Prevalence of sensitisation to ‘improver’ enzymes in UK
supermarket bakers.
Short title: Sensitisation to ‘improver’ enzymes in UK
bakers
M Jones, J Welch, J Turvey, J Cannon1, P Clark, J Szram, P
Cullinan
Department of Occupational and Environmental Medicine, Imperial
College and Royal Brompton and Harefield NHS Trust1, London,
UK.
Corresponding author:
Dr Meinir Jones,
Department of Occupational and Environmental Medicine,
Imperial College,
1B Manresa Rd,
London SW3 6LR
Telephone number: 0207 351 8355
Fax number: 0207 351 8336
E-mail: [email protected]
Abstract
Background: Supermarket bakers are exposed not only to flour and
alpha amylase but also other ‘improver’ enzymes, the nature of
which is usually shrouded by commercial sensitivity. We aimed to
determine the prevalence of sensitisation to ‘improver’ enzymes in
UK supermarket bakers.
Methods: We examined the prevalence of sensitisation to enzymes
in 300 bakers, employed by one of two large supermarket bakeries,
who had declared work related respiratory symptoms during routine
health surveillance. Sensitisation was determined using
radioallergosorbent assay to eight individual enzymes contained in
the specific ‘improver’ mix used by each supermarket.
Results: The prevalence of sensitisation to ‘improver’ enzymes
ranged from 5% to 15%. Sensitisation was far more likely if the
baker was sensitised also to either flour and or alpha amylase. The
prevalence of sensitisation to an ‘improver’ enzyme did not appear
to be related to the concentration of that enzyme in the mix.
Conclusions: We report substantial rates of sensitisation to
enzymes other than alpha amylase in UK supermarket bakers; in only
a small proportion of bakers was there evidence of sensitisation to
‘improver mix’ enzymes without sensitisation to either alpha
amylase or flour. The clinical significance of these findings need
further investigation but our findings indicate that specific
sensitisation in symptomatic bakers may not be identified without
consideration of a wide range of workplace antigens.
Introduction
Baker’s asthma represents a market failure; despite a thorough
understanding of its distributions and determinants (1) its
incidence remains obstinately high in many countries.
Immunologically, the disease reflects an immediate-type
hypersensitivity to one or more aeroallergens encountered at work,
chiefly one or more cereal flours but also, in most commercial
baking, one or more ‘improver’ enzymes, the best known of which is
fungal alpha-amylase. The standard diagnostic work up of bakers
includes the determination of specific IgE sensitisation to both
flour (usually wheat or rye) and alpha-amylase.
In the UK and other countries, modern industrial baking is
widely based on the Chorleywood Bread Process, developed in 1961 by
the British Baking Industries Research Association. The process
uses high-speed mixers and ‘improvers’ to accelerate, modify and
control dough proving and to increase the shelf-life of the final
product. The dough ‘improvers’ contain a product-specific set of
enzymes, including but not limited to fungal alpha-amylase, and are
generally provided to the bakery as a fine powder that is added
directly to the dry dough mix. The panel of ‘improver’ enzymes is
constantly reviewed and extended; and it appears (personal
communications) that many UK bakeries are unaware of the detailed
constituents of the mix.
There is evidence to suggest that bakers can develop
sensitisation to ‘improver’ enzymes other than fungal
alpha-amylase. Case reports of bakers in Spain have reported IgE
sensitisation and immediate responses to specific inhalation
challenge with cellulase (2), glucoamylase and hemicellulose (3);
and in a report from Germany, Baur et al reported sensitisation to
xylanase in a baker with work-related asthma (4). In a larger
survey (n=171) of bakers undergoing investigation for occupational
respiratory symptoms there was evidence of specific sensitisation
to glucoamylase (8%), cellulase (13%) and xylanase (11%) (5). In a
survey of small UK bakers (n=135) a small proportion (6%) of bakers
were sensitised to a mixture of cellulase, hemicellulose and
xylanase (6) although the contributions of these individually and
separately from flour and alpha-amylase were not reported. To date,
there has not been a study of sensitisation to the extensive panel
of enzymes contained within the ‘improver mix’ used in supermarket
scratch bakeries.
Our clinical service sees a large number of bakers with
work-related respiratory symptoms; routinely, we assay their serum
for specific IgE antibodies to wheat flour and to fungal
alpha-amylase. Aware that their enzyme exposure is complex, and of
the reports above, we wished to determine how frequently UK bakers
are sensitised to other ‘improver’ enzymes and by extension how
sensitive our routine immunological approach is.
Methods
Two large UK supermarket companies with in-store ‘scratch’
bakeries undertake regular health surveillance of their bakers.
Employees complete a short questionnaire designed to identify
work-related upper or lower respiratory symptoms (7); serum samples
from those with positive responses are assayed in our laboratory
for specific IgE to an in-house preparation of wheat flour and to
fungal alpha-amylase using ImmunoCAP (Phadia, Milton Keynes);
‘sensitisation’ is based on an IgE level ≥0.35 kU/l.
The two companies are provided with ‘improver’ mixes by a single
supplier from whom we obtained samples of each ‘improver’ enzyme
together with an estimate of its concentration in each mix. This
study uses serum samples collected between 2008 and 2012 from 122
bakers working for supermarket 1 (22% sensitised to wheat flour
and/or alpha amylase) and 147 bakers for supermarket 2 (40%
sensitised to flour and/or alpha amylase) - Table 1. In each of the
samples from bakers with flour or alpha-amylase sensitisation, and
in every third sample from bakers without such sensitisation, we
determined IgE sensitisation to the separate ‘improver’ enzymes
used by the relevant supermarket.
Specific IgE was measured using a radioallergosorbent assay
(RAST). Enzymes were extracted overnight in 0.01mol/L ammonium
carbonate at 40C, dialysed overnight against distilled water,
lyophilised and stored at -200C. Briefly, 3mg of enzyme was coupled
to 300mg of cyanogen bromide activated paper discs, according to
the method of Ceska et al (8). For the assay, serum (50 µl) was
added to an allergen-coupled disc. Following incubation at room
temperature for 16 hours, the disc was washed and 125I-anti-human
IgE (50 µl) added and after a further 16 hours, the disc was washed
again and the bound I125 anti-human IgE determined using a gamma
scintillation counter. The amount of antigen-specific IgE was
expressed as the percentage of the total counts that remained bound
to the disc (percentage binding). A level ≥2% binding was chosen as
a positive cut off point.
Enzymes were characterised on non-reduced NuPAGE® 4-12% Bis-Tris
Gel according to the manufacturer’s instructions (LifeTechnologies
Ltd, Paisley, UK) and subsequently stained using a silver kit
(LifeTechnologies Ltd, Paisley, UK) (Fig 1a, 1b). We observed a
variation in the number of proteins present within the enzymes,
some clearly have only one protein (bacterial xylanase) whereas
other enzymes such as fungal xylanase contain many proteins.
Results
Overall, about a third of the bakers were sensitised to either
wheat flour or alpha-amylase (Table 1); this proportion was higher
among employees of supermarket 2. A small minority of bakers were
sensitised to alpha-amylase alone.
Each of the other, ‘improver’ enzymes seemed capable of inducing
specific sensitisation (Table 2) with between 5% (maltogenic
amylase) and 15% (fungal xylanase) of all bakers sensitised. Bakers
who were sensitised to either wheat flour or fungal alpha-amylase
were far more likely to be sensitised also to an ‘improver’ enzyme.
The prevalence of sensitisation to an ‘improver’ enzyme did not
appear to be related to the concentration of that enzyme in the
mix. Among the 183 bakers who were not sensitised to either flour
or fungal alpha amylase, the prevalence of sensitisation to
‘improver mix’ enzymes was 5%; these bakers were for the most part
sensitised to just one ‘improver’ enzyme’ (Table 3).
Discussion
One quarter of this group of UK supermarket bakers with
work-related respiratory symptoms were sensitised to one or more
‘improver mix’ enzymes; most were also sensitised either to wheat
flour or to fungal alpha amylase. Testing to only these last two
antigens, as is our routine practice, would identify 91% of the
sensitised bakers in this population, suggesting a false-negative
rate of 9%.
Currently the clinical significance of sensitisation to
‘improver mix’ enzymes is not clear, although there is some
evidence that they can induce a positive inhalation challenge in
some bakers (2,3). Because this is a retrospective study we did not
undertake challenge testing in those bakers sensitised to ‘improver
mix’ enzymes without co-sensitisation to either flour or alpha
amylase; we followed up the bakers with their occupational health
physician, who reported no increase in their symptom severity or
frequency, despite daily exposures in the bakery. The clinical
significance of sensitisation to ‘improver mix’ enzymes requires
further investigation.
The determinants of sensitisation to ‘improver mix’ enzymes seem
complex and apparently unrelated to the concentration of the enzyme
in the mix. Data on airborne levels of the enzymes are limited and
aeroallergen measurements are not always comparable: xylanase was
measured in stationary samples taken from four typical Finnish
bakeries but reportedly at a much lower level than fungal alpha
amylase, with cellulase being undetectable (9). A more recent
survey reported that enzymes could be detected in a proportion of
personal air samples (n=195), collected from bakers working in UK
craft bakeries (n=55), namely amyloglucosidase (9%), fungal alpha
amylase (6%) and bacterial alpha amylase (7%), with no detectable
glucose oxidase in any of the personal samples (10). Further
studies are required to establish airborne levels of ‘improver mix’
enzymes in bakeries to establish whether differences in the ability
of enzymes to become airborne in the bakery are responsible for
differing levels of sensitisation. Enzymes probably have variable
antigenic potencies; for example, in a detergent factory the
prevalence of sensitisation to amylase and cellulase were similar
to protease, despite far higher quantities of protease being used
(11).
We were unable to determine the immunological similarity between
the ‘improver mix’ enzymes because of a lack of available serum.
Previous studies have reported that fungal alpha amylase is
immunologically distinct from either xylanase or cellulase, but
that xylanase and cellulase are immunologically similar (5, 12).
Merget et al were able to confirm this in a case report, where
allergy to xylanase and cellulase occurred without sensitisation to
fungal alpha amylase or flour 2001 (13).
In conclusion we have observed sensitisation to bakery enzymes
other than alpha amylase. In a small proportion of bakers, there
was sensitisation to ‘improver mix’ enzymes without
co-sensitisation to alpha amylase or flour. The clinical
significance of these findings needs further investigation, however
our findings highlight that specific sensitisation in symptomatic
bakers may not be identified without consideration of a wide range
of workplace antigens.
References
1. Verma DK, Purdham JT, Roels HA. Translating evidence about
occupational conditions into strategies for prevention. Occup
Environ Med 2002; 59: 205–213.
2. Quirce S, Cuevas M, Díez-Gómez M, Fernández-Rivas M, Hinojosa
M, González R, Losada E. Respiratory allergy to Aspergillus-derived
enzymes in bakers' asthma. J Allergy Clin Immunol. 1992;90(6 Pt
1):970-8.
3. Quirce S, Fernández-Nieto M, Bartolomé B, Bombín C, Cuevas M,
Sastre J. Glucoamylase: another fungal enzyme associated with
baker's asthma. Ann Allergy Asthma Immunol. 2002;89(2):197-202.
4. Baur X, Sander I, Posch A, Raulf-Heimsoth M. Baker's asthma
due to the enzyme xylanase - a new occupational allergen. Clin Exp
Allergy. 1998;28(12):1591-3.
5. Sander I, Raulf-Heimsoth M, Siethoff C, Lohaus C, Meyer HE,
Baur X. 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.
1998;102(2):256-64.
6. Elms J, Fishwick D, Walker J, Rawbone R, Jeffrey P, Griffin
P, Gibson M, Curran AD. Prevalence of sensitisation to cellulase
and xylanase in bakery workers. Occup Environ Med.
2003;60(10):802-4.
7. Brant A, Nightingale S, Berriman J, Sharp C, Welch J, Newman
Taylor AJ, Cullinan P. Supermarket baker's asthma: how accurate is
routine health surveillance? Occup Environ Med. 2005
Jun;62(6):395-9.
8. Ceska M, Eriksson R, Varga JM. Radioimmunosorbent assay of
allergens. J Allergy Clin Immunol. 1972;49:1–9
9. Vanhanen M, Tuomi T, Hokkanen H, Tupasela O, Tuomainen A,
Holmberg PC, Leisola M, Nordman H. Enzyme exposure and enzyme
sensitisation in the baking industry. Occup Environ Med.
1996;53(10):670-6.
10. Elms J, Robinson E, Mason H, Iqbal S, Garrod A, Evans GS.
Enzyme exposure in the British baking industry. Ann Occup Hyg. 2006
Jun;50(4):379-84.
11. Cullinan P, Harris JM, Newman Taylor AJ, Hole AM, Jones M,
Barnes F, Jolliffe G. An outbreak of asthma in a modern detergent
factory. Lancet. 2000 Dec 2;356(9245):1899-900.
12. Merget R, Sander I, Raulf-Heimsoth M, Baur X. Baker's asthma
due to xylanase and cellulase without sensitization to
alpha-amylase and only weak sensitization to flour. Int Arch
Allergy Immunol. 2001 Apr;124(4):502-5.
13.
Table 1. Sensitisation to flour or alpha amylase in serum from
bakers with work-related respiratory symptoms
Bakers tested
Specific IgE positive to flour and/or α-amylase
Specific IgE positive to flour only
Specific IgE positive to α-amylase
only
Specific IgE negative to either flour or α-amylase
Supermarket 1
122
27 (22%)
27 (22%)
0 (0%)
95 (78%)
Supermarket 2
147
59 (40%)
55 (37%)
4 (3%)
88 (60%)
Total
269
86 (32%)
82 (30%)
4 (1%)
183 (68%)
Table 2. Sensitisation to ‘improver mix’ enzymes stratified by
sensitisation to flour and/or alpha amylase
‘Improver’ enzyme
(CAS number)
Quantity of enzyme used in ‘improver’ mix
Supermarket
Sensitisation to ‘improver mix’ enzyme
All bakers
Bakers co-sensitised to flour and/or fungal alpha amylase
Bakers not sensitised to flour or fungal alpha amylase
Maltogenic amylase
(160611-47-2)
40-50 ppm
1+2
13/268
(5%)
11/81
(14%)
2/187
(1%)
Cellulase
(9012-54-8)
25 ppm
1
12/122
(10%)
9/27
(33%)
3/95
(3%)
Fungal xylanase
(9025-56-3)
20 ppm
1+2
41/268
(15%)
37/82
(45%)
4/186
(2%)
Lipase
9043-29-2
20 ppm
1
8/121
7
7/26
27
1/95
1
Lipase
9001-62-1
20 ppm
2
17/147
12
17/55
31
0/92
0
Bacterial xylanase
(9025-57-4)
10 ppm
1+2
36/269
(13%)
36/86
(42%)
0/183
(0%)
Glucose oxidase
(9001-37-0)
5-10 ppm
1+2
30/268
(11%)
28/85
(33%)
2/183
(1%)
Bacterial α-amylase
(9000-85-5)
0.5 ppm
1+2
16/269
(6%)
15/86
(17%)
1/183
(0.5%)
Any of the above enzymes
1+2
66/269
(25%)
57/86
(66%)
9/187
(5%)
Table 3. Sensitisation to ‘improver mix’ enzyme in bakers not
sensitised to either flour or alpha amylase
Supermarket
Baker
Maltogenic amylase
Cellulase
Fungal Xylanase
Lipase
Bacterial Xylanase
Glucose oxidase
Bacterial alpha amylase
1
1
2
3
4
5
2
6
7
8
9
Shaded areas represent sensitisation
Fig 1. Silver stain of ‘improver’ enzymes on reduced NuPAGE®
4-12% Bis-Tris Gel
Legend Fig 1: Analysis of ‘improver/enzymes on reduced NuPAGE®
4-12% Bis-Tris Gel. The enzymes (total amount 5µg per lane) were
maltogenic amylase in lane 2, cellulase in lane 3, fungal xylanase
in lane 4, lipase (9043-29-2) in lane 5, lipase (9001-62-1) in lane
6, bacterial xylanase in lane 7, glucose oxidase in lane 8 and
bacterial alpha amylase in lane 9. Molecular weight markers were
run in lane 1.
Statement of contribution by each named author.
M Jones acquired and interpreted data, wrote the manuscript and
was involved in the conception and design of the study
J Welch acquired and interpreted the data
J Turvey acquired and interpreted the data
J Cannon was involved with the design of the study and
critically reviewed the manuscript
P Clark was involved with the design of the study
J Szram was involved with the design of the study
P Cullinan conceived and designed the study and critically
reviewed the manuscript
Statement of conflicts of interest for each named author
M Jones - no conflict of interest
J Welch - no conflict of interest
J Turvey - no conflict of interest
J Cannon - no conflict of interest
P Clark - no conflict of interest
J Szram - no conflict of interest
P Cullinan - no conflict of interest
