, a metal chelator showed a decrease in radical Cu 3.30 ± 0.492 0.155 ± 0.004 signal strength in a concentration dependent manner Mild Steel with a complete abolishment of the DMPO/ OH signal Fe 776 ± 9.8 80.6 ± 0.17 at 2 mM deferoxamine. Additon of catalase, an H O 2 2 Mn 142 ± 2.0 14.7 ± 0.11 catalyst, to the cellular reactions decreased the radical signal as well (data not shown). Si 26.6 ± 2.9 2.75 ± 0.28 Figure 6 shows the results of total fumes drawn Cu 17.2 ± 0.20 1.79 ± 0.02 directly from the welding surface and bubbled through PBS in the presence of H O and DMPO, to act as a 2 2 stainless steel and mild steel samples. The number of spin trap. Results demonstrate the total sample potential radicals generated is reduced over time as shown at the to generate OH radicals when reacted with H O . Stain- 2 2 24 hour and 1 week post generation. Using the 0.32 μm less steel fumes showed a significantly higher generation mean aerodynamic diameter as a reference; stainless of OH radicals than was an equal mass of fume gener- steel showed a 68% reduction in OH radical signal ated by mild steel. between1 hour to 24 hours post generation which The results of RAW 264.7 cellular exposure to weld- became a 77% reduction at the 1 week time point. Mild ingfumebysizegroup andtypeoffumeare shownin steel showed similar reduction over time with a 64% figure 7. Stainless steel fumes incubated with cells drop in OH signal from 1 hour to 24 hours and a 67% showed a significant increase at all three size groups drop at the 1 week time point. The strongest OH radi- when compared to mild steel fumes. There was also an cal signals were measured at the 0.32 μmto0.056 μm increase in radical production within each steel type as size ranges; however, it should be noted this closely cor- particle size decreased. It should also be noted that the responds to the observed higher mass collected at those Cr (V) electron spin resonance signal (figure 7 inset) Figure 4 The generation of short-lived OH radicals upon reaction of H O with individual filter sizes using different wire type and 2 2 time periods after generation of fume. Open symbols represent stainless steel while filled symbols are mild steel wire. ESR spectrum recorded 3 min after reaction was initiated in PBS (pH 7.4), 1 mM H O and 100 mM DMPO. ESR settings were; center field, 3385 G; scan width, 100 G; 2 2 time constant, 40 msec; scans, 5; modulation amplitude, 1 G; receiver gain, 2.5 × 10 ; frequency, 9.793 GHz; and power, 63 mW. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 5 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 steel also showed a significant increase over mild steel when comparing fine and ultrafine size groups. H O production 2 2 Figure 9 shows the effects of welding fume on H O 2 2 generation after RAW 264.7 cells were exposure to welding fume. All welding fume sizes and steel types at equal mass showed a significant rise in H O produc- 2 2 tion in exposed RAW 264.7 cells when compared to control. A significant difference was also observed between stainless steel and mild steel fumes in the fine and ultrafine size groups. A size dependent increase in H O production was also observed in both types of 2 2 Figure 5 Mass normalized radical activity, OH radical peak metals. heights of individual filters sizes adjusted for radicals/mg on 1 hour post generation samples. ESR settings were; center field, O consumption 3385 G; scan width, 100 G; time constant, 40 msec; scans, 5; A significant rise in oxygen consumption was observed modulation amplitude, 1 G; receiver gain, 2.5 × 10 ; frequency, 9.793 GHz; and power, 63 mW. Asterisks indicate a significant increase in in RAW 264.7 cells in all stainless steel size exposures stainless steel fume compared to mild steel fume (P < 0.05). and in the mild steel ultrafine size as shown in figure 10. A significant difference was observed between stain- less steel fumes and mild steel fumes at equal mass in was observed only in spectra from stainless steel fume the ultrafine size group. A size dependent increase was exposure to cells. also observed within steel types. Lipid peroxidation DNA damage Stainless steel fumes showed a significant increase in Comet assay results for DNA damage are shown in fig- lipid peroxidation in exposed RAW 264.7 cells com- ure11. Thedatademonstrates total groupedsizefilters pared to control at all three sizes groups at equal mass for both stainless and mild steel at equal mass showed a as seen in figure 8. Mild steel fumes caused a significant significant increase in DNA damage in exposed RAW increase at the ultrafine particle size group. Stainless 264.7 cells. Furthermore, the stainless steel fumes caused Figure 6 OH radical peak heights of midget impinger bubbled samples. Flow rate 3.80 L/min, 20 min collection time. Welding fume was bubbled directly through PBS (pH 7.4), 1 mM H O and 100 mM DMPO. ESR settings were; center field, 3385 G; scan width, 100 G; time 2 2 constant, 40 msec; scans, 1; modulation amplitude, 1 G; receiver gain, 2.5 × 10 ; frequency, 9.793 GHz; and power, 63 mW. Asterisks indicate a significant increase in stainless steel fume compared to mild steel fume (P < 0.05). Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 6 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Figure 7 Radicals generated from RAW 264.7 cells (1 × 10 /ml) exposed to grouped filter sample welding fume (250 μg/ml) for 10 min at 37°C in a shaking water bath. ESR settings were; center field, 3385 G; scan width, 100 G; time constant, 40 msec; scans, 5; modulation amplitude, 1 G; receiver gain, 2.5 × 10 ; frequency, 9.793 GHz; and power, 63 mW. Asterisks indicate a significant increase in radicals compared between steel types; (+) indicate significant difference between fume sizes. significantly more DNA damage when compared to mild potential cellular damage from ROS generation and to steel fumes. determine if the biological effects were size, time and surface area dependent. Our results determined that Discussion both stainless steel and mild steel welding fumes are Thepresent studywas undertaken to investigatepossi- able to generate OH radicals in a Fenton-like system. ble radical generation of fume generated from two of These OH radials are precursors and initiators of many the most common types of welding processes to access forms of ROS that can produce damage to cellular membranes, proteins and DNA as well as initiate further downstream damage and signaling associated with respiratory burst and inflammation. Our results also determined that freshly generated fume samples, 1 hour post, were significantly more reactive in generating OH radials than samples aged for 24 hours and 1 week. This reactivity is attributed to the change in transition states of the freshly generated metals that are produced during the welding process due to the high energy involved. Transition metals have the ability to accept and donate single electrons. Metals used in welding may temporarily attain a different and possibly more reactive, transition or valence state and in this changed transition state they are able to overcome the spin restriction on direct reac- Figure 8 Welding fume induced lipid peroxidation in tion of O with non-radicals. These reactions can lead incubation mixture containing 250 μg/ml welding fume 2 sample and 5 × 10 RAW 264.7 cells. Data presented are means to the generation of ROS and further damage to biologi- of ± S.D. for 4 sets of experiments. (*) indicate a significant increase cal systems. Addition of metal clelators and catalase in lipid peroxidation compared to control. (+) indicate significant confirmed the involvement of metals and H O in the 2 2 difference between metal types at the same sizes. (P < 0.05) radical generation observed. A consistent result Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 7 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Figure 9 H O production in incubation mixtures containing 250 μg/ml welding fume sample and 1 × 10 RAW 264.7 cells.Data 2 2 presented are means of ± S.D. for 4 sets of experiments. (*) indicate a significant increase in lipid peroxidation compared to control. (+) indicate significant difference between metal types at the same sizes. (P < 0.05) throughout our various ESR measurements (H O , cellu- their ability to cycle through transition states once in a 2 2 lar and direct whole fume) was the ability of stainless cellular system can lead to extensive and ongoing ROS steel fumes to generate higher amounts of free radicals production. This metal cycling ability was demonstrated than mild steel. Animal studies indicate that stainless in our investigation by the ESR spectra results showing steel welding fumes induce more lung inflammation and the production of Cr(V) in stainless steel fume exposed injury compared to mild steel fumes [17] Elemental ana- cells. This Cr(V) radical signal came from the reduction lysisofthe fumesshowedthatstainless steelcontained of Cr(VI) in the welding fume in a cellular system. This two transition metals not found in mild steel, chromium Cr(VI) ® Cr(V) system has been researched and shown and nickel. Both chromium and nickel have had exten- to be toxic and to responsible for the generation ROS sive research performed on their reactivity and abilities associated damage [37]. to produce ROS in biological systems [36]. These metals The relationship between free radical generation and have been shown to be highly toxic and result in gen- different sizes at equal mass was also investigated to eration of ROS and associated damage. Furthermore, determine the effect on free radical generation after exposure to H O . The greatest free radical generation 2 2 was observed when analyzing the ultrafine particles demonstrating that reactivity of the particles is depen- dent on available surface area with equal mass. Gener- ated welding fumes are long chains of agglomerated particles which are trapped on filters in size groups. However, the large chains are actually made up of many smaller primary particles linked together, creating a much higher surface area at equal mass than spherical particles that are more commonly observed in the envir- onment and workplace. The welding fume chains formed under the high heat conditions of welding acts synergistically to create fresh particles of a high surface area which generate reactive species. Freshly generated Figure 10 Oxygen consumptioninwelding fume stimulated welding fume has been previously reported to be more RAW 264.7 cells. Incubation mixtures contain 3 × 10 RAW 264.7 cells + 500 μg/ml welding fume sample. Data presented are means biologically active in an animal model [16]. of ± S.D. for 4 sets of experiments. (*) indicate a significant increase Welding fumes were also found to cause lipid perox- in oxygen consumption compared to control. (+) indicate significant idation, an indicator of cell membrane damage and difference between metal types at the same sizes. (P < 0.05) precursor for other radical generation systems. Lipid Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 8 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 Figure 11 Comet assay results after incubation of RAW 264.7 cells for 1 h exposed to 250 μg/ml welding fume.Datapresented are means of ± S.D. for 4 sets of experiments. (*) indicate a significant increase in DNA damage compared to control. (+) indicate significant difference between metal types at the same sizes. (P < 0.05) peroxidation results in the release of lipid-derived radi- containing Cr causing significantly greater DNA . . . cals (R ,RO and ROO ) [38], which can lead to a cas- damage. In addition, H O and OH generated by cells 2 2 cade effect and release reactive iron [39] causing the exposed to welding fume may cause other cellular generation of more ROS. Lipid peroxidation and its damage via mechanisms associated with reactions effects have been found to cause DNA damage [40] initiated by ROS, for example, dG hydroxylation and and may function as tumor initiators [41]. Once again protein-DNA cross-links. Furthermore these ROS may the stainless steel fumes cause significantly greater also cause activation of nuclear transcription factors, damage to the cells than mild steel. This same trend, such as NF-B, over-expression of certain oncogenes as well as a trend in smaller size i.e. greater surface and induction of p53 mutation [44,45]. area was also demonstrated in H O generation, O 2 2 2 consumption and DNA damage in exposed RAW 264.7 Conclusions cells. H O generation and O consumption are indica- The significant increase in oxidative damage observed in 2 2 2 tors of a respiratory burst in the exposed cells creating the cellular exposures correlates well with the results as downstream effects and activation of inflammatory determined by ESR of OH generation in size and type pathways and signaling. The significantly higher DNA of welding fumes, indicating the influence of metal type damage observed in the stainless steel fumes may be and transition state on radical production as well as due to the different metals present. Hydroxyl radicals associated damage. Our results demonstrate that both generated with certain metals, such as nickel and cop- types of welding fumes are able to generate ROS and per, exhibit less reactivity with DNA. Possible reasons ROS-related damage over a range of particle sizes; how- for this non-reactivity include; that OH radicals are ever, the stainless steel fumes consistently showed a sig- generated within the domain of certain macromole- nificantly higher reactivity and radical generation cules and, therefore, are not able to exhibit significant capacity. The chemical composition of the steel had a reactivity and the structural contribution of metal significant impact on the ROS generation capacity with toward DNA-binding and metal interactions with the the stainless steel, but not mild steel, containing Cr and DNA [42,43]. The results from our study show that Ni causing more damage than the mild steel. Both mate- OH radicals generated from both welding fumes have rials contained Fe, a known Fenton radical producer, the potential to cause DNA damage with stainless steel Mn and Cu. Our results suggest that the marked Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 9 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 difference between the radicals produced in the two Electric), a water cooled arc welding torch (WC 650 types of fumes is due to the presence of Cr found in the amp, Lincoln Electric), a wire feeder that supplied the stainless steel. Cr has been shown in many investiga- wire to the torch at a programmed rate up to 300 tions to a cause a number of toxicities. It is also noted inches/min, and an automatic welding torch cleaner that that the smaller particle size of the fume the greater the kept thewelding nozzlefreeof debrisand spatter. Gas ROS potential at equal mass with the 0.180 - 0.032 μm metal arc welding was performed using a mild steel sizes showing the most reactivity in both types of weld- electrode (carbon steel ER70S-6, Lincoln Electric) or a ing fume which can penetrate deepest into the lung of stainless steel electrode (Blue Max E308LSi wire, Lin- an exposed welder. Our results further demonstrated coln Electric). Welding took place on A36 carbon steel that the freshly generated fumes were more reactive and plates for fume collection times of 20 min at 25 V and caused more oxidative damage than the aged particles, 200 amps. During welding, a shielding gas combination indicating that metal transition state also plays an of 95% Ar and 5% CO (Airgas Co., Morgantown, WV) important role in welding fume reactivity. Therefore, was continually delivered to the welding nozzle at an air our results suggest that welding fumes may cause acute flow of 20 L/min. lung injury. Since type of fume generated, particle size, Figure 1 shows an outline of the generation and col- and elapsed time after generation of the welding expo- lection system. The system is contained in three rooms; sure are significant factors in radical generation and par- the control room, the robotic arm welding fume genera- ticle deposition these factors should be considered when tor and the fume collection chamber. All three rooms developing protective strategies. were separated from each other. Aerodynamically size-selected aerosol samples were Methods collected with the Micro-Orifice Uniform Deposit Impac- Reagents tor (MOUDI), model # 110 and Nano-MOUDI MSP Chelex 100 resin was purchased from Bio-Rad Labora- Model #115, with rotator (MSP, Inc., Minneapolis, MN, tories (Richmond, CA, USA). Phosphate-buffered saline USA). This MOUDI/Nano-MOUDI arraignment pro- (PBS), (KH PO (1.06 mM), Na HPO (5.6 mM), NaCl vided a 15-stage research-grade cascade impactor. Each 2 4 2 4 (154 mM), pH 7.4), was purchased from Biowhittaker filter stage has a cutoff size with particles collected at Inc. (Walkersville, MD, USA). The PBS was treated with each stage being aerodynamically size-selected between Chelex 100 to remove transition metal ion contaminants. stage sizes. Cutoff sizes were; 18, 10, 5.6,3.2, 1.8, 1.0, 0.56, Dulbecco’s modified eagles medium (DMEM), 5,5- 0.32, 0.18, 0.10, 0.006, 0.03, 0.02, 0.01 μmand thefinal dimethyl-1-pyroline-oxide (DMPO), fetal bovine serum filter. Filters used for ESR and cellular exposures were (FBS), FeSO H O and penicillin/streptomycin were from Millipore Corp. (Billerica, MA, USA) 47 mm, 0.8 4, 2 2, purchasedfromSigma Chemical Company(St.Louis, μm, PVC model PVC0847600. PVC was selected because MO, USA). The spin trap, DMPO, was purified by char- it has previously been demonstrated to have no effect in coal decolorization and vacuum distillation and was free the free radical analysis. The MOUDI substrates are nor- of ESR detectable impurities. Quartz sample tubes were mally coated with grease to ensure adherence of depos- purchased from Wilmad Glass (Buena, NJ, USA). ited particles and to avoid bounce of large particles to lower stages of the impactor. However, grease can alter Cell culture the surface of collected aerosol particles and is not suita- RAW 264.7 mouse peritoneal monocytes were pur- ble for use in collecting samples for free radical analysis. chased from American Type Culture Collection (Rock- Therefore, the cascade impactor was operated without ville, MD). RAW 264.7 cells are commonly used and grease substrates to collect and fractionate the welding have been found to respond to particle exposure in a fume. Filters were collected for 20 min. Filters were then manner similar to primary alveolar macrophages splitinto3groups foreither 1 hour,24houror1week [46-49]. RAW 264.7 cells were cultured in DMEM with post generation analysis. 10% FBS, 2 mM L-glutamine, and 50 mg/ml pen/strep Filter suspensions, which were used for H O , lipid 2 2 at 37°C in a 5% CO incubator. Cells were split after peroxidation, oxygen consumption and DNA damage confluence approximately every 3 days. analysis, were prepared by splitting the filters into three size groups. For our study the size groups were defined Generation and collection of welding fumes as ultrafine (0.01 - 0.056 μm), fine (0.1 - 1.0 μm) and The welding fume generation system was similar to that coarse (1.8 - 18 μm) particles. The groups consisted of previously outlined [22]. Briefly, the generation system five filters each of which were placed in PBS. Pre- and consisted of a welding power source (Power Wave 455, post filter weights were used to prepare a final concen- Lincoln Electric, Cleveland, OH), an automated, pro- tration of 1 mg/ml welding fume suspension. The slurry grammable six-axis robotic arm (Model 100 Bi, Lincoln was centrifuged, and the fume suspension was decanted Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 10 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 from the filter pellet. A clean control suspension was due to its specificity and sensitivity. All ESR measure- also prepared at a ratio of 4 filters/1 ml PBS. ments were conducted using a Bruker EMX spectro- meter (Bruker Instruments Inc. Billerica, MA 01821, Welding particle morphology, transmission electron USA) and a flat cell assembly. Hyperfine couplings microscopy (TEM) were measured (to 0.1 G) directly from magnetic field Welding fume samples, collected separately from the separation using potassium tetraperoxochromate fumes used in free radial analysis, were collected at 30- (K CrO ) and 1,1-diphenyl-2-picrylhydrazyl (DPPH) as 3 8 minute intervals during 3 h of welding directly onto for- reference standards [51,52]. The relative radical con- mvar-coated TEM grids and viewed using a JEOL 1220 centration was estimated by multiplying half of the transmission electron microscope (JOEL, Inc.). peak height by (ΔH ) ,where ΔH represents peak- pp pp to peak width. The Acquisit program was used for Welding particle size distribution data acquisitions and analyses (Bruker Instruments Inc. Particle size distribution was determined by using a Billerica, MA 01821, USA). Micro-Orifice Uniform Deposit Impactor (MOUDI) model # 110 which is intended for the general purpose Radicals trapped by bubbler aerosol sampling, and a Nano-MOUDI (MSP Model Freshly generated total welding particle reactivity was 115) that is specifically designed for sampling aerosols measured using a midget bubbler (Ace Glass, Vineland in the size range down to 0.010 μm. NJ) containing a reaction mixture made up of H O (10 2 2 mM) DMPO (100 mM), and PBS to instantly trap gen- Welding particle metal content analysis erated radicals for ESR measurement. The midget bub- Mild and stainless steel welding particles were collected bler was attached to a collection tube with a flow rate of onto 5.0 mm polyvinyl chloride membrane filters in 37 1 L/min positioned approximately 18 inches from the mm cassettes during 30 min of welding. Metal analysis welding surface in order to directly collect the freshly samples were collected separately from the samples used generated fume. Fume was collected for 20 min. The for free radical analysis. The particle samples were reaction mixture was then immediately measured using digested and the metals analyzed by inductively coupled ESR in order to determine OH generation from freshly plasma atomic emission spectroscopy (ICP-AES) by generated whole fume. Clayton Group Services (A Bureau Vertis Company, Novi, MI) as coordinated with the Division of Applied Grouped filters Research and Technology (DART, Cincinnati, OH) Grouped filter suspensions, were used for H O , lipid 2 2 according to NIOSH method 7300 modified for micro- peroxidation, O consumption and DNA damage ana- wave digestion [50]. Metal content of blank filters also lysis. Suspensions were prepared by splitting the filters was analyzed for control purposes. into three size groups. For this study the size groups were defined as ultrafine (0.01 - 0.056 μm), fine (0.10 - Free radical measurements 1.0 μm) and coarse (1.8 - 18 μm) particles. The groups ESR spin trapping was used to detect short-lived free consisted of five filters each of which were placed in radical intermediates. Hydroxyl radicals were measured PBS and blended on ice into a fine slurry using a Tis- using the addition-type reaction of a short-lived radical sueTearor(BiospecProductsInc.Racine,WI).Pre- with a compound (spin trap) to form a relatively long- andpostfilter weightswereusedtoprepareafinal lived paramagnetic free radical product (spin adduct), suspension of 1 mg/ml welding fume. The slurry was which can then be studied using conventional ESR. For centrifuged, and the welding fume suspension was dec- hydroxyl radical measurements on individual filter anted from the filter pellet. A clean control suspension sizes, reactants were mixed in test tubes at a final was also prepared at a ratio of 5 filters/1 ml PBS. Sus- volume of 1.0mlofPBS in thepresenceof1 mM pension not used immediately was aliquoted and fro- H O . The reaction mixture was then transferred to a zen at -70°C. 2 2 quartz flat cell for ESR measurement. Experiments were performed at room temperature and under ambi- Lipid peroxidation ent air. The concentrations given in the figure legends Lipid peroxidation of RAW 264.7 mouse peritoneal are final concentrations. The intensity of the signal is monocytes was measured by using a colormetric assay used to measure the relative amount of short-lived for malondialdehyde (LPO-586 Oxis International Inc. radicals trapped, and the hyperfine couplings of the Portland, OR, USA). A reaction mixture contained var- spin adduct are characteristic of the original trapped ious size groups (refer to filter methods) of welding par- radicals. Spin trapping is the method of choice for ticle samples [250 μg/ml], H O (1 mM) and 1 × 10 2 2 detection and identification of free radical generation cells in a total volume of 1.0 ml PBS (pH 7.4). A Fenton Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 11 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 reaction, FeSO (1 mM), H O (1 mM) and 1 × 10 Cranberry Township, PA). A minimum of 50 cells were 4 2 2 cells, was also carried out as a positive control. The scored for each sample at 400× magnification. The dis- mixtures were exposed for 1 h in a shaking water bath tance between the edge of the head and the end of the at 37°C. The measurement of lipid peroxidation is based tail was measured using an automated image analysis on the reaction of a chromogenic reagent with malonal- system (Optimas 6.51, Media Cybernetics Inc., Silver dehyde at 45°C [53,54]. The absorbance of the supernate Spring, Md) [55,56]. was measured at 586 nm. Statistics H O production Data were expressed as mean ± standard error of the 2 2 H O production was monitored using a Bioxytech mean (SEM) (n = 4) for each group. One-way ANOVA 2 2 quantitative hydrogen peroxide assay kit (H O -560, test was performed using SigmaStat statistical software 2 2 Oxis International Inc. Portland, OR, USA). Measure- (Jandel Scientific, San Rafael, CA, USA) to compare the ments were made on a system containing 5 × 10 RAW responses between treatments. Statistical significance 264.7 mouse peritoneal monocytes/ml in pH 7.4 PBS was set at p < 0.05. and exposing them to the size groupings (refer to filter methods) of welding particle solution. Cells were Abbreviations exposed to welding particle solution [250 μg/ml], for 30 Defined in text. minutes in a 37°C incubator. Absorbance was monitored Acknowledgements at a wavelength of 560 nm using a Spectra Max 250 Disclaimer: The findings and conclusions of this paper have not been formally multi-well plate reader (Molecular Devices, Sunnyvale, dissemination by NIOSH and should not be construed to represent any agency CA, USA). determination or policy. Authors’ contributions O consumption SSL, BTC, DF and JMA contributed to the idea and design of the study. SSL, Oxygen consumption measurements were carried out BTC and SGS carried out welding generation and particle collection. SSL and AJK carried out all electron spin resonance and toxicity measurements. DSB using a Gilson oxygraph (Gilson Medical Electronics, carried out EM analysis of particles. All authors read and approved the final Middleton, WI). Measurements were made on a system manuscript. containing 3 × 10 RAW 264.7 mouse peritoneal mono- Competing interests cytes/mL and welding particle solution [500 μg/ml], in The authors declare that they have no competing interests. pH 7.4 phosphate buffer. The oxygraph was calibrated with media and equilibrated with known concentrations Received: 25 June 2010 Accepted: 3 November 2010 Published: 3 November 2010 of oxygen. References DNA damage, Comet assay 1. Bureau of Labor Statistics. [http://www.bls.gov/oes/current/oes514121. The Comet assay was performed using methods htm]. 2. Moulin JJ, Wild P, Haguenoer JM, Faucon D, De Gaudemaris R, Mur JM, described in the Trevigen (Gaithersburg, Md., U.S.A.) Mereau M, Gary Y, Toamain JP, Birembaut Y: A mortality study among assay kit. A typical reaction mixture contained welding mild steel and stainless steel welders. Br J Ind Med 1993, 50(3):234-43. particle solution [250 μg/ml], H O (1 mM), and 1 × 3. Simonato L, Fletcher AC, Andersen A, Anderson K, Becker N, Chang- 2 2 Claude J, Ferro G, Gérin M, Gray CN, Hansen KS: A historical prospective 10 RAW 264.7 mouse peritoneal monocytes cells study of European stainless steel, mild steel, and shipyard welders. Br J brought to a total volume of 1.0 mL in PBS (pH 7.4). Ind Med 1991, 48(3):145-54. Briefly, (all steps performed in the dark or low light con- 4. Zimmer AT, Biswas P: Characterization of the aerosols resulting from arc welding processes. J Aerosol Sci 2001, 32:993-1008. ditions) RAW264.7 mouse cells were exposed to the var- 5. Contreras GR, Chan-Yeung M: Bronchial reactions to exposure to welding ious sized welding fume and incubated in media at 37 ° fumes. Occup Environ Med 1997, 54(11):836-9. C for 1 h. The cells were washed (2×) with PBS and 6. Hannu T, Piipari R, Kasurinen H, Keskinen H, Tuppurainen M, Tuomi T: Occupational asthma due to manual metal-arc welding of special combined with LMAgarose, then pipetted onto a Comet stainless steels. Eur Respir J 2005, 26(4):736-9. slide. The slide was placed in a refrigerator for 30 min, 7. d’Errico A, Pasian S, Baratti A, Zanelli R, Alfonzo S, Gilardi L, Beatrice F, immersed in lysis solution, chilled for 60 min, and then Bena A, Costa G: A case-control study on occupational risk factors for sino-nasal cancer. Occup Environ Med 2009, 66(7):448-55. immersed in alkaline solution for 55 min. Slides were 8. Langård S: Nickel-related cancer in welders. Sci Total Environ 1994, 148(2- placed in a horizontal electrophoresis chamber for 40 3):303-9, Review. min at 300 mA. Slides were washed and SYBR green 9. Antonini JM, Taylor MD, Millecchia L, Bebout AR, Roberts JR: Suppression in lung defense responses after bacterial infection in rats pretreated with stain was added to each. Slides were visualized using different welding fumes. Toxicol Appl Pharmacol 2004, 200(3):206-18. fluorescence microscopy, with an image capturing sys- 10. Anderson SE, Meade BJ, Butterworth LF, Munson AE: The humoral immune tem (Olympus AX70 and sample PCI, Compix, response of mice exposed to manual metal arc stainless steel-welding fumes. J Immunotoxicol 2007, 4(1):15-23. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 12 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 11. Antonini JM, Lewis AB, Roberts JR, Whaley DA: Pulmonary effects of free radical production, DNA damage, and apoptosis induction. Mol Cell welding fumes: review of worker and experimental animal studies. Am J Biochem 2005, 279(1-2):17-23. Ind Med 2003, 43(4):350-60. 32. Edmé JL, Shirali P, Mereau M, Sobaszek A, Boulenguez C, Diebold F, 12. Sferlazza SJ, Beckett WS: The respiratory health of welders. Am Rev Respir Haguenoer JM: Assessment of biological chromium among stainless steel Dis 1991, 143(5 Pt 1):1134-48. and mild steel welders in relation to welding processes. Int Arch Occup 13. Solano-Lopez C, Zeidler-Erdely PC, Hubbs AF, Reynolds SH, Roberts JR, Environ Health 1997, 70(4):237-42. Taylor MD, Young SH, Castranova V, Antonini JM: Welding fume exposure 33. Antonini JM, Roberts JR: Chromium in stainless steel welding fume and associated inflammatory and hyperplastic changes in the lungs of suppresses lung defense responses against bacterial infection in rats. J tumor susceptible a/j mice. Toxicol Pathol 2006, 34(4):364-72. Immunotoxicol 2007, 4(2):117-27. 14. Sobaszek A, Boulenguez C, Frimat P, Robin H, Haguenoer JM, Edme JL: 34. Alexopoulos EC, Cominos X, Trougakos IP, Lourda M, Gonos ES, Acute respiratory effects of exposure to stainless steel and mild steel Makropoulos V: Biological monitoring of hexavalent chromium and welding fumes. J Occup Environ Med 2000, 42(9):923-31. serum levels of the senescence biomarker apolipoprotein j/clusterin in 15. Liu HH, Wu YC, Chen HL: Production of ozone and reactive oxygen welders. Bioinorg Chem Appl 2008, 420578. species after welding. Arch Environ Contam Toxicol 2007, 53(4):513-8. 35. Serageldin M, Reeves DW: Development of welding emission factors for 16. Antonini JM, Clarke RW, Krishna Murthy GG, Sreekanthan P, Jenkins N, Cr and Cr(VI) with a confidence level. J Air Waste Manag Assoc 2009, Eagar TW, Brain JD: Freshly generated stainless steel welding fume 59(5):619-26. induces greater lung inflammation in rats as compared to aged fume. 36. Leonard SS, Harris GK, Shi X: Metal-induced oxidative stress and signal Toxicol Lett 1998, 98(1-2):77-86. transduction. Free Radic Biol Med 2004, 37(12):1921-1942. 17. Taylor MD, Roberts JR, Leonard SS, Shi X, Antonini JM: Effects of welding 37. Leonard SS, Wang S, Castranova V, Vallyathan V, Shi X: Role of Molecular fumes of differing composition and solubility on free radical production Oxygen in the Generation of hyrdroxyl and superoxide anion radicals and acute lung injury and inflammation in rats. Toxicol Sci 2003, during enzymatic Cr (VI) reducation and its implication to (CrVI)-induced 75(1):181-91. carcinogenisis. J Environ Path Toxicol Oncol 2000, 19:49-60. 18. Stark M, Zubareb J, Jacovovitz R, Schwartz Y, Lerman Y, Grinberg N, 38. Shi X, Dalal NS, Vallyathan V: ESR evidence for the hydroxyl radical Fireman E: HO-1 and VEGF gene expressions are time dependant during formation in aqueous suspension of quartz particles and its possible exposure to welding fumes. Cytokine 2009, 46(2):290-5. significance to lipid peroxidation in silicosis. J Toxicol Environ Health 1988, 19. Seel EA, Zaebst DD, Hein MJ, Liu J, Nowlin SJ, Chen P: Inter-rater 25:237-245. agreement for a retrospective exposure assessment of asbestos, 39. Vladimirov YA: In Free Radicals, Aging and Degenerative Diseases. Edited by: chromium, nickel and welding fumes in a study of lung cancer and Johnson JE Jr, Walford R, Harman D, Miquel J. New York: Liss; 1986:141-195. ionizing radiation. Ann Occup Hyg 2007, 51(7):601-10. 40. Vaca CE, Wilhelm J, Harms-Ringdahl M: Interaction of lipid peroxidation 20. Tessier DM, Pascal LE: Activation of MAP kinases by hexavalent products with DNA. A review. Mutat Res 1988, 195:137-149. chromium, manganese and nickel in human lung epithelial cells. Toxicol 41. Comporti M: Lipid peroxidation and cellular damage in toxic liver injury. Lett 2006, 167(2):114-21. Lab Invest 1985, 53(6):599-623. 21. Hewett P: Estimation of regional pulmonary deposition and exposure for 42. Keyhani E, Abdi-Oskouei F, Attar F, Keyhani J: DNA strand breaks by metal- fumes from SMAW and GMAW mild and stainless steel consumables. induced oxygen radicals in purified Salmonella typhimurium DNA. Ann N Am Ind Hyg Assoc J 1995, 56(2):136-42. Y Acad Sci 2006, 1091:52-64. 22. Antonini JM, Afshari AA, Stone S, Chen B, Schwegler-Berry D, Fletcher WG, 43. Valko M, Morris H, Cronin MT: Metals, toxicity and oxidative stress. Curr Goldsmith WT, Vandestouwe KH, McKinney W, Castranova V, Frazer DG: Med Chem 2005, 12(10):1161-208. Design, construction, and characterization of a novel robotic welding 44. Chen F, Ding M, Lu Y, Leonard SS, Vallyathan V, Castranova V, Shi X: fume generator and inhalation exposure system for laboratory animals. J Participation of MAP kinase p38 and IkB kinase in chromium(VI)-induced Occup Environ Hyg 2006, 3(4):194-203. NF-kB and AP-1 activation. J Environ Pathol Toxicol Oncol 2000, 19:231-238. 23. Jenkins NT, Pierce WM-G, Eagar TW: Particle Size Distribution of Gas Metal 45. Shi X, Chiu A, Chen CT, Halliwell B, Castranova V, Vallyathan V: Reduction of and Flux Cored Arc Welding Fumes. Welding J 2005, 84:156s-163s. chromium(VI) and its relationship to carcinogenesis. J Toxicol Environ 24. Leonard SS, Castranova V, Chen BT, Schwegler-Berry D, Hoover M, Health 1998, 2:101-118. Piacitelli C, Gaughan D: Particle size dependent radical generation from 46. Hiura TS, Kaszubowski MP, Li N, Nel AE: Chemicals in diesel exhaust Wildland fire smoke. Toxicology 2007, 236:103-113. particles generate reactive oxygen radicals and induce apoptosis in 25. Antonini JM, Taylor MD, Zimmer AT, Roberts JR: Pulmonary responses to macrophages. J Immunol 1999, 163(10):5582-91. welding fumes: role of metal constituents. J Toxicol Environ Health A 2004, 47. Jalava PI, Salonen RO, Hälinen AI, Penttinen P, Pennanen AS, Sillanpää M, 67(3):233-49, Review. Sandell E, Hillamo R, Hirvonen MR: In vitro inflammatory and cytotoxic 26. du Plessis L, Laubscher P, Jooste J, du Plessis J, Franken A, van Aarde N, effects of size-segregated particulate samples collected during long- Eloff F: Flow cytometric analysis of the oxidative status in human range transport of wildfire smoke to Helsinki. Toxicol Appl Pharmacol peripheral blood mononuclear cells of workers exposed to welding 2006, 215(3):341-53. fumes. J Occup Environ Hyg 2010, 7(6):367-74. 48. Leonard SS, Roberts JR, Antonini JM, Castranova V, Shi X: PbCrO mediates 27. Vallyathan V, Castranova V, Pack D, Leonard S, Shumaker J, Hubbs AF, cellular responses via reactive oxygen species. Mol Cell Biochem 2004, Shoemaker DA, Ramsey DM, Pretty JR, McLaurin JL, Khan A, Teass A: Freshly 255(1&2):171-9. fractured quartz inhalation leads to enhanced lung injury and 49. Zhang Y, Fong CC, Wong MS, Tzang CH, Lai WP, Fong WF, Sui SF, Yang M: inflammation: potential role of free radicals. Am J Respir Crit Care Med Molecular mechanisms of survival and apoptosis in RAW 264.7 1995, 152:1003-1009. macrophages under oxidative stress. Apoptosis 2005, 10(3):545-56. 28. Antonini JM, Lawryk NJ, Murthy GG, Brain JD: Effect of welding fume 50. NIOSH: Elements (ICP): Method 7300. Manual of Analytical Methods. 4 solubility on lung macrophage viability and function in vitro. J Toxicol edition. US Dept of Health and Human Services publication No. 98-119. Environ Health A 1999, 58(6):343-63. Washington, D.C.: NIOSH; 1994, 2. 29. McNeilly JD, Heal MR, Beverland IJ, Howe A, Gibson MD, Hibbs LR, 51. Janzen EG, Blackburn BJ: Detection and identification of short-lived free MacNee W, Donaldson K: Soluble transition metals cause the pro- radicals by an electron spin resonance trapping technique. J Am Chem inflammatory effects of welding fumes in vitro. Toxicol Appl Pharmacol Soc 1968, 90:5909-10. 2004, 196(1):95-107. 52. Buettner GR: Spin trapping: ESR parameters of spin adducts. Free Rad Biol 30. Zeidler-Erdely PC, Kashon ML, Battelli LA, Young SH, Erdely A, Roberts JR, Med 1987, 3:259-303. Reynolds SH, Antonini JM: Pulmonary inflammation and tumor induction 53. Brambilla G, Martelli A, Marinari UM: Is lipid peroxidation associated with in lung tumor susceptible A/J and resistant C57BL/6J mice exposed to DNA damage? Mutat Res 1989, 214(1):123-7. welding fume. Part Fibre Toxicol 2008, 5:12. 54. Imamoglu N, Yerer MB, Donmez-Altuntas H, Saraymen R: Erythrocyte 31. Antonini JM, Leonard SS, Roberts JR, Solano-Lopez C, Young SH, Shi X, antioxidant enzyme activities and lipid peroxidation in the erythrocyte Taylor MD: Effect of stainless steel manual metal arc welding fume on membrane of stainless-steel welders exposed to welding fumes and gases. Int J Hyg Environ Health 2008, 211(1-2):63-8. Leonard et al. Particle and Fibre Toxicology 2010, 7:32 Page 13 of 13 http://www.particleandfibretoxicology.com/content/7/1/32 55. Collins AR: The comet assay for DNA damage and repair: principles, applications, and limitations. Mol Biotechnol 2004, 26(3):249-61. 56. Botta C, Iarmarcovai G, Chaspoul F, Sari-Minodier I, Pompili J, Orsière T, Bergé-Lefranc JL, Botta A, Gallice P, De Méo M: Assessment of occupational exposure to welding fumes by inductively coupled plasma- mass spectroscopy and by the alkaline Comet assay. Environ Mol Mutagen 2006, 47(4):284-95. doi:10.1186/1743-8977-7-32 Cite this article as: Leonard et al.: Comparison of stainless and mild steel welding fumes in generation of reactive oxygen species. Particle and Fibre Toxicology 2010 7:32. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit

Journal

Particle and Fibre Toxicology – Springer Journals

Published: Nov 3, 2010

There are no references for this article.