oxins 2013, 5, 605-617; doi:10.3390/toxins5040605
Detection of Mycotoxins in Patients with Chronic Fatigue
Joseph H. Brewer 1,*, Jack D. Thrasher 2, David C. Straus 3, Roberta A. Madison 4 and
Dennis Hooper 5
1. Plaza Infectious Disease and St. Luke’s Hospital, 4320 Wornall Road, Suite 440, Kansas City,
MO 64111, USA
2. Citrus Heights, CA 95610, USA; E-Mail: firstname.lastname@example.org
3. Department of Immunology and Molecular Microbiology, Texas Tech University Health Sciences
Center, Lubbock, TX 79430, USA; E-Mail: David.Straus@ttuhsc.edu
4. California State University, Northridge, CA 91330, USA;
5. RealTime Laboratories, Carrollton, TX 75010, USA; E-Mail: email@example.com
* Author to whom correspondence should be addressed; E-Mail: firstname.lastname@example.org;
Tel.: +1-816-531-1550, Fax: +1-816-531-8277.
Received: 18 March 2013; in revised form: 1 April 2013 / Accepted: 3 April 2013 /
Published: 11 April 2013
Abstract: Over the past 20 years, exposure to mycotoxin producing mold has been
recognized as a significant health risk. Scientific literature has demonstrated mycotoxins as
possible causes of human disease in water-damaged buildings (WDB). This study was
conducted to determine if selected mycotoxins could be identified in human urine from
patients suffering from chronic fatigue syndrome (CFS). Patients (n = 112) with a prior
diagnosis of CFS were evaluated for mold exposure and the presence of mycotoxins in
their urine. Urine was tested for aflatoxins (AT), ochratoxin A (OTA) and macrocyclic
trichothecenes (MT) using Enzyme Linked Immunosorbent Assays (ELISA). Urine
specimens from 104 of 112 patients (93%) were positive for at least one mycotoxin (one in
the equivocal range). Almost 30% of the cases had more than one mycotoxin present. OTA
was the most prevalent mycotoxin detected (83%) with MT as the next most common
(44%). Exposure histories indicated current and/or past exposure to WDB in over 90% of
cases. Environmental testing was performed in the WDB from a subset of these patients.
This testing revealed the presence of potentially mycotoxin producing mold species and
mycotoxins in the environment of the WDB. Prior testing in a healthy control population
Toxins 2013, 5 606
with no history of exposure to a WDB or moldy environment (n = 55) by the same
laboratory, utilizing the same methods, revealed no positive cases at the limits of detection.
Keywords: mycotoxin; mold exposure; chronic fatigue syndrome; Stachybotrys
Chronic fatigue syndrome (CFS), also called myalgic encephalitis, has been widely studied over the
past 25 years. Numerous mechanisms and theories have been proposed to explain its pathophysiology,
epidemiology, clinical features and causation [1–4]. Possible causations include infections (particularly
by viruses), oxidative stress, immune aberrations and toxic exposures, among others. However, no
single etiology has been confirmed to fully explain this syndrome. In many circumstances, these
patients remain chronically ill despite varying attempts at treatment [1–4].
During the same time frame, there has been a growing body of scientific literature indicating that
mycotoxins and exposure to mycotoxin producing molds has become hazardous to the health of
occupants of water-damaged buildings (WDB) (homes, schools and places of business).
Water-damaged environments contain a complex mixture of biocontaminants produced by both mold,
Gram-negative and Gram-positive bacteria . Secondary metabolites of molds and bacteria have been
identified in the dust, carpeting, wallpaper, heating, ventilation and air-conditioning (HVAC) systems
and respirable airborne particulates [6–16]. In addition, mycotoxins have been identified in clinical
isolates from corneal keratitis, aspergillosis and from body fluids and tissues of individuals exposed to
moldy environments [17–25]. Interestingly, patients with mycotoxin exposure in WDB frequently
have clinical features similar to CFS [5,26–29].
In this study, urine specimens were tested by ELISA-based assay to look for the presence of
mycotoxins in a group of patients with CFS. These results were compared to healthy control subjects
previously reported by the same testing laboratory. Additionally, in several cases, the WDB that were
the source of exposure were investigated for environmental mold and/or mycotoxins. A hypothesis of
possible mitochondrial damage in CFS is presented following review of the literature.
2. Materials and Methods
The study was conducted for 6 months from 1 February 2012 to 31 July 2012. Patients with chronic
illnesses, many of whom were previously diagnosed with CFS, were seen in a private practice (JHB)
which is a consultative outpatient infectious disease clinic in Kansas City, Missouri. Out of
approximately 300 patients with chronic illness that were seen for routine follow up clinic visit,
112 met the criteria for a diagnosis of CFS as outlined by Fikuda, et al. in 1994 . These patients
were from diverse geographic areas in the United States however, the majority resided in Midwestern
states. The patient ages ranged from 15 to 72 years with 84 (75%) females and 38 (25%) males. The
duration of symptoms ranged from 2 to 36 years with an average duration of 7.8 years. The illness was
so severe that 76 (68%) of the patients were either unable to work, receiving disability or unable to
Toxins 2013, 5 607
attend school. A past history of mold allergy and/or chronic sinusitis was present in 50 (48%) of the
patients. These patients failed to respond to treatments and their CFS symptoms lingered.
Common symptoms in this patient population included fatigue, headache, flu-like symptoms,
cognitive complaints, myalgia, arthralgia, gastrointestinal problems and various neurologic symptoms.
Other previous diagnoses included fibromyalgia, Lyme disease, peripheral neuropathy, orthostatic
intolerance (including postural orthostatic tachycardia syndrome and neural-mediated hypotension),
migraine, chronic dermatitis, gastroparesis, chronic abdominal pain, irritable bowel syndrome,
interstitial cystitis, anxiety, depression, chemical sensitivity, vertigo, chronic sinusitis, gluten
intolerance, tremor, myoclonus and cognitive dysfunction.
Routine laboratory parameters including complete blood count and chemistry panels were usually
normal. Immune testing had been performed previously in most of these patients. The most common
abnormality was diminished natural killer cell (NK) function. Other immune abnormalities were
occasionally noted (e.g., hypogammaglobulinemia).
Since these chronic conditions have been reported to be associated with exposure to mold and
bacteria in WDB and previous studies have shown an association between CFS and sick building
syndrome (SBS), it was decided to carry out an environmental history and discuss urine mycotoxin
testing . During these follow up visits, over 90% of the 112 patients confirmed exposure to a WDB
and frequently the presence of a moldy environments in the home, workplace or both.
2.2. Control Subjects
Healthy control patients with no known toxic mold exposures in water-damaged buildings were
previously reported . These controls (n = 55) consisted of 28 males and 27 females, ages 18 to
72 years. These patients were also from diverse geographic areas and resided in various areas of the
United States. Urine specimens from these individuals were used to develop reference data for the
control group used in this study. Furthermore, the same control subjects were also asked about
complaints and/or symptoms related to mold exposure as documented in the peer reviewed literature at
the time of this study . Symptoms that were screened included rhinitis, cough, headache,
respiratory symptoms, central nervous system symptoms, and fatigue. They did not give a history of
water-intrusion or mold growth in the workplace or at home. It was assumed that the controls had
exposure to foods and airborne mold spores that occur in their daily activity.
2.3. Mycotoxin Testing
Mycotoxin determination was conducted in similar fashion as described earlier with
modifications . Competitive direct enzyme linked immunosorbant assays (ELISAs) were
conducted on all groups of mycotoxins studied (AT, OTA, MT). Validated, competitive direct ELISA
tests for MT and AT/OTA (private communication, RealTime Laboratories, Inc, Carrollton, TX) were
conducted on all urines submitted . Validations have demonstrated that urine is the best fluid for
evaluation. However, the variability of urine matrix components such as organic compounds, pH and
electrolytes can affect antibody binding and assay performance in ELISA tests. To account for these
matrix effects, standard sample diluents for plasma serum, cell culture and other biological specimens
have been developed. No standard diluent has been developed for urine and many other biological
Toxins 2013, 5 608
fluids. Instead, phosphate buffered saline (PBS) was used as a “urine-like” diluent. In the validations
and continued testing of mycotoxins, a 10% methanol in PBS +10% methanol dilution (pH 7.2–7.4)
was used to compensate for the matrix effect in urine.
MT ELISA test: Urines were diluted at 1:5 to compensate for the matrix affect. Coated roridin A
(an MT) antibody ELISA wells (Beacon Analytic Inc., Maine) were inoculated with 100 microliters
(μL) of controls, calibrators, or patients specimens. One hundred microliters of diluted urine were
placed in each test well. One hundred μL of known MT (roridin A, Sigma Inc., St. Louis, MO)
calibrators (diluted: 10.0 μL/dL, 1.0 μL/dL, and 0.1 μL/dL, 0 μL/dL) and high, low, and negative
controls were placed in specific wells. Samples were incubated for 15 minutes at 21–25 degrees C
under continual rotation. One hundred μL of 1:1800 dilution of roridin A HRP-conjugate (Beacon
Analytic Inc., Maine) were added to each well and incubated for 15 minutes at 21–25 degrees C, under
continual mild rotation. All plates were washed 4 times with deionized water and tapped until
deionized water was removed. One hundred μL of substrate (Beacon Analytical Labs, Inc.) were
placed in each well and incubated 30 minutes at 21–25 degrees C under continual rotation. One
hundred μL of 1 N HCl were added to stop the reaction. The reactions were read on a Spectra Max 190
Spectrophotometer (Molecular Devices, Sunnyvale, CA) at 450 nm. Results were tabulated and
entered into a semilog software program (Beacon Analytical Labs, Inc.). Results were tabulated and
reported as ng/dL or parts per billion (ppb). All controls and calibrators met regulatory conditions as
specified in the standard operating procedures.
AT and OTA ELISA procedures: Urines were diluted at 1:7 to compensate for the matrix affect.
Coated wells (Neogen Corporation, Michigan) with either polyclonal antibodies to AT or OTA were
used in the separate mycotoxin procedures. Procedures for AT and OTA determinations were identical
except for the specific antibody coating the ELISA plates. Antibody ELISA wells (Neogen
Corporation) were inoculated with 100 μL of calibrators, controls or patients specimens. Initially,
100 μL of AT-HS conjugate (Neogen) and 100 μL of OTA conjugate (Neogen) were placed in the
respective antibody wells. One hundred μL of known antigen (AT or OTA, Trilogy Inc, MO)
calibrators for AT were 0, 1, 2, 4, and 8 ng/dL (ppb) and calibrators for OTA were 0, 2, 5, 10,
25 ng/dL (ppb). High, low, and negative controls for each mycotoxin were also placed in specific
wells. One hundred μL of diluted urine were placed in each test well. Plates were incubated at
21–25 degrees C for 10 minutes under continual mild rotation. All plates were washed 4 times with
deionized water and tapped until deionized water was removed. One hundred μL of substrate (Neogen)
were placed in each well and incubated 10 minutes at 21–25 degrees C under continual mild rotation.
One hundred 100 μL of 1N H2SO4 were added to stop the reaction. The reactions were read on a
Spectra Max 190 Spectrophotometer (Molecular Devices, Sunnyvale, CA) at 650 nm. Results were
tabulated and entered into a semilog software program (Neogen Inc). Results were tabulated and
reported as ng/dL or ppb. All controls and calibrators met regulatory conditions as specified in the
standard operating procedures.
Statistics were performed on the patient data and controls for each of the three mycotoxins (AT,
OTA and MT). Two-sided independent t-tests were performed on OTA and the MT. Two-sided
Toxins 2013, 5 609
Fischer exact test was performed on the AT because the control data were negative (zero) for
3. Results and Discussion
Mycotoxin testing revealed the presence of at least one of the toxins in the urine of 104 out of 112
(93%) patients. This included 103 with positive results and one that was in the equivocal range for
OTA. The frequency of the various mycotoxins in the urine of CFS patients (based upon the suggested
detection limits of RealTime Laboratories) is summarized Table 1. OTA was most commonly detected
mycotoxin comprising 83% of the patients. This was followed by MT (44%) and AT (12%). The
presence of combinations of mycotoxins in the urine were follows: OTA + MT (23%), AT + MT (4%),
and all three (8%).
Table 1. This table summarizes the detection of the mycotoxins in the urine of chronic
fatigue syndrome (CFS) patients individually or in combinations. The ranges and averages
are based upon the actual number of individual positives for each mycotoxin.
Mycotoxin Positive (N, %) Range (ppb) Average (ppb)
ATa 13, 12% 1.1–9.4 4.67
OTAa 87, 83% 2–14.6 6.2
MTa 46, 44% 0.21–5.72 0.85
OTA + MT 24, 23% N/Ab N/Ab
AT + MT 4, 4% N/Ab N/Ab
AT, OTA, MT 8, 8% N/Ab N/Ab
a: Limits of Detection: AT (1 ppb); OTA (2.0 ppb); MT (0.2 ppb). b: N/A: Not applicable.
The CFS patients were compared to a previously published group of healthy control subjects that
had no history of exposure to a WDB or moldy environment. The frequency of detection of these three
mycotoxins in the CFS patients compared to controls is seen in Table 2.
Table 2. Detection of mycotoxins in CFS patients compared to healthy controls.
Patient Group Number Tested ATa,b OTAa,b MTa,b Any Mycotoxinb
CFS 112 12 (12%) 87 (83%) 46 (44%) 104 (93%)
Controlc 55 0 0 0 0
a: Limits of Detection: same as Table 1. b: Number positive, percent positive; c: Control group
previously published .
The concentration of mycotoxins in the urine of patients and controls were statistically analyzed to
determine if a difference existed between the two groups. These data are summarized in Table 3. The
concentrations were significantly elevated in the patients compared to controls as follows: AT
(0.43 ± 1.36 vs. 0 ± 0 ppb, p = 0.0007), OTA (5.26 ± 3.65 vs. 0.355 ± 0.457 ppb, p < 0.0001), and MT
(0.422 ± 0.714 vs. 0.0169 ± 0.0265 ppb, p < 0.001).
Toxins 2013, 5 610
Table 3. This table summarizes the independent two tailed t-tests performed on the
patients and controls with respect to ochratoxin A (OTA) and macrocyclic trichothecenes
(MT). The Fisher 2-Sided Exact test was performed on the aflatoxins (AT) because the
control group had non-detection of AT. The mean and standard deviations are listed in ppb
for each mycotoxin (patients and controls).
Mycotoxin Patients (N = 104)
Controls (N = 55)
AT 0.43 ± 1.36 0 ± 0 —– 0.0007a
OTA 5.26 ± 3.65 0.355 ± 0.457 13.5 <0.0001
MT 0.422 ± 0.714 0.0169 ± 0.0265 5.78 <0.001
a: Fisher 2-Sided Exact Test Matrix: Controls (55 and 0); Patients (87 and 17).
Environmental histories of these patients were positive for exposure to WDB (many with visible
mold) in over 90% of the cases tested, including residential and/or workplace. In the residential group,
water damage to the basement was a common finding. However, other sources of water intrusion were
noted during history taking, which included water pipe leaks, roof leaks, window leaks and plugged
drains. In 24 patients, symptoms, which eventually became chronic, started within one year of the
exposure in the WDB.
Environmental tests (air spore counts, tape lifts and the examination of dust for mycotoxins) were
performed in 10 of the situations of the 104 patients (data not shown). In addition, two families
discussed below also conducted environmental testing. In the 10 cases mold genera associated with the
potential for mycotoxin production were found. In 8 of the situations, Stachybotrys was identified in
the WDB. In each of these 8 patients, MT was detected in the urine assay. In addition,
Aspergillus/Penicillium-like spores were detected in 8 buildings to which these patients were exposed.
The urine mycotoxin assays identified OTA in 5 patients and AT was present in 2 subjects.
Additionally, dust specimens collected from 5 homes and one office building were sent to RealTime
Laboratories for mycotoxin testing on environmental dust. MT was found in the dust samples from all
6 of these buildings. Small amounts of OTA were detected in 4 of the dust samples. There were
7 patients that had been exposed to mold in these buildings. Of these 7 patients, 6 had tested positive
for MT in the urine assay, with the levels ranging from 0.21 ppb to 5.72 ppb. Additionally, 4 of the
7 patients had tested positive for OTA with values ranging from 3.7 ppb to 10.2 ppb.
The two families that conducted environmental tests on their homes are presented below. The
families consisted of four individuals per household (Tables 4 and 5).
Family #1: The parents moved into a new home in 1991 and the family has lived there since. The
father began to develop symptoms of fatigue, muscle aches and cognitive problems, which was
subsequently diagnosed as CFS, within 4 months of moving into their new home. Within 3 years, the
mother developed CFS. Neither of the parents had any history to suggest occupational exposure
outside of the home. The two daughters were born and raised in that home. Both children developed
chronic illness (CFS) while living in the home. All four remain chronically ill (CFS). Urine AT, OTA
and MT concentrations (ppb) for each family member were as follows: father 0, 0, 0.59; mother 0, 3.6,
0.19; daughter 0, 4.2, 0.13 and another daughter 0, 3.6 and 0.17. Table 4 summarizes the results from
Toxins 2013, 5 611
Table 4. Detection of various mycotoxins in the urine of members of Family #1.
Age Sex ATa
Father 50 M 0 0 0.59
Mother 49 F 0 3.6 0.19
Child 19 F 0 4.2 0.13
Child 16 F 0 3.6 0.17
a: Limits of detection same as Table 1.
Standard air testing from their home revealed high numbers of Aspergillus spores in one area of the
home. Dust samples were collected on three different occasions from top of doorway jambs, kitchen
cabinets, bedroom, living room, kitchen, and home office and sent to Mycometrics LLC, (Monmouth
Junction NJ) for MSQPCR-36 (ERMI) testing . The moldiness indices for these samples were
8.61, 16.54 and 16.7. Mycotoxin producing molds identified in dust samples were as follows:
Aspergillus (flavus, fumigatus, niger, ochraceus and versicolor); Penicillium (brevicompactum,
purpurogenum, crustosum, corylophilum and chrysogenum); Chaetomium globosum, Stachybotrys
chartarum and Trichoderma viride. A dust sample from under the refrigerator sent to RealTime
Laboratories for mycotoxin testing revealed MT (0.42 ppb) and OTA (0.6 ppb). The family had no
idea there was a “mold problem” in their home until 2012 when the environmental testing
Family #2: All members of this family had chronic illness (CFS, celiac disease, chemical
hypersensitivity) which had developed after living in this home. The family moved into a home in
1997 and within months discovered problems with the exterior drainage which led to water intrusion.
There was subsequent flooding of the lower level of the home on multiple occasions. Environmental
air sampling of the home in 2005 revealed Aspergillus/Penicillium-like spores and
Stachybotrys-spores. The family moved to a different home in 2005 (within months of the testing
results) but all remained ill. The father had no history of occupational exposure outside the home and
the mother did not work outside the home during this time frame. The urine mycotoxin levels (ppb) for
AT, OTA, and MT in this family were as follows: father 0, 4.6, 0.02; mother 0, 6.8, 0.01; son 0.5, 6.1,
0.48 and daughter 0, 2.3, 0.03. The results for this family are seen in Table 5.
Table 5. Detection of various mycotoxins in the urine of members of Family #2.
Age Sex ATa
Father 49 M 0 4.6 0.02
Mother 54 F 0 6.8 0.01
Child 23 M 0.5 6.1 0.48
Child 15 F 0 2.3 0.03
a: Limits of detection same as Table 1.
The etiology of CFS has been studied for several decades and numerous proposed etiologies have
been suggested [1–4]. Studies of CFS patients have demonstrated evidence of increased viral
activation, oxidative stress, immune abnormalities, neurocognitive features and endocrine
abnormalities [1,2]. In addition, CFS patients have mitochondrial dysfunction with impaired oxidative
Toxins 2013, 5 612
phosphorylation, low ATP stores and increased lactic acid with exercise [32–34]. Individuals that have
been exposed to WDB frequently have clinical features similar to CFS [5,22–29]. One study reported
concurrent sick building syndrome and CFS . However, it should be recognized that additional
symptoms in CFS patients include fibromyalgia, headaches, loss of balance, neurocognitive
difficulties, flu-like symptoms, irritable bowel syndrome, anxiety, depression, among others
In this study, patients with a prior diagnosis of CFS were evaluated for the presence of mycotoxins
utilizing a sensitive and specific ELISA-based urine assay for three common mycotoxins. Ninety-three
percent of the cases demonstrated the presence of at least one of the mycotoxins in the urine. Over
90% of the patients gave a history of exposure to WDB. Additionally, mycotoxin-producing mold
species, mycotoxins or both were demonstrated in WDB that were associated with exposures in 18 of
these patients. The demonstration of the actual toxins in dust samples from the buildings in which the
patients either lived or worked is of considerable interest. This is because the same mycotoxins were
recovered from the urine of these patients. Trichothecene mycotoxins can be found in small fragments
as well as in conidia . Furthermore, a variety of mycotoxins and bacterial exotoxins are present in
the dust and building materials of WDB [8–16]. Therefore, exposure to microbial toxins is most likely
underestimated, particularly since mold and bacteria shed fine respirable particulates less than 1 micron
in diameter in water-damaged conditions that contain toxins and other by-products [5,6,35–40].
The common denominators in these patients included CFS, additional symptoms, a water-damaged
environment, indoor mold and urine specimens positive for mycotoxins. OTA was the most common
mycotoxin detected in 83% of subjects followed by MT (44%) and AT (13%). Interestingly, more than
one of the mycotoxins was also present ranging from 8% (all three mycotoxins) to 23% (MT and
OTA) (Table 1). Moreover, the major mycotoxin in the urine of the two families (Tables 4 and 5) was
OTA, while MT were positive in two of the subjects. The question that arises is what is the probable
role of these mycotoxins in the symptoms experienced by the patients in this study? The Mitochondrial
Disease Foundation lists several masquerader health problems that are associated with mitochondrial
deficiency, which include the following organs: central nervous system, heart, peripheral nerves,
muscles, liver, ears, eyes, pancreas, digestive system and endocrine system. Manifestations of
mitochondrial deficiency can include autoimmune disorders, chronic fatigue, neurodegenerative
disorders (amyotrophic lateral sclerosis, multiple sclerosis, Parkinson’s disease), depression, other
psychiatric disorders, glycogen storage disorders, among others .
In vivo and in vitro studies have demonstrated that mycotoxins cause mitochondrial dysfunction.
Aflatoxins alter mitochondria as follows: mitochondrial DNA adducts, inhibition of protein synthesis,
pleomorphism, disruption of cristae, membrane damage and induction of apoptosis [42–45].
Trichothecenes have multiple inhibitory effects that include oxidative stress, apoptosis, inhibition of
protein, RNA and DNA synthesis, opening of phosphorescent Pt(II)-coporporphyrin (PtCP) and loss of
transmembrane potential and mitochondrial translation [46–50]. With respect to OTA the primary
thrust has been detecting its role in urinary tract and kidney diseases. However, the research into
kidney diseases has shown that OTA is also a mitochondrial poison. Mitochondrial abnormalities
resulting from OTA include membrane swelling, disarray of cristae, loss of transmembrane potential,
inhibition of succinate cytochrome c reductase and succinate dehydrogenase and inhibition of
succinate-supported electron transfer, and activities of the respiratory chain. The toxicity of OTA
Toxins 2013, 5 613
appears to result from oxidative stress leading to nuclear DNA damage, cytotoxicity and
apoptosis [51–54]. Thus, it appears that mitochondrial dysfunction may be correlated with the presence
of CFS and other symptoms in these patients.
The patients presented in this study had multiple symptoms including those consistent with CFS as
reported by others [22–29,55,56]. They had increased concentrations of AT, OTA and MT in their
urine samples compared to a group of previously published healthy controls. Their health conditions
and symptoms in these patients were suggestive of mitochondrial dysfunction as reported in subjects
with CFS [32–34]. In addition, the symptom complex of these patients was suggestive of
mitochondrial disease as reported by the Mitochondrial Disease Foundation . Moreover, AT, OTA
and MT can cause mitochondrial damage [42–54].
Mycotoxins can be detected in the urine in a very high percentage of patients with CFS. This is in
contrast to a prior study of a healthy, non-WDB exposed control population in which no mycotoxins
were found at the levels of detection. The majority of the CFS patients had prior exposure to WDB.
Environmental testing in a subset of these patients confirmed mold and mycotoxin exposure. We
present the hypothesis that mitochondrial dysfunction is a possible cause of the health problems of
these patients. The mitochondrial dysfunction may be triggered and accentuated by exposure
Conflict of Interest
Dr. Brewer and Madison declare no conflict of interest. Drs. Straus, Hooper and Thrasher have
served as expert witnesses in mold and mycotoxin exposure litigation.