α-Rhamnosidase activity in the marine isolate Novosphingobium sp. PP1Y and its use in the bioconversion of flavonoids

Article abstract of DOI:10.1016/j.molcatb.2014.04.002

Crude protein extracts of Novosphingobium sp. PP1Y, a microorganism isolated from polluted marine waters in Pozzuoli (Italy), were analyzed for the presence of glycosidase activities. Particular attention was devoted to a α-L-rhamnosidase activity able to hydrolyze several flavonoids of interest for the pharmaceutical and food industries. This activity had an alkaline pH optimum and a moderate tolerance to the presence of organic solvents, appealing features for its possible biotechnological uses. An increase of the α-L-rhamnosidase activity in PP1Y crude extracts was induced by adding naringin to the growth medium, suggesting the possibility to use material from Citrus industrial waste to induce the glycosidase activity expressed by strain PP1Y and produce simultaneously high-added-value molecules from the hydrolysis of their flavonoids. In order to investigate on the enzymatic mechanism of PP1Y α-L-rhamnosidase activity, hydrolysis products of PNP-α-L- rhamnopyranoside were analyzed by ¹H-NMR experiments. The kinetic behaviour clearly indicated an inverting mechanism of hydrolysis for this novel enzymatic activity.

Full text of DOI:10.1016/j.molcatb.2014.04.002

Journal of Molecular Catalysis B: Enzymatic 105 (2014) 95–103
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
␣-Rhamnosidase activity in the marine isolate Novosphingobium sp.
PP1Y and its use in the bioconversion of flavonoids
Viviana Izzoa,c, Pietro Tedescoa, Eugenio Notomistaa, Eduardo Pagnottab,
Alberto Di Donatoa, Antonio Trinconeb, Annabella Tramiceb,∗
a Dipartimento di Biologia, Università di Napoli Federico II, Complesso Universitario di Monte S. Angelo, Via Cinthia 4, 80126 Naples, Italy
b Istituto di Chimica Biomolecolare, Consiglio Nazionale delle Ricerche, Via Campi Flegrei 34, 80072 Pozzuoli, Naples, Italy
c Dipartimento di Medicina e Chirurgia, Università degli Studi di Salerno, Via S. Allende, 84081 Baronissi, Salerno, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Crude protein extracts of Novosphingobium sp. PP1Y, a microorganism isolated from polluted marine
waters in Pozzuoli (Italy), were analyzed for the presence of glycosidase activities. Particular attention
was devoted to a ␣-L-rhamnosidase activity able to hydrolyze several flavonoids of interest for the phar-
maceutical and food industries. This activity had an alkaline pH optimum and a moderate tolerance to the
presence of organic solvents, appealing features for its possible biotechnological uses. An increase of the
␣-L-rhamnosidase activity in PP1Y crude extracts was induced by adding naringin to the growth medium,
suggesting the possibility to use material from Citrus industrial waste to induce the glycosidase activity
expressed by strain PP1Y and produce simultaneously high-added-value molecules from the hydrolysis of
their flavonoids. In order to investigate on the enzymatic mechanism of PP1Y ␣-L-rhamnosidase activity,
hydrolysis products of PNP-␣-L-rhamnopyranoside were analyzed by 1H-NMR experiments. The kinetic
behaviour clearly indicated an inverting mechanism of hydrolysis for this novel enzymatic activity.
© 2014 Elsevier B.V. All rights reserved.
Received 3 January 2014
Received in revised form 6 March 2014
Accepted 5 April 2014
Available online 18 April 2014
Keywords:
␣-l-rhamnosidase
Novosphingobium sp. PP1Y
Flavonoids bioconversion
1. Introduction
the focus of an increasing scientific interest. As an example, the use
of ␣-RHAs to improve the bioavailability, and thus the biological
Biocatalysis represents nowadays a versatile and valuable tool
for industrial biotechnologies. The use of enzymes as biocatalysts
can have significant benefits compared to conventional chemical
technologies, for achieving high reaction selectivity, higher reac-
tion rate, improved product purity, and a significant decrease in
chemical waste production. A wide variety of chemical substances
is already produced in industrial processes through the use of
enzymes [1,2].
Within this framework, ␣-l-rhamnosidases [E. C. 3.2.1.40] have
attracted great attention in the last decade due to their application
as biocatalysts in a variety of food, pharmaceutical and chemical
industrial processes [3].
␣-l-Rhamnosidases (from now on indicated as ␣-RHAs) are a
class of glycosyl hydrolases that specifically cleave terminal ␣-l-
rhamnose from a large number of natural products, which include
flavonoids, some terpenyl glycosides [4,5] and many other natural
glycosides containing a terminal rhamnose, such as glycopeptides
antibiotics and glycolipids [6]. These enzymes have recently been
activity, of flavonoids beneficial for human health as direct drugs
or as nutritional supplements, has been reported [3].
Moreover, l-rhamnose has a central role in the organic synthesis
as a chiral intermediate of pharmaceutically important compounds.
Its production using the enzymatic activity of ␣-RHAs in hydrolysis
reactions of glycosylated compounds that might be recovered from
waste material of food processing industry (e.g. citrus peel), repre-
sents an interesting and useful biotechnological application these
glycosidases [7].
Fungi are the main source of ␣-RHAs. However, these enzymes
have been isolated also from animal tissues, such as the liver of both
the marine gastropod Turbo cornutus [8] and pig [9], and from plants
such as Rhamnus daurica [10] and Fagopyrum esculentum [11].
Bacteria represent a yet unexplored reservoir of ␣-RHAs, which
might show novel interesting properties. An analysis of several ␣-
RHAs isolated from various microbial sources reveals, for example,
that one of the main differences found between fungal and bacte-
rial enzymes is their optimal pH, with the fungal enzymes showing
more acidic pH optima compared to bacterial counterparts, for
which neutral and alkaline optimal pH values have generally been
reported. This characteristic suggests different potential applica-
tions for fungal and bacterial enzymes, making bacterial ␣-RHAs
∗
Corresponding author. Tel.: +39 0818675070; fax: +39 0818041770.
E-mail address: atramice@icb.cnr.it (A. Tramice).
http://dx.doi.org/10.1016/j.molcatb.2014.04.002
1381-1177/© 2014 Elsevier B.V. All rights reserved.

96
V. Izzo et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 95–103
suitable in biotechnological processes requiring good activity in
more basic solutions such as the production of l-rhamnose by
hydrolysis of naringin or hesperidin, compounds whose solubility
strongly increases at higher pH values [3,12].
without any further purification. Silica gel, reverse-phase silica
gel C-18 and TLC silica gel plates were from E. Merck (Darmstadt,
Germany). All other chemicals were of analytical grade. Compounds
on TLC plates were visualized under UV light or charring with ␣-
naphthol reagent.
Despite the very limited number of bacterial ␣-RHAs that has
been described so far, data in literature suggest that this enzymatic
activity is widely distributed over a diverse range of ecological
niches. ␣-RHAs have been detected and characterized in the human
intestinal bacterium Bacteroides JY-6 [13], in cold-adapted Pseu-
doalteromonas species and Ralstonia pickettii isolated in the sea
water of sub Antarctic environment [14], also in soil bacteria such as
Bacillus sp. GL1 [15,16], Sphingomonas paucimobilis FP2001 [17] and
Sphingomonas sp. R1 [18] and finally in wine strains of Oenococcus
oeni [19].
␣-RHAs from Lactobacillus species were identified and investi-
gated for their potential biotechnological use to de-rhamnosylate
flavonoids present in frequently consumed food commodities [20].
Moreover, ␣-RHAs genes have been cloned and expressed from
thermophilic bacteria such as Clostridium stercorarium [21] and
from the bacterium PRI-1686, which is a member of the phylum
of Thermomicrobia [22].
Recently, the crystal structure of Streptomyces avermitilis
␣-l-rhamnosidase (SaCBM67) was reported [23]; its structure pre-
sented a novel catalytic carbohydrate-binding module and resulted
different from the only two structures of ␣-l-rhamnosidases
(GH78) previously reported: the BsRhaB isolated from Bacillus sp.
GL1 [24] and the putative ␣-l-rhamnosidase BT1001 from Bac-
teroides thetaiotaomicron VPI-5482 [25].
TLC solvent systems used were (A) EtOAc:MeOH:H2O, 70:20:10
and (B) EtOAc:AcOH:2-Propanol:HCOOH:H2O, 25:10:5:1:15.
1H-, 13C- and 2D-NMR spectra were acquired by the NMR Service
of the Istituto di Chimica Biomolecolare of the National Research
Council of Italy (C.N.R.-Pozzuoli, Naples, Italy) and recorded on a
Bruker DRX-600 spectrometer, equipped with a TCI CryoProbeTM
,
fitted with a gradient along the Z-axis, and on other Bruker
instruments with fields at 400 and/or 300 MHz. Samples for NMR
analysis were dissolved in the appropriate solvent; spectra in D2O
were referenced to internal sodium 3-(trimethyl-silyl)-(2,2,3,3-
2H4) propionate (Aldrich, Milwaukee, WI); for other solvents
downfield shift of the signal of the solvent was used as internal
standard.
In the bioconversion studies followed by 1H-NMR experiments,
reported in paragraph 2.8 and 2.9, the selected signals integral
values of products and reagents resulted affected by an error in
integration (<5%) which depended upon instrument optimization.
Protein concentration was routinely estimated using the Bio-
Rad Protein System [28]; bovine serum albumin was used as
standard.
2.2. Bacterial growth
It goes without saying that the biotechnological potential of bac-
terial ␣-RHAs, whose structural, functional and molecular biology
aspects have not been sufficiently investigated, is strictly related
to the acquisition of new information on the enzymatic systems
isolated from new bacterial sources.
Novosphingobium sp. PP1Y was routinely grown in minimal
medium. Potassium phosphate minimal medium (PPMM) con-
tained 20 mM potassium phosphate pH 6.9, 1 g/L NH4Cl and
100 mM NaCl. After autoclaving, 5 mL of a trace element solution
was added to each litre of cooled PPMM. The trace element solu-
tion contained in a 0.92% solution of HCl: 30.1 g/L MgSO4, 4.75 g/L
FeSO4 × 7H2O, 5.4 g/L MgO, 1.0 g/L CaCO3, 0.72 g/L ZnSO4 × 7
H2O, 0.56 g/L MnSO4 × H2O, 0.125 g/L CuSO4 × 5 H2O, 0.14 g/L
CoSO4 × 7 H2O, 0.03 g/L H3BO3, 0.004 g/L NiCl2 × 6 H2O, 0.006 g/L
Na2MoO4 × 2H2O. When using PPMM as growth medium, 0.4%
(w/v) of glutamic acid, prepared in deionized water and steril-
ized by filtration with a 0.22 ␮m Millipore membrane, was used
as unique carbon and energy source and added to the autoclaved
media. Cultures were incubated at 30 ◦C with orbital shaking
(220 rpm).
Bacterial growth was monitored by measuring the optical
density at 600 nm (OD600). A pre-inoculum in rich medium was pre-
pared by transferring 50 ␮L from a glycerol stock stored at −80 ◦C
to a 50 mL Falcon tube containing 12.5 mL of sterile LB medium. LB
was prepared according to Sambrook et al. [29]. The pre-inoculum
was allowed to grow at 30 ◦C for at least 14 h at 220 rpm and then
used to inoculate 1 L of a preparative growth in PPMM medium at
an initial cell concentration of 0.01–0.02 OD600/mL. Cells were usu-
ally harvested after 24 h at a final optical density of ca. 1 OD600/mL.
It is worthy to note that, when following the growth for more than
24 h, the formation of flocks was observed, thus hampering a correct
estimation of the turbidity of the cell suspension.
Recently, Novosphingobium sp. PP1Y, an organic solvent-
resistant, biofilm-forming marine microorganism, was isolated
from the surface waters of a docking bay in the harbour of Poz-
zuoli (Naples, Italy), an area heavily polluted with hydrocarbons
[26]. The analysis of its genome, confirmed the presence of several
genomic features of interest for the biotechnological potential of
this microorganism, and more specifically for its exploitation as a
source of enzymes active on carbohydrates. Strain PP1Y shows in
fact a unique abundance among Sphingomonadales of genes encod-
ing for glycosyl hydrolases (53 orfs) [27], which are distributed
among 27 different families. This prompted our interest in investi-
gating the presence of ␣-RHA activities in the crude protein extract
of strain PP1Y.
In this work, cell extracts obtained from Novosphingobium sp.
PP1Y cells grown in minimal medium were tested for the presence
of glycosidase activities; in particular ␣-RHA and ␤-glucosidase
activities were detected and partially characterized.
This ␣-RHA activity was successfully used in bioconversion
studies of flavonoidic compounds, thus suggesting its potential use
as an eco-friendly tool to modulate the biological and pharmacolog-
ical properties of these molecules. It was further demonstrated by
1H-NMR experiments that ␣-RHA activity from Novosphingobium
sp. PP1Y functioned with an inverting mechanism, in agreement
with the enzymatic mechanism already proposed for other ␣-RHAs
previously described.
2.3. Preparation of Novosphingobium sp. PP1Y crude protein
extract
Cell cultures were harvested by centrifugation at 5,524 × g for
30 min at 4 ◦C. Cell pellets were suspended in 25 mM MOPS at pH
7.0 and disrupted by sonication (30ꢀ total, 30 s ON and 30 s OFF) in
an ice-water bath. Cell suspension was centrifuged at 17,418 × g at
4 ◦C for 60 min. The soluble fraction obtained after centrifugation
was filtered on a 0.45 mm Millipore membrane, divided in aliquots
and stored at −80 ◦C. When used for the screening of glycosidase
2. Materials and methods
2.1. General
Novosphingobium sp. PP1Y was isolated as previously reported
[26]. All reagents were purchased from Sigma-Aldrich and used

V. Izzo et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 95–103
97
Table 1
Rutinose hydrolysis was also investigated. In this case, 0.5 mL of
a solution containing 6 mM rutinose, 0.25 mL of PP1Y crude extract
in 50 mM of Tris/HCl buffer at pH 8.5 was incubated at 35 ◦C under
magnetic stirring and monitored by TLC analysis (solvent system
A) for 24 h.
Evaluation of different substrates bioconversion by using glycosidase activities in
the crude extract of Novosphingobium sp. PP1Y cells.
Substratesa
Crude extract of PP1Y cells grown in PPMM medium
Hydrolysisb
3 h
Transglycosylationc
In all experiments, TLC standard solutions of pure reagents and
products were used for comparison.
24 h
3 h
24 h
PNP-␣-d-Glcp
PNP-␤-d-Glcp
PNP-␣-d-Galp
PNP-␤-d-Galp
PNP-␣-d-Manp
PNP-␤-d-Manp
PNP-␣-d-Xylp
PNP-␤-d-Xylp
PNP-␣-l-Fucp
PNP-␤-d-Fucp
PNP ␣-d-NAGp
PNP ␤-d-NAGp
PNP-␣-l-Rhap
PNP-␣-l-Araf
+
+
+
++
+
−
+/−
++
+/−
++
−
+
2.5. ˛-l-rhamnosidase activity assay
+/−
−
+
++
−
+
␣-RHA activity was determined using p-nitrophenyl-␣-l-
rhamnopyranoside (pNPR) as substrate. The assay was performed
in 0.5 mL of 50 mM potassium acetate buffer pH 5.5, containing a
variable amount of PP1Y crude extract and pNPR at a final concen-
tration of 0.28 mM. The assay mixture was incubated for 30 min
at room temperature; afterwards, 0.5 mL of a 1 M sodium carbon-
ate solution were added and the sample absorbance was recorded
spectrophotometrically at 405 nm (ε405 = 0.0182 ␮M−1cm−1). One
milliunit of enzymatic activity was defined as the amount of
enzyme that releases 1 nmol of p-nitrophenol per min.
−
−
−
−
−
−
−
−
−
−
+
++
−
+/−
−
+
−
−
+
++
−
+
++
−
−
−
+/−
++
−
+
+/−
−
+/−
+/−
−
+++
−
−
Note: −/+: percentage of products below 10%; +: 10–30% of products; ++: 30–70% of
products; +++: percentage of products higher than 70%.
a
Araf, arabinofuranose; Fucp, fucopyranose; Galp, galactopyranose; Glcp, glu-
2.6. Induction of ˛-l-rhamnosidase activity
copyranose; Manp, mannopyranose; NAGp, N-acetylglucosaminopyranose; Rhap,
Rhamnopyranose; Xylp, xylopyranose.
b
Hydrolysis products were analyzed by comparing Rf values with appropriate
To evaluate the possible influence of naringin on the expres-
sion of the intracellular ␣-rhamnosidase activity of the strain PP1Y,
we compared the specific activity in cell extracts obtained from
growths in PPMM to which increasing concentrations of naringin
were added. To this purpose, four 250 mL-Erlenmeyer flasks were
prepared, each containing 250 mL of PPMM medium and 0.4% glu-
tamic acid as unique carbon and energy source. Variable amounts
of naringin were added to autoclaved media, in order to have final
concentrations of 0.1–0.2–0.3 mM. The four flasks were inoculated
with an overnight culture of PP1Y in LB, as described in paragraph
2.2. The initial cell concentrations in the flasks were comparable
and in the range 0.02–0.04 OD600/mL.
standards in TLC solvent system A.
c
Due to self-transglycosylation reactions, products showed positive UV
absorbance and lower Rf than the corresponding aryl substrates in the selected TLC
solvent system A.
activities (Section 2.4), the soluble fraction was dialyzed against
20 mM potassium phosphate at pH 7, and then divided in aliquots
and stored at −80 ◦C. PP1Y crude extracts were assayed for their
total protein content.
Without any different indication, all experiments reported in the
following sections, were performed using a crude extract obtained
from a 2 L culture of PP1Y cells grown in PPMM medium and with
a protein concentration of 3.72 mg total protein/mL. Cells were
recovered after 24 h at an optical density of 1.02 OD600/mL.
2.7. Influence of pH and organic solvents on ˛-l-rhamnosidase
and ˇ-glucosidase activities
In experiments of pH monitoring, 2 mM solutions of naringin
were prepared in HCl/KCl, Na-acetate, Na-phosphate, Tris/HCl, Na-
carbonate buffers, in order to cover the pH range between 2.2 and
10.9 (Table 2). 100 ␮L of PP1Y crude extract were added to 0.5 mL
of each solution and incubated at 35 ◦C under magnetic stirring.
Reactions were checked every 20 min by TLC analyses (solvent sys-
tem A) and naringin consumption was evaluated by using solutions
of pure naringin, narigenin, rhamnose and glucose as chromato-
graphic standards. Blank reactions at different pHs were performed
to evaluate the chemical degradation of the flavanone glycoside,
occurring preferentially in strongly basic conditions.
Similarly, solutions of 2 mM naringin in 50 mM Tris/HCl buffer
pH 8.5 containing 1, 10 and 50% v/v of DMSO or CH3CN (Table 2)
were placed under constant agitation at 35 ◦C in the presence of
200 ␮L of PP1Y crude enzymatic extract (0.74 mg total protein)/mL
of reaction. As previously reported, reactions were monitored by
TLC analysis (solvent system A).
2.4. Glycosidase activities screening
PP1Y crude protein extract, prepared as described in the previ-
ous paragraph, was tested for the presence of glycosidase activities
by using several pNP-␣- and pNP-␤-substrates (Table 1) at a 20 mM
concentration in 0.6 mL of 50 mM Na-phosphate buffer pH 7, at
35 ◦C and under magnetic stirring. All reactions were carried out
using 4.42 mg of total protein/mmol of reagent. Reactions were
monitored over time (0–24 h) by TLC analysis (system solvent A).
Moreover, hydrolysis reactions with maltose, lactose, cellobiose,
sucrose, raffinose, laminaripentaose, laminarin, curdlan, ␤-glucan
from barley, xylan from birch wood, pullulan, amylopectin and
starch as substrates were carried out using crude extracts of PP1Y
cells grown in PPMM. Each substrate (10 mg) was suspended in
1 mL of 50 mM Na-phosphate buffer pH 7, containing 41.6 ␮g of
total protein/mL of reaction mixture. Reactions were performed at
35 ◦C. Hydrolysis products were monitored by TLC analysis (system
solvents A and B).
2.8. Stereochemical course of ˛-l-rhamnosidase activity in
Flavonoidic substrates such as naringin, diosmin, rutin, hes-
peridin, neohesperidin dihydrochalcone, quericitrin, were tested.
To this purpose, a 2 mM solution of each compound prepared in
a final volume of 1 mL of 50 mM Na-phosphate buffer pH 7 was
incubated at 35 ◦C in the presence of 100 ␮L of PP1Y crude extract
obtained from cells grown in PPMM (3.72 mg of total protein/mL)
for 24 h. Reactions were checked over time by TLC analysis (solvent
system A).
hydrolysis reactions
To the purpose, 800 ␮L of a 6 mM pNPR solution prepared
in 50 mM Tris/HCl buffer pH 7 was freeze-dried, re-suspended
three times in D2O and re-freeze-dried to exchange the 1H atoms
involved in labile linkages for 2H ones. A similar procedure of proton
isotope exchange was applied to 200 ␮L of PP1Y crude extract. Just
prior to the 1H-NMR experiment, the previously exchanged pNPR

98
V. Izzo et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 95–103
sample was incubated at 35 ◦C, under magnetic stirring. Reaction
was stopped after 72 h by heating the sample at 100 ◦C for 2 min
and prunin was isolated by silica gel purification with EtOAc:MeOH
9:1 (v/v) as mobile phase. 1D and 2D NMR analyses confirmed its
structure.
Table 2
The influence of pH and organic solvents presence on naringin hydrolysis by ␣-RHA
activity in PP1Y crude protein extract.
pH values
Naringin hydrolysis
Time (min)
60
120
140
180
2.2
3.6
5.4
5.9
6.1
7.2
8.1
8.3
8.5
8.8
9.1a
9.9a
10.9a
−
−
−
−
−
−/+
−
+
+
+
−
−
−
−
−
−
3. Results and Discussion
−
−
−
−
−
−
3.1. Glycosidase activities screening
−
−
−/+
+
−/+
+
−/+
++
++
+++
+++
+++
+
++
++
+++
+++
+++
+
Glycosidase activities in crude protein extract of Novosphingob-
ium sp. PP1Y cells collected after a 24 h growth in PPMM medium
(Section 2.2) were preliminary investigated for their hydrolytic and
transglycosylation potentials, as described in Section 2.4. Reaction
products were analyzed over time by TLC analysis using solvent
system A Section 2.1, and the results were reported in Table 1.
Data indicated that ␣-l-rhamnosidase and ␤-glycosyl hydrolases
activities resulted the most abundant ones. After 3 h, 30–70% of
PNP-␣-l-Rhap and 10–30% of PNP-␤-d-Glcp, PNP-␤-d-Galp, and
PNP-␤-d-Xylp were hydrolyzed. Hydrolysis reactions were allowed
to proceed for 24 h. At this time, an almost complete conversion of
PNP-␣-l-Rhap was observed.
++
++
+++
++
−
−
+
+
+
+
−
% of organic solvent
DMSO 1%
DMSO 10%
DMSO 50%
CH3CN 1%
CH3CN 10%
CH3CN 50%
+
+
−
−
−
−
++
+
−
−
−
−
++
++
−
+++
+++
−
−
−/+
−
−
−
−
Transglycosylation reactions resulted more interesting by using
PNP-␤-glycosides as substrates, suggesting
a better attitude
Note: − no hydrolysis; −/+: percentage of hydrolysis below 10%; +: 10–30% of hydrol-
ysis; ++: 30–70% of hydrolysis; +++: percentage of hydrolysis higher than 70%.
of ␤-glycosyl hydrolases to transfer monosaccharidic units in
self-condensation reactions than the ␣-l-rhamnosidase activity;
aryl-␤-oligosaccharides were present in the reaction media up to
24 h.
a
In these conditions a partial chemical degradation of naringin was observed
which was suggested by the presence on TLC, along with the disappearance of the
substrate, of a product more polar than naringin and the absence of rhamnose and
glucose spots, thus confirming the absence of an enzymatic hydrolysis.
Polysaccharides and oligosaccharides were also tested as pos-
sible substrates for the enzymatic activities present in PP1Y crude
extracts. Products formation was monitored by TLC analysis (sys-
tem solvents A and B, Section 2.1), using pure compounds as
chromatographic standards. After 24 h of reaction, curdlan and
laminarin were only slightly consumed whereas laminaripentaose
resulted the best hydrolyzed oligosaccharide, resulting in the major
production of glucose; cellobiose, instead, was not a good substrate
(data not shown).
␣-glucans like pullulan, amylopectin, starch and maltose were
partially or almost totally hydrolyzed to glucose, maltose or larger
oligosaccharides, suggesting the possible presence of ␣-glucanase
activities (data not shown).
solution was dissolved in 0.75 mL of D2O, equilibrated at 35 ◦C in a
NMR tube inside the spectrometer probe, and an initial spectrum of
solution without enzyme was recorded (T0). The aliquot of freeze-
dried crude extract was suspended in 50 ␮L of D2O and added to the
solution. Starting from that moment (T0 with enzyme), the stereo-
chemical course of hydrolysis was followed by recording 1H-NMR
spectra at close intervals during the incubation time (from 0 to
60 min).
A 6 mM solution of free rhamnose dissolved in 50 mM Tris/HCl
buffer pH 7 was treated and monitored by 1H-NMR experiments as
previously reported.
These preliminary data confirmed the biotechnological poten-
tial of PP1Y strain as a source of a wide array of enzymatic activities
that can be used for the modification of carbohydrates and glyco-
conjugates.
2.9. Bioconversion studies followed by 1H-NMR analyses
pNPR, naringin, rutin and neohesperidin dihydrochalcone
bioconversions were performed and compared by 1H-NMR spec-
troscopic analyses.
In particular, 11.4 ␮mol of each substrate were dissolved in
1.9 mL of Na-phosphate buffer (50 mM, pH 8.5 or pH 7) contain-
ing 1 mL of PP1Y crude extract. Reactions were incubated at 35 ◦C,
under constant magnetic stirring for 3 h. Aliquots of 380 ␮L were
withdrawn from each reaction at different time intervals from 0
to 173 min. Enzymatic reactions were stopped by boiling at 100 ◦C
for 2 min and samples were freeze-dried. Collected aliquots were
dissolved in MeOD or D2O, as for pNPR, and analyzed by 1H-NMR
experiments. These reactions were also monitored by TLC analy-
sis (system solvent A). These reactions were also monitored by TLC
analysis (system solvent A).
3.2. Hydrolytic behaviour of PP1Y crude extract enzymes on
flavonoids: description, pH dependence and tolerance to organic
solvents
The simultaneous presence in PP1Y crude extract of ␣-RHA and
␤-glucosidase activities prompted us to investigate the possibil-
ity to hydrolyze selected flavonoidic compounds, having in their
chemical structure both ␣-rhamnose and ␤-glucose units, such as
naringin, diosmin, rutin, hesperidin, neohesperidin dihydrochal-
cone and quericitrin (Fig. 1). It is worth to note that chemical
modification of these compounds, which are endowed with ther-
apeutic potential [12] is important to modulate their biological
activity.
2.10. Prunin production
In the reaction conditions reported in Section 2.4, after 3 h of
incubation, complete conversion of naringin, rutin and neohes-
peridin dihydrochalcone was detected by TLC analysis (system
solvent A, Fig. 2): spots of rhamnose, glucose and the corresponding
aglycone were observed for each reaction.
The reaction was performed using 2 mL of PP1Y crude extract
which were previously lyophilized and then resuspended in a 1 mL
solution of Na-phosphate buffer (50 mM, pH 8.5) containing 10%
of DMSO. 72.5 mg of naringin (0.125 mmol) were added and the

V. Izzo et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 95–103
99
Fig. 1. Chemical structure of flavonoidic compounds used as substrates. Chemical shift values of signals that are integrated in bioconversion reactions are reported.
On the other hand, diosmin, hesperidin and quericitrin were
not hydrolyzed; after 24 h as observed by TLC investigations (sys-
tem solvent A) rhamnose and glucose spots, at Rf 0.56 and 0.3
respectively, were absent on TLC plates of diosmin and hesperidin
reactions. Quericitrin was so poorly consumed (yield below 3%) that
its hydrolysis was considered negligible; indeed, a very scarce pres-
ence of quercetin and rhamnose, at Rf 0.88 and 0.56 respectively,
was observed.

100
V. Izzo et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 95–103
Free rutinose disaccharide was also examined as a possible sub-
strate, resulting in a very low yield of products (data not shown),
thus suggesting the importance of the aromatic portion of the
flavonoidic moiety in the recognition mechanism of the enzymes
involved.
Moreover, quercitrin, the quercetin 3-␣-l-rhamnoside, in which
l-rhamnose is directly linked to the aglycon, was not hydrolyzed.
It should be noted that a marked preference for the of PNP-␣-l-
Rhap (from now on indicated as pNPR) in comparison to quercitrin,
robonin or rutin has been reported also for other characterized ␣-
RHAs [3].
The solubility of flavonoids (typically polyphenols) is pH-
dependent and generally increases in alkaline media, which are
often not compatible with the enzymatic activities used in bio-
processes. Using naringin as the most water-soluble available
flavonoid (0.5 g/L at 20 ◦C) [31], the influence of pH and organic sol-
vents on the hydrolytic behaviour of the enzymes present in PP1Y
crude extracts was investigated.
Results of these experiments were reported in Table 2. A higher
yield of naringin hydrolysis products was obtained at alkaline pH
values ranging from 7.2 to 8.8 (Na-phosphate and Tris/HCl buffers)
and an optimum pH value of 8.5 was determined. In this case, after
3 h of reaction, substrate spots were absent on TLC plates, thus
suggesting a substrate complete depletion.
Within a reaction time of 3 h, TLC analyses of reaction systems in
the pH range 8.2–8.8 showed the presence of a product with an Rf
value higher than that of naringin. Free glucose and the narigenin
aglycone were almost absent.
These data suggested the occurrence at these basic pH values of
a marked decrease of the ␤-glucosidase activity, which should be
responsible for the production of free glucose and narigenin agly-
cone, and the permanence of the ␣-RHA activity. This hypothesis
was confirmed by the concomitant accumulation of a compound
which was identified as prunin (Fig. 1), the de-rhamnosylated
product of naringin, a molecule of biotechnological importance
endowed with anti-inflammatory and antiviral activity against
DNA/RNA viruses [32,33].
Fig. 2. Flavonoids hydrolysis followed by TLC (system solvent A). TLC standard
solutions at a concentration of 2 mM were used. Nar: naringin, R-nar: naringin
reaction; rut: rutin, R-rut: rutin reaction; neohe: nehoesperidin dihydrochalcone;
R-neohe: nehoesperidin dihydrochalcone reaction; glu: glucose, rha: rhamnose; es:
enzymatic solution.
It is worth to note that bacterial ␣-RHAs having a pH optimum
value at 8–8.5 have not been often reported in literature [30].
Prunin structure (Fig. 1) was confirmed by a 2D NMR spectro-
scopic investigation. The presence of the ␤-glucose residue was
established by the value of the anomeric position signal in 13C NMR
spectrum (in MeOD) at 101.25 ppm which resulted correlated in
HSQC spectrum to the diastereoisomeric anomeric protons H-1ꢀ at
5.01 (J = 6.90 Hz), and 4.99 (J = 6.90 Hz) ppm. The following signals
(in ppm): C-2ꢀ 74.66 (H-2ꢀ: 3.48), C-3ꢀ 77.82 (H-3ꢀ: 3.47), C-4ꢀ 71.16
(H-4ꢀ: 3.41), C-5ꢀ 78.28 (H-5ꢀ: 3.42), C-6ꢀ 62.34 (H-6ꢀ-H-6ꢀꢀ: 3.90,
3.72) confirmed the glucose structure of prunin saccharidic portion.
Aglycone structure was confirmed by the 13C spectrum signals at
80.69 ppm (C-2), 44.19 ppm (C-3), 166.98 and 167.05 ppm (C-7, the
glycosylation site), 129.12 ppm (C-2ꢀꢀ-C6ꢀꢀ), 116.33 ppm (C-3ꢀꢀ-C5ꢀꢀ)
and in 1H-NMR spectrum by the proton signals at 7.35 ppm (H-
2ꢀꢀ-H6ꢀꢀ) and 6.85 ppm (H-3ꢀꢀ-H5ꢀꢀ). The other signals in the spectra
were in agreement with data previously reported [34]. The increase
of the ␣-RHA activity present in PP1Y protein extracts observed at
alkaline pH values was also confirmed by the pNPR bioconversion
study followed by NMR analysis reported in Section 3.3.
The pH variation is not the unique strategy that has been used to
overcome the poor hydrosolubility of flavonoids; the use of water-
miscible organic co-solvents, in fact, has also been suggested [31].
However, it should be noted that this strategy has several draw-
backs, such as toxicity problems derived from the use of an organic
solvent in a bioprocess, and the possible decrease in the enzymatic
activities employed [35].
These data gave some indication on the substrate preference of
the ␣-RHA activity expressed by strain PP1Y under our experimen-
tal conditions.
Diosmin and hesperidin (Fig. 1), which contained the rutinose
disaccharide (␣-l-Rhamnopyranosyl-(1 → 6)-␤-d-glucopyranose)
linked to the phenolic portion of the molecules, were not
hydrolyzed. Rutin, whose rutinoside portion was linked to the
quercetin 3-enol position, was instead converted. Under similar
reaction conditions, flavonoids with the neohesperidose disac-
charide 2-O-(6-deoxy-␣-l-mannopyranosyl)-␤-d-glucopyranose)
linked to the phenolic portion (naringin and neohesperidin dihy-
drochalcone) were completely transformed.
These data suggested that the ␣-RHA activity was able to
hydrolyze both ␣,1-2 and ␣,1-6 interglycosidic linkages. Most of
the ␣-RHAs reported in literature are mainly active on ␣-1, 2 gly-
cosidic linkages, to a smaller extent on ␣-1,6 linkages, and even
less on other glycosidic bonds [30]. Interestingly, the enzymatic
activity expressed by strain PP1Y was specific for ␣,1-2 intergly-
cosidic bonds when the disaccharide unit of rutinose was linked
to the phenolic hydroxyl groups. In fact, diosmin and hesperidin,
which shared the presence of a ␣,1-6 interglycosidic bond in the
disaccharide units linked to a phenolic site, were not hydrolyzed.
The ␣,1-6 interglycosidic bond was instead hydrolyzed when the
rutinose unit was linked to 3-enol position of rutin. This hydrolytic
behaviour might be a consequence of a different steric effect deriv-
ing from the attachment of the disaccharidic units to different sites
of the aglycons [3].
In this study, the influence of the presence of organic solvents
on the hydrolysis of naringin was evaluated by TLC analysis of

V. Izzo et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 95–103
101
On the other hand, in 1H-NMR spectra at 173 min, anomeric
protons region was more crowded with signals due to the pres-
ence of free glucose (␣-proton at 5.13 ppm J = 3.91 Hz; ␤-proton at
4.50 ppm, J = 7.8 Hz) and a signal at 5.023 ppm of free rhamnose
␣-proton (with the ␤-anomeric proton overlapped by the solvent
residual water signal).
These data suggested a fast hydrolysis of the interglycosidic link-
age inside the rutinose unit (␣-l-Rha-(1 → 6)-␤-d-Glu) with the
production of free rhamnose at the beginning of the reaction, as
confirmed by the presence of a further methyl signal at 1.27 ppm
after 10 min of reaction, corresponding to protons at position 6 of
free rhamnose, and the subsequent release of flavonol quercetin
and glucose.
The Neohesperidin dihydrochalcone bioconversion was eval-
uated by 1H-NMR using the aromatic signals at 6.049 ppm,
corresponding to two overlapped aromatic protons H2 and H6
(Fig. 1) linked to the saccharidic portion, as the reagent reference
signals, with a integral value of 1, and a signal at 6.077 ppm, belong-
ing to the reaction product.
An evaluation of the anomeric signals in the range 4.45:5.2 ppm
of 1H-NMR experiment showed, after 173 min of reaction, the pres-
ence of the ␣-proton signal of linked rhamnose at 5.28 ppm and the
anomeric signal of ␤-glucose at 5.065 ppm (d, J = 7.63 Hz) which
belonged to the substrate, the ␣- and ␤-anomeric protons sig-
nals of free glucose at 5.13 ppm (d, J = 3.81 Hz) and 4.50 ppm (d,
J = 7.63 Hz) respectively. Moreover, an intense signal at 5.03 ppm
of the ␣-anomeric proton of free rhamnose, a signal at 4.95 ppm
(d, J = 7.31 Hz) almost as intense as the previous signal, which was
partially covered by residual H2O signal in MeOD, corresponding
to ␤-glucose H1 signal of neohesperidin dihydrochalcone without
rhamnose were also observed. On the contrary, the free rhamnose
␤-anomeric signal was totally overlapped by residual H2O signal in
MeOD.
reaction mixtures containing from 1 to 50% v/v DMSO or CH3CN
(Table 2). After 90 min, in the presence of 1% and 10% of DMSO,
naringin was almost totally consumed and traces of prunin were
detected, suggesting a tolerance of the ␣-RHA activity to the pres-
ence of organic solvents along with a detrimental effect of DMSO
on the ␤-glucosidase activity, as free glucose was almost absent in
the reaction mixtures.
Finally, in the presence of either 50% DMSO or any concentration
of CH3CN, de-rhamnosylation and de-glucosylation reactions did
not proceed even after 10 h.
Using the apparent best experimental conditions reported in
Table 2 (pH value of 8.5 and 10% DMSO), and starting from a 125 mM
naringin solution (Section 2.10), prunin was produced with a yield
of 32.1%, corresponding to a final concentration of 0.02 M and an
amount of 17.4 mg of recovered product.
Furthermore, free l-rhamnose was produced as a secondary
product at a concentration of 6 g/L. The yield of products obtained
in this experiment was comparable to that reported for a recombi-
nant ␣-RHA from Clostridium stercorarium used for the hydrolysis
of citrus peel waste naringin [7] and to other similar processes [30].
3.3. Bioconversion experiments followed by NMR
To confirm and better detail the observations on the substrate
preference of the ␣-RHA activity in Novosphingobium sp. PP1Y
crude protein extracts described in the previous paragraph, bio-
conversion reactions of pNPR, naringin, rutin and neohesperidin
dihydrochalcone were performed using PP1Y cell extracts (Section
2.9), and monitored over time by 1H-NMR.
Based on diagnostic signals of reagents and products for each
reaction, substrate conversions were calculated by measuring the
percentage ratio between the integrals of diagnostic signals of the
products and the sum of the integration values of selected signals
of reagents and final products. Results were reported in Fig. 3.
The highest yield of bioconversion in a shortest time was
obtained with rutin. After 20 min 76.2% of the substrate was con-
sumed. At the same time, 30–35% of pNPR, 26% of naringin, and 7%
of neohesperidin dihydrochalcone were transformed.
After 173 min, naringin and neohesperidin dihydrochalcone
reached bioconversions yields (95.6 and 86%, respectively) similar
to rutin (89.4%). Under these conditions, 3.1–3.3 g/L of flavonoids
were converted to products and, although the flavonoidic sub-
strates were not totally hydrolyzed to aglycone, free l-rhamnose
was released into the reaction media at a concentration of about
0.77–0.86 g/L.
It should be added that pNPR bioconversion was performed not
only at pH 8.5 but also at pH 7, looking for a possible pH influence on
PP1Y ␣-RHA activity (see Section 3.2). In the experiment in which
naringin was used as a substrate, the reaction course was evalu-
ated by comparing in 1H-NMR spectra the integral values of two
aromatic protons singlets (Fig. 1) H6–H8 in ring B at 6.19 ppm and
6.28 ppm, which were attributed a value of 2, to the integration of
the corresponding protons of the final product narigenin (aglycon)
at 5.78 and 5.8 ppm. 1H-NMR spectra in MeOD of naringin, narin-
genin and prunin (H6–H8 signals at 6.22–6.24, respectively) of pure
solutions confirmed the identity of narigenin integrated signals.
When rutin was used as substrate, aromatic proton signal H6
of the reagent at 6.18 ppm was selected as reference for the inte-
gration (Fig. 1). As for the reaction products, an aromatic signal at
6.13 ppm, which increased over time, was selected for evaluating
rutin bioconversion.
The high intensity of the ␣-anomeric proton of free rhamnose
and the scarce presence of anomeric signals of free glucose, led us to
suppose that the main reaction product was the de-rhamnosylated
neohesperidin dihydrochalcone.
In experiments using pNPR as substrate, the integral value of
aromatic proton signal at 7.23 ppm was set to 1 and the signal of p-
nitrophenol at 6.69 ppm produced by the hydrolysis of glycosidic
linkage was selected for the assessment of pNPR bioconversion.
This bioconversion was performed at two pH values; as reported in
Fig. 2, after 60 min, the conversion of pNPR was of about 61% at pH
7, whereas at higher pH it reached 97%, thus confirming the trend of
the experiments in which naringin was used as substrate described
in Section 3.2.
3.4. Induction of ˛-RHA activity in PP1Y cell extracts
␣-RHA activity can be specifically induced by several flavonoids
[36,37] among which naringin is of particular interest because this
compound can be abundantly recovered in citrus solid wastes.
These wastes are quite interesting because they can be used either
as an energetic source for growing microorganisms, due to their
content in carbon and other nutrient components, and/or as a
specific inducer for the biosynthesis of glycosidases due to the pres-
ence of flavonoids. It goes without saying that the use of citrus
waste as starting material in bioconversion processes is an advan-
tage for industrial companies because, among others, it decreases
the expenses for waste disposal. The possible role of naringin as
an inducer of the ␣-RHA activity [37] in PP1Y protein extracts
was investigated using PP1Y cells grown in PPMM medium sup-
plemented with naringin at different initial concentrations up to
0.3 mM, as described in detail in Section 2.6.
After 10 min of reaction, the anomeric proton of ␤-glucose in
rutin at 5.07 ppm (d, J = 7.61 Hz) almost disappeared (72% of con-
version) and an anomeric proton signal at 5.035 ppm (d, J = 8.07 Hz),
possibly corresponding to the ␤-anomeric proton of quercetin 3-␤-
glucoside, appeared; anomeric signals of free glucose were absent.

102
V. Izzo et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 95–103
Fig. 3. Bioconversions of naringin ᭹, rutin ꢀ, neohesperidin dihydrochalcone —, p-nitrophenyl-␣-l-rhamnopyranoside at pH 7 ꢁ and 8.5 ꢂ.
The effect of naringin could not be evaluated at concentrations
higher than 0.3 mM due to its poor solubility under the experimen-
tal conditions used.
␣-1(P), and ␤-1(P), were plotted as % of total anomeric (H-1) signals.
At t0, the only ␣-anomeric proton signal present was that of pNPR
at 5.68 ppm (␣-1 (S)). During the first minutes of reaction a signal
at 4.83 ppm, corresponding to the ␤-l-Rhap anomeric proton (␤-1
(P)), appeared only slightly earlier than the appearance of a further
signal at 5.06 ppm, corresponding to the ␣-l-Rhap anomeric proton
(␣-1 (P)).
These signals increased in intensity over time, while the
anomeric signal of pNPR decreased concomitantly, as it is evident
from Fig. 5. NMR analysis did not gave evidence for the onset of
mutarotation equilibration during the first minutes of reaction.
Later in the incubation, mutarotation of the initially released ␤-
l-Rhap increased the ratio between the intensities of the ␣- and
␤-anomer protons signals. After 7 min, at a conversion percentage
of 7.5%, the relative intensities of the ␣- and ␤-anomer resonances
(calculated from peaks integration) were in a ratio of about 1:3
(25% of the ␣- and 75% for the ␤-anomer). After 20 min, at a con-
version percentage of 29%, the ␣-/␤-anomer ratio was of about 1:2
(32% of the ␣- and 68% for the ␤-anomer). After 60 min, at a con-
version percentage of 61.5%, the ␣/␤ ratio value increased to about
1:1. This ratio was kept constant also after 80 min, when the inte-
gration values showed that the yield of pNPR consumption was of
The effect of naringin could not be investigated at concentra-
tion values higher than 0.3 mM due to its poor solubility under the
experimental conditions used. Results were reported in Fig. 4 and
confirmed that naringin acts as an inducer leading, at a concentra-
tion of 0.3 mM, to a maximum 5-fold increase of the ␣-RHA activity
(15.2 mU/mg) detected in the crude extract of PP1Y cells. These
data encourage a future biotechnological use of PP1Y strain and of
its ␣-RHA activity for the use of agro-industry vegetable residues.
3.5. Stereochemical course of ˛-RHA activity in hydrolysis
reactions
To shed light on the type of enzymatic mechanism of PP1Y
␣-RHA activity, hydrolysis products of pNPR were analyzed by
recording 1H-NMR spectra over time. The integration of the
anomeric proton signals of reagent and products was investigated;
the analysis was performed taking into account that the anomeric
proton of free rhamnose shows peaks at 5.06 ppm (␣-anomeric
form) and 4.83 ppm (␤-anomeric form). Results were reported in
Fig. 5, where the reaction time-dependent changes in amplitude
of H-1 signals from substrate, ␣-1(S) (5.68 ppm), and products,
PNP-α- Rha hydrolysis
100
90
80
70
60
50
40
30
20
10
0
2
7
10
12
20
40
60
80
time (min
)
Fig. 4. ␣-RHA specific activity in the cell extract of strain PP1Y is expressed as
mU/mg of total proteins (Y-axis). Specific activity is reported as a function of the
concentration of naringin initially added to the growth medium.
Fig. 5. % of total anomeric signals in p-nitrophenyl-␣-l-rhamnopyranoside hydrol-
ysis reaction: signals from substrate, ␣-1(S) ꢁ, and products, ␣-1 (P) ꢂ, and ␤-1 (P)
ꢀ, are plotted as % of total anomeric (H-1) signals.

V. Izzo et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 95–103
103
about 86%, and was maintained up to 18 h, thus representing the
equilibrium value of the ratio between both anomeric forms of l-
rhamnopyranose. A solution of free rhamnose used in the same
experimental conditions and monitored by 1H-NMR reached the
mutarotational equilibrium after 1–2 min and the concentrations
ratio of ␣- and ␤-rhamnose was 1:1.
[7] A. Kaur, S. Singh, R.S. Singh, W.H. Schwarz, M. Puri, J. Chem. Technol. Biotechnol.
85 (2010) 1419–1422.
[8] Y. Kurosawa, K. Ikeda, F. Egami, J. Biochem. 73 (1973) 31–37.
[9] S. Qian, H. Yu, C. Zhang, M. Lu, H. Wang, F. Jin, Chem. Pharm. Bull. 53 (2005)
911–914.
[10] H. Suzuki, Arch. Biochem. Biophys. 99 (1962) 476–483.
[11] R. Bourbouze, F. Pratviel-Sosa, F. Percheron, Phytochemistry 14 (1975)
1279–1282.
The kinetic behaviour suggested by these experiments for PP1Y
␣-RHA activity was an inverting mechanism of hydrolysis in which
␤-rhamnose was formed from the ␣-rhamnose via a single dis-
placement mechanism and was then spontaneously converted by
mutarotation to the ␣-anomeric form.
[12] M. Puri, Appl. Microbiol. Biotechnol. 93 (2012) 49–60.
[13] I.S. Jang, D.H. Kim, Biol. Pharm. Bull. 19 (1996) 1546–1549.
[14] A.G. Orrillo, P. Ledesma, O.D. Delgado, G. Spagna, J.D. Breccia, Enzyme Microb.
Technol. 40 (2007) 236–241.
[15] W. Hashimoto, H. Nankai, N. Sato, S. Kawai, K. Murata, Arch. Biochem. Biophys.
368 (1999) 56–60.
[16] W. Hashimoto, O. Miyake, H. Nankai, K. Murata, Arch. Biochem. Biophys. 15
(2003) 235–244.
[17] T. Miyata, N. Kashige, T. Satho, T. Yamaguchi, Y. Aso, F. Miake, Curr. Microbiol.
51 (2005) 105–109.
4. Conclusions
[18] W. Hashimoto, K. Murata, Biosci. Biotechnol. Biochem. 62 (1998)
1068–1074.
[19] A. Grimaldi, E. Bartowsky, V. Jiranek, Int. J. Food Microbiol. 105 (2005)
233–244.
[20] J. Beekwilder, D. Marcozzi, S. Vecchi, R. deVos, P. Janssen, C. Francke, R.D. Hall,
Appl. Environ. Microbiol. 75 (2009) 3447–3454.
[21] V.V. Zverlov, C. Hertel, K. Bronnenmeier, A. Hroch, J. Kellermann, W.H. Schwarz,
Mol. Microbiol. 35 (2000) 173–179.
Bacterial glycosyl hydrolases are the focus of an increasing num-
ber of researches because of their key role in fundamental biological
processes and their biotechnological applications. Here, the atten-
tion was focused on a novel ␣-RHA activity from Novosphingobium
sp. PP1Y, which resulted interesting for its alkaline pH optimum
and moderate tolerance to organic solvents.
This ␣-RHA was used, even without any further purification of
PP1Y crude protein extract, for the bioconversion of flavonoids use-
ful for the food and pharmaceutical industries. However, a future
detailed biochemical characterization of this enzyme is now cru-
cial for a better understanding of its effective biotechnological
potential.
[22] H. Birgisson, G.O. Hreggvidsson, O.H. Fridjónsson, A. Mort, J.K. Kristjánsson, B.
Mattiasson, Enzyme Microb. Technol. 34 (2004) 561–571.
[23] Z. Fujimoto, A. Jackson, M. Michikawa, T. Maehara, M. Momma, B. Henrissat,
H.J. Gilbert, S. Kaneko, J. Biol. Chem. 288 (2013) 12376–12385.
[24] Z. Cui, Y. Maruyama, B. Mikami, W. Hashimoto, K. Murata, J. Mol. Biol. 374
(2007) 384–398.
[25] J.B. Bonanno, S.C. Almo, A. Bresnick, M.R. Chance, A. Fiser, S. Swaminathan, J.
Jiang, F.W. Studier, L. Shapiro, C.D. Lima, T.M. Gaasterland, A. Sali, K. Bain, I.
Feil, X. Gao, D. Lorimer, A. Ramos, J.M. Sauder, S.R. Wasserman, S. Emtage, K.L.
D’Amico, S.K. Burley, J. Struct. Funct. Genomics 6 (2005) 225–232.
[26] E. Notomista, F. Pennacchio, V. Cafaro, G. Smaldone, V. Izzo, L. Troncone, M.
Varcamonti, A. Di Donato, Microb. Ecol. 61 (2011) 582–594.
[27] V. D’Argenio, M. Petrillo, P. Cantiello, B. Naso, L. Cozzuto, E. Notomista, G.
Paolella, A. Di Donato, F. Salvatore, J. Bacteriol. 193 (2011) 4296.
[28] M.M. Bradford, Anal. Biochem. 72 (1976) 248–254.
[29] J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning. A Laboratory Manual,
2nd ed., Cold Spring Harbor Laboratory Press, New York, 1989.
[30] V. Yadav, P.K. Yadav, S. Yadav, K.D.S. Yadav, Process Biochem. 45 (2010)
1226–1235.
Acknowledgments
We thank D. Melck, E. Castelluccio and A. Esposito (NMR service
of Istituto di Chimica Biomolecolare, CNR, Pozzuoli, Italy) for their
skilful assistance.
References
[31] L. Chebil, C. Humeau, J. Anthoni, F. Dehez, J.-M. Engasser, M. Ghoul, J. Chem.
Eng. Data 52 (2007) 1552–1556.
[1] A. Schmid, J.S. Dordick, B. Hauer, A. Kiener, M. Wubbolts, B. Witholt, Nature 409
(2001) 258–268.
[32] G. Celiz, M. Daz, M.C. Audisio, J. Appl. Microbiol. 111 (2011) 731–738.
[33] T. Kaul, E. Middleton, P. Ogra, J. Med. Virol. 15 (1985) 71–79.
[34] N.H. Tung, J.-H. Son, K. Cho, J.-A. Kim, J.-H. Hyun, H.-K. Kang, G.Y. Song, C.J. Park,
Y.H. Kim, Food Sci. Biotechnol. 19 (1) (2010) 271–274.
[35] L. Weignerova, P. Marhol, D. Gerstorferova, V. Kren, Bioresource Technol. 115
(2012) 222–227.
[36] D. Monti, A. Pisvejcová, V. Kren, M. Lama, S. Riva, Biotechnol. Bioeng. 87 (6)
(2004) 763–771.
[37] V.V. Kumar, Afr. J. Biotechnol. 9 (45) (2010) 7683–7686.
[2] U.T. Bornscheuer, G.W. Huisman, R.J. Kazlauskas, S. Lutz, J.C. Moore, K. Robins,
Nature 485 (2012) 185–194.
[3] P. Manzanares, S. Vallés, D. Ramón, M. Orejas, in: J. Polaina, A.P. MacCabe (Eds.),
Industrial Enzymes, Springer, The Netherlands, 2007, pp. 117–140.
[4] K. Habelt, F. Pittner, Anal. Biochem. 134 (1983) 393–397.
[5] M. Roitner, T. Schalkhammer, F. Pittner, Monatsh. Chem. 115 (1984)
1255–1267.
[6] N.L. Rojas, C.E. Voget, R.A. Hours, S.F. Cavalitto, J. Ind. Microbiol. Biotechnol. 38
(9) (2011) 1515–1522.