Monitoring hydroxycinnamic acid decarboxylation by lactic acid bacteria using high-throughput UV-Vis spectroscopy

Article abstract of DOI:10.3390/molecules25143142

Hydroxycinnamic acid (HCA) decarboxylation by lactic acid bacteria (LAB) results in the production of 4-vinylplenols with great impact on the sensorial characteristics of foods. The determination of LAB decarboxylating capabilities is key for optimal strain selection for food production. The activity of LAB strains from the Ohio State University-Parker Endowed Chair (OSU-PECh) collection potentially capable of synthesizing phenolic acid decarboxylase was evaluated after incubation with HCAs for 36 h at 32 °C. A high-throughput method for monitoring HCAs decarboxylation was developed based on hypsochromic shifts at pH 1.0. Out of 22 strains evaluated, only Enterococcus mundtii, Lactobacillus plantarum and Pediococcus pentosaceus were capable of decarboxylating all p-coumaric, caffeic and ferulic acids. Other strains only decarboxylated p-coumaric and caffeic acid (6), only p-coumaric acid (2) or only caffeic acid (1), while 10 strains did not decarboxylate any HCA. p-Coumaric acid had the highest conversion efficiency, followed by caffeic acid and lastly ferulic acid. Results were confirmed by HPLC-DAD-ESI-MS analyses, showing the conversion of HCAs into their 4-vinylphenol derivatives. This work can help improve the sensory characteristics of HCA-rich foods where fermentation with LAB was used during processing.

Full text of DOI:10.3390/molecules25143142

molecules
Article
Monitoring Hydroxycinnamic Acid Decarboxylation
by Lactic Acid Bacteria Using High-Throughput
UV-Vis Spectroscopy
Gonzalo Miyagusuku-Cruzado , Israel García-Cano , Diana Rocha-Mendoza
Rafael Jiménez-Flores and M. Monica Giusti *
,
Department of Food Science and Technology, The Ohio State University, 2015 Fyffe Ct., Columbus,
OH 43210-1007, USA; miyagusukucruzado.1@osu.edu (G.M.-C.); garciacano.1@osu.edu (I.G.-C.);
rochamendoza.1@osu.edu (D.R.-M.); jimenez-flores.1@osu.edu (R.J.-F.)
* Correspondence: giusti.6@osu.edu; Tel.: +1-614-247-8016
Academic Editors: Franco Biasioli, Vittorio Capozzi, Iuliia Khomenko, Luigimaria Borruso,
Patrick Silcock and Igor Jerkovic´
ꢀꢁꢂꢀꢃꢄꢅꢆꢇ
ꢀꢁꢂꢃꢄꢅꢆ
Received: 18 May 2020; Accepted: 3 July 2020; Published: 9 July 2020
Abstract: Hydroxycinnamic acid (HCA) decarboxylation by lactic acid bacteria (LAB) results
in the production of 4-vinylplenols with great impact on the sensorial characteristics of foods.
The determination of LAB decarboxylating capabilities is key for optimal strain selection for
food production. The activity of LAB strains from the Ohio State University—Parker Endowed
Chair (OSU-PECh) collection potentially capable of synthesizing phenolic acid decarboxylase was
evaluated after incubation with HCAs for 36 h at 32 ◦C. A high-throughput method for monitoring
HCAs decarboxylation was developed based on hypsochromic shifts at pH 1.0. Out of 22 strains
evaluated, only Enterococcus mundtii, Lactobacillus plantarum and Pediococcus pentosaceus were capable
of decarboxylating all p-coumaric, caffeic and ferulic acids. Other strains only decarboxylated
p-coumaric and caffeic acid (6), only p-coumaric acid (2) or only caffeic acid (1), while 10 strains did
not decarboxylate any HCA. p-Coumaric acid had the highest conversion efficiency, followed by
caffeic acid and lastly ferulic acid. Results were confirmed by HPLC-DAD-ESI-MS analyses, showing
the conversion of HCAs into their 4-vinylphenol derivatives. This work can help improve the sensory
characteristics of HCA-rich foods where fermentation with LAB was used during processing.
Keywords: p-hydroxystyrene; 3,4-dihydroxystyrene; 4-hidroxy-3-mthoxystyrene; flavor compounds;
lactic acid bacteria
1. Introduction
Hydroxycinnamic acids (HCAs) are secondary metabolites in plants and fungi characterized by
a C6–C3 structure [
1]. At high concentrations, these compounds exert an antibacterial effect over
a wide range of Gram-positive and Gram-negative bacteria [
2–4]. Further studies have shown that
some bacterial strains can decarboxylate HCAs as a mechanism for detoxification of its environment,
given that the antimicrobial activity of the aforementioned acids depends on the presence of a double
bond in the side chain of the structure [
5]. However, it also has been reported that not all bacterial [6] or
all yeast strains [ ] are capable of decarboxylating HCAs that lead to the formation of 4-vinylphenols
7
(4VPs) or their reduced form, 4-ethylphenols. In fermented products such as wine, HCA degradation by
yeast results in the formation of 4VPs; which are responsible for imparting distinct phenolic off-flavors
in the final product [
cases such as wheat beer, where the reduction in 4VP and 4-vinylguaiacol (4VG) production resulted
in a less pronounced wheat beer aroma [ ]; or artisanal bread, where decarboxylation of ferulic acid
7]. However, depending on the application, these compounds may be desirable for
8
(FA) generated 4VG, an important component in its flavor profile [9]. In addition, 4VPs do not only
Molecules 2020, 25, 3142; doi:10.3390/molecules25143142
www.mdpi.com/journal/molecules

Molecules 2020, 25, 3142
2 of 11
affect the flavor of foods, but also their color, as they can react with other components in the food
matrix such as anthocyanins. This interaction can result in the formation of pyranoanthocyanins [10],
compounds partially responsible for imparting color to aged red wines with an enhanced resistance
to bleaching [11
and yogurt, as nowadays it is not uncommon to find these products fortified with HCA-rich plant
extracts [13 15]. Moreover, the increasing use of pigment-rich plant extracts as food colorants can also
result in an increased HCA content [16 17]. Depending on the LAB strain used during production,
,12]. In dairy foods, HCA decarboxylation by LAB is of special importance in cheese
–
,
these HCAs could be biotransformed into 4VPs, affecting the sensorial characteristics of the final
product. Therefore, the study of the decarboxylating capabilities of LAB strains isolated from dairy
products is a subject of increasing interest in the food industry. The most common methodologies for
the screening of these decarboxylated products have been HPLC-DAD-MS [7,10,18], GC-FID [18] or
GC-MS [19]. These methodologies, although precise and reliable, require expensive equipment, trained
personnel, and high costs of operation. A simpler spectrophotometric method for the detection of 4VPs
in a mixture was first reported using quartz cuvettes and pH 6.0 buffer [20], showing hypsochromic
shifts in the λmax from 285–300 nm for HCAs to 260 nm for 4VPs. This hypsochromic shift can be the
basis for a modified high-throughput method that could aid in the screening and selection of LAB
capable–or incapable, depending on the application–of decarboxylating HCAs. These bacteria could
be used to modulate the release of compounds that affect the sensory characteristics of fermented
products elaborated from HCA-rich sources.
The objective of this study was to develop a fast and simple methodology for high-throughput
screening of LAB that are capable of HCA biotransformation, in order to determine the decarboxylating
activity of LAB strains from the Ohio State University—Parker Endowed Chair (OSU-PECh) collection.
2. Results and Discussion
2.1. UV-Vis Spectra of HCAs, Decarboxylated HCAs and Growth Medium
The first goal was to monitor the decarboxylation of HCAs using a fast, high-throughput method
based on UV-Vis light absorption changes. Figure 1a shows the UV-Vis light absorption spectra of
p-coumaric acid (pCA), 4VP, a chemically defined medium (CDM), and de Man, Rogosa and Sharpe
broth (MRS) all at pH 7.0. The wavelength of maximum absorption for pCA is near 290 nm (blue line),
while the one registered for 4VP is near 260 nm (red line). These results are consistent with previously
reported values for chromophores with similar structural characteristics (200–280 nm for phenolic
acids and 300–350 nm for HCAs) [21]. Therefore, this UV-Vis light wavelength range (200–350 nm)
was a critical factor for the selection of a culture medium in which the decarboxylation experiments
could be conducted, and that will allow a direct measurement of UV-Vis light absorption changes
in response to structural modifications in the chromophore after incubation with LAB. Figure 1b
shows the absorption of all samples at 260, 300 and 340 nm. Results showed that CDM (green bar)
had a significantly lower absorption than MRS broth (black bar) at all three wavelengths analyzed
(p < 0.01). This significantly lower absorption makes CDM an ideal medium to conduct decarboxylation
experiments due to a much lower interference with the UV-Vis light absorption of the compounds of
interest. The significantly lower absorption of CDM in the UV-Vis wavelength range of interest when
compared to MRS broth can be attributed to a lower concentration of protein-rich components in its
formulation (5 g/L in CDM and 25 g/L in MRS). Therefore, MRS broth was used for initial cell growth
until the log phase was reached, and CDM was used for decarboxylation experiments and subsequent
UV-Vis light absorption measurements.
2.2. Determination of the Optimal pH for Monitoring HCA Decarboxylation
After selecting an appropriate culture medium to conduct decarboxylation experiments, the next
step was the determination of the optimal pH for monitoring changes in the UV-Vis light absorption
in response to HCA decarboxylation by LAB. It was previously noticed that absorption increased

Molecules 2020, 25, 3142
3 of 11
significantly in wavelengths shorter than 230 nm, and no valuable information was obtained from
that region of the spectra due to the high interference of the microplate used for assays. Therefore,
subsequent experiments were focused on the light absorption changes in wavelengths longer than
230 nm. To determine the best pH for detection of changes in light absorption after incubation with
LAB, the spectra of CDM containing HCAs or 4VPs were measured at pH 1.0, pH 3.0, pH 5.0 and pH
7.0. As shown in Figure 2, the λ230–500 max of HCAs registered a bathochromic shift when pH decreased.
At pH 1.0 (red line) and 3.0 (orange line), the λ230–500 max was 309 nm for pCA, 320 nm for caffeic acid
(CA), and 320 nm for FA. At pH 5.0 (light green line) and 7.0 (green line), the absorption reading
was 285 nm for pCA, 284 nm for CA, and 283 nm for FA. The similarity between the λ230–500 max of
acidic pH (pH 1.0 and pH 3.0) and near-neutral pH (pH 5.0 and pH 7.0) can be explained by the
pKa of HCAs, which were reported to be between 4.0 and 4.5 [22
than 4.0, the acid is expected to be in its undissociated form, while at pH values larger than the pKa,
they are expected to be in their dissociated form [ ], which ultimately changes the light absorption
–24]. Therefore, at pH values lower
5
pattern of the chromophore. No major changes in the λ230–500 max of decarboxylated products were
observed in response to changes in the pH. Further measurements were conducted at pH 1.0 because it
provided a more distinguishable hypsochromic shift between the λ230–500 max of the HCAs and their
decarboxylated products than buffers at pH 5.0 and pH 7.0. Although measurements in acidic pH
(pH 1.0 and 3.0) showed similar absorbance values for the λ230–500 max, pH 1.0 buffer was selected due
to the lower absorbance of the medium at wavelengths below 260 nm. This provided a more defined
peak in the spectra, thereby allowing for a better visualization of hypsochromic changes that occur as a
result of HCA decarboxylation. Moreover, the light absorption of MRS broth and CDM at different pH
values were also monitored. No significant differences were found in light absorption as a response
to pH changes (p > 0.05, data not shown); however, it is clear that the absorption of MRS broth is
higher than that of CDM at all four pH values. With the data obtained, we proceeded to initially grow
bacteria at their optimal pH in MRS broth, conducted decarboxylation experiments at their optimal pH
in CDM, and diluted an aliquot 25 times with pH 1.0 buffer prior to measurements.
a. Absorption spectra (200-500nm)
b. Absorption at select wavelengths
2.0
CDM
2.0
CDM
MRS
d
MRS
pCA
4VP
c
1.5
1.0
0.5
0
pCA
4VP
b
1.5
1.0
0.5
0
c
a
b
b
d
c
a
a
a
300
400
500
260 nm
300 nm
340 nm
Wavelength (nm)
Figure 1. Measurements of UV-Vis light absorption at pH 7. (
and Sharpe broth (MRS, black line), p-coumaric acid (pCA, blue line) and 4-vinylphenol (4VP, red line)
in chemically defined medium (CDM) and CDM alone (green line). ( ) Absorption at 260, 300 and
a) Absorption spectra of Man, Rogosa
b
340 nm showed that MRS had a significantly higher absorption than CDM did, between 200–500 nm.
Different letters indicate significant differences (p < 0.01). Data represent means of n = 3, error bars
represent SD.
2.3. HPLC-DAD-ESI-MS Analysis of Decarboxylated Products
Results in Figure 3 showed that the degradation products from pCA, CA and FA had longer
retention times, and the absorption spectra of the product peak registered by the DAD, showed
a hypsochromic shift in the λ230–500 max when compared to the precursor HCA (>50 nm shift),
consistent with the production of decarboxylated products [20,25]. Under our experimental conditions,
no sinapic acid (SA) decarboxylation was observed after incubation with LAB. This was denoted
by the absence of peaks corresponding to common products from SA decarboxylation such as

Molecules 2020, 25, 3142
4 of 11
4-vinylsyringol or 4-ethlysyringol. To better identify decarboxylated products, mass spectrometry
analyses were conducted at time 0 and after 36 h of incubation at 32 ◦C in the dark. After incubation,
HPLC-DAD-ESI-MS analyses confirmed the absence of pCA (M− 163 m/z) and CA (M− 197 m/z),
and showed their transformation into their corresponding decarboxylated vinyl derivatives, 4-VP
(MH+ 121 m/z) and 4-vinylcatechol (4VC, M− 135 m/z), respectively. The product from the partial
degradation of FA (M− 193 m/z) was consistent with that of 4VG (M− 149 m/z). Previous studies
have reported the activity of a vinylphenol reductase resulting in the appearance of 4-ethylphenols
derivatives from HCAs [
6
,
20
,
26]. However, in this study, the HPLC-DAD◦-ESI-MS analyses showed
no evidence of 4-ethylphenol production. The 36-h incubation at 32 C allowed for complete
decarboxylation of both pCA and CA into their 4VP derivatives with no 4-ethylphenols, making
them ideal substrates for monitoring the activity of phenolic acid decarboxylase in this proposed
methodology. The characteristic m/z of decarboxylated products from SA (225 m/z), 4-vinylsyringol
(179 m/z) and 4-ethylsyringol (181 m/z) were also monitored, but no signal was found by the MS detector.
4VP
pCA
CA
FA
4VC
pH1
pH3
pH5
pH7
MRS
CDM
400
300
300
400
Wavelength (nm)
Figure 2. UV-Vis absorption spectra of pCA: p-coumaric acid, CA: caffeic acid, FA: ferulic acid, 4VP:
4-vinylphenol, 4VC: 4-vinylcatechol, MRS: De Man, Rogosa and Sharpe broth, CDM: chemically defined
medium, diluted 25 times in buffers at different pH values. Bathochromic shifts were observed in the
λ230–500 max of HCAs when pH decreased. No changes in the λ230–500 max of 4VPs were observed.
Light absorption of MRS was higher than CDM at all pH values. Data represent means of n = 3,
error bars represent SD.
In Figure 4, the absorption spectra of the individual peaks for each HCA and its respective
decarboxylated derivatives (upper) are compared to the UV-Vis absorption of the bacterial supernatant
measured with a plate reader directly after dilution (lower). Both methodologies were able to
consistently detect hypsochromic shifts in pCA, CA and FA after incubation with LAB. These shifts
occurred due to a decrease in the number of conjugated double bonds in the chromophore, as
decarboxylation results in the loss of a double bond in the side chain of the HCA structure. This is
consistent with previous literature showing the formation of decarboxylated products by phenolic acid
decarboxylase [6,18,20]. With the plate reader, the incubated samples displayed hypsochromic shifts of
51 nm for pCA, 61 nm for CA, and 60 nm for FA. These hypsochromic shifts were similar to the ones
registered by the HPLC-DAD.

Molecules 2020, 25, 3142
5 of 11
O
OH
p-Coumaric acid
λmax = 307 nm
Time 0
36 h
Ferulic acid
M- 193 m/z
p-Coumaric acid
M- 163 m/z
4-vinylphenol
λ
max = 256 nm
H3CO
OH
Caffeic acid
λmax = 321 nm
Time 0
36 h
CH2
4-Vinylphenol
MH+ 121 m/z
4-Vinylguaiacol
M- 149 m/z
4-vinylcatechol
λmax = 259 nm
OH
Ferulic acid
λmax = 321 nm
Time 0
36 h
Sinapic acid
MH+ 225 m/z
Caffeic acid
M- 179 m/z
4-vinylguaiacol
λ
max = 280 nm
Sinapic acid
max = 322 nm
Time 0
36 h
λ
4-Vinylcatechol
M- 135 m/z
4-Vinylsyringol
M- 179 m/z
0
5
10 min
Figure 3. Ultra-high-pressure liquid chromatography-mass spectrometry representative chromatograms
and identification of the initial hydroxycinnamic acids and their respective decarboxylated products
after incubation for 36 h at 32 ◦C in the dark with Lactobacillus plantarum (OSU-PECh-BB).
λ
max
:
maximum absorption in the 230–500 nm range.
Spectra read by DAD after chromatographic separation
p-Coumaric acid
Caffeic acid
Ferulic acid
4-vinylphenol
4-vinylcatechol
4-vinylguaiacol
300
400
500
300
400
500
300
400
500
Wavelength (nm)
Spectra read on CDM diluted with pH1 buffer using a plate reader
p-Coumaric acid,
Caffeic acid,
time 0
Ferulic acid,
time 0
time 0
p-Coumaric acid,
Caffeic acid,
36 h
Ferulic acid,
36 h
36 h
300
400
500
300
400
500
300
400
500
Wavelength (nm)
Figure 4. Representative UV-Vis light absorption spectra for hydroxycinnamic acids and decarboxylated
products using a diode-array detector (upper) and a plate reader (lower) after incubation with
Lactobacillus plantarum (OSU-PECh-BB). Spectra for pCA and 4VP (blue), CA and 4VC (purple), and FA
and 4VG (orange). Spectra read on CDM diluted with pH 1.0 buffer using a plate reader.
2.4. Screening of the OSU-PECh LAB Collection for Their Ability to Decarboxylate HCAs
LAB strains isolated from dairy products from the OSU-PECh collection were screened for their
ability to decarboxylate HCAs using the high-throughput UV-Vis spectrophotometric method described
in Section 3.5. Results from the initial bioinformatic analysis, showed that 22 strains from a total
of 137 were potentially capable of synthesizing PAD based on their reported genome annotations.
Lactobacillus plantarum (OSU-PECh-A) was used as a positive control for PAD activity, and Staphylococcus
epidermidis (ATCC1222) and Lactobacillus casei (OSU-PECh-C) were used as negative controls. A total
of 24 bacterial strains were grown in CDM and incubated at 32 ◦C for 36 h to determine their ability
to decarboxylate HCAs. Results in Table 1 show that three strains were able to decarboxylate
pCA, CA and FA (Enterococcus mundtii, OSU-PECh-39B; Lactobacillus plantarum, OSU-PECh-BB

Molecules 2020, 25, 3142
6 of 11
and Pediococcus pentosaceus, OSU-PECh-27B). Six strains were able to decarboxylate pCA and CA
(Lactobacillus helveticus, OSU-PECh-LH7; Lactobacillus pentosus, OSU-PECh-LP6C; Lactobacillus plantarum,
OSU-PECh-A; Pediococcus acidilactici, OSU-PECh-PA3A; Pediococcus pentosaceus, OSU-PECh-PP13
and OSU-PECh-PP6A).Two strains were able to decarboxylate only pCA (Lactobacillus acidophilus,
OSU-PECh-LA5 and Lactobacillus helveticus, OSU-PECh-LH1B); and one strain was able to decarboxylate
only CA (Lactobacillus helveticus, OSU-PECh-LH19). No strain from the OSU-PECh collection showed
the ability to decarboxylate SA under these conditions. However, this behavior was not unexpected, as
literature shows that LAB may not be capable of SA decarboxylation [18]. Moreover, this selectivity
against SA is apparently not unique to bacterial PAD, as yeast-derived PAD seems to have a similar
behavior [27]. Although Enterococcus mundtii, OSU-PECh-39B; Pediococcus pentosaceus, OSU-PECh-27B;
and Lactobacillus plantarum, OSU-PECh-BB were capable of decarboxylating all pCA, CA and FA;
none was able to completely decarboxylate FA as they did with pCA and CA. These decarboxylation
patterns for two of the three strains aforementioned are consistent with previous literature, where
Lactobacillus plantarum [5,25,26] and Pediococcus pentosaceous [6] were able to metabolize pCA and
CA with high efficiency, but FA only partially. However, no information on the decarboxylating
capabilities of Enterococcus mundtii was found in published literature. This report showed the ability of
the Enterococcus mundtii strain to decarboxylate HCAs, resulting in the production of 4VPs, strongly
suggesting that this strain can in fact, synthesize PAD. This finding may promote the use of this
strain for targeted bioconversion of hydroxycinnamic acids, with special interest in its capability
to decarboxylate FA into 4-VG in products such as sourdough bread [28] and functional beverages
obtained from lactic acid fermentation [29].
Table 1. Ability of lactic acid bacteria strains to degrade hydroxycinnamic acids. Results based
on UV-Vis light absorption changes. (++): Hypsochromic changes in λ230–500 max indicating
complete decarboxylation of a given HCA, (+): hypsochromic changes inλ230–500 max indicating
partial decarboxylation of a given HCA, (
CA: caffeic acid, FA: ferulic acid, SA: sinapic acid. Results from 3 repetitions.
−
): no changes in absorption spectra. pCA: p-Coumaric acid,
Lactic Acid Bacteria Strain
Enterococcus mundtii
Accession Code
OSU-PECh-39B
OSU-PECh-LA5
pCA
(++)
(++)
CA FA
SA
(−)
(−)
(++) (+)
Lactobacillus acidophilus
(−)
(
−)
OSU-PECh-25
OSU-PECh-26
OSU-PECh-33
OSU-PECh-40
OSU-PECh-57B
OSU-PECh-60
OSU-PECh-LH1B
OSU-PECh-LH4A
OSU-PECh-LH7
OSU-PECh-LH15A
OSU-PECh-LH19
(−)
(−)
(
(
(
(
(
(
(
(
−
−
−
−
−
−
−
−
)
)
)
)
)
)
)
)
(
(
(
(
(
(
(
(
−
−
−
−
−
−
−
−
−
−
−
)
)
)
)
)
)
)
)
)
)
)
(−)
(−)
(−)
(−)
(−)
(−)
(−)
(−)
(−)
(−)
(−)
(−)
(−)
(−)
Lactobacillus helveticus
(−)
(++)
(−)
(++)
(−)
(++) (
(−
)
(
(−)
(++) (
(++) (
(++) (
Lactobacillus pentosus
Lactobacillus plantarum
Lactobacillus rhamnosus
Pediococcus acidilactici
OSU-PECh-LP6C
(++)
−)
(−)
OSU-PECh-A
OSU-PECh-BB
(++)
(++)
−)
(−)
(++) (+)
(−)
OSU-PECh-24
(−)
(++)
(−)
(++)
(++)
(++)
(−
)
(
−
)
(−)
OSU-PECh-PA3A
OSU-PECh-PAL
(++) (
−
)
)
(−)
(−
)
(
−
(−)
OSU-PECh-27B
OSU-PECh-PP6A
OSU-PECh-PP13
(++) (+)
(−)
(−)
(−)
Pediococcus pentosaceus
(++) (
(++) (
−
)
)
−
Lactobacillus casei
OSU-PECh-C
ATCC 1222
(−)
(−)
(
−
)
)
(
(
−)
)
(−)
(−)
Staphylococcus epidermidis
(−
−

Molecules 2020, 25, 3142
7 of 11
Moreover, in order to investigate if the light absorption changes were time dependent, samples
were taken after 30 min of incubation and every hour thereafter, and their light absorption spectra
were analyzed. Figure 5 shows that the changes in the absorption spectra are in fact, time dependent
and further analysis showed that pCA is decarboxylated at a faster rate than CA, with this trend
being independent of the bacterial strain used. This is similar to a previous report using purified
PAD, where at high concentrations of HCAs, pCA and CA were decarboxylated much faster than
FA [30]. A possible explanation for this behavior may be related to the antibacterial effect of each
HCA. Reports show that against E. coli, S. aureus and B. cereus [2], and LAB [5], pCA had a stronger
antibacterial activity than a more polar CA and a less polar FA. Moreover, it was hypothesized that
this stronger antibacterial activity may be related to the solubility of pCA in both aqueous and lipid
phases [
2]. This particular solubility pattern allows pCA to bind to the outer bacterial membrane [4],
facilitating its interaction and subsequent decarboxylation by PAD.
Enterococcus mundtii
OSU-PECh-39B
Pediococcus pentosaceus Lactobacillus plantarum
OSU-PECh-27B OSU-PECh-BB
Lactobacillus casei
OSU-PECh-C
2.25
1.50
0.75
p-Coumaric
acid
0 h
30 min
1 h
2 h
3 h
0
1.6
4 h
5 h
6 h
1.2
0.8
0.4
0
Caffeic
acid
300
400
500
300
400
500
300
400
500
300
400
500
Wavelength (nm)
Figure 5. UV-Vis light absorption changes for p-coumaric acid (upper) and caffeic acid (lower)
during incubation with Enterococcus mundtii (OSU-PECh-39B), Pediococcus pentosaceus (OSU-PECh-27B),
Lactobacillus plantarum (OSU-PECh-BB) and Lactobacillus casei (OSU-PECh-C) for up to six hours at 32 ◦C
in the dark. Data represent means of n = 3, bars represent SD.
3. Materials and Methods
3.1. Chemicals and Reagents
CA, FA, SA, dipotassium hydrogen phosphate (K2HPO4), monopotassium phosphate (KH2PO4),
yeast extract, magnesium sulfate (MgSO4) and glucose were purchased from Sigma-Aldrich (St. Louis,
MO, USA); pCA was purchased from MP Biomedicals (Solon, OH, USA); sodium chloride (NaCl) was
obtained from Fisher Scientific (Pittsburgh, PA, USA); MRS lactobacilli broth was purchased from
Difco Laboratories (Detroit, MI, USA). 4VP was purchased from Santa Cruz Biotechnology (Santa Cruz,
CA, USA). All other reagents and solvents were of analytical grade.
3.2. Growth Medium Selection for HCA Decarboxylation Experiments
To develop a rapid, high-throughput spectrophotometric method, a medium with minimal light
absorption interference at the wavelengths of interest had to be selected. The light absorption spectra
of two growth media were compared; commercially available MRS broth, commonly used as a LAB
growth medium; and CDM, designed to provide LAB with the minimum required nutrients for normal
growth [31]. CDM was prepared using 0.1% K2HPO4, 0.1% KH2PO4, 0.5% yeast extract, 0.025% MgSO4,
0.0005% NaCl and 0.5% glucose in distilled water. Absorption spectra in the wavelengths of interest
(200–500 nm) were obtained using a SpectraMax M2 plate reader (Molecular Devices, Sunnyvale, CA,
USA) after diluting the growth medium 25 times with 0.1 M phosphate buffer pH 7.0.

Molecules 2020, 25, 3142
8 of 11
3.3. Bacterial Strain Selection
LAB strains were isolated from dairy products such as milk, milk powder, buttermilk powder,
cheese and yogurt [32]. Initial selection from the 137 LAB strains in the OSU-PECh collection was
conducted using bioinformatic tools from the National Center for Biotechnology Information (NCBI,
https://www.ncbi.nlm.nih.gov/) and genome annotations for each species [33]. Bacterial strains
were selected based on their potential capability to synthesize phenolic acid decarboxylase (PAD),
an enzyme responsible for metabolizing HCAs, which has been previously purified, characterized
and sequenced [34]. Positive and negative controls for PAD synthesis were initially selected using the
aforementioned methodology, and the PAD activity of these strains was verified using a previously
published spectrophotometric method [20] and pCA as a decarboxylation substrate. Based on these
preliminary studies, Staphylococcus epidermidis (ATCC1222) was selected as a non-LAB strain, with no
PAD activity (non-LAB negative control). Lactobacillus casei (OSU-PECh-C) isolated from whey protein
isolate produced by Hilmar Ingredients (Hilmar, CA, USA), was selected as a LAB strain showing
no PAD activity (LAB negative control). Lactobacillus plantarum (OSU-PECh-A) isolated from natural
yogurt produced by Superior Dairy (Canton, OH, USA) showing PAD activity and was selected as
a LAB positive control.
3.4. Decarboxylation Experiments
Preparation of stock solutions and incubation in CDM was conducted using a modification of
a previously reported spectrophotometric method [20]. Briefly, stock solutions of pCA, CA, FA and
SA were prepared at a concentration of 5 g/L in aqueous ethanol (50% v/v). Stock solutions of HCAs
were diluted in CDM for a final concentration of 500 mg/L. This allowed for a clear identification of
absorption changes, as the spectrum is initially dominated by the HCA and after incubation, by the
degradation product. Bacterial strains potentially capable of producing PAD were initially grown
in MRS broth in sterile centrifuge tubes under microaerobic conditions, and incubated overnight at
32 ◦C. The culture was collected by centrifugation at 8000 g for 10 min at 4 ◦C (Centrifuge 5804R,
×
Eppendorf, Boulder, CO, USA), the supernatant was then discarded, and pellets were washed twice
with sterile saline solution (0.085% NaCl, pH 7.0). Washed pellets were reconstituted in sterile saline
solution and seeded on 96-well plates with CDM containing HCAs. LAB strains were seeded at an
optical density of 0.1 (O.D. 600nm) calculated using a SpectraMax M2 plate reader (Molecular Devices,
Sunnyvale, CA, USA). Plates were covered with a sterile lid, sealed with parafilm to avoid evaporation,
and were then incubated at 32 ◦C for 36 h. Each HCA used in this study was incubated individually
with every LAB strain.
3.5. Rapid, High-Throughput Screening of Hydroxycinnamic acid Degradation by LAB Using UV-Vis
Spectroscopy and Selection of Optimal pH for Measurement
Spectrophotometric analyses were used to monitor the ability of bacteria to decarboxylate HCAs.
Briefly, 96-well plates were centrifuged at 4000
5804 benchtop centrifuge, Eppendorf, Boulder, CO, USA, Eppendorf A-2-DWP microplate rotor).
Supernatants were then immediately collected and frozen until further analysis. To select the best pH for
×
g for 30 min at room temperature (Eppendorf
monitoring absorption changes, a 10 µL aliquot of the CDM containing HCAs or their decarboxylated
products, was diluted 25 times with either KCl buffer (0.25 M) at pH 1.0, citrate buffer (0.1 M) at
pH 3.0 or 5.0, or phosphate buffer (0.1 M) at pH 7.0. Subsequently, the diluted aliquots were placed
in a UV-transparent 96-well microplate (Thermo Scientific Nunc®, Wilmington, DE, USA) for the
measurement of their absorbance between the range of 230–500 nm with a SpectraMax M2 plate reader
(Molecular Devices, Sunnyvale, CA, USA) at 25 ◦C.

Molecules 2020, 25, 3142
9 of 11
3.6. uHPLC-DAD-ESI-MS Analysis of HCAs and Degradation Products
A Nexera-i-LC-2040 3D ultra-high-pressure liquid chromatograph was used for reverse-phase
chromatographic separation and UV-Vis light absorption characterization. This system consisted of
4 pumps, a refrigerated autosampler, a column oven, and a diode-array detector (DAD). Tentative
identification of HCAs and their degradation products was conducted using a LCMS-8040 triple
quadrupole mass spectrometer with electrospray interface (Shimadzu, Columbia, MD, USA). Briefly,
30
µ
L of supernatant were injected into a Restek Pinnacle DB C18 column (50
×
2.1 mm, 1.9 µm,
Restek, Bellefonte, PA, USA) using a binary solvent system consisting of 0.1% aqueous formic (A) and
acetonitrile (B). The elution profile was 0–10 min, 3–20% B in A, 10–11 min, 20–40% B in A, 11–13 min, 40%
B (isocratic), with a flow rate of 0.3 mL/min at 40 ◦C. Identical ionizing conditions for the electrospray
interface were used for tentative identification of degradation products: 1.5 L/min nebulizing gas,
230 ◦C desolvation line temperature, 300 ◦C heat block temperature, 15 L/min drying gas flow. MH+
and M− of intact structures were analyzed under positive and negative mode, respectively, using the
Q1 scan function from 100 to 1000 m/z and single ion monitoring function for the respective 4VPs and
4-ethylphenols at their characteristic m/z.
3.7. Statistical Analysis
All experiments were conducted in triplicate. Figure design, one-way ANOVA analyses and
Tukey post hoc tests were conducted using GraphPad Prism 8.3.0 (GraphPad Software, San Diego, CA,
USA). A value of p < 0.05 was considered significant.
4. Conclusions
The decarboxylating capabilities of a collection of LAB were successfully studied with a fast, simple,
and reproducible methodology based on UV-Vis light absorption changes. Results were consistent with
the formation of 4VPs, decarboxylated products from pCA, CA and FA, while no decarboxylation was
observed for SA. In this study, we reported the ability of the Enterococcus mundtii strain to decarboxylate
pCA, CA and FA; strongly suggesting that this strain is capable of synthesizing PAD. This rapid
and inexpensive method has an advantage of using reagents, materials, and equipment that can be
found in most laboratories. Moreover, the selection of a growth medium with low interference and
the correct pH for the detection of hypsochromic changes, allow for a clearer visualization of the
results. This high-throughput method will facilitate the screening of LAB strains or fermentation
starters capable of enzymatic decarboxylation of different HCAs, as it does not require further
preparations or extraction procedures. Furthermore, LAB grown in CDM can potentially be used for
the production of 4VPs as it does not seem to promote the activity of a phenyl reductase, evidenced
by the absence of 4-ethylphenol after incubation. This method can help in the selection of bacterial
strains that capable or incapable, depending on the application, of decarboxylating HCAs into 4VPs,
for their use in industrial fermentation processes. Ultimately, the simple, rapid method and the
information conveyed in this report can promote the selection of better fermentation starters capable
of targeted HCA biotransformation into valuable flavor compounds. This could result in significant
quality improvements in the flavor and color of fermented products and an increased consumer
acceptance. Future experiments will focus on the effect of the matrix composition and physicochemical
parameters on the decarboxylating capabilities of Lactobacillus plantarum, Pediococcus pentosaceus
and Enterococcus mundtii in different food matrices. Moreover, future studies will also focus on the
expression, purification, and characterization of the PAD from Enterococcus mundtii.
Author Contributions: Conceptualization, G.M.-C., I.G.-C. and M.M.G.; Formal analysis, G.M.-C., I.G.-C. and
M.M.G.; Funding acquisition, R.J.-F. and M.M.G.; Investigation, G.M.-C., I.G.-C. and D.R.-M.; Methodology,
G.M.-C. and I.G.-C.; Project administration, G.M.-C. and M.M.G.; Resources, I.G.-C., D.R.-M., R.J.-F. and
M.M.G.; Supervision, M.M.G.; Validation, G.M.-C., I.G.-C. and M.M.G.; Visualization, G.M.-C. and M.M.G.;
Writing—original draft, G.M.-C. and M.M.G.; Writing—review & editing, G.M.-C., I.G.-C., D.R.-M., R.J.-F. and
M.M.G. All authors have read and agreed to the published version of the manuscript.

Molecules 2020, 25, 3142
10 of 11
Funding: This research was supported in part by FONDECYT—CONCYTEC, grant number 225-2015-FONDECYT;
the Parker Endowment at OSU, grant number 00100; and the USDA National Institute of Food and Agriculture,
Hatch Project OHO01423, Accession Number 1014136.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
Pereira, D.M.; Valent
14, 2202–2211. [CrossRef]
ão, P.; Pereira, J.A.; Andrade, P.B. Phenolics: From chemistry to biology. Molecules 2009,
Herald, P.J.; Davidson, P.M. Antibacterial activity of selected hydroxycinnamic acids. J. Food Sci. 1983, 48,
1378–1379. [CrossRef]
Borges, A.; Ferreira, C.; Saavedra, M.J.; Simões, M. Antibacterial activity and mode of action of ferulic and
gallic acids against pathogenic bacteria. Microb. Drug Resist. 2013, 19, 256–265. [CrossRef] [PubMed]
Lou, Z.; Wang, H.; Rao, S.; Sun, J.; Ma, C.; Li, J. p-Coumaric acid kills bacteria through dual damage
mechanisms. Food Control 2012, 25, 550–554. [CrossRef]
Sánchez-Maldonado, A.F.; Schieber, A.; Gänzle, M.G. Structure-function relationships of the antibacterial
activity of phenolic acids and their metabolism by lactic acid bacteria. J. Appl. Microbiol. 2011, 111, 1176–1184.
[CrossRef] [PubMed]
6.
7.
8.
Couto, J.A.; Campos, F.M.; Figueiredo, A.R.; Hogg, T.A. Ability of lactic acid bacteria to produce volatile
phenols. Am. J. Enol. Vitic. 2006, 57, 166–171.
Shinohara, T.; Kubodera, S.; Yanagida, F. Distribution of phenolic yeasts and production of phenolic off-flavors
in wine fermentation. J. Biosci. Bioeng. 2000, 90, 90–97. [CrossRef]
Langos, D.; Granvogl, M. Studies on the simultaneous formation of aroma-active and toxicologically relevant
vinyl aromatics from free phenolic acids during wheat beer brewing. J. Agric. Food Chem. 2016, 64, 2325–2332.
[CrossRef]
9.
Wang, H.E.; Hwang, C.F.; Tzeng, Y.M.; Hwang, W.Z.; Mau, J.L. Quality of white bread made from lactic acid
bacteria-enriched dough. J. Food Process. Preserv. 2012, 36, 553–559. [CrossRef]
10. Benito, S.; Morata, A.; Palomero, F.; González, M.C.; Suárez-Lepe, J.A. Formation of vinylphenolic
pyranoanthocyanins by Saccharomyces cerevisiae and Pichia guillermondii in red wines produced following
different fermentation strategies. Food Chem. 2011, 124, 15–23. [CrossRef]
11. Farr, J.E.; Giusti, M.M. Investigating the interaction of ascorbic acid with anthocyanins and
pyranoanthocyanins. Molecules 2018, 23, 744. [CrossRef] [PubMed]
12. De Freitas, V.; Mateus, N. Formation of pyranoanthocyanins in red wines: A new and diverse class of
anthocyanin derivatives. Anal. Bioanal. Chem. 2011, 401, 1467–1477. [CrossRef] [PubMed]
13. Oh, N.S.; Lee, J.Y.; Joung, J.Y.; Kim, K.S.; Shin, Y.K.; Lee, K.W.; Kim, S.H.; Oh, S.; Kim, Y. Microbiological
characterization and functionality of set-type yogurt fermented with potential prebiotic substrates
Cudrania tricuspidata and Morus alba L. leaf extracts. J. Dairy Sci. 2016, 99, 6014–6025. [CrossRef] [PubMed]
14. Ruggeri, S.; Straniero, R.; Pacifico, S.; Aguzzi, A.; Virgili, F. French marine bark extract pycnogenol as
a possible enrichment ingredient for yogurt. J. Dairy Sci. 2008, 91, 4484–4491. [CrossRef] [PubMed]
15. Khan, I.T.; Nadeem, M.; Imran, M.; Ajmal, M.; Ali, S. Antioxidant activity, fatty acids characterization and
oxidative stability of Gouda cheese fortified with mango (Mangifera indica L.) kernel fat. J. Food Sci. Technol.
2018, 55, 992–1002. [CrossRef] [PubMed]
16. Karaaslan, M.; Ozden, M.; Vardin, H.; Turkoglu, H. Phenolic fortification of yogurt using grape and callus
extracts. Lwt Food Sci. Technol. 2011, 44, 1065–1072. [CrossRef]
17. Coïsson, J.D.; Travaglia, F.; Piana, G.; Capasso, M.; Arlorio, M. Euterpe oleracea juice as a functional pigment
for yogurt. Food Res. Int. 2005, 38, 893–897. [CrossRef]
18. Curiel, J.A.; Rodríguez, H.; Landete, J.M.; de las Rivas, B.; Muñoz, R. Ability of Lactobacillus brevis strains to
degrade food phenolic acids. Food Chem. 2010, 120, 225–229. [CrossRef]
19. Thurston, P.A.; Tubb, R.S. Screening yeast strains for their ability to produce phenolic off-flavours: A simple
method for determining phenols in wort and beer. J. Inst. Brew. 1981, 87, 177–179. [CrossRef]
20. Cavin, J.F.; Andioc, V.; Etievant, X.; Divi
ès, C. Ability of wine lactic acid bacteria to metabolize phenol
carboxylic acids. Am. J. Enol. Vitic. 1993, 44, 76–80.

Molecules 2020, 25, 3142
11 of 11
21. Robbins, R.J. Phenolic acids in foods: An overview of analytical methodology. J. Agric. Food Chem. 2003, 51,
2866–2887. [CrossRef]
22. Ong, B.Y.; Nagel, C.W. High-pressure liquid chromatographyic analysis of hydroxycinnamic acid-tartaric
acid esters and their glucose esters in vitis vinifera. J. Chromatogr. A 1978, 157, 345–355. [CrossRef]
23. Erdemgil, F.Z.; S¸anli, S.; S¸anli, N.; Özkan, G.; Barbosa, J.; Guiteras, J.; Beltrán, J.L. Determination of pKa
values of some hydroxylated benzoic acids in methanol-water binary mixtures by LC methodology and
potentiometry. Talanta 2007, 72, 489–496. [CrossRef]
24. Sanli, N.; Fonrodona, G.; Barrón, D.; Özkan, G.; Barbosa, J. Prediction of chromatographic retention, pKa
values and optimization of the separation of polyphenolic acids in strawberries. J. Chromatogr. A 2002, 975,
299–309. [CrossRef]
25. De Las Rivas, B.; Rodríguez, H.; Curiel, J.A.; Landete, J.M.; Muñoz, R. Molecular screening of wine lactic
acid bacteria degrading hydroxycinnamic acids. J. Agric. Food Chem. 2009, 57, 490–494. [CrossRef] [PubMed]
26. Fras, P.; Campos, F.M.; Hogg, T.; Couto, J.A. Production of volatile phenols by Lactobacillus plantarum in wine
conditions. Biotechnol. Lett. 2014, 36, 281–285. [CrossRef]
27. Edlin, D.A.N.; Narbad, A.; Gasson, M.J.; Dickinson, J.R.; Lloyd, D. Purification and characterization of
hydroxycinnamate decarboxylase from Brettanomyces anomalus. Enzym. Microb. Technol. 1998, 22, 232–239.
[CrossRef]
28. Ripari, V.; Bai, Y.; Gänzle, M.G. Metabolism of phenolic acids in whole wheat and rye malt sourdoughs.
Food Microbiol. 2019, 77, 43–51. [CrossRef]
29. Nsogning Dongmo, S.; Sacher, B.; Kollmannsberger, H.; Becker, T. Key volatile aroma compounds of lactic
acid fermented malt based beverages–impact of lactic acid bacteria strains. Food Chem. 2017, 229, 565–573.
[CrossRef] [PubMed]
30. Rodríguez, H.; Angulo, I.; De Rivas, B.L.; Campillo, N.; Páez, J.A.; Muñoz, R.; Mancheño, J.M. p-Coumaric
acid decarboxylase from Lactobacillus plantarum: Structural insights into the active site and decarboxylaction
catalytic mechanism. Proteins 2010, 78, 1662–1676. [CrossRef] [PubMed]
31. Rocha-Mendoza, D.; Kosmerl, E.; Miyagusuku-Cruzado, G.; Giusti, M.M.; Jiménez-Flores, R.; García-Cano, I.
Growth of lactic acid bacteria in milk phospholipids enhances their adhesion to Caco-2 cells. J. Dairy Sci.
2020, in press. [CrossRef]
32. García-Cano, I.; Rocha-Mendoza, D.; Ortega-Anaya, J.; Wang, K.; Kosmerl, E.; Jiménez-Flores, R. Lactic
acid bacteria isolated from dairy products as potential producers of lipolytic, proteolytic and antibacterial
proteins. Appl. Microbiol. Biotechnol. 2019, 103, 5243–5257. [CrossRef] [PubMed]
33. Pruitt, K.D.; Tatusova, T.; Maglott, D.R. NCBI reference sequences (RefSeq): A curated non-redundant
sequence database of genomes, transcripts and proteins. Nucleic Acids Res. 2007, 35, 61–65. [CrossRef]
[PubMed]
34. Cavin, J.F.; Dartois, V.; Diviès, C. Gene cloning, transcriptional analysis, purification, and characterization of
phenolic acid decarboxylase from Bacillus subtilis. Appl. Env. Microbiol. 1998, 64, 1466–1471. [CrossRef]
Sample Availability: Samples of the compounds are not available from the authors.
©
2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).