Vol. 58, No. 6APPLIED AND ENVIRONMENTAL MICROBIOLOGY, June 1992, P. 1823-1831
0099-2240/92/061823-09$02.00/0
Copyright C) 1992, American Society for Microbiology
Biodegradation of Bisphenol A and Other Bisphenols by
a Gram-Negative Aerobic Bacterium
JOHN H. LOBOS,’* TERRY K. LEIB,’ AND TAH-MUN SU2
General Electric Corporate Research and Development, Schenectady, New York 12301-0008,’
and Taiwan Biotech Corporation, Taoyun, Taiwan2
Received 24 October 1991/Accepted 26 March 1992
A novel bacterium designated strain MV1 was isolated from a sludge enrichment taken from the wastewater
treatment plant at a plastics manufacturing facility and shown to degrade 2,2-bis(4-hydroxyphenyl)propane
(4,4’-isopropylidenediphenol or bisphenol A). Strain MV1 is a gram-negative, aerobic bacillus that grows on
bisphenol A as a sole source of carbon and energy. Total carbon analysis for bisphenol A degradation
demonstrated that 60%o of the carbon was mineralized to CO2, 20%o was associated with the bacterial cells, and
20%o was converted to soluble organic compounds. Metabolic intermediates detected in the culture medium
during growth on bisphenol A were identified as 4-hydroxybenzoic acid, 4-hydroxyacetophenone, 2,2-bis(4-
hydroxyphenyl)-1-propanol, and 2,3-bis(4-hydroxyphenyl)-1,2-propanediol. Most of the bisphenol A degraded
by strain MV1 is cleaved in some way to form 4-hydroxybenzoic acid and 4-hydroxyacetophenone, which are
subsequently mineralized or assimilated into cell carbon. In addition, about 20%o of the bisphenol A is
hydroxylated to form 2,2-bis(4-hydroxyphenyl)-1-propanol, which is slowly biotransformed to 2,3-bis(4-
hydroxyphenyl)-1,2-propanediol. Cells that were grown on bisphenol A degraded a variety of bisphenol
alkanes, hydroxylated benzoic acids, and hydroxylated acetophenones during resting-cell assays. Transmission
electron microscopy of cells grown on bisphenol A revealed lipid storage granules and intracytoplasmic
membranes.
Bisphenol A, also called 2,2-bis(4-hydroxyphenyl)propane
or 4,4’-isopropylidenediphenol, is an industrially important
compound used in the production of polycarbonates and
other plastics at many chemical manufacturing plants
throughout the world. The annual production of bisphenol A
exceeds 930 million pounds (1). Manufacturing facilities
generate significant quantities of waste containing bisphenol
A, some of which is discharged into the terrestrial, aquatic,
and marine environments (4).
The health effects from exposure to bisphenol A have been
investigated elsewhere (1, 9). One study used mice and
guinea pigs as models in an attempt to determine the effects
of external exposure to bisphenol A on humans (9). Photo-
allergic contact dermatitis related to bisphenol A was in-
duced in the mouse model, but there was no response in the
guinea pig model. A toxicological study on the developmen-
tal toxicity of bisphenol A in rats and mice found that
bisphenol A treatment at maternally toxic doses during
organogenesis produced fetal toxicity in mice but not in rats.
No alteration in the morphologic development of the fetus in
either species occurred (10).
In 1988, the ad hoc Bisphenol A Task Group from the
Society of Plastics Industry reviewed the data available for
the health and ecological effects of bisphenol A with the U.S.
Environmental Protection Agency (1). According to stan-
dard evaluation procedures published by the U.S. Environ-
mental Protection Agency, bisphenol A was determined to
be slightly to moderately toxic for fish and invertebrates.
Lethal and effective concentrations of bisphenol A for 50%
of the experimental groups were from 1.1 to 10 mg/liter.
Very little is known about the chemical fate of bisphenol A
in the environment. A recent study reported that most
bisphenol A in a waste stream was degraded rapidly in a
* Corresponding author.
chemical plant’s biotreatment facility or in surface waters
that received a continuous discharge of bisphenol A (4).
However, there are no reports on the microorganisms that
degrade bisphenol A or the chemistry of bisphenol A bio-
degradation.
The present investigation reports on the isolation and
characterization of a gram-negative, aerobic bacterium
(strain MV1) that utilizes bisphenol A as the sole carbon
source. The identification of several metabolic intermediates
has led to a partial understanding of the biochemical path-
way of bisphenol A biodegradation.
MATERIALS AND METHODS
Chemicals. Bisphenol A, bis-(4-hydroxyphenyl)methane,
and compounds tested for the ability to support the growth
of strain MV1 as the sole carbon source (see Results) were
purchased from Aldrich Chemical Co., Inc., Milwaukee,
Wis. 2,2-Bis(4-hydroxyphenyl)butane was purchased from
TCI American, Inc., Portland, Oreg. 4,4’-Dihydroxystilbene
was purchased from Spectrum Chemical Corp., Gardena,
Calif. All other bisphenols tested were chemically synthe-
sized at General Electric Corporate Research and Develop-
ment, Schenectady, N.Y. Bis(trimethylsilyl)trifluoroaceta-
mide was purchased from Supelco, Inc., Bellefonte, Pa. All
the components of Luria agar except NaCl were purchased
from Difco Laboratories, Detroit, Mich.
Media and culture conditions. The PAS mineral salts basal
medium used for the isolation and growth of the bisphenol
A-degrading microorganism contained the following (in
grams per liter of distilled water): K2HPO4, 4.35; KH2P04,
1.7; NH4Cl, 2.1; MgSO4, 0.2; MnSO4, 0.05; FeSO4- 7H20,
0.01; and CaCl2- 2H20, 0.03. The medium was prepared by
adding a concentrated solution of phosphates and ammo-
nium salts (PA) to distilled water. PA was autoclaved for 20
min at 121 C and cooled to room temperature before the
1823

APPL. ENVIRON. MICROBIOL.
addition of a concentrated, filter-sterilized solution of min-
eral salts. The pH of the medium was 6.8. Unless specified,
bisphenol A (0.5%) was added to PA before it was auto-
claved. Since the solubility of bisphenol A is -1.5 mM in
PAS medium, crystals of bisphenol A were present in the
medium during growth to maintain the concentration of
bisphenol A near saturating levels. Liquid cultures contained
25 ml of PAS medium in 50-ml Erlenmeyer flasks stoppered
with polyurethane foam plugs. Unless indicated otherwise,
an inoculum (5% [vol/vol]) of strain MV1 grown on bisphe-
nol A was added to each flask and incubated at 30 C in a
rotary incubator-shaker at 200 rpm (model G25; New Bruns-
wick Scientific Co., Inc., Edison, N.J.). Cells grown on
glucose or 4-hydroxybenzoic acid (4-HBA) were cultured in
PAS medium containing 20 mM glucose or 4-HBA. When
strain MV1 was grown on bisphenol A in a 1-liter stirred
bioreactor, the pH and temperature were maintained at 6.5
and 30 C, respectively. Oxygen was supplied by sparging
with air, and the dissolved oxygen was monitored continu-
ously with a dissolved-oxygen probe and analyzer (New
Brunswick Scientific).
For solid PAS medium, 16 g of purified agar was added per
liter of PA, and this mixture was autoclaved for 20 min and
cooled to 50 C before the addition of a concentrated solution
of mineral salts. Solid PAS medium containing bisphenol A
was prepared by the addition of powdered bisphenol A (2
g/liter) to PAS medium cooled to 50 C and immediately
distributed into petri dishes. Cultures grown on solid PAS
medium were incubated at 30 C for 1 to 2 weeks.
Enrichment and isolation. Ten milliliters of sludge from a
wastewater treatment plant at a plastics manufacturing facil-
ity was added to a 125-ml foam-stoppered shake flask
containing 50 ml of PAS medium and 100 mg of bisphenol A
crystals as the carbon source. A subculture was obtained by
transferring 2 ml of the grown enrichment into 50 ml of PAS
medium containing 100 mg of bisphenol A. Subsequent
cultures grown on bisphenol A were used to obtain the
pure-culture isolate, bacterial strain MV1.
Strain MV1 was isolated on solid PAS medium containing
bisphenol A crystals dispersed throughout the agar. Several
colonies that grew on the agar containing bisphenol A were
streaked onto Luria agar containing the following (in grams
per liter of distilled water): tryptone, 10; yeast extract, 5;
NaCl, 5; glucose, 1; and agar, 15. Colonies that grew in 24 to
48 h were restreaked on Luria agar several times to ensure
culture purity.
Strain MV1 was submitted to the American Type Culture
Collection (Rockville, Md.) for identification. A variety of
biochemical and physiological tests routinely used to taxo-
nomically classify microorganisms were performed. The
strain was compared with several Pseudomonas strains and
CDC group Ve (5, 7).
Growth on other carbon sources. A culture grown on either
bisphenol A or 4-HBA was inoculated into PAS medium
containing one of several bisphenols tested (see Fig. 3) or
one of the compounds tested for the ability to support the
growth of strain MV1 as the sole carbon source (see Re-
sults). When the solubilities permitted, the potential sub-
strates were tested at concentrations of 1 and 5 mM. The
compounds were filter sterilized and aseptically added to
sterile PAS medium. The cultures were incubated at 30 C for
up to 5 days. Growth was determined by measuring the
increase in turbidity at A600 on a Perkin-Elmer Lambda 5
UV-VIS spectrophotometer. High-pressure liquid chroma-
tography (HPLC) analyses of the culture supernatants were
performed before and after incubation as described below.
Inhibition studies. Inhibition of cell growth by the meta-
bolic intermediates 4-HBA and 4-hydroxyacetophenone (4-
HAP) was tested in 20-ml screw-cap tubes containing 4 ml of
PAS-glucose medium. Sterile solutions of the metabolites
were aseptically added to the tubes to achieve concentra-
tions of 1 to 20 mM. A 5% inoculum of an overnight culture
grown on glucose was added to each of the tubes. Growth
was measured as described below. The lowest concentration
of 4-HBA or 4-HAP that did not produce at least a twofold
increase in turbidity at A600 was the MIC.
Growth studies. Growth rates of strain MV1 on three
different carbon sources (bisphenol A, 4-HBA, and glucose)
were determined in triplicate 50-ml Erlenmeyer flasks con-
taining 20 ml of PAS medium. Unless indicated otherwise, a
5% inoculum was added and the cultures were agitated at 250
rpm and 30 C. Growth was monitored by measuring turbid-
ity atA600 in a 1-cm-wide cuvette on a Perkin-Elmer Lambda
5 UV-VIS spectrophotometer. Turbidity of 1.0 at A600 rep-
resented 1.6 x 109 cells per ml, as determined from the
number of CFU per milliliter after plating on PAS-glucose
agar.
The rate of growth of strain MV1 on bisphenol A was
determined in shake flasks containing PAS medium and
bisphenol A crystals larger than 2 mm. Cell growth was
measured in a 1-cm-wide cuvette as an increase in turbidity
at A600. The large crystals of bisphenol A settled to the
bottom of the flasks before sampling and did not interfere
with the cell turbidity measurements. The bisphenol A
concentration remained near saturation (-1.5 mM) for the
growth period.
The pH optimum was determined by adjusting the pH of
PAS medium from 5.0 to 8.0 by altering the relative propor-
tions of mono- and dibasic potassium phosphates. A culture
grown at pH 6.5 was inoculated into each flask and incubated
at 30 C for 3 days. Either glucose or bisphenol A was added
as the sole carbon source, and growth was monitored.
Carbon balance analysis. Cultures of strain MV1 were
grown in triplicate 120-ml serum bottles crimp sealed with
black rubber stoppers. Each bottle contained 19.2 ml of
C02-free PAS medium and the quantity of each substrate
indicated in Table 1. The culture used as an inoculum was
grown on the same substrate to which it was added. Cells
were collected by centrifugation in an Eppendorf microcen-
trifuge and resuspended in C02-free PA before being added
to bottles. The bottles were pressurized with air to 1.5
atmospheres (ca. 151.9 kPa) and incubated for 48 h at 30 C
on a rotary incubator-shaker at 250 rpm.
The inorganic carbon (IC) and total carbon (TC) analyses
for the carbon balance were performed on a total organic
carbon analyzer (model TOC-500; Shimadzu Corp., Kyoto,
Japan). A 1,000-ppm organic carbon standard for TC analy-
sis was prepared by dissolving 2.125 g of reagent-grade
potassium hydrogen phthalate in 1 liter of glass-distilled
water. Appropriate dilutions were prepared to produce a
standard curve from 10 to 200 ppm of organic carbon. A
1,000-ppm IC standard was prepared by dissolving 3.5 g of
reagent-grade sodium hydrogen carbonate in 1 liter of glass-
distilled water. A standard curve from 10 to 200 ppm of IC
was obtained by use of the appropriate dilutions. Gas
standards for CO2 gas analysis were prepared by the addition
of the appropriate volumes of N2 or CO2 gas to evacuated
serum bottles to prepare 10, 20, and 30% CO2 gas standards.
Gas-phase CO2 was determined by use of multiple injec-
tions of 50 ,ul of the gas phase into the IC port of the
TOC-500 analyzer. The CO2 dissolved in 20 ml of culture
medium was measured directly by use of multiple injections
1824 LOBOS ET AL.