molecules

Article
Production of Pyracantha Polysaccharide-Iron(III) Complex and
Its Biologic Activity
Wan-Fen Li † , Hao-Hai Ma † , Shuai Yuan and Xi-Feng Zhang *

College of Veterinary Medicine, Qingdao Agricultural University, Qingdao 266109, China;

liwfqd1981@[Link] (W.-F.L.); mahaohai741@[Link] (H.-H.M.); arbeit19960413@[Link] (S.Y.)
* Correspondence: zhangxf106@[Link]
† These authors contributed equally to this work.

Abstract: In this study, the optimum synthetic process of the Pyracantha polysaccharide-iron (PPI)
complex was studied via response surface methodology (RSM). Its antioxidant and anti-cancer
activities were also investigated. It was demonstrated that the optimal conditions for the synthetic
process of the complex were as follows: a pH of 8.9, a reaction temperature of 70 ◦ C and a trisodium
citrate:polysaccharide ratio of 1:2. PPI were analysis by UV, FTIR, SEM, CD, XRD, TGA and NMR.
PPI was able to scavenge the metal ion, ABTS and free radicals of the superoxide anion, demonstrating
its potential antioxidant activity. PPI was found to display cytotoxicity to Skov3 cells, as shown by its
ability to induce apoptosis and alter gene expression in Skov3 cells. These findings show than PPI
may represent a novel antioxidant and chemotherapeutic drug.

Keywords: polysaccharides; polysaccharide iron; response surface optimisation; antioxidant







Citation: Li, W.-F.; Ma, H.-H.; Yuan, 1. Introduction

S.; Zhang, X.-F. Production of
Polysaccharides are ubiquitous biological macromolecules that are involved in myriad
Pyracantha Polysaccharide-Iron(III)
physiological human processes. These molecules have been documented to enhance the
Complex and Its Biologic Activity.
body’s immunity, prevent tumours, and possess antiviral properties. Modern day formu-
Molecules 2021, 26, 1949. https://
lations of polysaccharides in combination with metals and non-metals have been shown
[Link]/10.3390/molecules26071949
to possess good functional qualities [1]. Iron is an essential trace element in the human
body and participates in several enzymatic reactions in the body, with its most well-known
Academic Editor: M. Gilles Mailhot
role being oxygen transportation as well as the maintenance of cellular metabolism [2].
Received: 4 March 2021
A polysaccharide iron complex can be used as a supplement that has very stable properties.
Accepted: 23 March 2021 Valent iron complexes that enter the body produce free radicals, which contribute to cell
Published: 30 March 2021 membrane damage [3]. Oral consumption of iron complexes also leads to a host of side
effects, such as nausea and vomiting [4]. Polysaccharide iron complexes are usually com-
Publisher’s Note: MDPI stays neutral plexes of ferric ions [5,6]. These compounds have a relatively high iron concentration and
with regard to jurisdictional claims in can be soluble and non-toxic at physiological pH values, making them an efficacious iron
published maps and institutional affil- supplement in the body [7].
iations. Many studies have shown that the polysaccharide iron complex possesses high sta-
bility, water solubility and a good absorption rate [8–10]. It also has a safer adverse event
profile compared to ferrous sulphate. Given these excellent properties, the polysaccha-
ride iron complex has gained recognition as a potential treatment for anaemia. Chemical
Copyright: © 2021 by the authors.
synthesis and simulated biomineralisation are the main methods for the preparation of
Licensee MDPI, Basel, Switzerland. polysaccharide iron complexes, with the former method being more common. The polysac-
This article is an open access article charide iron complex is a chelation consisting of polysaccharides and iron that does not
distributed under the terms and cause gastrointestinal discomfort, as the ferric iron is bound and not in its free state. In the
conditions of the Creative Commons body, ferric iron is absorbed and reduced to ferrous ion. Polysaccharide chelated iron exists
Attribution (CC BY) license (https:// mostly in a ferric state. The polysaccharide is primarily extracted from plants and has
[Link]/licenses/by/ immunomodulatory effects. The Pyracantha fortuneana fruit is a high-quality medicinal and
4.0/).

Molecules 2021, 26, 1949. [Link] [Link]

Molecules2021,
Molecules 2021,26,
26,1949
x 22 of
of 16
17

is a high-quality
edible natural plant medicinal
resourceand thatedible natural
contains plant
several resource
active that contains
ingredients. several
It is rich active
in carbohy-
ingredients. It is rich in carbohydrates and the total soluble sugar content
drates and the total soluble sugar content of the flesh of the fruit has been reported to be of the flesh of
the fruit has been reported to be between 10.59–13.40% [11]. Polysaccharides
between 10.59–13.40% [11]. Polysaccharides derived from the Pyracantha fortuneana fruit derived from
thepromising
are Pyracanthacompounds
fortuneana fruit to beare promising
used compounds
in the formulation of to be used in theiron
polysaccharide formulation
complexes. of
polysaccharide iron complexes. We have completed the extraction
We have completed the extraction and structural characterization of polysaccharides of and structural
characterization
Pyracantha fortuneanaof polysaccharides
(PSPF) [12]. of Pyracantha fortuneana(PSPF) [12].
Inthis
In thisstudy,
study, response
response surface
surface methodology
methodology (RSM) was(RSM)usedwas usedthetopreparation
to study study the
preparation
of PPI, and the of antioxidant
PPI, and theeffect antioxidant effect of the
of the Pyracantha Pyracantha polysaccharide-iron
polysaccharide-iron(III) complex was (III)
complex was also studied. Human ovarian carcinoma cells (Skov3) was
also studied. Human ovarian carcinoma cells (Skov3) was used as a model cell to investigate used as a model
cellanti-cancer
the to investigate the anti-cancer
activity activity of this compound.
of this compound.

2.2. Results
Results and
and Discussion
Discussion
2.1.
2.1. EstablishmentofofaaStandard
Establishment StandardCurve
Curvefor
forIron
Iron
Based
Basedononthethe
absorbance of different
absorbance concentration
of different gradients
concentration at 510 nm,
gradients at the
510regression
nm, the
equation is given as the following equation: y = 0.0982x + 0.0041 (R2 = 0.999). There was a
regression equation is given as the following equation: y = 0.0982x + 0.0041 (R2 = 0.999).
good
Therelinear
was arelationship
good linearfrom 0.4–2.8 µg/mL.
relationship from 0.4–2.8 μg/mL.
2.2. Response Surface Optimization of PPI Complexsynthesis Conditions
2.2. Response Surface Optimization of PPI Complexsynthesis Conditions
Response surfaces were plotted using Design Expert software [13]. As shown in
Response surfaces were plotted using Design Expert software [13]. As shown in
Figure 1, response surface plots and the respective contour plots demonstrated the effects
Figure 1, response
of two factors surface
and were plots
used and thethe
to obtain respective
optimumcontour plotsfor
conditions demonstrated
the productiontheof
effects
PPI.
of two
The factors and
interactions were used
amongst to obtain
the factors theresponse
in the optimum conditions
surface can befor the production
intuitively shown byof PPI.
the
The interactions
degree of contour amongst
density inthe
thefactors
contourinmap
the response surface can
and the steepness beresponse
of the intuitively shown
surface mapby
the degree of contour density in the contour map and the steepness of the response
in Figure 1. Figure 1A,B represents the effects of different pH (X1 ) and different extraction surface
map in Figure
temperatures (X1. Figure 1 A,B represents the effects of different pH (X1) and different
2 ) on complex synthesis at a given amount of catalyst. It shows that the
3-D plot and the contour (X
extraction temperatures 2) on complex synthesis at a given amount of catalyst. It shows
plot described the effect of pH surface as relatively steep with a
that the
more 3-D plot
intensive and the contour plot described the effect of pH surface as relatively steep
contour.
with a more intensive contour.

(A) (B)

(C) (D)
Figure 1. Cont.
Molecules 2021,26,
Molecules2021, 26,1949
x 33 of
of 16
17

(E) (F)
Figure1.1. Response
Figure Response surface
surface plots
plots showing
showing the
the effects
effects of
of variables
variables on
on the
the yield
yield of
of PPI.
PPI. (A,B)
(A,B) Plots
Plots of
of the
the effect
effect of
of pH,
pH,
temperatureand
temperature andtheir
theirreciprocal
reciprocalinteractions
interactionsononthe
theiron
ironcontent
content(%).
(%).(C,D)
(C,D)Plots
Plotsof ofthe
theeffect
effectof
ofpH,
pH,ratio
ratioofoftrisodium
trisodium
citrateto
citrate topolysaccharide
polysaccharideandandtheir
theirreciprocal
reciprocalinteractions
interactionsononthe
theiron
ironcontent
content(%).
(%).(E,F)
(E,F)Plots
Plotsofofthe
theeffect
effectofoftemperature,
temperature,
ratio of trisodium citrate to polysaccharide and their reciprocal interactions on the iron content
ratio of trisodium citrate to polysaccharide and their reciprocal interactions on the iron content (%).
(%).

Comparatively,the
Comparatively, thecontour
contourplotplotfor
foreffect
effectofofreaction
reactiontemperature
temperatureof ofthe
thesurface
surfacewaswas
smooth with
smooth withrelatively
relativelysparse
sparsecontours,
contours,indicating
indicating that
thatpH
pHlevels
levelshadhadaamoremoresignificant
significant
effect on
effect onPPI
PPIsynthesis.
synthesis. Figure
Figure 1C,D
1C,D shows
shows the the 3-D
3-D plot
plot and
and thethecontour
contourplot plotat atratio
ratioof
of
catalyst to Pyracantha fortuneana polysaccharides (PSPF) and pH. The
catalyst to Pyracantha fortuneana polysaccharides (PSPF) and pH. The effect of catalyst surfaceeffect of catalyst
issurface is relatively
relatively steep,and
steep, dense dense and possesses
possesses obviousobvious
contourscontours
while the while the contour
contour plot of plot
the
of the
pH pHis effect
effect is relatively
relatively smooth withsmooth
sparsewith sparse contours,
contours, indicatingindicating that theused
that the catalysts catalysts
has
aused
morehas a more significant
significant effect on PPIeffect on PPI synthesis.
synthesis. The 3-D plot Theand 3-Dtheplot and the
contour contour
plot based plot
on
based on independent
independent variable ratiovariable ratio
of catalyst of catalyst to polysaccharides
to polysaccharides and reaction temperatureand reaction are
temperature
shown are shown
in Figure in Figure
1E,F. The contour 1E,F.
plot The
of contour
the effectplot
of of
the the effect of
catalyst the catalyst
surface surface
is relatively
is relatively
steep, steep,
with more with more
intensive intensive
contours, whereascontours,
that ofwhereas
the effectthat of the effect
of reaction of reaction
temperature has
relatively
temperature smooth
has sparse contours,
relatively smooth also indicating
sparse the more
contours, also significant
indicating effect
the more of catalysts on
significant
PPI synthesis.
effect As shown
of catalysts on PPI in Figure 1,As
synthesis. theshown
highest inpoint
Figurein1,the
thegraph
highestis also
point theincentre pointis
the graph
of thethe
also smallest
centre ellipse
point ofinthe
thesmallest
contourellipse
line [14]. As contour
in the shown in linethe[14].
figure
As above,
shown the contour
in the figure
diagram
above, the ofcontour
polysaccharide
diagram iron shows strongiron
of polysaccharide interactions
shows strong amongst the various
interactions amongstfactors.
the
This conclusion is consistent with the ANOVA results in Table 1.
various factors. This conclusion is consistent with the ANOVA results in Table 1.

Table [Link] 1. Analysis

Analysis of variance
of variance of the experimental
of the experimental results results of the
of the BBD BBD
(*** p < (*** p < 0.001).
0.001).

Degree ofSum of Degree of Mean

Source Sum of Squares Source Mean Square F Value
Square F Value pp Value
Value
FreedomSquares Freedom
Model 60.38 Model 9 60.38 6.710 9 6.710 89.09 89.09 <0.0001
<0.0001 ***
***
X1 4.96 X 1 1 4.96 4.961 1 4.961 65.88 65.88 <0.0001 ***
<0.0001 ***
X2 13.65 X 2 1 13.65 13.6501 13.650 181.26 181.26 <0.0001 ***
<0.0001 ***
X3 19.75 X3 1 19.75 19.7501 19.750 262.25 262.25 <0.0001 ***
<0.0001 ***
X1×X2 6.55 X 1 × X 2 1 6.55 6.550 1 6.550 87.02 87.02 <0.0001 ***
<0.0001 ***
X1 × X3 0.06 1 0.058 0.76 0.4108
X1×X3 0.06 1 0.058 0.76 0.4108
X2 × X3 6.79 1 6.790 90.11 <0.0001 ***
X2×X3 6.79 1 6.790 90.11 <0.0001 ***
X1 2 4.98 1 4.980 66.15 <0.0001 ***
X12 4.98 1 4.980 66.15 <0.0001
<0.0003 ***
X2 2 3.19 1 3.190 42.34 ***
X22 3.19 X3 2 1 0.02 3.190 1 0.015 42.34 0.20 <0.0003
0.6685***
X32 0.02 Residual 1 0.53 0.015 7 0.075 0.20 0.6685
Residual 0.53 Lack of Fit 7 0.27 0.0753 0.091 1.43 0.3586
Lack of Fit 0.27 Pure Error 3 0.25 0.0914 0.064 1.43 0.3586
Pure Error 0.25 Cor Total 4 60.91 16
0.064
Cor Total 60.91 R2 = 0.99 16 R2 Adj = 0.98 R2 pred = 0.92 Adeq Precisior = 33.86
R2 = 0.99 R Adj = 0.98
2 R2pred = 0.92 Adeq Precisior = 33.86
Through the response surface software, the optimal reaction conditions for the pro-
Through
duction the response
of PPI were 70 ◦ C pHsurface
8.9 andsoftware, the optimal
ratio of catalyst reaction conditions
to polysaccharide for the
0.5. Under the
production
optimal of PPI were
conditions, 70 °Cable
we were pH to
8.9achieve
and ratio
an of catalyst
iron contentto of
polysaccharide
PPI of 30.76%0.5.
(n =Under the
5), close
optimal conditions, we were able to achieve an iron content of PPI of 30.76% (n = 5), close
Molecules 2021, 26, 1949 4 of 16

to the predicted iron content 30.81%. The model was proven to be valid and can be used as
an established protocol for PPI synthesis.
The ANOVA results are demonstrated in Table 1. The significance analysis of exper-
iment results were obtained using the F-value and p-value [14]. A large F-value and a
small p-value indicate a more significant effect on the response variable [15]. According
to Table 2, the F-value was 89.09, and the p-value of lack of fit was <0. 0001, indicating
that the lack of fit was not significant. Values of ‘Prob. > F’, more than 0.0500 indicate not
significant model terms. In Table 2, the variables with a significant effect on the yield of PPI
were the linear coefficients (X1 , X2 , X3 ), the quadratic term coefficient (X1 2 , X2 2 ), and the
cross product coefficients (X1×2 , X2 X3 ).

Table 2. Box-Behnken experimental design and the results for the Iron content of polysaccharide
iron complex.

Level
Run
X1 X2 X3 Fe (%)
1 0 0 0 26.27
2 1 1 0 27.90
3 0 1 1 23.88
4 0 1 −1 29.86
5 1 0 −1 27.55
6 1 0 1 24.40
7 0 0 0 26.80
8 −1 0 1 23.35
9 0 0 0 26.20
10 0 −1 1 23.84
11 1 −1 0 22.76
12 0 −1 −1 24.61
13 0 0 0 26.2
14 −1 1 0 23.48
15 −1 −1 0 23.46
16 0 0 0 26.32
17 −1 0 −1 26.02

The model fit was checked against the determination coefficient (R2 ) [16]. In this
study, the value of R2 Adj (98.02%) and the R2 pred of 92.18% were in reasonable agreement
with the R2 Adj of 98.02%. This indicates that the fitted quadratic model accounts for more
than 98.02% of the variations in the experimental data. The coefficient of determination
R2 (99.18%) shows that the model has a good fitting degree and can be used to analyse
and predict the absorbance value of PPI. ‘Adeq. Precision’ was used to measure the signal
to noise ratik. An ‘Adeq. Precision’ of 33.858 is high enough to indicate that the model is
predictive regarding experimental results.

2.3. Characterisation of PPI

Figure 2A depicts reactions between PSPF and PPI. The absorption of PPI was in accor-
dance with the previous literature and ranged from 450 nm to 200 nm, suggesting that the
iron core in PPI was Fe-OOH [17]. At the same concentration, compared with indicates that
PSPF reacts with iron to form a PSPF, the absorption intensity of PPI in ultraviolet region is
obviously increased, which complex without free substance [18]. Figure 2B demonstrates
the infrared spectra of PSPF and PPI. The infrared spectra of the two samples are relatively
similar, with both displaying strong absorption peaks near 3500–4000 cm−1 , which is due
to the strong hydrogen bonding of multiple -OH in polysaccharide molecules. They belong
to the absorption peaks of light stretching vibrations in the PSPF. As shown in Figure 2B,
PPI also possessed another peak at 672.47 cm−1 . Taken together, this suggests that the core
of the polysaccharide iron complex was a β-FeOOH nucleus, which is consistent with the
report of Marshall [19].
Molecules 2021, 26, x 5 of 17

Molecules 2021, 26, 1949 5 of 16

this suggests that the core of the polysaccharide iron complex was a β-FeOOH nucleus,
which is consistent with the report of Marshall [19].

(A) (B)

(C) (D)

(E) (F)

Figure 2. Cont.
Molecules 2021,26,
Molecules2021, 26,1949
x 66 ofof1617

(G) (H)
[Link]
Figure AnalysisofofPSPF
PSPFand
andPPI
PPI(A).
(A).UV
UVanalysis
analysisofofPSFP
PSFPand
andPPI.
PPI.(B)
(B)FT-IR
FT-IRspectrum
spectrumofofPSFP
PSFPand
andPPI.
PPI.(C)
(C)SEM
SEMimages
images
of the PPI and PSPF. (D).CD spectra of the PPI and PSPF. (E) X-ray diffraction (XRD) pattern of PSPF and PPI.
of the PPI and PSPF. (D).CD spectra of the PPI and PSPF. (E) X-ray diffraction (XRD) pattern of PSPF and PPI. (F) TG curves (F) TG
curves of PSPF under nitrogen condition.
1 1H NMR spectra, (G) PSPF, (H) PPI.
of PSPF under nitrogen condition. H NMR spectra, (G) PSPF, (H) PPI.

TheSEM
The SEMimagesimagesofofPPI PPIandandPSPF
PSPF at at different
different magnifications
magnifications (2000×
(2000 × and and500500×) are
×) are
illustratedininFigure
illustrated Figure2C. [Link]
surfacesofofPPIPPIand andPSPFPSPFhad hadobvious
obviousvariations
variationsininsize sizeand
and
[Link]
shape. Theimage
imageatataalow lowmagnification
magnificationshows showsthat thatthethesurface
surfacemorphology
morphologyofofPSPF PSPFwaswas
characterisedby
characterised byaasmall
smallsize sizeand
andhomogeneous
homogeneousdispersal. [Link]
surfaceofofPPI PPIpresents
presents
smoothand
smooth andirregular
irregularflakyflakystructure,
structure,and andthe theslightly
slightlyuneven
unevensurface
surfaceofofPSPF PSPFisismoremore
obviousthan
obvious [Link].
CDwas
CD wasanan effective
effective method
method to investigate
to investigate the three-dimensional
the three-dimensional structurestructure
of com- of
pounds.
compounds. A new A new positive
positiveCotton
Cottoneffect appeared
effect appeared at at
215215
nm nm with
withincreasing
increasingintensity
intensity
(Figure
(Figure2D),2D),which
whichmightmightsuggest
suggestthatthatthethestructural
structuralasymmetry
asymmetryininPPI PPIwaswasenhanced,
enhanced,
which
whichmight
mightbe beattributed
attributedtotothe thecharge
chargetransfer
transferinteraction
interactionbetween
betweeniron ironions
ionsand andthe
the
carboxyl
carboxylgroup
groupininpolysaccharide
polysaccharidechains chains[20].[20].
XRD
XRDisisan aneffective
effectivemethod
methodfor foranalysing
analysingthe themicrostructure
microstructureofofcrystalline
crystallinematerials
materials
and
and some non-crystalline materials. X-ray diffraction patterns of PPI andPSPF
some non-crystalline materials. X-ray diffraction patterns of PPI and PSPFare areshown
shown
ininFigure 2E.2E.
TheTheXRDXRD pattern of PPIofpresented strong diffraction peaks at 20–45 ◦
Figure pattern PPI presented strong diffraction peakscompared
at 20–45°
tocompared
PSPF. Theto diffraction peaks of PPI were ◦ , 28.81◦ , 33.12◦ , and 43.24◦ . Additionally,
PSPF. The diffraction peaksatof 20.51
PPI were at 20.51°, 28.81°, 33.12°, and 43.24°.
the XRD resultsthe
Additionally, confirmed
XRD resultsthat aconfirmed
new compound,that a newPPI,compound,
was formedPPI, and was
withformed
a large and
amount
with
ofa iron(III) [21].
large amount of iron (III) [21].
The
Thethermal
thermalgravimetric
gravimetriccurve curveofofthethePPIPPIisispresented
presentedininFigure
Figure2F,2F,which
whichshowsshowsthe the
degradation pattern of PPI. As shown in Figure 2F, there were three
degradation pattern of PPI. As shown in Figure 2F, there were three weight loss events weight loss events for
the ◦ C and
forPPI.
the The
PPI. first
The weight loss of
first weight PPIofoccurred
loss PPI occurred in theintemperature
the temperaturerangerange
of 25–200
of 25–200 °C
the loss ◦ C and the loss rate
and therate
losswasrate18%.
was The
18%.nextTheweight loss ofloss
next weight PPIofoccurred at 200–380
PPI occurred at 200–380 °C and the loss
was ◦ C and the loss rate was
rateupwasto up
20%. to The
20%.third
The weight loss ofloss
third weight PPIofoccurred at 380–650
PPI occurred at 380–650 °C and the loss rate
15%.
was1Therefore, it is relatively
15%. Therefore, [Link].
it is relatively
H
1H NMR
NMR spectroscopy
spectroscopy were were performed
performedfor foranalysis
analysisofof thethe structure
structure of PSPF
of PSPF andandPPI.
PPI. The 1 HNMR spectrum of PSPF (Figure 2G) revealed different linkage patterns at
The 1HNMR spectrum of PSPF (Figure 2G) revealed different linkage patterns at 5.0–4.0
5.0–4.0 ppm. However, the complete signal pattern of PSPF was missing in the spectrum of
ppm. However, the complete signal pattern of PSPF was missing in the spectrum of PPI
PPI (Figure 2H), This might be due to the relatively large, blinded region around high-spin
(Figure 2H), This might be due to the relatively large, blinded region around high-spin (S
(S = 5/2) iron(III) centres, which was the same of the reports of the Bertini group [22]. In the
= 5/2) iron(III) centres, which was the same of the reports of the Bertini group [22]. In the
metal centre, ligand signals were too broad to be detectable due to shortened relaxation
metal centre, ligand signals were too broad to be detectable due to shortened relaxation
times [23].
times [23].
2.4. Antioxidant Activity of PPI
2.4. Antioxidant Activity of PPI
As shown in Figure 3, PPI exhibited apparent antioxidant activity. The antioxidant ac-
AsPPI
tivity of shown in Figure
through TEAC 3,(onPPI exhibited
ABTS apparent
+ radical antioxidant
scavenging activity.
activity) is The
depicted in antioxidant
Figure 3A.
activity
The of PPI through
scavenging activityTEAC (on ABTS
increased with radical
+ scavenging
increasing activity) of
concentrations is depicted
PPI. Thein Figure 3A.
maximum
The scavenging activity increased with increasing concentrations of PPI. The maximum
Molecules 2021,
Molecules 26, x1949
2021, 26, 7 7of
of 17
16
Molecules 2021, 26, x 7 of 17

clearance
clearance
clearancerate was
raterate
waswas up up to 98.5%, when
to 98.5%, the the
when concentration
concentration of PPI reached
reached
of PPI reached 10 mg/mL.
10 mg/mL.
10 mg/mL. Metal
Metal
chelating
chelating activity
activity has
has long
long been
been regarded
regarded as
as a strong antioxidant
chelating activity has long been regarded as a strong antioxidant mechanism given mechanism
mechanism given
given its its
ability
ability to
to reduce
reduce lipid peroxidation in metal catalysts. The
The antioxidant
antioxidant
ability to reduce lipid peroxidation in metal catalysts. The antioxidant activity of PPI activity
activity of
of PPI
through
through
through metalmetalionscavenging
ion scavenging
ion scavenging activity
activity is is shown
shown
activity in Figure
in Figure
is shown 3B. 3B.
3B. PPI
in Figure PPIPPIexhibited
exhibited apparent
apparent
exhibited bind-
apparent
binding
ingbindingcapacity
capacity in a in a concentration-dependent
concentration-dependent [Link].
The The maximum
maximum
capacity in a concentration-dependent manner. The maximum clearance rate clearance
clearance rate rate
was waswas
63%,
63%,
when63%,when
thewhenthe the
concentration
concentration of PPI
concentration of reached
PPI reached
of PPI 10 mg/mL.
10 mg/mL.
reached TheThe
10 mg/mL. scavenging
scavenging
The activities
activities
scavenging of PPI
of PPI
activities onPPI
of
on superoxide
superoxide anion anion
are are presented
presented in in
Figure Figure
3C. The 3C. The superoxide
superoxide
on superoxide anion are presented in Figure 3C. The superoxide radical scavengingradical radical
scavenging scavenging
activities
of PPI were
activities of concentration-dependent
activities PPI
of PPIwerewere when the mass
concentration-dependent
concentration-dependent when concentration
the the
when mass of PPI was 1–8
concentration
mass concentration mg/mL
of PPI waswas
of PPI
and1–8
1–8 themg/mL
mg/mL maximum
andand clearance
the the
maximum rate was
clearance 42.69%.
rate was On the
42.69%. other
On thehand,
other super-oxide-radical
hand,
maximum clearance rate was 42.69%. On the other hand, super-oxide- super-oxide-
scavenging
radical
radical activities
scavenging
scavenging of PPI of
activities were
activities PPI decreased
were
of PPI when when
decreased
were decreased thewhen
mass concentration
the the
mass concentration
mass of PPI
concentration of was
PPI
of PPI
10 mg/mL.
waswas 10 mg/mL. After
10 mg/mL. combining
AfterAfter with
combining
combining iron
with ions,
with PPI
ironiron
ions,enhanced
PPIPPI
ions, the
enhancedABTS
enhanced scavenging
the theABTSABTS activity
scavenging
scavenging
much
activity more
much
activity than
much morePSPF,
morethan but there
PSPF,
than PSPF, wasbut
but nothere
there significant
waswas no nochange
significantin ferrous
change
significant changeion and superoxide
in ferrous
in ferrousion ion
andand
anion scavenging
superoxide anion activity.
scavenging activity.
superoxide anion scavenging activity.

(A) (A) (B) (B) (C) (C)

Figure 3. Antioxidant
Figure Antioxidant activity
activity
3. Antioxidant of
of PPI
activity PPI (A). Scavenging
(A).(A).
of PPI Scavenging effects
effects
Scavenging of PPI
effects on ABTS
of PPI radical.
on ABTS (B). Scavenging
(B).(B).
radical. Scavenging effects
effects
Scavenging of
of PPI
effects PPI on on
of PPI
ferrous
ferrous ion
ferrous scavenging.
ion ion
scavenging. (C). Scavenging
(C).(C).
scavenging. Scavenging
Scavengingeffects
effects of
of PPI
effectsPPI on
on superoxide
of PPIsuperoxide anions.
anions.
on superoxide anions.

2.5. Effect
[Link].
Effect of
of PPI
Effect PPI on
on Cell
of PPI Cell Viability
Viability
on Cell and Cell
andand
Viability Morphology
CellCell
Morphology
Morphology
As
As shown
shown
As shownin
inFigure
Figure
in Figure4A,B,
4A,B, the viability
thethe
4A,B, viability of
viabilityofSkov3
Skov3
of Skov3cells
cells was significantly
waswas
cells significantly decreased
decreased
significantly after
decreasedafter
after
24
24 hh of
of exposure
exposure to
to PPI.
PPI. Cell
Cell viability
viability was
was affected
affected in
in a
a dose-dependent
dose-dependent
24 h of exposure to PPI. Cell viability was affected in a dose-dependent manner, with the manner,
manner, with the
viability
viability of
of Skov3
viability Skov3
of Skov3 cells
cells decreasing
decreasing
cells decreasingsignificantly
significantly
significantlyas
as concentrations
concentrations
as concentrations increased
increased from
from
increased 100 to to
100100
from
400 μg/mL
400400µg/mL PPI
PPI (p
(p <<0.05;
0.05; Figure
Figure 4B).
4B). Compared
Compared with
with PSPF,
PSPF, PPI
PPI
μg/mL PPI (p < 0.05; Figure 4B). Compared with PSPF, PPI has weaker inhibitoryhas
has weaker
weaker inhibitory
inhibitory
activity
activity on
on cancer
activity cancer
on cell
cancer proliferation.
cellcell
proliferation. The
proliferation. viability
TheThe
viability of
of human
viability human
of human ovarian
ovarian cancer
cancer
ovarian cells
cells
cancer decreased
decreased
cells decreased
significantly
significantly
significantlyexposure
exposure
exposure to 100 andand
to 100 200200
μg/mL
µg/mL
μg/mL PSPF
PSPF
PSPF[12].
[12].
[12].

(A) (A) (B) (B)

Figure 4. Cont.
Molecules 26, 1949
2021,2021,
Molecules 26, x 8 of8 16
of 17

(C) (D)
Figure
Figure 4. Effects
4. Effects of PPIofonPPI
cell on cell viability
viability and evaluation
and evaluation of (A).
of ROS level. ROSPhase
[Link]
(A). Phase contrast
microscopy microscopy
showing showing the
the morphology
morphology of Skov3 cells after PPI treatment for 24 h. Scale bars = 200 μm. (B). Data for Skov3 cells. (C).
of Skov3 cells after PPI treatment for 24 h. Scale bars = 200 µm. (B). Data for Skov3 cells. (C). Intracellular ROS levels Intracellular
were
ROS levels were measured with fluorescence imaging using the DCFH-DA probe in cells cultured in
measured with fluorescence imaging using the DCFH-DA probe in cells cultured in the presence of PPI (0, 100, 200 and the presence of PPI
(0, 100, 200 and 400 μg/mL) for 24 h. Scale bars = 100 μm. (D). The average intensity of fluorescence in Skov3 cells. Data
400 µg/mL) for 24 h. Scale bars = 100 µm. (D). The average intensity of fluorescence in Skov3 cells. Data are expressed as
are expressed as mean ± SD of three independent experiments performed in triplicate. **p < 0.01.
mean ± SD of three independent experiments performed in triplicate. ** p < 0.01.

2.6. PPI Increased Levels of Reactive Oxygen Species

2.6. PPI Increased Levels of Reactive Oxygen Species
Studies
Studies havehave shown
shown thatthat
ROSROSareare closely
closely related
related to apoptosis,
to apoptosis, as increasing
as increasing thethe ROS
ROS
level leads to a decrease in the level of cellular antioxidants, resulting in overall
level leads to a decrease in the level of cellular antioxidants, resulting in overall oxidative oxidative
damage
damage to cellular
to cellular components
components [24,25].
[24,25]. ROS ROSwaswas detected
detected using
using a fluorescent
a fluorescent DCFH-DA
DCFH-DA
probe.
probe. As As shown
shown in Figure
in Figure 4C,D,
4C,D, thethe results
results indicate
indicate thatthat
the the intercellular
intercellular ROSROS induced
induced
by PPI in Skov3 cells was upregulated in a dose-dependent manner. Simultaneously,the
by PPI in Skov3 cells was upregulated in a dose-dependent manner. Simultaneously,
theresults
resultsshow
showthat
thatPPI
PPIwas
wasresponsible
responsibleforfor cell
cell death
death and
and enhanced
enhanced ROS ROS production.
production.

[Link].
LossLoss of Mitochondrial
of Mitochondrial Membrane
Membrane Potential
Potential (∆ψm)
(∆ψm) andand Apoptosis
Apoptosis Induction
Induction
TheThe decline in ∆Ψm
decline in ∆Ψm is early
is an an early marker
marker of apoptosis.
of apoptosis. Cells
Cells havehave
beenbeen shown
shown to enter
to enter
an irreversible apoptosis process if ∆Ψm is lost. JC-1, the ideal fluorescent probe for ∆Ψm
an irreversible apoptosis process if ∆Ψm is lost. JC-1, the ideal fluorescent probe for ∆Ψm
detection,
detection, cancan easily
easily be be detected
detected by by
thethe transition
transition fromfrom
redred to green
to green fluorescence
fluorescence [26,27].
[26,27].
TheThe intensity
intensity of of JC-1
JC-1 was
was significantlydecreased
significantly decreased(p(p<< 0.01)
0.01) in the groups
groups treated
treatedwith
withPPI
PPIininaa dose-dependent
dose-dependent manner manner (Figure 5A,B).5A,B). DNADNAdegradation
degradationininthe theearly stage
early of of
stage
apoptosis
apoptosis waswasevaluated
evaluatedusing thethe
using TUNEL
TUNEL assay [28].
assay Compared
[28]. Compared with thethe
with control group,
control group,
TUNEL-positive
TUNEL-positive cellscells
and fluorescence
and fluorescenceintensity were high
intensity were in high
the groups
in thetreated
groupswith higher
treated with
concentrations of PPI (Figure 5C,D), indicating more pronounced
higher concentrations of PPI (Figure 5C,D), indicating more pronounced [Link]. Furthermore,
theFurthermore,
expression of the proteins relatedof
expression toproteins
apoptosisrelated
were detected
to apoptosisby Western blot analysis
were detected (see
by Western
Section
blot 2.9). The expression
analysis (see Section of Bax
2.9).protein was significantly
The expression of Baxup-regulated
protein waswhile Bcl-2 wasup-
significantly
significantly
regulated down-regulated
while Bcl-2 wasin significantly
PPI-treated Skov3 cells. Additionally,
down-regulated the expression
in PPI-treated Skov3 of cells.
γ-H2AX and RAD51,
Additionally, the which were detected
expression by immunofluorescence,
of γ-H2AX and RAD51, which were also
weresignificantly
detected by
up-regulated (Figure 6). were
immunofluorescence, Compared with PSPF, up-regulated
also significantly PPI has weaker ability
(Figure 6).toCompared
induce cell with
apoptosis [12].
PSPF, PPI has weaker ability to induce cell apoptosis [12].
Molecules 2021, 26, x 9 of 17
Molecules 2021, 26,
Molecules 2021, 26, 1949
x 99 of
of 16
17

(A) (B)
(A) (B)

(C) (D)
(C) (D)
Figure 5. Loss
Figure of mitochondrial
5. Loss of mitochondrialmembrane
membranepotential (∆ψm)
potential and TUNEL
(∆ψm) [Link].
and TUNEL (A). The The ∆ψm
(A).∆ψm was was
Figure
evaluated 5. Loss
using of
JC-1 mitochondrial
in treated cells. membrane
Red potential
fluorescence (∆ψm)
indicates and
JC-1 TUNEL assay.
aggregates (A).
within
evaluated using JC-1 in treated cells. Red fluorescence indicates JC-1 aggregates within mitochondria The ∆ψm was
evaluated
mitochondria using JC-1 in treated cells. Red fluorescence indicates JC-1 aggregates within
in healthyincells,
healthy cells, whereas
whereas green fluorescence
green fluorescence indicatesindicates JC-1 monomers
JC-1 monomers in the
in the cytoplasm and loss
mitochondria
cytoplasm and lossinofhealthy cells,bars
∆ψm. Scale whereas
= 100green fluorescence
μm. (B). indicates
Ratio of JC-1 JC-1 monomers
monomers in the
to JC-1 aggregates.
of ∆ψm. Scale bars = 100 µm. (B). Ratio of JC-1 monomers to JC-1 aggregates. (C). TUNEL assay
cytoplasm
(C). TUNEL andinloss
assay of ∆ψm.
treated cells,Scale
scalebars
bars= =100
100μm.
μm.(B).
(D).Ratio of JC-1
Average monomers
intensity to JC-1 aggregates.
of TUNEL
in treated
(C). TUNEL cells, scale
assay in bars = 100
treated µm.
cells, (D).
scale Average
bars = 100 intensity
μm. (D). of TUNEL
Average fluorescence
intensity of TUNELin Skov3 cells.
fluorescence in Skov3 cells. * p <0.05, ** p < 0.01.
p < 0.05, ** pin< Skov3
*fluorescence 0.01. cells. * p <0.05, ** p < 0.01.

Figure 6. Western blotting. Bax, Bcl2, RAD51, γH2AX and p53 expression in Skov3 cells. Cells
Figure
wereFigure [Link]
treated6. Western blotting.
PPI at
Western Bax, for
400 μg/mL
blotting. Bax, Bcl2, RAD51,
24RAD51,
Bcl2, h. γH2AX and
γH2AX and p53
p53 expression
expressionin
inSkov3
Skov3cells.
[Link]
Cellswere
were treated with PPI at 400 μg/mL
treated with PPI at 400 µg/mL for 24 h. for 24 h.
2.8. PPI Induces Nuclear DNA Breakage
2.8. PPI Induces Nuclear DNA Breakage
Apoptosis is always accompanied by DNA damage. γ-H2AX and Rad51, both
markers of Apoptosis
DNA damage,isisalways
playaccompanied
always major roles inbythe
accompanied DNA
by DNAdamage.
repair damage.
process γ-H2AX
γ-H2AX
after DNA anddamage
Rad51, both mark-
and Rad51,
[29,30].both
ers
Levels of
markers DNA damage,
of DNA damage,
of phosphorylated play major
play in
γ-H2AX roles
major in
cellsroles the repair
in thewith
exposed process
repair
PPIprocess after
for 24 hafter DNA
wereDNA damage
damage
measured [29,30].
using
Levels of
of phosphorylated
phosphorylated γ-H2AX
γ-H2AX inin cells
cells exposed
exposed with
with PPI PPI
for
immunofluorescent images (Figure 7A). Green fluorescence, which represents γ-H2AX, was for
24 h 24
wereh were measured
measured using
using immunofluorescent images 7A).
(Figure 7A). Green fluorescence, which represents γ-
remarkably enhanced in PPI-treated samples in a dose-dependent manner (Figure 7B, p < was
immunofluorescent images (Figure Green fluorescence, which represents γ-H2AX,
H2AX, was enhanced
remarkably PPI-treated
enhanced insamples
PPI-treated samples in a dose-dependent manner
0.01).remarkably
The presence of DNAin damage was further in a dose-dependent
confirmed by RAD51 stainingmanner (Figure
(Figure 7B, p <
7C,D).
(Figure
0.01). The p < 0.01).
7B,presence of The
DNA presence
damage ofwas
DNA damage
further was further
confirmed by confirmed
RAD51 by(Figure
staining RAD517C,D).
stain-
We further analysed their expression using WB. The expression levels of γ-H2AX and
ing (Figure
We further 7C,D). We further analysed their expression using WB. The expression levels
RAD51 all wereanalysed
up-regulatedtheir in
expression using WB.
the 400 μg/mL PPI The expression
treatment grouplevels of γ-H2AX
compared to the and
of γ-H2AX
RAD51 all and RAD51
were all were
up-regulated in up-regulated
the 400 μg/mL inPPI
the treatment
400 µg/mL PPI treatment
group compared group
to the
control group (Figure 6).
compared to the control
control group (Figure 6). group (Figure 6).
Molecules 2021, 26, x 10 of 17
Molecules 2021, 26, 1949 10 of 16

(A) (B)

(C) (D)
Figure
Figure7. 7.
Nuclear
Nuclear DNADNAdamage in in
damage Skov3
Skov3cells after
cells CPPI
after CPPItreatment
treatment using
usingimmunocytofluorescence
immunocytofluorescence with γ-H2AX
with γ-H2AX and
and
Rad51 antibody. (A). Immunocytofluorescence with γ-H2AX in treated cells. (B). Average intensity of green
Rad51 antibody. (A). Immunocytofluorescence with γ-H2AX in treated cells. (B). Average intensity of green fluorescence fluorescence
in Skov3
in Skov3cells. (C) Immunocytofluorescence
cells. (C) Immunocytofluorescencewith with
Rad51Rad51
in treated cells. (D).
in treated Average
cells. intensityintensity
(D). Average of red fluorescence in Skov3 in
of red fluorescence
cells. Scale bars = 100 μm.* p < 0.05, ** p < 0.01.
Skov3 cells. Scale bars = 100 µm. * p < 0.05, ** p < 0.01.

[Link].
PPIPPI
Affected Gene
Affected Expression
Gene in Skov3
Expression in Skov3 Cells
Cells
In In
order to reveal the influence of
order to reveal the influence of PPI PPI exposure
exposure onongene expression
gene expression in in
Skov3
Skov3cell,
cell,
Skov3 cell mRNA was examined by high-throughput sequencing.
Skov3 cell mRNA was examined by high-throughput sequencing. As shown in Figure As shown in Figure 8A,8A,
a total
a totalof of
29,078
29,078genes
geneswere
were detected.
detected. There
Therewere
were 873873
genes
geneswhich
which were
weresignificantly
significantly
different
differentbetween
between control
controland
andPPI-treated
PPI-treatedgroups.
groups. A A total
total of 455 genes
of 455 geneswereweresignificantly
significantlyup-
up-regulated
regulated and and418418
werewere down-regulated
down-regulated afterafter PPI treatment.
PPI treatment. The scatter
The scatter plot is
plot is presented
presented as a log2 of fold change of signal intensity and blue spots
as a log2 of fold change of signal intensity and blue spots represent genes expressed represent genes at
expressed
similar levels; the up-regulated genes in the PPI-treated group are marked in red, andinthe
at similar levels; the up-regulated genes in the PPI-treated group are marked
red, and the down-regulated
down-regulated genes are
genes are marked marked
in green. Theinhierarchical
green. The clustering
hierarchical clustering
analyses of the
analyses of the 863 differentially-expressed
863 differentially-expressed genes (DEGs) isgenesshown(DEGs) is shown
in Figure 8B. Sterolin biosynthetic
Figure 8B. Sterol
process,
biosynthetic
secondaryprocess,
alcohol secondary
biosynthesisalcohol biosynthesis
process and sterol process and sterol
metabolic metabolic
process were theprocess
mainly
were the mainly
enriched DEGsenriched DEGs
in BP; the DEGs in in
BP;
CC the DEGs
were in CCtowere
related related to the90s
the preribosome, preribosome,
preribosome
90sandpreribosome
membrane and membrane
region, and snoRNA region, and and
binding snoRNA binding
coenzyme and were
binding coenzyme binding
significant DEGs
were
in MFsignificant
betweenDEGs in MFand
the control between the control
PPI-treated groupand PPI-treated
(Figure group (Figure
8C,D). Through KEGG 8C,D).
analysis,
steroid biosynthesis,
Through KEGG analysis, carbon metabolism
steroid and terpenoid
biosynthesis, carbon backbone
metabolism biosynthesis were the
and terpenoid
most significantly
backbone biosynthesis altered signalling
were the pathways between
most significantly the controlpathways
altered signalling and treated groups
between
the(Figure
control8E,F). PPI alsogroups
and treated causes (Figure
cell ferroptosis,
8E,F). PPIa new
also cell death
causes cellmode discovered
ferroptosis, in recent
a new cell
years
death (Figure
mode 8E,F). in recent years (Figure 8E,F).
discovered
 Molecules 2021, 26, x 11 of 17
Molecules 2021, 26, 1949 11 of 16

(A) (B)

(C) (D)

(E) (F)

Figure Gene expression

Figure8.8. Gene expression ininPPI-treated
PPI-treatedcells. Scatter
cells. plotplot
Scatter of PPJ-treated Skov3
of PPJ-treated cell gene
Skov3 expression
cell gene profiling.
expression (A) The
profiling. (A)455
The
up-regulated genes and 418 down-regulated genes in the treated group are plotted in green and red, respectively. (B)
455 up-regulated genes and 418 down-regulated genes in the treated group are plotted in green and red, respectively.
Hierarchical clustering analyses of the differentially expressed genes in two replicates of the two groups. (C) Dot plots of
(B) Hierarchical clustering analyses of the differentially expressed genes in two replicates of the two groups. (C) Dot plots of
overrepresented gene ontology (GO) terms in molecular function (MF), biological process (BP) and cellular component
overrepresented
(CC) [Link]
(D) ontology
Bar plots(GO) terms in molecular
of overrepresented GOfunction (MF),
terms in MF,biological
BP and process (BP) and(E)
CC categories. cellular component
Dot plots (CC)
of Kyoto
categories. (D) Bar plots of overrepresented GO terms in MF, BP and CC categories. (E) Dot plots of Kyoto Encyclopaedia of
Genes and Genomes (KEGG) terms. (F) Bar plots of KEGG terms. Top ten significant terms are shown in each category
(p < 0.05).
Molecules 2021, 26, 1949 12 of 16

3. Conclusions
In this study, PPI was synthesised and studied in detail. The synthesis of PPI was
successfully optimised using RSM. The optimal conditions that achieved maximum com-
plex yield were a reaction temperature of 70 ◦ C, pH 8.9 and a catalyst to polysaccharide
ratio of 0.5. Under the optimal conditions, the PPI produced was a reddish-brown powder
with an iron content of PPIC that reached 30.76%, which suggests that the model was
satisfactory and accurate. In addition, the antioxidant study found that PPIs have a clear
role in scavenging superoxide radicals, metal ions and ABTS radicals. The results indicate
that PPI had antioxidant activities in the in vitro assays. Additionally, we also verified that
PPI could induce significant cytotoxicity, as evidenced by an increase in intracellular ROS,
mitochondrial membrane potential disruption and consequent DNA damage. Moreover,
PPI exposure significantly altered the cancer-related gene expression pattern of Skov3 cells
in vitro, which suggests that PPI could be a potential drug for anticancer therapy.

4. Materials and Methods

4.1. Drugs and Reagents
Pyracantha fortuneana polysaccharide (PSPF) were extracted and purified in our labo-
ratory [12], sodium citrate tribasic, 2,20-azino-bis (3-ethylbenzthiazoline-6-sulfonic) acid
(ABTS) was provided by Sigma (St. Louis, MO, USA). All other chemicals and reagents
were purchased locally and were of analytical grade.

4.2. Preparation of Polysaccharide

Pyracantha fortuneana polysaccharide (PSPF) was extracted with water and alcohol
precipitation under the following conditions: extraction time of 2.1 h, 61.5 ◦ C, ratio of water
to raw material = 36.3 [12]. D101 macroporous adsorption resin was used for pigment
removal with protein removed by the Sevage method. The resultant polysaccharide was
collected, concentrated, dialysed and lyophilised for purification.

4.3. Single Factor Experimental Data Analysis

Based on a literature review of the synthesis process of polysaccharide iron, the main
factors affecting polysaccharide iron are as follows: citric acid, the ratio of catalyst to
polysaccharide, temperature, pH and reaction time [31]. In this study, pH, tempera-
ture and the ratio of catalyst to polysaccharide were selected as three independent vari-
ables. As shown in Table 3, pH = 8, temperature = 60 ◦ C and the ratio of catalyst to
polysaccharide = 1.25 were each at their optimal values.

Table 3. Levels and code of extraction variables used in Box–Behnken design.

Symbols Coded Levels

Variable
Coded −1 0 1
pH X1 7 8 9
Reaction temperature X2 50 60 70
Sodium Citrate
X3 0.5 1.25 2
tribasic/Polysaccharide ratio

4.4. Synthesis of Pyracantha Polysaccharide-Iron(III) Complex (PPI)

PPI was synthesised chemically. Other methods of PPI synthesis that have previously
been documented include the ammonium sulphate method and the composite membrane
method; Tang and Liu have prepared the polysaccharide iron complex using a chemical
synthesis method [32].

4.5. Response Surface Design

In this experiment, the iron content of PPI was used as an evaluation index. The PPI at
points based on the experimental design are shown in Table 2. The whole design consisted
Molecules 2021, 26, 1949 13 of 16

of 17 experimental points carried out at random. The best fitting model was the result of
RSM as determined via regression using Design Expert software (v.[Link] trial, Stat-Ease
Inc., Minneapolis, MN, USA). At the bottom of Table 1, the fitted quadratic polynomial
equation is given as follows: Y = 0.52 + 0.0.015X1 + 0.026X2 − −0.031X3 + 0.025X1 X2 −
2.35*10−3 X1 X3 − 0.026X2 X3 − 0.022X1 2 − 0.017X2 2 + 9.1*10−4 X3 2 .

4.6. Determination of PPI

4.6.1. Establishment of Standard Curve for Iron
The two valent iron standard solution configuration is as follows: a standard iron
solution was prepared by dissolving 0.7025 g of (NH4 )Fe(SO4 )2 ·H2 O in 1000 mL of distilled
water with the addition of 2 mL of hydrochloric acid iron. Then, 0, 1, 2, 3, 4, 5, 6 and 7 mL of
iron standard solution was added to 1.75 mL of ascorbic acid and 2.5 mL of phenanthroline.
After 10 min, the UV absorbance of sample was assessed at 510 nm. Finally, the corresponding
standard curve was obtained. The calibration curve was y = 0.0982x + 0.0041, R2 = 0.9999.

4.6.2. Determination of the Iron Content in PPI

The iron content was measured using the phenanthroline colorimetric method [33].
For this, 0.05 g of PPI was dissolved in 50 mL of water, then 1.0 mL of the sample was mixed
with 2.5 mL of 10% ascorbic acid solution and 5 mL of 0.1% o-phenanthroline solution.
Finally, the solution was adjusted to 50 mL with water. After 3 h, the UV absorbance
of the sample was assessed at 510 nm. The iron content of PPI was calculated using the
calibration curve.

4.6.3. Experimental Design

RSM is a mathematical and statistical technique to optimise the available parameters
through the least number of experiments and to analyse the interaction between these
parameters [34,35]. In this study, RSM was adopted to optimise the process conditions for
PPI production. The three-level Box–Behnken design (BBD) was used to evaluate three
independent variables on the basis of single factor experiments: pH (X1 ), temperature (X2 )
and the ratio of catalyst to PSPF (X3 ). The iron content of PPI was selected as the response.
The variables and their respective levels are presented in Table 1.
ANOVA was used for the statistical analysis. p-values less than 0.05 were considered
statistically significant. ANOVA, regression analysis and the response surface rendering
method were used to fit the quadratic polynomial equation of all response variables in
order to obtain the optimum conditions for complex formation.

4.6.4. Antioxidant Activities of PPI

The antioxidant activities of PPI were assessed based on previously established meth-
ods [36–39]. For this, 1, 2, 4, 8 and 10 mg/mL PPI was prepared in water. The scavenging
ability for pyrogallol autoxidation was calculated: inhibition rate (%) = (1 − (A1 − A2 )/A0 )
× 100%, where A0 is the absorbance of the control (water instead of sample), A1 is the
absorbance of the sample and A2 is the absorbance of the sample with anhydrous ethanol
instead of pyrogallol.

4.7. Characterisation of PPI

4.7.1. UV and FTIR Analysis
PSPF and PPI powder were dissolved in 50 mL of water. The absorbance was measured
from 200 to 450 nm using an ultraviolet spectrum scan. Absorptions of PSPF and PPI
were identified by the FTIR spectrum. Approximately 2 mg of PSPF and PPI powder were
weighed and mixed with KBr powder, ground and pressed for FTIR measurements through
spectrometry (PerkinElmer, Spectrum 400, Waltham, MA, USA).
Molecules 2021, 26, 1949 14 of 16

4.7.2. SEM Analysis

The surface morphology of PPI and PSPF were analysed by scanning electron mi-
croscopy (Hitachi, Tokyo, Japan). With the help of double-sided tape, the sample was placed
on a brass stub and observed after coating with gold in a vacuum by a sputter coater.

4.7.3. Circular Dichroism (CD) Analysis

The CD spectra of PSPF and PPI solutions (1.0 mg/mL) were measured on a Chirascan
V100 CD (Applied Photophysics, Leatherhead, UK) spectropolarimeter at 25 ◦ C. Each CD
spectrum was the accumulation of three scans at 100 nm/min with a 1 nm slit width and a
time constant of 1 s. Data were collected from 185 to 300 nm at 1 nm intervals.

4.7.4. Structural Characteristics of PSPF and PPI

X-ray diffraction (XRD) analysis was performed with an X-ray diffractometer (D8ADVANCE,
Bruker, Karlsruhe, Germany). The X-ray diffractometer was operated at 40 kV and 30 mA
produced with a Cu-κα radiation. The XRD patterns were recorded over 5–90◦ (θ being
the angle of diffraction) at a rate of 4◦ /min.

4.7.5. Thermogravimetric Analysis (TGA)

A simultaneous thermal analyser (STA449F3, Netzsch Corporation, Selb, Germany)
was applied to determine the thermal stability of PPI. Approximately 10 mg of PPI powder
was placed in a platinum crucible and heated from 30 to 650 ◦ C at a rate of 10 ◦ C/min in a
nitrogen atmosphere with a flow rate of 50 mL/min.

4.7.6. Nuclear Magnetic Resonance (NMR) Analysis

For NMR measurements, PPI and PSPF powder were completely dissolved in 0.5 mL
of dimethyl-d6 sulfoxide (D-DMSO) and transferred into an NMR tube. The 1 H spectra
experiments were recorded at room temperature at 500 MHz on a spectrometer (AVIII500,
Bruker, Fällanden, Switzerland).

4.7.7. Cytotoxicity of PPI Assay and ROS Assay

Skov3 (human ovarian carcinoma cells) were cultured in DMEM/F12 medium + FBS
(10%) and antibiotics (100 U/mL penicillin together with 100 µg/mL streptomycin) in a 5%
CO2 atmosphere at 37 ◦ C in humidified incubator. Different concentrations of PPI (0, 100,
200, 400 µg/mL) were used to treat cells for 24 h, and the MTT assay was used to evaluate
the cytotoxicity of PPI [40]. The S0033 detection kit (Beyotime, Haimen, Jiangsu, China)
was used to measure the intracellular ROS levels in accordance with this method.

4.7.8. Jc-1 Assays, TUNEL Assay and Immunohistochemistry

The JC-1 Mitochondrial Membrane Potential Kit (Beyotime, Shanghai, China) and
TUNEL BrightRed Apoptosis Detection Kit (Vazyme, Piscataway, NJ, USA) were used to
detected the quantity of JC-1 and apoptotic cells, respectively. Cell that were treated with
PPI for 24 h were subjected to both the Jc-1 assay and TUNEL assay. Primary antibod-
ies against anti-γH2AX and anti-Rad51 were used for immunohistochemistry based on
published methods [12].

4.7.9. Western Blot Analysis

Cells were treated with 400 µg/mL PPI for 12 h in order to produce cell pellets for
western blot analysis [41]. The following primary antibodies were used: anti-Rad51 (Abcam
ab88572, London, UK), anti-γH2AX (Abcam ab26350, London, UK), anti-Bcl2 (ImmunoWay
YT0470, Suzhou Jiangsu, China), anti-BAX (Cell Signaling Technology, #2772, Boston, MA,
USA) and β-action (Abcam ab8227, London, UK).
Molecules 2021, 26, 1949 15 of 16

4.7.10. Bioinformatics Analysis of the mRNA Expression Profile

RNA extraction and RNA expression profiling were performed based on pre-existing
methods [42]. GenePix 4.1 software (Molecular Devices, Sunnyvale, CA, USA) was used to
analyse to microarray signal intensity of each spot. Three replicates in one group were used.
The expressions of mRNAs with log2 |fold change| > 1 (absolute |fold change| > 2)
and p < 0.05 were defined as differentially expressed mRNAs. Hierarchical clustering
analysis combined with a heatmap was applied to evaluate the three samples within
each group and the differences between the two groups. An average linkage hierarchical
clustering was generated using the R program in the heatmap.2 package, a function
of g-plots package that is used for hierarchical clustering. Gene ontology analysis was
performed using DAVID, a database for gene functional annotation.

4.7.11. Statistical Method

The differences between mean values were statistically tested using Student’s t test or
one-way ANOVA followed by the Tukey test for multiple comparisons. Comparisons were
considered significant at p < 0.05 and p < 0.01 (asterisk).

Author Contributions: X.-F.Z. designed the study. W.-F.L., H.-H.M., S.Y. and X.-F.Z. collected data.
All authors have read and agreed to the published version of the manuscript.
Funding: This work was supported by the High level talents research fund project of Qingdao
Agricultural University in China (1120043) to X.-F.Z.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in this study are available in the article.
Conflicts of Interest: The authors declare no conflict of interest.
Sample Availability: Samples of the compounds are not available from the authors.

References
1. Wang, X.H.; Bai, H.Q.; Wulangerile. Progress in the Studies of Polysaccharide Metal Complex. J. Inn. Mong. Univ. Natl. 2014, 29,
516–519. [CrossRef]
2. Danielson, B.G. Structure, chemistry, and pharmacokinetics of intravenous iron agents. J. Am. Soc. Nephrol. 2004, 15, 93–98.
3. Eichbaum, Q.; Foran, S.; Dzik, S. Is iron gluconate really safer than iron dextran? Blood 2003, 101, 3756. [CrossRef]
4. Johnson, C.A.; Rosowski, E.; Zimmerman, S.W. A prospective open-label study evaluating the efficacy and adverse reactions of
the use of Niferex-150 in ESRD patients receiving EPOGEN. Adv. Perit. Dial. Conf. Perit. Dial. 1992, 8, 444–447.
5. Floyd, R.A.; Lewis, C.A. Hydroxyl free radical formation from hydrogen peroxide by ferrous iron-nucleotide complexes.
Biochemistry 1983, 22, 2645–2649. [CrossRef] [PubMed]
6. Oates, P.S.; Morgan, E.H. Defective iron uptake by the duodenum of Belgrade rats fed diets of different iron contents. Am. J.
Physiol. Liver Physiol. 1996, 270, G826–G832. [CrossRef] [PubMed]
7. Kudasheva, D.S.; Lai, J.; Ulman, A.; Cowman, M.K. Structure of carbohydrate-bound polynuclear iron oxyhydroxide nanoparticles
in parenteral formulations. J. Inorg. Biochem. 2004, 98, 1757–1769. [CrossRef] [PubMed]
8. Chen, Z.W.; Zhang, Q.; Wang, S.H.; Shi, L. Study on the synthesis technique of angelica polysaccharides iron complex by
orthogonal test. Med. J. Defending Forces Southwest China 2003, 13, 374–376. [CrossRef]
9. He, R.X.; Gao, W.H.; Zhu, S.M. Preparation and characters of soluble soybean polysaccharide-iron (III) complex. Sci. Technol.
Food Ind. 2012, 2, 326–330, doiorg/101155/2019/6416941.
10. Wu, G.J.; Li, Y.P.; Hu, L.; Chen, Z.Q.; Hong, L.M. Synthesis of Portulace oleracea L. polysaccharide iron complex and its physical-
chemical properties. Food Sci. Technol. 2012, 37, 250–252. [CrossRef]
11. Li, S.J. Study on Chemical and Antioxidation of Fruit of Pyracantha Resources; Huazhong University of Science and Technology:
Wuhan, China, 2011.
12. Yao, Y.-L.; Shu, C.; Feng, G.; Wang, Q.; Yan, Y.-Y.; Yi, Y.; Wang, H.-X.; Zhang, X.-F.; Wang, L.-M. Polysaccharides from Pyracantha
fortuneana and its biological activity. Int. J. Biol. Macromol. 2020, 150, 1162–1174. [CrossRef]
13. Yang, T.; Zhang, S.; Wang, R.; Li, D.; Hu, Y.; Nie, J.; Zhao, X.; Wang, Q.; Chen, Y.; Zheng, Y.; et al. Polysaccharides from Rhizoma
Panacis Majoris and its anti-oxidant activity. Int. J. Biol. Macromol. 2016, 86, 756–763. [CrossRef] [PubMed]
14. Zhao, Z.-Y.; Zhang, Q.; Shu-Lin, L.; Dong, L.-L.; Liu, S.-L. Optimization of ultrasound extraction of Alisma orientalis polysaccha-
rides by response surface methodology and their antioxidant activities. Carbohydr. Polym. 2015, 119, 101–109. [CrossRef]
Molecules 2021, 26, 1949 16 of 16

15. Mehmood, T.; Ahmed, A.; Ahmad, A.; Ahmad, M.S.; Sandhu, M.A. Optimization of mixed surfactants-based β-carotene
nanoemulsions using response surface methodology: An ultrasonic homogenization approach. Food Chem. 2018, 253, 179–184.
[CrossRef] [PubMed]
16. Celebi, N.; Yildiz, N.; Demir, A.S.; Çalımlı, A.; Calimli, A.; Yıldız, N. Optimization of benzoin synthesis in supercritical carbon
dioxide by response surface methodology (RSM). J. Supercrit. Fluids 2008, 47, 227–232. [CrossRef]
17. Chang, H.Y.; Ho, Y.L.; Sheu, M.J.; Lin, Y.H.; Tseng, M.C.; Wu, S.H.; Huang, G.J.; Chang, Y.S. Antioxidant and free radical
scavenging activities of Phellinus merrillii extracts. Bot. Stud. 2007, 48, 407–417.
18. Pitarresi, G.; Tripodo, G.; Cavallaro, G.; Palumbo, F.S.; Giammona, G. Inulin–iron complexes: A potential treatment of iron
deficiency anaemia. Eur. J. Pharm. Biopharm. 2008, 68, 267–276. [CrossRef]
19. Marshall, P.; Rutherford, D. Physical investigations on colloidal iron-dextran complexes. J. Colloid Interface Sci. 1971, 37, 390–402.
[CrossRef]
20. Park, J.; Chakrabarti, B. Optical characteristics of carboxyl group in relation to the circular dichroic properties and dissociation
constants of glycosaminoglycans. Biochim. et Biophys. Acta (BBA)—Gen. Subj. 1978, 544, 667–675. [CrossRef]
21. Chi, Y.; Li, Y.; Zhang, G.; Gao, Y.; Ye, H.; Gao, J.; Wang, P. Effect of extraction techniques on properties of polysaccharides from
Enteromorpha prolifera and their applicability in iron chelation. Carbohydr. Polym. 2018, 181, 616–623. [CrossRef]
22. Bertini, I.; Luchinat, C.; Parigi, G.; Pierattelli, R. NMR Spectroscopy of Paramagnetic Metalloproteins. ChemBioChem 2005, 6,
1536–1549. [CrossRef]
23. Kästele, X.; Sturm, C.; Klüfers, P. 13 C NMR spectroscopy as a tool for the in situ characterisation of iron-supplementing
preparations. Eur. J. Pharm. Biopharm. 2014, 86, 469–477. [CrossRef]
24. Bai, D.-P.; Zhang, X.-F.; Zhang, G.-L.; Huang, Y.-F.; Gurunathan, S. Zinc oxide nanoparticles induce apoptosis and autophagy in
human ovarian cancer cells. Int. J. Nanomed. 2017, 12, 6521–6535. [CrossRef]
25. Carmody, R.; Cotter, T. Signalling apoptosis: A radical approach. Redox Rep. 2001, 6, 77–90. [CrossRef]
26. Ishikawa, K.; Takenaga, K.; Akimoto, M.; Koshikawa, N.; Yamaguchi, A.; Imanishi, H.; Nakada, K.; Honma, Y.; Hayashi, J.-I.
ROS-Generating Mitochondrial DNA Mutations Can Regulate Tumor Cell Metastasis. Science 2008, 320, 661–664. [CrossRef]
[PubMed]
27. Turrens, J.F. Mitochondrial formation of reactive oxygen species. J. Physiol. 2003, 552, 335–344. [CrossRef] [PubMed]
28. Darzynkiewicz, Z.; Galkowski, D.; Zhao, H. Analysis of apoptosis by cytometry using TUNEL assay. Methods 2008, 44, 250–254.
[CrossRef]
29. Redon, C.E.; Dickey, J.S.; Bonner, W.M.; Sedelnikova, O.A. γ-H2AX as a biomarker of DNA damage induced by ionizing radiation
in human peripheral blood lymphocytes and artificial skin. Adv. Space Res. 2009, 43, 1171–1178. [CrossRef] [PubMed]
30. Rothkamm, K.; Horn, S. gamma-H2AX as protein biomarker for radiation exposure. Annali dell’Istituto Superiore di Sanità 2009, 45,
265–271.
31. He, D.; Yuan, X.; Lin, G.; Chen, W. Determination of iron ion in polysaccharide iron complex pellets. Chin. J. Hosp. Pharm. 2006,
26, 107–108. [CrossRef]
32. Tang, C.N. Research Progress of Polysaccharide-iron Complex(PIC). Anhui Chem. Ind. 2012, 38, 1–36. [CrossRef]
33. Saywell, L.G.; Cunningham, B.B. Determination of Iron: Colorimetric o-Phenanthroline Method. Ind. Eng. Chem. Anal. Ed. 1937,
9, 67–69. [CrossRef]
34. Azargohar, R.; Dalai, A. Production of activated carbon from Luscar char: Experimental and modeling studies. Microporous
Mesoporous Mater. 2005, 85, 219–225. [CrossRef]
35. Adegoroye, A.; Wang, L.; Omotoso, O.; Xu, Z.; Masliyah, J. Characterization of organic-coated solids isolated from different oil
sands. Can. J. Chem. Eng. 2010, 88, 462–470. [CrossRef]
36. Du, X.; Mu, H.; Zhou, S.; Zhang, Y.; Zhu, X. Chemical analysis and antioxidant activity of polysaccharides extracted from Inonotus
obliquus sclerotia. Int. J. Biol. Macromol. 2013, 62, 691–696. [CrossRef] [PubMed]
37. Gao, C.; Wang, Y.; Wang, C.; Wang, Z. Antioxidant and immunological activity in vitro of polysaccharides from Gomphidius
rutilus mycelium. Carbohydr. Polym. 2013, 92, 2187–2192. [CrossRef] [PubMed]
38. Hu, Y.; Sheng, Y.; Yu, M.; Li, K.; Ren, G.; Xu, X.; Qu, J. Antioxidant activity of Inonotus obliquus polysaccharide and its
amelioration for chronic pancreatitis in mice. Int. J. Biol. Macromol. 2016, 87, 348–356. [CrossRef]
39. Ben Jeddou, K.; Chaari, F.; Maktouf, S.; Nouri-Ellouz, O.; Helbert, C.B.; Ghorbel, R.E. Structural, functional, and antioxidant
properties of water-soluble polysaccharides from potatoes peels. Food Chem. 2016, 205, 97–105. [CrossRef]
40. Paździoch-Czochra, M.; Wideńska, A. Spectrofluorimetric determination of hydrogen peroxide scavenging activity. Anal. Chim.
Acta 2002, 452, 177–184. [CrossRef]
41. Gurunathan, S.; Zhang, X.-F.; Choi, Y.-J.; Han, J.W.; Kim, E.; Park, J.H.; Kim, J.-H. Differential nanoreprotoxicity of silver
nanoparticles in male somatic cells and spermatogonial stem cells. Int. J. Nanomed. 2015, 10, 1335–1357. [CrossRef] [PubMed]
42. Li, L.; Liu, J.-C.; Zhao, Y.; Lai, F.-N.; Yang, F.; Ge, W.; Dou, C.-L.; Shen, W.; Zhang, X.-F.; Chen, H. Impact of diethylhexyl phthalate
on gene expression and development of mammary glands of pregnant mouse. Histochem. Cell Biol. 2015, 144, 389–402. [CrossRef]
[PubMed]