Antimicrobial Effects of Quillaja saponaria

Extract Against Escherichia coli O157:H7

and the Emerging Non-O157 Shiga
Toxin-Producing E. coli
Snigdha Sewlikar and Doris H. D’Souza

Abstract: Natural alternate methods to control the spread of Shiga toxin-producing Escherichia coli (STEC) are important
to prevent foodborne outbreaks. Quillaja saponaria aqueous bark extracts (QE), cleared by the U.S. Food and Drug Admin-
istration as a natural flavorant, contain bioactive polyphenols, tannins, and tri-terpenoid saponins with anti-inflammatory
and antimicrobial activity. The objective of this study was to determine the effects of commercial QE against E.
coli O157:H7 and non-O157 strains over 16 h at 37 °C and RT. Overnight cultures of 4 E. coli O157:H7 strains and
6 non-O157 STECs in Tryptic Soy Broth (TSB) were washed and resuspended in phosphate-buffered saline (PBS, pH
7.2), and treated with QE and controls including citric acid (pH 3.75), sodium benzoate (0.1% w/w), acidified sodium
benzoate (pH 3.75) or PBS for 6 h or 16 h. Recovered bacteria were enumerated after plating on Tryptic Soy Agar, from
duplicate treatments, replicated thrice and the data were statistically analyzed. The 4 QE-treated E. coli O157:H7 strains
from initial 7.5 log CFU had remaining counts between 6.79 and 3.5 log CFU after 16 h at RT. QE-treated non-O157
STECs showed lower reductions with remaining counts ranging from 6.81 to 4.55 log CFU after 16 h at RT. Incubation
at 37 °C caused reduction to nondetectable levels within 1 h, without any significant reduction in controls. Scanning
electron microscopy studies revealed damaged cell membranes of treated bacteria after 1 h at 37 °C. QE shows potential
to control the spread of STECs, and further research in model food systems is needed.

Keywords: control, E. coli O157:H7, non-O157 STEC, Quillaja saponaria, reduction

Practical Application: Quillaja saponaria aqueous bark extracts (QE) have U.S. Food and Drug Administration clearance
as natural flavoring substances. The objective of this study was to determine the effects of commercial QE against E.
coli O157:H7 and the emerging non-O157 shigatoxin-producing E. coli (STEC) strains over 16 h at 37 °C and RT.

Food Microbiology &

The 4 E. coli O157:H7 strains and non-O157 STECs treated with QE at RT showed significant reductions after 16 h.
Incubation temperature of 37 °C caused reduction to nondetectable levels within 1 h, without any significant reduction

Safety
in the untreated controls. Scanning electron microscopy studies revealed damaged cell membranes of treated bacterial
cells after 1 h at 37 °C. QE shows potential to control the spread of STECs.

Introduction
Escherichia coli O157:H7 was first identified as a human pathogen
in 1982 and associated with 2 food-related outbreaks (consump-
tion of undercooked ground beef) of hemorrhagic colitis in Ore-
gon (26 cases) and Michigan (21 cases) (Riley and others 1983;
Wells and others 1983). E. coli O157:H7 causes hemorrhagic coli-
tis, that is, bloody diarrhea, hemolytic uremic syndrome (renal fail-
ure) (HUS), thrombotic thrombocytopenic purpura (TTP), and
is also referred to as enterohaemorrhagic E. coli (EHEC) (Doyle
1991; Siegler 1995; Kim and others 2016). The symptoms include
abdominal cramping, watery diarrhea which later becomes bloody
that can last from 2 to 9 d (Doyle 1991).
These pathogenic organisms have been isolated from animals,
including cattle, deer, sheep, goats, horses, dogs, as well as birds,
and flies (Griffin and Tauxe 1991; Chapman and others 1997;
JFDS-2016-1348 Submitted 8/19/2016, Accepted 2/21/2017. Authors are O121, and O145 are responsible for 70% to 83% of confirmed
with Dept. of Food Science and Technology, The Univ. of Tennessee-Knoxville, 2600
River Drive, Knoxville, Tenn. 37966, U.S.A. Direct inquiries to author D’Souza
(Email: ddsouza@utk.edu).
Hancock and others 1998; Schmidt and others 2014). Under-
cooked ground beef, raw milk, poultry, leafy vegetables, fruits,
nuts, dairy, sprout, and game meat are some of the implicated
food items in outbreaks in the U.S. by Shiga toxin–producing E.
coli (STEC) O157:H7 between 1998 and 2008, with beef being
the most common food source (Doyle 1991; CDC 2012: CDC
2013; Foodborne Disease Outbreak Surveillance System, United
States 1998–2008). Thorough cooking of food and proper han-
dling (following GMP and proper handwashing) are the important
inactivation and control measures to be taken in order to prevent
E. coli O157:H7 infections (Doyle 1991; Beier and others 2016;
Kim and others 2016).
The emerging non-O157 STEC serotypes cause 64% of the
265,000 STEC infections illnesses annually in the U.S. (Brooks
and others 2005; Scallan and others 2011). The active surveillance
system for STEC infections as part of FoodNet since 2001 indicates
that infections caused by serogroups O26, O45, O103, O111,
STEC illnesses (CDC 2010, 2011; 2012; 2013). The U.S. Dept. of
Agriculture (USDA) declared that STEC O26, O45, O103, O111,
O121, and O145 (termed as “Big Six”) in addition to STEC

C 2017 Institute of Food Technologists

 R

doi: 10.1111/1750-3841.13697 Vol. 82, Nr. 5, 2017 r Journal of Food Science 1171
Further reproduction without permission is prohibited
 Quillaja saponaria to control Escherichia coli . . .
O157, would also be considered as adulterants when found in raw
or nonintact beef (USDA, FSIS 2011). Given the incidence of
STEC infections and their severity of illness, natural plant extracts
are gaining popularity as control options in the food industry to
prevent food recalls and outbreaks.
The aqueous extract of the Chilean soapbark tree
(Quillaja saponaria Molina), composed of many bioactive triter-
penoid saponins (Guo and Kenne 2000), has the U.S. Food and
Drug Administration clearance (21 CFR 172.510) as a natural
flavoring substance and is used in food (as a foaming agent) and
beverages (as an emulsifier) (Chino and Wako 1992; Murakami and
Watanabe 1988a, 1988b; Naknukool and others 2011). In addi-
tion to Chile, this soapbark tree is native to China and many South
American countries including Bolivia and Peru. This quillaja ex-
tract is also cleared by the EU for use in water-based nonalcoholic
drinks under the code E999 (CAS number: 68990-67-0) (Tam
and Roner 2011). Additionally, in Japan, it is used as an emulsifier
and foaming agent and for application in cosmetics (FAO 2004).
In order to avoid deforestation, and to have a sustainable, relatively
inexpensive source of the food-additive, the aqueous extracts are
obtained from the bark of the tree or wood of the branches (Tam
and Roner 2011; FAO 2004).
The aqueous extract of Quillaja (QE) is stable over a wide
pH range, that is pH 2 to 11 (Roner and others 2007). QE is
known to have anti-viral activity against vaccinia virus, herpes
simplex virus type 1 (HSV-1), varicella zoster virus (VZV), human
immunodeficiency viruses 1 and 2 (HIV-1, HIV-2), reovirus and
rhesus rotavirus (RRV) (both in vitro and in vivo) (Roner and others
2007, 2010). QE was not found to be cytotoxic at 0.1 mg/mL to all
the tested cell lines in the reported study (Roner and others 2007).
Saponins in Quillaja extracts are also known to have hemolytic
and antibacterial activity (against S. aureus, S. Typhimurium and
E. coli with the minimum inhibitory concentration (MIC) of
Food Microbiology &
0.1 mg/mL). In Quillaja saponin (QS), quillaic acid is the central
main aglycone attached with varying ratios of glucuronic acid,
rhamnose, hexose, and a fatty acyl chain (Higuchi and others
Safety
1987). These saponins are reported to act against eukaryotes due
to their ability to permeate the cell membrane by complexing
with cholesterol (Gogelein and Huby 1984). Research litera-
ture and reported studies suggest that the hemolytic activity of
saponins could be due to the affinity of their aglycone moiety
to sterols within membranes, in particular cholesterol (Gee and
others 1998). The hemolytic activity could also be attributed to
the effect of saponin effects by increasing cell membrane per-
meability through either forming pores in membranes, or alter-
ing sodium–potassium and calcium–magnesium ATPase activity
(Choi and others 2001), and/or also by insertion of the hydropho-
bic saponin nucleus into the lipid bilayer of membranes (Hu and
others 1996).
Based on this information, the aims of this study were: (a)
to determine the effect of commercially available aqueous QE
against E. coli O157:H7 and emerging non-O157 STECs at room
temperature (RT) and 37 °C over 16 h, and (b) to understand the
mechanism of action of QE using Scanning Electron Microscopy
(SEM).
Materials and Methods
Bacterial strains
The bacterial strains used in this study were obtained from Statistical analysis
ATCC (Manassas, Va., U.S.A.), Dr. Pina Fratamico, USDA- The recovered bacterial counts from all the treatment times
APHIS (Beltsville, Md., U.S.A.) and from the culture collections and controls (means ± S.D.) were statistically analyzed using
1172 Journal of Food Science r Vol. 82, Nr. 5, 2017
of the Animal Science and the Food Science and Technology
Depts. at the Univ. of Tennessee, Knoxville. A total of 4 Shiga
toxin-producing E. coli (STEC) O157:H7 strains (Jack in the box
outbreak strain, ATCC 43894 strain, ATCC 43888 strain, and
CIDER outbreak strain), and the “Big Six” non-O157 STEC
serotypes (E. coli O145:H18, E. coli O26:H11, E. coli O121:H19,
E. coli O103:H11, E. coli O45:H2, and E. coli O111:H8) were
used in this study.
Treatment of bacterial strains with QE: Cultures of the 4 E. coli
O157:H7 and 6 non-O157 STECs were grown overnight in
Tryptic Soy Broth at 37 °C (TSB, Difco, Becton Dickinson). One
milliliter of the homogenous bacterial culture was centrifuged for
3 min at 5000 × g to obtain a pellet which was then washed twice
and resuspended in phosphate-buffered saline (PBS, Difco, Bec-
ton Dickinson; pH 7.2), diluted 1:10 with PBS and treated with
equal amounts of filter sterilized commercially available undiluted
QE (pH 3.75) (obtained as a gift from Desert King, San Diego,
Calif., U.S.A.) or PBS (untreated control) for 6 h or 16 h at room
temperature (longer times were used compared to those at 37 °C,
since initial studies did not show any reduction after shorter peri-
ods at RT) or for up to 1 h at 37 °C (Arabski and others 2012;
Su and others 2012). The commercially available QE has a pH
of 3.75 and contains sodium benzoate (0.1% w/w). The bacterial
cultures were therefore treated with equal amounts of citric acid
(pH control, pH 3.75) or sodium benzoate (0.1% w/w) for 16 h to
determine if either the pH of QE or the sodium benzoate content
(individually or when used alone) contributes to its antibacterial
activity. Sodium benzoate, acidified to pH 3.75 was also used as
a control to test the synergistic effect of pH and preservative on
the bacterial cells. All treatments were carried out in duplicate and
replicated thrice.
Enumeration of recovered bacterial counts
Treated or control bacteria were serially diluted and plated
on Tryptic Soy Agar (TSA; Difco, Becton Dickinson), and
incubated overnight at 37 °C. The recovered counts of bacteria
were enumerated as colony forming units (CFU) per mL (Su and
others 2012).
Scanning electron microscopy (SEM) sample preparation
and observation
E. coli O157:H7 ATCC 43894 and E. coli O145:H18 were
chosen as representative strains for E. coli O157:H7 and non-O157
STECs, respectively. Washed overnight cultures of the strains,
resuspended in PBS (to achieve initial counts of 7 log CFU/ml),
were mixed with equal volumes undiluted QE or PBS and in-
cubated at 37 °C for 1 h. Primary fixation of samples was carried
out with 3% glutaraldehyde in 0.1 M cacodylate and secondary
fixation with 2% osmium tetroxide in 0.1 M cacodylate as de-
scribed before and per the guidelines of the Advanced Microscopy
and Imaging Center at the Univ. of Tennessee-Knoxville (Su
and others 2012). The fixed samples were dehydrated through
graded ethanol series (25% ethanol to 100% dry ethanol) and
then subjected to critical point drying (as described earlier by Su
and others 2012). One sample was used per treatment for SEM
visualization at the Advanced Microscopy and Imaging Center at
the Univ. of Tennessee-Knoxville with the assistance of Dr. John
Dunlap.
Quillaja saponaria to control Escherichia coli . . .
analysis of variance (ANOVA) for each bacteria separately. The
Tukey’s test was used to determine differences in recovered counts
over incubation time (fixed at 6 h and 16 h) between control
and treatments at P  0.05 with the Statistical Analysis Soft-
ware (SAS, version 9.2, SAS Institute, Cary, N.C., U.S.A.) us-
ing Proc GLM. The null hypothesis being that there would
be no significant difference between the recovered counts (that
were converted to log CFU) from treatments and controls over
time.
Results and Discussion
Antibacterial effects of QE
Treatment of E. coli O157:H7 at initial 7.5 log CFU/mL with
undiluted QE at RT resulted in remaining 6.79 ± 0.09 and 6.04
± 0.05 log CFU for the ATCC 43888 strain, 6.19 ± 0.04 and
6.24 ± 0.07 log CFU for the CIDER strain, 6.41 ± 0.33 and 4.55
± 0.32 log CFU for Jack in the Box strain, and at least 3 ± 0.17
and 3 ± 0.09 log CFU for ATCC 43894 strain after 6 h and 16 h,
respectively (Figure 1).
Treatment of the non-O157 STECs with undiluted QE at RT
resulted in remaining counts of 6.65 ± 0.12 and 6.81 ± 0.19 log
CFU for O26, 7 ± 0.12 and 5.93 ± 0.01 log CFU for O45, 6.19 ±
0.04 and 6.24 ± 0.07 log CFU for O111, 7.11 ± 0.02 and 5.22 ±
0.23 log CFU for O121, and 7.23 ± 0.01 and 5.75 ± 0.02 log CFU
for O103 and 6.49 ± 0.33 and 4.55 ± 0.32 log CFU for O145 after
6 and 16 h, respectively (Figure 2). However, only 1 strain of each
non-O157 STEC was used in this study, as the main purpose was
to determine if QE would have antibacterial effect against the big
6 non-O157 STECs namely O26, O45, O103, O111, O121, and
O145. It is likely that different strains of these big 6 non-O157
STECs will respond differently to the same QE treatment-time
combinations.
At 37 °C, significantly increased reduction was obtained for
all the tested strains of at least 6 log CFU (to undetectable lev-
els) after 1 h of treatment with undiluted QE. A gradual in-
crease in reduction was observed for all tested O157 strains
at 37 °C over 1 h (Figure 3 and 4). All the 6 non-O157
STEC were reduced by 6 log CFU (to undetectable levels) after
30 min.
Since the tested pH of the provided commercial QE was 3.75,
citric acid with a pH of 3.75 as a pH control was used to treat the
tested STECs at similar temperatures and times as QE. In addition,
0.1% sodium benzoate was tested against the STECs as a control
for the same temperatures and times, as the commercial QE is re-
ported to contain 0.1% sodium benzoate. Using these controls, it
could be deduced if pH alone or the 0.1% sodium benzoate alone
showed any reduction in bacterial counts, and if the antibacte-
rial activity of QE was mainly due to pH, sodium benzoate, or
bioactive components of the QE or potentially combinations of
all 3.
The pH control showed no significant reduction for any tested
bacterial strain at RT even after 16 h, with merely 1 log differ-
ence for all strains at 37 °C only after 16 h. The sodium benzoate
control showed no significant reduction for any of the strains both
at RT and 37 °C. Sodium benzoate acidified to pH 3.75 also
showed no significant reduction for any of the strains at both RT
and 37 °C (P > 0.05). Thus, it appeared that the bioactive com-
ponents of QE played a major role in inactivating the tested STEC
strains.
Figure 1–Reduction of E. coli O157:H7 strains
Food Microbiology &
with undiluted QE at RT over 6 h and 16 h. QE:
Quillaja extract; PBS: phosphate buffered saline.
Different letters indicate significant differences
Safety
when compared within each bacterial strain over
time (P < 0.05).
Figure 2–Reduction of non-O157 Shiga
toxin-producing E. coli strains with undiluted QE
at RT over 6 h and 16 h. QE: Quillaja extract; PBS:
phosphate buffered saline Different letters
indicate significant differences when compared
within each bacterial strain over time (P < 0.05).
Vol. 82, Nr. 5, 2017 r Journal of Food Science 1173
 Quillaja saponaria to control Escherichia coli . . .
Food Microbiology &
Scanning electron microscopy (SEM) of treated
and untreated bacterial cells
Safety
SEM images of the tested STEC strains revealed the formation
of pores in cells treated with QE that could lead to leakage of
inner cell material with collapse of cells (Figure 5 and 6).
QE treatment resulted in significant reductions in bacterial
counts/numbers of the tested STEC strains. Temperature played a
significant role, where all the tested strains treated at RT showed
lesser reduction compared to treatments at 37 °C. Although, a
remainder of 3.5 log CFU was obtained at RT (for E. coli
O157:H7 ATCC 43894) after 6 h, lower remaining counts to
almost nondetectable levels (based on assay detection limits) were
obtained for all strains at 37 °C within 1 h. This difference could
be attributed to the difference in bacterial cell membrane perme-
ability at the 2 temperatures. Also, longer incubation time resulted
in higher reductions. No significant reductions were obtained for
all the controls used (pH control, sodium benzoate control, and
acidified sodium benzoate (pH 3.75)). This suggests that the an-
tibacterial activity demonstrated by QE could be attributed to its
inherent bioactive components and not the pH or presence of
preservative alone. The incubation with the extract at RT resulted
in the remainder of 3.5 log CFU from initial 7.5 log CFU only
after 6 h, which shows that it may not be practical for use at RT.
However, if it is used as a coating during transport of consumables
for long distances and times of 6 h, it could potentially prevent
surface contamination after 6 h contact.
Three commercially available Quillaja saponins (Sigma, Roth
and Nor-feed) were reported to decrease the growth of E. coli
1174 Journal of Food Science r Vol. 82, Nr. 5, 2017
Figure 3–Reduction of E. coli O157:H7 strains
with undiluted QE at 37 °C over 60 min. QE:
Quillaja extract; PBS: phosphate buffered saline
over 60 min. Different letters indicate significant
differences when compared within each bacterial
strain over time (P < 0.05).
Figure 4–Reduction of non-O157 Shiga
toxin-producing E. coli strains with undiluted QE
at 37 °C over 60 min. QE: Quillaja extract; PBS:
phosphate buffered saline over 60 min. Different
letters indicate significant differences when
compared within each bacterial strain over time
(P < 0.05).
K-12 at higher concentrations of saponins (>1 % for Roth and
Nor-feed saponins and >0.25% for Sigma saponins) (Sen and oth-
ers 1998). Therefore, these researchers suggested that at higher
concentrations, the saponins could cause destruction or alteration
of the cellular semi-permeability and ultimately cause cell death,
leading them to conclude that the antibacterial effect of QE against
E. coli was concentration-dependent. Other studies also showed
that glycyrrhetinic acid (aglycone of the saponin glycyrrhizin from
liquorice roots) used in the treatment of gastric ulcer had bacterio-
static effects (Hostettmann and others 1991), while medicagenic
acid (aglycone of alfalfa saponin) was reported to be bacteriostatic
against Trichoderma viride (Lu and others 1987). Arabski and others
(2012) reported that the activity of Quillaja saponin was bacteri-
cidal against all tested strains of STEC at 37 °C within 1 h. Other
bioactives like tannins and polyphenols present in QE could also
contribute to the antibacterial activity of the extract used in this
study. Therefore, it remains essential to evaluate the activity of in-
dividual bioactive components of commercial QE to understand
the primary components responsible for the antibacterial activity
of the extract.
Other plant extracts and polyphenols have also been stud-
ied for antibacterial activity against E. coli. Taguri and others
(2004) tested the effect of ten plant polyphenols (epigallocate-
chin, epigallocatechin-3-O-gallate, punicalagin, tannic acid, casta-
lagin, prodelphinidin, geraniin, procyanidins, a theaflavin mixture
of black tea, and green tea polyphenols treated with loquat
polyphenol oxidase) against 23 strains of E. coli (10 strains of en-
terohemorrhagic E. coli, 4 strains of nonpathogenic E. coli, 2 strains
Quillaja saponaria to control Escherichia coli . . .
Figure 5–Scanning electron microscopy images of E. coli O145 before (A and B) treatment and after (C) treatment with undiluted QE for 1 h at 37 °C.
of enteroinvasive E. coli (EIEC), and 5 strains of enterotoxigenic
E. coli with reported average minimum inhibitory concentration
(MIC) of all the polyphenols of 1519 ± 949 mg/mL, which was
greater than that obtained for S. aureus (20 strains) and Vibrio (27
strains). Therefore, they suggested that there was greater resistance
of E. coli to reductions by polyphenols than the tested S. aureus
or Vibrio strains (Taguri and others 2004). Voravuthikunchai and
others (2006) reported MICs and minimum bactericidal concen-
trations (MBC) of ethanolic extract of P. granatum (pomegranate)
Figure 6–Scanning electron microscopy images
of E. coli O157:H7 ATCC 43894 before (A and
B) and after (C and D) treatment with undiluted
QE for 1 h at 37 °C.
Food Microbiology &
Safety
against E. coli O157:H7 of 0.49 to 1.95 mg/mL and 1.95 to
3.91 mg/mL, respectively. Medicinal plants such as Acacia catechu,
Peltophorum pterocarpum, Holarrhena antidysenterica, Punica granatum,
Psidium guajava, Uncaria gambir, Quercus infectoria, and Walsura ro-
busta was reported to show antibacterial activity against E. coli
(including 3 strains of E. coli O157:H7, E. coli O26:H11, E. coli
O111:NM, and E. coli O22) with inhibition zones of 7 to 17 mm,
using the paper disc agar diffusion assay (Voravuthikunchai and
others 2004). They reported that Quercus infectoria (ethanolic and
Vol. 82, Nr. 5, 2017 r Journal of Food Science 1175
 Quillaja saponaria to control Escherichia coli . . .

aqueous extracts) and pomegranate (aqueous extract) were very analysis at the Advanced Microscopy and Imaging Center at the
effective against E. coli O157:H7. Thus, it is likely that the com- Univ. of Tennessee-Knoxville.
ponents of QE namely saponins, tannins, and polyphenols work
together to contribute towards antibacterial activity.
To understand the mechanism of action QE against the STECs, References
SEM analysis was undertaken which revealed alterations of the Arabski M, Wegierek-Ciuk
˛ A, Czerwonka G, Lankoff, A, Kaca, W. 2012. Effects of saponins
against clinical E. coli strains and eukaryotic cell line. J Biomed Biotechnol Article ID:286216.
cellular structures that are in line with the proposed mechanisms Beier RC, Franz E, Bono JL, Mandrell RE, Fratamico PM, Callaway TR, Andrews K, Poole TL,
Crippen TL, Sheffield CL, Anderson RC, Nisbet DJ. 2016. Disinfectant and antimicrobial
of action against the bacterial cell membrane (Gogelein and Huby susceptibility profiles of the Big Six non-O157 Shiga toxin-producing Escherichia coli strains
1984; Hu and others 1996; Gee and others 1998; Choi and others from food animals and humans. J Food Prot 79(8):1355–70.
2001). This study showed that QE causes the formation of pores Brooks JT, Sowers EG, Wells JG, Greene KD, Griffin PM, Hoekstra RM, Strockbine NA. 2005.
Non-O157 Shiga toxin–producing Escherichia coli infections in the United States, 1983–2002.
in the cell membrane leading to leakage of the inner cell material J Infect Dis 192:1422–9.
that could result in cell death. This study therefore demonstrated Centers for Disease Control and Prevention (CDC). 2011. Vital signs: incidence and trends
of infection with pathogens transmitted commonly through food—foodborne diseases active
the antibacterial effect of QE against pathogenic E. coli O157:H7 surveillance network, 10 U.S. sites, 1996–2010. Morb Mortal Wkly Rep 60:749–55.
and the “Big Six” non-O157 STECs. CDC. 2012. National Shiga toxin-producing Escherichia coli (STEC) Surveillance Overview.
Atlanta, Georgia: US Department of Health and Human Services.
Regarding the application of QE in the food industry, the “Gen- CDC. 2013. Surveillance for foodborne disease outbreaks — United States, 1998–2008. Morb
eral Standard on Food Additives (GSFA) of the Codex Alimentarius Mortal Wkly Rep Surveillance Summaries 62:1–34
CDC. 2010. Preliminary FoodNet data on the incidence of infection with pathogens transmitted
Commission lists QE as suitable for use as a foaming agent in ‘water-based commonly through food—10 States, 2009. Morb Mortal Wkly Rep 59:418–22.
flavored drinks’, including ‘sport’ or ‘electrolyte’ drinks and particulated Chapman PA, Siddons CA, Gerdan Malo AT, Harkin MA. 1997. A 1-year study of Escherichia
coli O157 in cattle, sheep, pigs and poultry. Epidemiol Infect 119:245–50.
drinks (GSFA category 14.1.4, 50 mg/kg maximum use level, at Step Chino Y, Wako M. 1992. Manufacture of transparent emulsions for foods and beverages. Patent
6 in the Codex process)”; (FAO 2004), and also used as antioxi- 04 51:83.
dants (Hisayuki and Takashi 1987; Kooryama and Chiba 1996). Choi S, Jung S, Kim C, Kim H, Rhim H, Kim S, Nah S. 2001. Effect of ginsenosides on voltage-
dependent Ca2+ channel subtypes in bovine chromaffin cells. J Ethnopharmacol 74:75–81.
QE at up to 200 mg/kg concentration is commonly used in soft Doyle MP. 1991. Escherichia coli O157:H7 and its significance in foods. Intl J Food Microbiol
drinks, and is also reported to be used in whipping cream, can- 12:289–301.
Food Additive Organization (FAO). 2004. Quillaja extracts Type 1 and Type 2. 61st JECFA,
dies, frozen dairy products, gelatin and puddings, low-cholesterol Chemical and Technical Assessment.
dairy food products, mayonnaise production, soy sauce, root beer, Gee JM, Pricet KR, Ng K, Johnson IT, Rhodes MJ. 1998. The relationship between saponin
structure and bioactivity—a preliminary study. In: Bardocz S, Muzquiz M, Pusztai A, ed-
cider, cream soda, cocktail mixes, and in baked goods as leavening itors. Cost 98: effects of antinutrients on the nutritional value of legume diets. European
agents (Watanabe and others 1989) and as a red coloring material Commission, Luxemburg, p 8–14.
(Oono and Higashimura 1995; Mukai and others 1993, Nayyar Gogelein H, Huby A. 1984. Interaction of saponin and digitonin with black lipid membranes
and lipid monolayers. Biochim Biophys Acta 773:32–8.
and others 1998; Richardson and Jimenez-Flores 1991; Sundfeld Griffin PM, Tauxe RV. 1991. The epidemiology of infections caused by Escherichia coli O157:H7,
and others 1994; Toya and others 1994; Murakami and Watan- other enterohemorrhagic E. coli, and the associated hemolytic uremic syndrome. Epidemiol
Rev 1(13):60–98.
abe 1988a, 1988b; Maeda and others 1989; FAO 2004). It is also Guo S, Kenne L. 2000. Structural studies of triterpenoid saponins with new acyl components
reported to be used in microemulsions (Kudo and Nishi 1992), from Quillaja saponaria Molina. Phytochemistry 55:419–28.
Hancock DD, Besser TE, Rice DH, Ebel ED, Herriott DE, Carpenter LV. 1998. Multiple
and in dissolving propolis (Kawai and others 1994; FAO 2004). sources of Escherichia coli O157 in feedlots and dairy farms in the northwestern United States.
Food Microbiology &

Therefore, further work will focus on the application of QE in Prevent Vet Med 35:11–9.
Hassan SM, Byrd JA, Cartwright AL, Bailey CA. 2010. Hemolytic and antimicrobial activities
model food systems such as apple juice (rich in carbohydrates) and differ among saponin-rich extracts from Guar, Quillaja, Yucca and Soybean. Appl Biochem
2% milk (rich in protein and lipids) to determine the retention of Biotechnol 162:1008–17.
Safety

Higuchi R, Tokimitsu Y, Fujioka T, Komori T, Kawasaki T, Oakenful, DG. 1987. Structure of

antibacterial effect in the presence of food matrices, and also as
micro-emulsions or an encapsulated product (for targeted delivery) Hisayuki K, Takashi I. 1987. Antioxidant. Patent 62243681 A.
Hostettmann K, Hostettmann M, Marston A. 1991. Saponins. In: Dey, PM, Harborne, JB,
to determine antibacterial effects against the STECs.
Conclusions
The results from this study demonstrate that Quillaja extract,
an economical and sustainable product, shows potential for use as
an antimicrobial agent against O157 STECs and the big 6 non-
O157 STECs (perhaps potentially as a marinade mix for meats or
coating ingredient in other food products or upon consumption to
reduce or prevent illness symptoms as it has increased antibacterial
activity at 37 °C over 16 h corresponding to the amount of time
for digestion). However, as with any novel natural product, before
any claims can be made (antibacterial activity in this case), studies
to test the efficacy of QE using in vitro model food systems (in apple
juice and milk), under simulated gastric conditions, followed by in
vivo animal feeding studies and clinical trials must be undertaken
and regulatory approval should be obtained as reported before
(Joshi and others 2015).
Acknowledgments
We gratefully acknowledge the support provided by the UT-
Institute of Agriculture (Ag Experiment Station) and the Multi-
state Project S1026 to carry out this research. We also acknowledge
the guidance and help provided by Dr. John Dunlap for the SEM
desacylsaponins obtained from the bark of Quillaja saponaria. Phytochemis 26:229–35.
editors. Methods in plants biochemistry. . Vol. 7Academic Press, New York, USA. pp. 435–
71.
Hu M, Konoki K, Tachibana K. 1996. Cholesterol-independent membrane disruption
caused by triterpenoid saponins. Biochim Biophys Acta (BBA) - Lipids Lipid Metabol
1299:252–8.
Joshi SS, Su X, D’Souza DH. 2015. Antiviral effects of grape seed extract against feline calicivirus,
murine norovirus, and hepatitis A virus in model food systems and under gastric conditions.
Food Microbiol 52:1–10
Kawai M, Hirashita N, Kanae J. 1994. Manufacture of propolis compositions for health foods.
Patent 94197734.
Kim GH, Fratamico P, Breidt F, Oh DH. 2016. Survival and expression of acid resistance genes
in Shiga toxin-producing Escherichia coli acid adapted in pineapple juice and exposed to
synthetic gastric fluid. J Appl Microbiol 121(5):1416–26.
Kooryama T, Chiba I. 1996. Oxidation-stable emulsions of oil and fats with transparent appear-
ance and food containing them. Patent 9651928 A2.
Kudo S, Nishi E. 1992. Surfactant and method for producing the same. U.S. Patent 5 (1992)
152928.
Lu CD, Schaefer DM, Jorgensen NA. 1987. Alteration of fermentation in continuous culture of
mixed rumen bacteria by isolated alfalfa saponins. J Dairy Sci 70:799–805.
Maeda K, Tanada T, Akita S. 1989. Manufacture of improved mayonnaise. Patent 01, 108:
956.
Mukai S, Matusuda T, Kanazawa Y. 1993. Manufacture of foaming beverages with saponins,
saccharides and carbon dioxide. Patent 05:275.
Murakami F, Watanabe T, 1988a. Manufacture of stabilized cream using Quillaja saponin. Patent
63 (1988):735.
Murakami F, Watanabe T, 1988b. Production of emulsified fat and oil-containing soy sauce.
Patent 63:23371.
Naknukool S, Horinouchi I, Hatta H. 2011. Stimulating macrophage activity in mice
and humans by oral administration of Quillaja saponin. Biosci Biotechnol Biochem 75:1889–
93.
Nayyar D, Schulok W, Saleeb F, Modesta E. 1998. Dry mix for producing a slush beverage.
U.S. Patent 5853785.
Oono T, Higashimura Y. 1995. Method for manufacturing of red coloring material. Patent
9590188 A2.

1176 Journal of Food Science r Vol. 82, Nr. 5, 2017

Quillaja saponaria to control Escherichia coli . . .
Richardson T, Jimenez-Flores R. 1991. Saponin extraction of cholesterol from dairy products.
PCT International Application W, O 91 05:836.
Riley LW, Remis RS, Helgerson SD, McGee HB, Wells JG, Davis BR, Hebert RJ, Olcott, ES,
Johnson LM, Hargrett NT, Blake PA, Cohen ML. 1983. Hemorrhagic colitis associated with
a rare Escherichia coli serotype. N Engl J Med 308:681–5.
Roner MR, Sprayberry J, Spinks M, Dhanji S. 2007. Antiviral activity obtained from aqueous
extracts of the Chilean soapbark tree (Quillaja saponaria Molina). J Gen Virol 88:275–85.
Roner, MR, Tam KI, Kiesling-Barrager M. 2010. Prevention of rotavirus infections in vitro
with aqueous extracts of Quillaja saponaria Molina. Future Med Chem 2:1083–97.
Scallan E, Hoekstra RM, Angulo FJ, Tauxe RV, Widdowson MA, Roy SL, Jones JL, Griffin
PM. 2011. Foodborne illness acquired in the United States—major pathogens. Emerg Infect
Dis 17:7–15.
Schmidt JW, Bosilevac JM, Kalchayanand N, Wang R, Wheeler TL, Koohmaraie M. 2014.
Immersion in antimicrobial solutions reduces Salmonella enterica and Shigatoxin-producing
Escherichia coli on beef cheek meat. J Food Prot 77(4):538–48.
Sen S, Makkar HP, Muetzel S, Becker K. 1998. Effect of Quillaja saponaria saponins and Yucca
schidigera plant extract on growth of Escherichia coli. Lett Appl Microbiol 27:35–8.
Siegler RL. 1995. The hemolytic uremic syndrome. Pediatr Clin North Am 42:1505–29.
Su X, Howell AB, D’Souza DH. 2012. Antibacterial effects of plant-derived extracts on
methicillin-resistant Staphylococcus aureus. Foodborne Pathog Dis 9(6):573–8.
Sundfeld E, Krochta J, Richardson T. 1994. Aqueous process to remove cholesterol from food
products. US005370890A.
Taguri T, Tanaka T, Kouno I. 2004. Antimicrobial activity of 10 different plant polyphenols
against bacteria causing food-borne disease. Biol Pharm Bull 27:1965–9.
Tam KI, Roner MR. 2011. Characterization of in vivo anti-rotavirus activities of saponin extracts
from Quillaja saponaria Molina. Antiviral Res 90:231–41.
Toya C, Fujita M, Toshio A. 1994. Enhancement of oil-soluble flavors in candies. J Patent 06
22:697.
U.S. Department of Agriculture (USDA), Food Safety and Inspection Service (FSIS). 2011.
USDA takes new steps to fight E. coli, protect the food supply. Designation extends zero
tolerance policy for E. coli O157:H7 to six additional E. coli serogroups. Available at:
http://www.usda.gov/wps/portal/usda/usdahome?contentidonly˜true&contentid˜2011/09/
0400.xml. Accessed July 30, 2014.
Voravuthikunchai SP, Limsuwan S. 2006. Medicinal plant extracts as anti-Escherichia coli O157:H7
agents and their effects on bacterial cell aggregation. J Food Protect 69(10):2336–41.
Voravuthikunchai S, Lortheeranuwat A, Jeeju W, Sririrak T, Phongpaichit S, Supawita T. 2004.
Effective medicinal plants against enterohaemorrhagic Escherichia coli O157:H7. J Ethnophar-
macol 94(1):49–54.
Watanabe T, Murakami F, Harada S, Watanabe Y. 1989. Leavening agents contain Quillaja
saponin and carrageenan for cake manufacture. J Patent 01 168:227.
Wells JG, Davis BR, Wachsmuth, IK, Riley LW, Remis RS, Sokolow R, Morris GK, 1983.
Laboratory investigation of haemorrhagic colitis outbreaks associated with a rare Escherichia
serotype. J Clin Microbiol 18:512–20.
Food Microbiology &
Safety
Vol. 82, Nr. 5, 2017 r Journal of Food Science 1177