Annals of Occupational Hygiene Advance Access originally published online on February 8, 2006
Annals of Occupational Hygiene 2006 50(4):379-384; doi:10.1093/annhyg/mei080
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original Article |
Enzyme exposure in the British baking industry
1 Health and Safety Laboratory, Harpur Hill, Buxton SK17 9JN, UK; 2 Health and Safety Executive, Stanley Precinct, Bootle, Merseyside L20 3QZ, UK
* Author to whom correspondence should be addressed. Tel: +44 1298 218 449; fax: +44 1289 218 172; e-mail: joanne.elms{at}hsl.gov.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Objectives: Enzymes are commonly used in the baking industry, as they can improve dough quality and texture and lengthen the shelf life of the final product. There is little published information highlighting exposure to enzymes (other than fungal alpha-amylase) in the baking industry, therefore the purpose of this study was to identify antibodies and develop assays for the measurement of a variety of such enzymes in samples of airborne flour dust.
Methods: Polyclonal antibodies to bacterial amylase, glucose oxidase and amyloglucosidase were identified and developed into ELISA assays. The assays showed limited cross-reactivity with other enzymes commonly used in the baking industry.
Results: We measured levels of airborne enzymes in 195 personal air samples taken from a sample of 55 craft baking establishments. We were able to detect amyloglucosidase in 9% (16/184) of the samples, fungal alpha-amylase in 6% (11/171), bacterial alpha-amylase in 7% (13/195). However, we were unable to detect glucose oxidase in any of the samples. Measurements for protease enzymes were not carried out. Median levels in detectable samples of amyloglucosidase, fungal alpha-amylase and bacterial amylase were similar at 10.3, 5.3 and 5.9 ng/m3, respectively. These figures represent the total enzyme protein (active and inactive) measured.
Conclusions: There are few data in the literature regarding sensitization and exposure–response relationships to these enzymes, and indeed there is often a lack of information within the industry as to the precise enzyme content of particular baking ingredients. As a precautionary measure, all enzymes are regarded as having the potential to cause respiratory sensitization. Consequently, exposures need to be controlled to as low a level as reasonably practicable, and future investigation may highlight the importance of measuring a variety of enzyme exposures and standardizing these methodologies to inform approaches to adequate control.
Keywords: baking • enzyme exposure • fungal alpha-amylase • bacterial amylase • glucose oxidase • amyloglucosidase
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Enzymes are now used widely by a variety of industries to effect chemical changes in a range of substrates (European Communities Collection of Information on Enzymes, 2002). Flour and baking industries first began adding enzymes to their products in the 1970s and this has increased over the last 30 years. Enzymes such as fungal alpha-amylase, bacterial alpha-amylase, amyloglucosidase, glucose oxidase, xylanase and protease are used in bread making to improve dough quality and lengthen the shelf life of the final product. There are many reports in the literature highlighting the potential of some of these enzymes (Baur, 1998
; Sander et al., 1998; Merget et al., 2001) to cause allergic sensitization and respiratory symptoms in susceptible bakers. In order to measure exposure to bakery enzymes, sensitive and specific assays are required. Specific ELISAs for measurement of airborne fungal alpha-amylase have been developed (Houba et al., 1997; Sander et al., 1997; Elms et al., 2001), however there is limited information about exposure to the other airborne bakery enzymes.
The development of specific monoclonal or polyclonal antibodies by different research groups is time consuming and costly, and it is argued that effective use of these reagents requires inter-laboratory assay standardization (Lillienberg et al., 2000). The purpose of this study was to identify commercially available antibodies to enzymes used in the baking industry, and to develop cost effective, sensitive assays for the measurement of these enzymes. Using these assays we determined airborne concentrations of fungal and bacterial amylase, glucose oxidase and amyloglucosidase in the British craft baking industry.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Bakery details
The Health and Safety Executive Specialist Inspectors together with the Health and Safety Laboratory (HSL) sampled 55 craft baking establishments between October 2002 and December 2003. Bakery size ranged from micro bakeries (one to nine employees) to large bakeries (>250 employees), however no automated plant bakeries were sampled. Details of the bakeries surveyed are reported elsewhere (Elms et al., 2005
). Workers were randomly selected from different areas of these bakeries where flour products were used. The workers' 8-h time weighted average (TWA) samples were classified into four main groups by job task (bakers, mixers/weighers and sievers, cleaners and others).
Sampling methods
Personal long-term samples were collected in the workers' breathing zone using IOM sampling heads with glass fibre filters (GF/A, 1.6 mm, Millipore) at a calibrated flow rate of 2 l/min. Field blanks were included for each sampling visit. The filters were weighed twice in a preconditioned room before and after sampling and the personal dust exposure (mg/m3) was calculated. Filters were eluted into 2 ml phosphate-buffered saline (PBS)/0.1% Tween 20 overnight by end-over-end mixing. The filters and supernatant were centrifuged at 600x g for 5 min to remove fine particle matter. The supernatant was removed and stored at –20°C until analysis. The limits of detection of the assays were calculated from the analysis of 10 laboratory blank filters (mean + 2 SD).
Monoclonal enzyme-linked immunosorbant assay for the determination of airborne fungal alpha-amylase
The inhalable fungal alpha-amylase levels were quantified using a monoclonal ELISA previously developed by the HSL (Elms et al., 2001).
Polyclonal antibody peroxidase conjugation
The bacterial amylase and glucose oxidase detection antibodies required a peroxidase-labelling step. These polyclonal (anti-rabbit) antibodies were conjugated using EZ-Link plusTM Activated Peroxidase Kit (Pierce, Rockford, USA) as described by the manufacturer's instructions. One milligram polyclonal antibody was dissolved into 1 ml of carbonate/bicarbonate buffer (pH 9–10). One milligram of lyophilized activated peroxidase was reconstituted into 100 ml of distilled water and added to the antibody solution and incubated for 1 h at room temperature. Following this, 10 ml of reductant solution was added to the antibody solution, which was incubated again at room temperature for 10 min. Subsequently, 20 ml of Quench Buffer (pH 9.0) was added. This mixture was allowed to react at room temperature for an additional 15 min, and then dialysed against PBS overnight. This was stored in aliquots at 4°C until used.
Polyclonal assays for the determination of airborne bacterial amylase, glucose oxidase and amyloglucosidase
Airborne levels of bacterial amylase, glucose oxidase and amyloglucosidase were quantified using commercially available polyclonal antibodies using a Rosys Plato automated liquid handling system (Qiagen AG, Switzerland). ELISA plates (Nunc, Maxisorb, Denmark) were coated with a polyclonal antibody to bacterial amylase (Bacillus amyloliquefaciens), glucose oxidase (Aspergillus niger) or amyloglucosidase (A. niger) (100 µl per well) diluted in carbonate/bicarbonate coupling buffer (pH 9–10), overnight at 4°C. The plates were washed and 100 µl per well of the standards, quality control samples, and test samples were added (diluted in assay buffer comprised of PBS/0.1% bovine serum albumin (BSA)/0.1% Tween20) and incubated for 1 h at room temperature. After washing, 100 µl per well of the polyclonal antibody diluted in Pierce Superblock (Pierce, Rockford, USA) was added and incubated for 1 h. In the case of bacterial amylase and glucose oxidase, the antibody was peroxidase labelled as described above (EZ-link plus Activated Peroxidase Kit, Pierce, Rockford, USA). The amyloglucosidase detection antibody was biotin labelled and required an additional step following washing, whereby 100 µl per well of 1 µg/ml horseradish peroxidase streptavidin (HRP; Vector Laboratories, UK) was incubated for 10 min. After a final washing step, 100 µl per well of tetramethylbenzidine was added and the plate incubated for 15 min at room temperature. The reaction was stopped by the addition of 2 M sulphuric acid (50 µl per well) and the absorbance at 450 nm for each well was measured. Table 1 lists sources and concentrations of antibodies used. A selection of 22 industrial enzymes was tested for cross reactivity to each of the assays (see Table 2). For these cross-reactivity tests, the enzyme was considered not to cross react when at 10 µg/ml, the subsequent substrate absorbance was below the limit of detection. The extent of any cross reactivity was described as a percentage. For this, comparisons were made between the observed concentrations at half the maximum absorbance for a particular assays standard and the other enzymes analysed. All protein concentrations were determined using the bicinchoninic acid protein assay (BCA) (Smith et al., 1985) using the Cobas Fara.
|
|
Statistics
All the data analysis for enzyme exposure levels was performed using SPSS software (Statistical Package for Social Scientists v10, SPSS Inc., Chicago, USA). Natural logarithms of exposures were tested for normality using the Shapiro–Wilk test. Descriptive statistical analysis was performed on the data in addition to cross tabulation with Fisher's exact test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Performance of ELISA assays and cross-reactivity
Fungal alpha-amylase
The fungal alpha-amylase assay exhibited an inter-assay variation of 4.8% at 3 ng/ml and 5.5% at 20 ng/ml (n = 46). The assay had a limit of detection of 0.39 ng/ml equating to 0.78 ng/m3 when collected over an 8-h shift at 2 l/min using a personal sampler.
Cellulase from A. niger and Trichoderma reesei demonstrated a limited cross reactivity of <4% and <0.2%, respectively, at 50% of the maximum optical density for the assay (Table 2). However, cellulase from Trichoderma viride did not cross react. Amylase from barley malt and amyloglucosidase (A. niger) (which are commonly found in bakeries) also exhibited a partial cross reactivity of 1% and <0.2% respectively. Protease from Aspergillus oryzae demonstrated a relatively high level of cross reactivity at 34%.
Bacterial alpha-amylase
The bacterial alpha-amylase assay (B. amyloliquefaciens) had a limit of detection of 0.31ng/ml (equating to 0.62 ng/m3 when collected over an 8-h shift at 2 l/min using a personal sampler), and an inter-assay variation of 7% at 6 ng/ml and 6% at 25 ng/ml (n = 34). Alpha-amylase from Bacillus subtilis demonstrated 91% cross reactivity at 50% of the maximum optical density for the bacterial alpha-amylase assay (Table 2), demonstrating suitability of the assay for measurement of alpha-amylase from B. subtilis and B. amyloliquefaciens. However, Bacillus licheniformis, exhibited no cross reactivity within the assay. Only Neutrase 0.8LTM (from B. amyloliquefaciens) demonstrated cross reactivity (34% at 50% of the maximum optical density) within the assay. Glucose oxidase (A. niger), cellulase (T. viride) and amyloglucosidase (A. niger) demonstrated cross reactivity of <1%.
Amyloglucosidase
The amyloglucosidase assay exhibited an inter-assay variation of 13.9% at 3 ng/ml and 8.3% at 43 ng/ml (n = 46) and a limit of detection of 0.74 ng/ml. Cellobiase (from A. niger, used in the fruit juice industry) demonstrated 38% cross reactivity at 50% of the maximum optical density. Glucose oxidase (A. niger), Cellulase (A. niger or T. viride) and protease (Bacillus species) demonstrated cross reactivities 
1.5% (Table 2).
Glucose oxidase
The glucose oxidase assay had a limit of detection of 0.65 ng/ml. The inter-assay variation was 11.1% at 3 ng/ml and 10.7% at 26 ng/ml (n = 42). Only cellulase (T. reesei), amyloglucosidase (A. niger) and protease (Bacillus polymyxa) demonstrated cross reactivity within the assay of <0.1% (Table 2).
Airborne levels of enzymes for British craft bakeries
A total of 208 personal, 8-h TWA samples were collected from workers. One hundred and ninety-five filters were available for further analysis for fungal alpha-amylase, bacterial amylase, glucose oxidase and amyloglucosidase (Table 3). However, due to insufficient sample volumes, samples that required repeat analysis (due to unacceptable coefficients of variation for duplicate samples, or quality control samples) could not be re-measured and were excluded from the analysis. We were able to detect amyloglucosidase in 9% (16/184) of the samples, fungal amylase in 6% (11/171) and bacterial alpha-amylase in 7% (13/195) of the samples (Table 3). No glucose oxidase was detectable in any of the samples. As there were limited numbers of samples with detectable levels of these enzymes, we calculated the mean and median of samples above the limit of detection of the appropriate assay. The median airborne concentrations of detectable bacterial alpha-amylase, fungal alpha-amylase and amyloglucosidase were broadly similar at 5.9, 5.3 and 10.3 ng/m3, respectively. (Table 3).
|
There was no association (Fisher's exact test) between an individual baker being exposed to detectable bacterial amylase and fungal alpha-amylase (P = 1.000), bacterial amylase and amyloglucosidase (P = 0.312), than was expected by random chance alone. There was however, an association between measurable exposure to fungal alpha-amylase and amyloglucosidase (P = 0.06), although this was not significant at the 5% level. The association between fungal alpha-amylase and amyloglucosidase is unlikely to have been due to the limited cross reactivity of the assays, as amyloglucosidase only cross reacted <0.2% at a 50% maximum optical density within the fungal alpha-amylase assay, and the fungal alpha-amylase did not cross react within the amyloglucosidase assay.
Workers were categorized as being exposed to one or more measurable enzymes or to no enzymes at all. 23% (23/100) of ‘bakers’, 22% (12/55) of ‘mixer/sievers/weighers’, 33% (2/6) of ‘cleaners’ and 9% (3/34) of ‘others’ were exposed to measurable levels of one or more enzyme (only a small number of workers sampled were defined as ‘cleaners’ and these workers were reported to be involved with job tasks of potentially higher exposure—sweeping machinery for example).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
During this study, we developed sensitive assays for the measurement of bacterial amylase, glucose oxidase and amyloglucosidase, which were used to measure enzymes in flour dust from 55 baking establishments. The assays exhibited low intra and inter-assay variation, low limits of detection and limited cross reactivity with the 22 industrial enzymes tested. There are limited numbers of research papers focusing on the potential for enzymes used in the baking industry to cause sensitization and respiratory health effects (Baur et al., 1988
; Merget et al., 2001; Quirce et al., 2002). With the exception of fungal alpha-amylase, there are few available data on exposure levels to such enzymes in the baking industry. This is one of the few studies that have identified exposure levels to a number of enzymes used by the baking industry, in particular, bacterial amylase and amyloglucosidase. In this study the median value for detectable fungal alpha-amylase was 5.3 ng/m3. The median levels of bacterial amylase and amyloglucosidase were broadly similar at 5.9 and 10.3 ng/m3, respectively; however, there were no other comparable data in the literature. We were unable to detect glucose oxidase in any of the samples, but do not know whether this relates to the quantity or form of the enzyme used.
In total, 21% (40/195) of the samples included in the analysis had measurable levels of one or more of these enzymes. The analysis of some samples could not be repeated due to insufficient sample volume. Therefore, these values may represent an underestimate (or overestimate) of exposure to these enzymes. Eleven of the samples contained measurable fungal alpha-amylase (above the limit of detection). If fungal alpha-amylase were the only enzyme that had been measured, then it follows from this data set that, exposure to enzymes in at least 15% (29/195) of the samples would have been missed. Additionally, as we have not measured the exposure to an exhaustive list of enzymes this percentage might possibly be higher (proteases and lipases are regularly used in the baking process).
There are few data in the literature regarding respiratory sensitization to other enzymes (for example, proteases), and to our knowledge, there are no available data on the exposure–response relationships or potency of these enzymes. As a precautionary measure, exposures need to be controlled to as low a level as reasonably practicable. For the baking industry, controls should extend to the supply of materials, such as pre-packaged and low dust enzyme mixes.
There is often a lack of information within the industry as to the enzyme content of particular baking ingredients. Further work needs to be undertaken to investigate the consequences of using such enzymes in the baking industry and to determine whether ‘adequate’ control of exposure to flour dust is likely to afford adequate control of exposure to enzymes in nearly all cases.
It is proposed that setting occupational exposure limits for all enzymes, with an associated duty to monitor exposure, may not be useful. Having straightforward methods that associate low exposure with good control and using indicative markers of exposure may be a more productive and efficient option and applicable to other industries that use enzymes.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We would like to thank the Health and Safety Executive and Local Authority Inspectors; Sandy Ritchie, Carole Keddie and Fiona MacNeill, our planning team and the field scientists who conducted surveys with the Health and Safety Laboratory sampling teams. We would also like to extend our gratitude to the participants of this survey.
Received June 10, 2005; in final form December 12, 2005
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Baur X, Weiss W, Sauer W et al. (1988) ‘Baking ingredients as a contributory cause of baker's asthma. [Article in German].’ Dtsch Med Wochenschr; 113: 1275–8.[Medline]
Elms J, Denniss S, Smith M et al. (2001) ‘Development and validation of a monoclonal based immunoassay for the measurement of fungal alpha-amylase: focus on peak exposures.’ Ann Occup Hyg; 45: 89–95.
Elms J, Robinson E, Rahman S et al. (2005) Exposure to flour dust in UK bakeries: Current use of control measures. Ann Occup Hyg; 49: 85–91.
European-Communities Collection of information on enzymes. Luxembourg: Office for Official Publications of the European Communities, 2002. ISBN 92-894-4218-2.
Houba R, van Run P, Doekes G et al. (1997) ‘Airborne levels of alpha-amylase allergens in bakeries.’ J Allergy Clin Immunol; 99: 286–92.[CrossRef][ISI][Medline]
Lillienberg L, Baur X, Doekes G et al. (2000) ‘Comparison of four methods to assess fungal alpha-amylase in flour dust.’ Ann Occup Hyg; 44: 427–33.
Merget R, Sander I, Raulf-Heimsoth M et al. (2001) ‘Baker's asthma due to xylanase and cellulase without sensitization to alpha-amylase and only weak sensitization to flour.’ Int Arch Allergy Immunol; 124: 502–5.[CrossRef][Medline]
Quirce S, Fernandez-Nieto M, Bartolome B et al. (2002) ‘Glucoamylase: another fungal enzyme associated with baker's asthma.’ Ann Allergy Asthma Immunol; 89:197–202.[Medline]
Sander I, Neuhaus-Schroder C, Borowitzki G et al. (1997) ‘Development of a two-site enzyme-linked immunosorbent assay for alpha-amylase from Aspergillus oryzae based on monoclonal antibodies.’ J Immunol Methods; 210: 93–101.[Medline]
Sander I, Raulf-Heimsoth M, Siethoff C et al. (1998) ‘Allergy to Aspergillus-derived enzymes in the baking industry: identification of beta-xylosidase from Aspergillus niger as a new allergen (Asp n 14)’ J Allergy Clin Immunol; 102: 256–64.[Medline]
Smith PK, Krohn RI, Hermanson GT et al. (1985) Measurement of protein using bicinchoninic acid. Anal Biochem; 150:76–85.[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. MEIJSTER, E. TIELEMANS, N. D. PATER, and D. Heederik Modelling Exposure in Flour Processing Sectors in The Netherlands: a Baseline Measurement in the Context of an Intervention Program Ann. Hyg., April 1, 2007; 51(3): 293 - 304. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||