Structure-function relationships of pectic polysaccharides from broccoli by-products with in vitro B lymphocyte stimulatory activity

Carbohydrate Polymers 303 (2023) 120432

Contents lists available at ScienceDirect

Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol

Structure-function relationships of pectic polysaccharides from broccoli
by-products with in vitro B lymphocyte stimulatory activity
´nia S. Ferreira a, b, *, Alexandra Correia c, d, e, Artur M.S. Silva a, Dulcineia Ferreira Wessel a, f, g,
So
Susana M. Cardoso a, Manuel Vilanova c, d, h, Manuel A. Coimbra a, **
a

LAQV-REQUIMTE, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal
CICECO - Aveiro Institute of Materials, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal
i3S, Instituto de Investigaỗ
ao e Inovaỗ
ao em Saỳde, Universidade do Porto, 4200-135 Porto, Portugal
d
IBMC, Instituto de Biologia Molecular e Celular, Universidade do Porto, 4150-180 Porto, Portugal
e
DGAOT, Faculdade de Ciˆencias, Universidade do Porto, Rua do Campo Alegre, 4169-007 Porto, Portugal
f
School of Agriculture, Polytechnic Institute of Viseu, 3500-606 Viseu, Portugal
g
CITAB, University of Tr´
as-os-Montes e Alto Douro, 5001-801 Vila Real, Portugal
h
ICBAS, Instituto de Ciˆencias Biom´edicas de Abel Salazar, Universidade do Porto, 4050-313 Porto, Portugal
b

c

A R T I C L E I N F O

A B S T R A C T

Keywords:
Brassica by-products
Sulfated pectic polysaccharides
Arabinogalactans
NMR
Methylation analysis
Enzyme modified pectin
Structure-function relationships

To study structure-function relationships of pectic polysaccharides with their immunostimulatory activity,
broccoli by-products were used. Pectic polysaccharides composed by 64 mol% uronic acids, 18 mol% Ara, and
10 mol% Gal, obtained by hot water extraction, activated B lymphocytes in vitro (25–250 μg/mL). To disclose
active structural features, combinations of ethanol and chromatographic fractionation and modification of the
polysaccharides were performed. Polysaccharides insoluble in 80 % ethanol (Et80) showed higher immunosti­
mulatory activity than the pristine mixture, which was independent of molecular weight range (12–400 kDa) and
removal of terminal or short Ara side chains. Chemical sulfation did not promote B lymphocyte activation.
However, the action of pectin methylesterase and endo-polygalacturonase on hot water extracted polysaccharides
produced an acidic fraction with a high immunostimulatory activity. The de-esterified homogalacturonan region
seem to be an important core to confer pectic polysaccharides immunostimulatory activity. Therefore, agri-food
by-products are a source of pectic polysaccharide functional food ingredients.

1. Introduction
Broccoli by-products are a source of pectic polysaccharides (Petkoư
ăfer et al., 2017), which information about

wicz & Williams, 2020; Scha
their structure and potential bioactivities is scarce (Busato et al., 2020;
Xu et al., 2015; Zhang et al., 2017). In general, pectic polysaccharides
are complex heteropolysaccharides with a high proportion of GalpA
(Waldron & Faulds, 2007). They include various fragments of linear and
ramified regions covalently connected. It is well established that the
linear region consists of (α1→4)-D-GalpA residues (homogalacturonan
region) carrying methyl ester groups and that can also be acetylated. A
backbone of alternating (α1→4)-D-GalpA and (α1→2)-L-rhamnopyr­
anosyl (L-Rhap) residues, ramified in the Rha by galactans, arabinoga­
lactans, and/or arabinans of varying structure is named type I

rhamnogalacturonan (Vincken et al., 2003; Yapo, 2011). The diversity
of pectic polysaccharide structures is dependent on plant source, stages
of maturity, plant part, or processing.
Pectic polysaccharides found in several plants have been related to
health effects, from anticancer to immunomodulatory activities (Jin
et al., 2021; Ramberg et al., 2010; Yin et al., 2019). Immunomodulatory
polysaccharides may strengthen innate and adaptive immunity by
directly interacting with distinct cellular and humoral components of
the immune system, or indirectly through complex reaction cascades
between immune components (Ferreira et al., 2015). The immunosti­
mulatory activity of pectic polysaccharides has been associated to the
branched regions (e.g. arabinans (Dourado et al., 2004; Dourado et al.,
2006; Popov & Ovodov, 2013), arabinogalactans (Hamed et al., 2022;
Zou et al., 2017), or their combination (Westereng et al., 2008;

* Correspondence to: S.S. Ferreira, LAQV-REQUIMTE, Department of Chemistry, University of Aveiro, 3810-193 Aveiro, Portugal.
** Corresponding author.
E-mail addresses: (S.S. Ferreira), (M.A. Coimbra).

/>Received 19 July 2022; Received in revised form 18 November 2022; Accepted 1 December 2022
Available online 7 December 2022
0144-8617/© 2022 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license ( />

S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Westereng et al., 2009). In fact, the removal of the linear regions of
pectic polysaccharides by endo-polygalacturonase has been shown to
improve their complement fixing activity (Togola et al., 2008) whereas
no effect or the opposite effect was observed by the removal of the
branching regions with exo-α-L-arabinofuranosidase and galactosidases
(Kiyohara et al., 2010; Nergard et al., 2005). Decrement of complement
fixing activity or splenocytes proliferation was also observed by removal
of arabinofuranosyl (Araf) residues with weak acid hydrolysis (Diallo,
ăck, & Michaelsen, 2001, 2003; Wang, Liu, & Fang, 2005;
Paulsen, Liljeba
Duan et al., 2010; Inngjerdingen et al., 2007). However, the removal of
Araf residues after enzymatic degradation of pectic polysaccharides
from Glinus oppositifolius (L.) Aug. DC. (Aizoaceae) aerial parts did not
affect their complement fixing activity or their ability to stimulate
Peyer’s patches cells to produce secreted factors able to induce bone
marrow cell proliferation (Inngjerdingen et al., 2007). On the other
hand, the removal of homogalacturonan after endo-polygalacturonase
treatment of pectic polysaccharides from Pterospartum tridentatum (L.)
Willk. inflorescences decreased the nitric oxide production by macro­
phages (Martins et al., 2017). These results show that different structural
features of pectic polysaccharides can be involved in the triggering or
modulation of immune responses. In addition, chemical sulfation of

polysaccharides has been shown to modulate their immunostimulatory
activity (Ferreira et al., 2015). However, considering pectic poly­
saccharides, contradictory data have been reported with studies
showing improvements (Du et al., 2010) or no effect on spleen cells
proliferation (Wang et al., 2018). Therefore, the establishment of
structure-function relationships of immunomodulatory pectic poly­
saccharides is far from being fully revealed.
Following the current state of knowledge about polysaccharides
immunostimulatory activity, the pectic polysaccharides from broccoli
by-products were extracted, fractionated and/or modified with 80 %
ethanol solution, size exclusion chromatography, enzymatic treatments,
chemical sulfation, and anion-exchange chromatography to study in
vitro the potential immunostimulatory activity and to establish
structure-function relationships.

separated from un-extracted residue by centrifugation (24,652g, 4 ◦ C,
for 30 min) followed by filtration in a fritted funnel (G-3). The soluble
material was concentrated in a rotary evaporator and dialysed using
membranes with 12 kDa cut-off to obtain HW. This sample and the
residue were freeze-dried for yield determination. HW carbohydrates
were analysed and further purified, fractionated, and derivatized
(Fig. 1).
2.4. Ethanol precipitation of HW
Polysaccharides from HW were fractionated according to their sol­
ubility in ethanol solutions. HW (100 mg) was hydrated by the addition
of 20 mL of water, heated at 40 ◦ C, for 5 min, and vortexed. Some
insoluble particles, <1 % of HW dry weight, were removed by centri­
fugation at 2500 g, for 10 min. Absolute ethanol was added to the soluble
material to obtain a solution of 50 % ethanol. After 2 h, at 4 ◦ C, as no
precipitate was observed, more absolute ethanol was added until

reaching a concentration of 80 % ethanol. This solution was kept for 2 h
at 4 ◦ C, allowing to obtain the compounds that precipitate in the 80 %
ethanol solution (Et80), which were separated by centrifugation
(24,652 g, 10 min, 4 ◦ C) from the soluble material (SnEt) (Fig. 1). Both
fractions were evaporated under reduced pressure and freeze-dried for
yield determination and carbohydrate analysis.
2.5. Size-exclusion chromatography of fraction Et80
To unveil the molecular weight of polysaccharides found in fraction
Et80, this was fractionated by a size-exclusion chromatography (SEC)
column (XK 16/100, Pharmacia) of Sephacryl S400 HR (10–2000 kDa
fractionation range for dextrans, Pharmacia–Biotech). The chromatog­
raphy was run using 200 mM phosphate buffer, pH 7, 200 mM NaCl, and
0.02 % sodium azide as eluent, with a flow rate of 1.0 mL/min, at room
temperature. The sample (10 mg) was dissolved in 1 mL of the same
elution buffer and then passed through the column. Fractions of 2 mL
were collected and assayed for total sugars according to phenolconcentrated sulfuric acid method (absorption at 490 nm) (Dubois
et al., 1956) and at 280 nm for UV-absorbing noncarbohydrate com­
pounds. The appropriate fractions were combined, dialysed and freezedried. The average molecular weight of each fraction was established
after a calibration of the column with dextrans of average molecular
weight 12 kDa (2.8 mg, Supelco-31418–100 mg Sigma), 80 kDa (2.45
mg, Supelco-31421–100 mg Sigma), and 270 kDa (2.84 mg, Supelco31423 Sigma). The void volume (V0) was determined with blue
dextran (2000 kDa, 5 mg) and the inner volume (Vt) of the column was
determined with Glc (5 mg).

2. Material and methods
2.1. Broccoli by-products samples and materials
Broccoli (Brassica oleracea var. Parthenon) by-product samples from
the frozen food industry were provided by Monliz SA, Alpiarỗa,
Portugal. They were kept frozen under 20 C until use. All reagents
used were of analytical grade or higher available purity.

2.2. Preparation of alcohol insoluble residue (AIR)

2.6. α-L-Arabinofuranosidase treatment of fraction Et80

To isolate polysaccharides from broccoli by-product cell walls,
inactivate enzymes, and remove low-molecular-weight compounds
(Knee, 1973), AIR was prepared. Briefly, ethanol was added to 150 g of
frozen material, previously blended, to obtain a final concentration of
80 % (v/v). After boiling for 10 min, the mixture was cooled in a coldwater bath and filtered in fritted funnel (G-1). The filtration residue
was dispersed again in 80 % (v/v) ethanol, boiled for 10 min, filtered,
washed with 500 mL of ethanol and 500 mL of acetone, and allowed to
dry at room temperature. The AIR obtained was further milled in a
coffee grinder and the moisture content was determined by freezedrying.

To study the influence of the terminal Ara of pectic polysaccharides
for the in vitro immunostimulatory activity, the Et80 fraction (30 mg)
was hydrolysed with 2 U of Clostridium thermocellum α-L-arabinofur­
anosidase 51 A (EC 3.2.1.55, purified from a recombinant Escherichia coli
strain, Nzytech), for 48 h at 37 ◦ C with continuous gentle stirring in 5 mL
of 100 mM sodium acetate buffer, pH 5.5, containing 0.02 % sodium
azide (Ferreira, Passos, Cepeda, et al., 2018). After enzymatic treatment
and enzyme denaturation by placing in boiling water for 5 min, the
hydrolysed material was dialysed and freeze-dried to get sample
Et80_Ara.

2.3. Extraction of hot water (HW) soluble polysaccharides from AIR

2.7. Chemical sulfation of fraction Et80

To extract pectic polysaccharides with green solvents, AIR was

submitted to a hot water extraction. Briefly, 1 g of AIR was hydrated
with 100 mL of distilled water for 2 h, under constant stirring. After­
wards, it was boiled for 1 h, using a condenser to avoid solvent loss. The
mixture was cooled in a cold-water bath and the soluble material was

The Et80 fraction was sulfated to study the influence of this func­
tional group in pectic polysaccharide immunostimulatory activity. Sul­
fation was performed using a methodology with chlorosulfonic acid and
pyridine, according to Du et al. (2010). Et80 (100 mg) was suspended in
formamide (10 mL) under agitation at room temperature for 15 min,
2


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Fig. 1. Scheme of fractionation and modification of hot water extract (HW) from broccoli by-product AIR.

followed by dropwise addition of sulfating reagent (286 μL chlor­
osulfonic acid and 1714 μL of pyridine). The reaction was stirred at room
temperature for 2 h and maintained at 30 ◦ C for 5 h with continuous
stirring. After cooling to room temperature, the solution was neutralized
with 15 % (w/v) NaOH solution, dialyzed (Mw cut-off 12 kDa) against
distilled water, concentrated, and freeze-dried to obtain the sulfated
polysaccharide (Sulf_Et80).

(Bastos et al., 2015; Blakeney et al., 1983; Selvendran et al., 1979) and
quantification of uronic acids (UA) by the 3-phenylphenol colorimetric
method (Blumenkrantz & Asboe-Hansen, 1973). The total sugars were

determined by the sum of the amount of the individual sugars.
Glycosidic-linkage composition was determined by methylation of
polysaccharides and analysis of partially methylated alditol acetates
(PMAA) (Ciucanu & Kerek, 1984; Coimbra et al., 1996; Harris et al.,
1984; Isogai et al., 1985; Passos & Coimbra, 2013). UA linkages were
detected after carboxyl reduction of methylated polysaccharides
(Dourado et al., 2004; Lindberg & Lă
onngren, 1978). The PMAA were
separated and analysed by gas chromatography quadrupole mass spec­
trometry (GCqMS) (GC-2010 Plus, Shimadzu, Japan) using a non-polar
column HT5 (30 m length, 0.25 mm internal diameter and 0.10 μm
stationary phase, Trajan, Australia) as described by Hamed et al. (2022).

2.8. Enzymatic treatment of HW
Ramified regions of pectic polysaccharides from HW were purified
by enzymatic treatment to remove protein and starch, and to cleave the
homogalacturonan backbone of the pectic polysaccharides. First, HW
(20 mg) was solubilized in 200 mL of 50 mM phosphate buffer, pH 7,
with 0.1 % sodium azide. Pronase, a mixture of proteases (2 mg, from
Streptomyces griseus, ROCHE), was added to this solution and incubated
for 24 h, at 37 ◦ C, under constant stirring. Enzymes were denatured by
placing in boiling water for 5 min. After cooling down, 10 μL of
α-amylase (1210 U/mg protein, 27 mg protein/mL, A-2643, Sigma), 10
μL of endo-polygalacturonase (5000 U/mL, M2, Megazyme), and 1 mL of
pectin methylesterase (75.2 U/mL, from Streptomyces avermitilis, PRO­
ZOMIX) were added and incubated for 24 h, at 37 ◦ C, also under con­
stant stirring. Enzymes were denatured as previous described and the
solution was dialysed, using membranes with 12 kDa cut-off, before
freeze-drying to obtain Enz_HW.

2.11. NMR experiments
NMR experiments were performed as described by Fernandes et al.
(2019). 1H and 13C NMR spectra of samples Et80 (25 mg), Enz_HWB (10
mg), and Et80_Ara (25 mg) was recorded in D2O (500 μL), at 60 ◦ C, on a
Bruker DRX 500 spectrometer operating at 500.13 and 125.77 MHz,
respectively; the chemical shifts were expressed in δ (ppm) values
relative to TSS as external reference. 2D (1H,13C) gHSQC (heteronuclear
single quantum coherence, using gradient pulses for selection), and 2D
(1H,13C) gHMBC (heteronuclear multiple quantum coherence, using
gradient pulses for selection; delays for one-bond and long-range JC/H
couplings were optimised for 145 and 7 Hz, respectively) spectra of
samples Et80 and Enz_HWB were also recorded.

2.9. Anion-exchange chromatography of Enz_HW
Anion-exchange chromatography on diethylaminoethyl cellulose
(DEAE)-Trisacryl M (1 mL of resin/7.5 μmol of uronic acids, Sigma) was
performed on a 40 × 16 mm column (XK 16/20, Pharmacia), at a flow
rate of 0.4 mL/min. Enz_AIR (10 mg) was dissolved in 1 mL of 50 mM
Tris-HCl buffer, pH 7.4, 0.02 % sodium azide without formation of
insoluble material after centrifugation (2490 g, 15 min). The sample was
eluted with 40 mL of Tris-HCl buffer, and then with 20 mL solutions of
buffer containing 0.125, 0.250, 0.500, and 1.00 M NaCl, successively,
and washed with 2.00 M and 4.00 M NaCl. Fractions of 1.9 mL were
collected and assayed for total sugars according to phenol-concentrated
sulfuric acid method (absorption at 490 nm) (Dubois et al., 1956) and at
280 nm for phenolics. The fractions of interest were combined, dialysed,
and freeze-dried.

2.12. In vitro immunostimulatory activity assays

2.10. Carbohydrate analyses

2.12.1. Mice
Female BALB/c mice were obtained from Charles River (Barcelona,
Spain) and were kept at i3S animal facilities (University of Porto). Ex­
periments followed the Portuguese rules (DL 113/2013), the European
Convention for the Protection of Vertebrate Animals used for Experi­
mental and Other Scientific Purposes (ETS 123), the directive 2010/63/
EU of the European parliament, and the council of 22 September 2010
on the protection of the animals used for scientific purposes. Experi­
ments were approved by the institutional board responsible for animal
welfare (ORBEA) at i3S (code 2019-5) and by the competent national
o Geral de Alimentaỗa
o e Veterina
ria), reference
authority (Direỗa
number 014036/2019-07-24.

Samples were analysed for their sugar content after acid hydrolysis,
derivatisation to alditol acetates, and analysis of individual neutral
sugars by gas chromatography with flame ionization detector (GC-FID)

2.12.2. In vitro lymphocyte stimulating effect by flow cytometry analysis
The preparation of spleen lymphocytes for stimulation tests and flow
cytometry analysis was performed according to Ferreira et al. (2020).
3


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Cells were stimulated with RPMI (negative control), 2.5 μg/mL of li­
popolysaccharides (LPS) from Escherichia coli O111:B4 (Sigma, St. Louis,
B cells positive control), 2.5 μg/mL of concanavalin A (T cell positive
control), or 25, 100 and/or 250 μg/mL of samples. Cell viability was
assessed by fluorescent staining with propidium iodide before flow
cytometry. Propidium iodide is a live cell impermeant red-fluorescent
nuclear and chromosome counterstain. Since propidium iodide does
not permeate live cells, it was used as a fluorescent staining to exclude
dead cells and assess cells viability by flow cytometry. The collected data
files were analysed using FlowJo v10.3 software (Tree Star Inc., Ash­
land, OR, USA).

3. Results and discussion
3.1. Broccoli hot water extractable pectic polysaccharides and evaluation
of their immunostimulatory activity
To evaluate the in vitro potential immunostimulatory activity of
pectic polysaccharides from broccoli, an alcohol insoluble residue (AIR)
was prepared (60 % yield on a dry weight basis) using broccoli byproducts, constituted by the remains of inflorescences, leaves, and
stalks, with a carbohydrate composition similar to the broccoli available
commercially (Ferreira, Passos, Cardoso, Wessel and Coimbra, 2018). A
fraction rich in pectic polysaccharides was obtained by extraction of AIR
with hot water, mimicking the boiling procedure when cooking
vegetables.
Hot water (HW) extraction yielded 12 % of high-molecular-weight
material. It was composed by 69 % of polysaccharides, rich in uronic
acids (UA, 64 mol%), Ara (18 mol%), and Gal (10 mol%), with a small
amount of Rha (1 mol%), together with Glc, Xyl, Man, and Fuc (Table 1),
indicating that pectic polysaccharides were the main polysaccharides.

The presence of (1→4)-GalpA residues in HW, determined by
carboxyl reduction with LiAlD4 of methylated polysaccharides, and the
identification of (1→2)-Rhap and (1→2,4)-Rhap residues (Table 2)
confirmed the presence of pectic polysaccharides. The high amount of
(1→5)-Araf indicates the existence of arabinans in the ramified domain
of pectic polysaccharides, as found in broccoli stems and other Brassica
ăfer et al., 2017; Stevens & Selvendran, 1980). The presence of
(Scha
(1→3)- and (1→3,6)-Galp residues, as well as t-Galp, (1→6)-Galp, and tGlcpA residues are diagnostic of type II (arabino)galactans. On the other

2.12.3. Detection of LPS
To evaluate possible endotoxin contamination in 5 mg/mL of sample,
LPS presence was evaluated after methanolysis, hexane extractions, and
acetylation of fatty acids for GC–MS analysis as 3-O-acetyl fatty acid
methyl esters (FAME), according to de Santana-Filho et al. (2012).
2.12.4. Statistical analysis
Statistical analysis was performed by Student’s t-test using GraphPad
prism, Version 6.0 (GraphPad Software, Inc. La Jolla, CA, USA). Results
were represented as mean ± SD from at least three experiments with
different animals (* p < 0.01).

Table 1
Yield, carbohydrate composition, and total carbohydrates of samples obtained by hot water extraction, fractionation, and enzymatic treatments of broccoli by-product
alcohol insoluble residue (AIR).
Sample

η (%)

Carbohydrate composition (mol%)

Total Carb. (mg/g)

Rha

Fuc

Ara

Xyl

Man

Gal

Glc

UA

60

1
1

tr
tr

13
16

4

5

4
2

9
9

46
33

23
34

384 ± 24
555 ± 4

Hot water (HW) extraction
HW
12
Residue
73

1
1

1
tr

18

16

1
6

tr
3

10
9

5
38

64
27

685 ± 5
539 ± 20

80 % ethanol fractionation of HW
Et80
85
Sn80
14

1
2

tr

tr

17
43

1
3

1
3

10
7

6
6

65
36

642 ± 25
336 ± 10

1

tr

10

1

1

14

9

64

609 ± 43

Size-exclusion chromatography of Et80
Et80-I >151 kDa
20
Et80-II 41–151 kDa
34
Et80-III 12–41 kDa
25
11
Et80-IV <12 kDa

2
1
1
2

1
tr
1
1

26
22
12
9

1
1
2
4

1
1
3
6

10
13
11
20

7
4
10
42

52
58
60
16

366 ± 22
729 ± 22
591 ± 16
23 ± 2

Chemical sulfation of Et80
Et80-Sulf
104

1

tr

17

1

2

11

9

59

515 ± 12

Enzymatic treatments of HW*
Enz_HW

32

2

tr

33

2

3

19

4

37

601 ± 14

Anion-exchange chromatography of Enz_HW
Enz_HW A
36
1
Enz_HW B
50
1
Enz_HW C
14
1

tr
tr
1

31
38
43

2
1
1

4
tr
1

25
19
11

3
1
4

34
40
38

677 ± 0

628 ± 12
17 ± 2

Broccoli by-products
AIR

α-L-Arabinofuranosidase treatment of Et80
Et80-Ara

79

AIR: alcohol insoluble residue; tr traces; *treatments with pronase, α-amylase, pectin methylesterase and endo-polygalacturonase. Data are expressed as mean ±
standard deviation of 3 replicates.
4


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Table 2
Heatmap of glycosidic linkage analysis of samples obtained by hot water extraction (HW) from AIR of broccoli by-products and following purification, fractionation,
sulfation, and enzymatic treatments. mol% of glycosidic linkages were balanced with the amount of UA determined by the colorimetric method.
Et80 Ara
removal

Et80-Sulf

HW
enzymatic

treated
Enz_HW

Enz_HWA

Enz_HWB

0.2
0.7

0.6
1.1

0.1
0.4

0.5
2.1

0.0

0.1

0.2

-

0.2

7.9

0.1
8.4
1.6
2.1
1.2

3.6
0.3
5.4
0.3
0.7
0.2

1.4
0.7
3.0
1.8
1.7
1.9

10.3
0.7
14.1
1.8
3.7
0.4

9.3
1.0
13.7

1.1
2.9
1.5

10.7
0.2
15.2
0.8
5.2
2.8

0.6
0.8

0.2
1.0

0.3
3.1

1.2
1.0

1.2
2.2

0.2
5.1

0.9

2.1

1.4
1.7
0.6
4.5
2.2
-

0.5
2.6
0.1
0.5
2.2
-

0.6
1.7
0.3
2.6
4.3
-

1.0
5.4
1.5
1.8
3.1
-

1.1
1.4
0.7
3.7
4.4
1.4
1.4
1.1

2.8
1.8
2.3
4.4
9.0
-

2.1
3.0
5.0
4.6
9.4
-

1.8
4.9
0.5
3.5
4.7
-

0.9
4.7
0.9

0.9
11.0
0.4

0.7
7.9
0.5

0.5
7.2
0.3

1.1
6.7
2.7

0.4
4.4
3.7

0.3
3.8
0.3

0.1
3.4

0.2

0.1
1.4
0.1

35.5

63.5

51.8

58.2

60.3

59.1

37.1

33.6

40.3

HW - 80% ethanol
fractionation

Et80 - Size-exclusion
chromatography

Deduced
linkages (mol%)

HW

Et80

Sn80

Et80-Ara

Et80-I

Et80-II

Et80-III

1,2-Rhap
1,2,4 Rhap

0.3
0.3

0.3
0.5

0.1
0.2

0.4

0.8

1.2
0.8

0.3
0.4

0.3
0.2

t-Fucp

0.1

0.1

-

0.1

0.4

0.1

t-Araf
1,3-Araf
1,5-Araf
1,2,5-Araf
1,3,5-Araf

1,2,3,5-Araf

6.0
0.1
7.1
0.8
1.6
1.4

5.0
0.1
6.4
0.8
1.4
0.9

11.4
1.2
17.8
2.2
7.0
1.8

1.8
0.1
6.2
0.2
0.1
0.8

8.0
0.2
13.0
0.5
2.5
1.2

t-Xylp
1,4-Xylp

0.5
0.6

0.6
0.7

0.5
3.3

1.3
0.9

t-Galp
1,3-Galp
1,4-Galp
1,6-Galp
1,3,6-Galp
1,4,6-Galp
1,2,4,6-Galp
1,3,4,6-Galp

0.8
2.1
0.6
1.7
3.4
-

1.1
1.6
1.0
2.2
4.2
-

0.7
1.0
1.9
1.5
5.8
-

t-Glcp
1,4-Glcp
1,4,6-Glcp

0.5
6.5
0.7

0.4
6.6
0.6

UA

64.2

65.0

mol% 0

5

Et80
sulfation

Enz - Anion-exchange
chromatography

>10

hand, (1→4)-Galp residues are diagnostic of type I arabinogalactans, and
t-Glcp, (1→4)-Glcp, and (1→4,6)-Glcp may be related with the occur­
rence of residual starch (Femenia et al., 2000), although t-Glcp and
(1→4)-Glcp may be components of pectic polysaccharides, as cellobiose
was reported in plum pectic polysaccharides (Nunes et al., 2012). Also, tXylp and (1→4)-Xylp may derive from pectic polysaccharides and/or
polysaccharides tightly associated to pectic polysaccharides (Schols
et al., 1990).
The potential immunostimulatory activity of HW extract, was

assessed in vitro in murine splenocyte cultures. Cells viability was not
affected by incubation with HW extract (100 mg/L). In addition, CD3+
cells (T cells) did not show significant up-regulated expression of the
early activation marker CD69 on their surface, indicating that the ex­
tracts did not stimulate T cells. In contrast, higher proportions of CD19+
cells (B cells) expressed CD69 upon stimulation by HW extract, as
compared to non-stimulated control (16 % vs 7.7 % of control, Fig. 2). As
no lipid A was detected in 5 mg/mL of HW (limit of quantification of
100 ng/mL), it was excluded a possible lipopolysaccharide contamina­
tion (de Santana-Filho et al., 2012). These results show that pectic
polysaccharides from broccoli have potential immunostimulatory ac­
tivity, as observed for pectic polysaccharides from other sources
(Dourado et al., 2004; Ferreira et al., 2015; Martins et al., 2017; Popov &
Ovodov, 2013; Togola et al., 2008; Westereng et al., 2008, 2009; Zou
et al., 2017). B cells can be found along the intestinal tract in Peyer’s
patches and can be reached by pectic polysaccharides through M cells or
dendritic cells (Huang et al., 2017; Komban et al., 2019). This could
constitute a route for direct B cell activation, despite the low absorption
of pectic polysaccharides. Nevertheless, indirect B cell activation may
occur through cytokines produced by polysaccharide-stimulated enter­
ocytes or phagocytes. Recognition of pectic polysaccharides may occur

through pattern recognition receptors (Beukema et al., 2020), expressed
on enterocytes and on dendritic cells and macrophages located in Pey­
er’s Patches or in the lamina propria (Farache et al., 2013). The detected
B cell stimulatory effect suggests a potential effect of these poly­
saccharides in strengthening antibody-mediated immunity in the intes­
tinal tract and their suitability to be used in functional food ingredients.
3.2. Purification and modification of broccoli hot water extractable pectic
polysaccharides

In order to evaluate the characteristics of pectic polysaccharides
from HW extract that may be involved in their immunostimulatory ac­
tivity, two strategies were defined: 1. Fractionation by ethanol precip­
α-Litation
followed
by
size-exclusion
chromatography,
arabinofuranosidase treatment or chemical sulfation. Ethanol precipi­
tation allows to obtain fractions with different proportions of neutral
sugars to uronic acids and size-exclusion chromatography allows to
obtain fractions with different molecular weight. The α-L-arabinofur­
anosidase treatment and chemical sulfation allow to modify the poly­
saccharide side chains by removal of terminally-linked arabinose
residues and incorporation of sulfate groups in the structure; 2. Purifi­
cation by enzymatic treatment with pronase and α-amylase followed by
pectin methylesterase and endo-polygalacturonase treatments. This
strategy allows to remove proteins, residual starch, and part of the
homogalacturonan backbone, resulting in fractions richer in pectic
polysaccharide side chains. This enzymatic treated fraction was sub­
mitted to an anion-exchange chromatography to separate the different
domains according to their charge density.

5


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Fig. 2. Percentage of B cells activated (CD69+) by treatment with polysaccharides from broccoli by-products at the concentrations of 25, 100, and 250 mg/L. Culture
medium alone (RPMI) was used as negative control and lipopolysaccharides (LPS) were used as positive control. Statistical differences between RPMI (negative
control) and different stimulus are indicated above bars (* p < 0.01). Statistical differences among samples are also highlighted (* p < 0.01).

3.2.1. Ethanol precipitation of HW extract
To find the immunostimulatory active structural motifs of HW
polysaccharides, a fractionation according to the solubility of the poly­
saccharides in ethanol solutions was followed. The HW extract was
dissolved in cold water, centrifuged, and ethanol was added to the su­
pernatant. The extract was soluble in 50 % ethanol but precipitated
when the solution reached 80 % ethanol, yielding a fraction insoluble in
80 % ethanol (Et80) that was separated from the 80 % ethanol soluble
material (SnEt). The Et80 fraction accounted for 85 % of HW and con­
tained 64 % of polysaccharides (Table 1). This fraction had UA (65 mol
%) as major sugar, followed by Ara (17 mol%), Gal (10 mol%), and Rha
(1 mol%), similar in carbohydrate content and composition with the
pristine HW extract. The SnEt fraction accounted for 14 % of HW mass
and had only 34 % of carbohydrates, mainly Ara (43 mol%) and UA (36
mol%), allowing to infer the presence of highly branched pectic poly­
saccharides, rich in Ara, in agreement with their high solubility in
ethanol (Fernandes et al., 2019), together with non-carbohydrate com­
pounds that are usually adsorbed to the polysaccharides and are solu­
bilized with ethanol (Fernandes et al., 2020; Gonỗalves et al., 2018).

When incubated with splenocytes, fraction Et80 stimulated 11 % to
37 % of B cells in a dose-response relationship from 25 to 250 mg/L,
respectively (Fig. 2), showing a higher immunostimulatory activity than
100 mg/L of initial HW. As the sugars and the glycosidic linkage
composition of fraction Et80 were comparable to those of the initial HW,
the higher immunostimulatory activity of fraction Et80 seems to be due

to the removal of non-carbohydrate compounds in the EtSn. Even in
small amounts, the EtSn compounds may prevent the expression of
polysaccharide immunostimulatory activity, similarly to the effect of
chlorogenic acids mixed with coffee arabinogalactans (Passos et al.,
2021).
3.2.1.1. Size-exclusion chromatography of Et80 fraction. Polysaccharides
from Et80 fraction were submitted to size-exclusion chromatography on
a Sephacryl 400-HR to characterize their molecular weight distribution
(Fig. 3.a) and evaluate the effect of this parameter in their immunosti­
mulatory activity. Although it was not possible to resolve poly­
saccharides by size, eluted compounds were separated in four fractions:
Et80-I consisted on the material with molecular weight higher than 151
6


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Fig. 3. Chromatograms of a) size exclusion chromatography of Et80, with indication of recovered fractions (FI, FII, FIII, and FIV), blue dextran (void volume),
dextrans (12 kDa, 80 kDa, and 270 kDa), and Glc (inner volume of the column) elution volumes and b) anion-exchange chromatography of Enz_HW with indication of
recovered fractions (A, B, and C).

kDa, accounting for 20 % of the eluted material, and had 37 % of
polysaccharides; Et80-II consisted on the material with molecular
weight between 41 and 151 kDa, accounting for 34 % of the eluted
material, and was the richest fraction in polysaccharides (73 %); Et80-III
consisted on the material with molecular weight between 12 and 41
kDa, accounting for 25 % of the eluted material, which contained 59 %
of sugars and represented the peak of material that absorbs at 280 nm,

ie, diagnostic of the presence of phenolic compounds; Et80-IV consisted
on the material with molecular weight lower than 12 kDa, yielded 11 %,
and had vestigial amounts of carbohydrates, only 2 %.
Carbohydrate composition and linkage analysis of fractions Et80-I,
Et80-II, and Et80-III (Tables 1 and 2) revealed their resemblance with
initial Et80. The higher molecular weight polysaccharides (Et80-I) were
richer in Ara (26 mol%) whereas the lower molecular weight poly­
saccharides (Et80-III) had higher amount of UA (60 mol%). Linkage
analysis showed that the branching degree of arabinans, estimated by
the ratio between the branching units (sum of (1→2,5)-, (1→3,5)-, and
twice (1→2,3,5)-Araf linkages) and total Ara, of fraction Et80-II (0.29)
was similar to the initial Et80 (0.27), lower in Et80-I and Et80-III (0.21
and 0.13, respectively). The branching degree of galactans, estimated by
the ratio between the branching units ((1→3,6)-Galp linkages) and total
Gal, was also lower for fractions Et80-I and Et80-III (0.37 and 0.24,
respectively) than for fraction Et80-II and initial Et80 (0.45 and 0.42,
respectively), indicating that the branching degree of arabinans and
galactans was correlated.
To evaluate the contribution of molecular weight to fraction Et80
immunostimulatory activity, fractions Et80 I, Et80 II, and Et80 III were
incubated with spleen cells. All fractions presented immunostimulatory
activity comparable to Et80 (Fig. 2), showing a lack of molecular weightimmunostimulatory activity relationship in Et80, despite the variation
of arabinans and galactans branching degree.

increase of the relative content of terminally- and 6-linked Gal residues.
These results show that the Ara residues were removed from positions 2,
3, and 5 of arabinans and 3 of galactans, in accordance with the data
reported for other arabinans and arabinogalactans (Ferreira, Passos,
Cepeda, et al., 2018; Peng et al., 2016; Taylor et al., 2006). The α-Larabinofuranosidase treatment resulted in arabinans and galactans with
a branching degree of 0.21 and 0.22, respectively, lower than the 0.27

and 0.42 of the initial Et80 fraction. The remaining 58 % Ara residues in
Et80_Ara should be part of longer arabinan chains, as α-L-arabinofur­
anosidase preferably acts on single and short chain Ara residues (Taylor
et al., 2006).
The sample Et80_Ara was incubated with spleen cells, resulting in
stimulation of B cells to the same extension of that observed for Et80
fraction. These results show that the removal of terminal Ara or small
Ara chains did not interfere with immunostimulatory activity of broccoli
pectic polysaccharides. These results were in accordance with the lower
contribution of Ara residues of pectic polysaccharides with immuno­
modulatory activity (Inngjerdingen et al., 2007), showing the impor­
tance of the core (1→3,6)- and (1→6)-β-D-galactan for the expression of
their activity, in accordance with the reported arabinogalactans anti­
complement activity (Peng et al., 2016; Yamada & Kiyohara, 2007; Zou
et al., 2019).
3.2.1.3. Chemical sulfation of Et80 fraction. To study the effect of
chemical modification with sulfate groups, Et80 fraction was incubated
with chlorosulfonic acid and pyridine, giving origin to Et80_Sulf sub­
fraction. Et80_Sulf had 52 % of total sugars and a sugar composition
resembling Et80 fraction. The sulfation was observed in the FTIR spec­
– O absorption band at 1260 cm− 1
trum by the presence of a strong S–
(Fig. 4, Castro et al., 2006). In addition, linkage analysis showed the
increase of 2,5-, 3,5-, and 2,3,5-linked Ara, indicating the presence of
sulfate groups in 2, 3 and 5 positions. Considering Gal residues, it was
possible to observe a slight decrease in 3- and 4-Gal with an increase of
all other residues, allowing to infer some extent of sulfation of the gal­
actan moiety. Similarly, the glucan moiety seems to have been also
sulfated at position 6, inferred by the decrease of 4-Glc and the increase
of 4,6-Glc.

When incubated with spleen cells, Et80_Sulf showed the same
immunostimulatory activity of Et80 fraction. Similar results were
observed for other sulfated pectic polysaccharides that retained the
same spleen cell proliferation activity (Wang et al., 2018). It is possible
that the sulfated polysaccharides contribute positively to the

3.2.1.2. α-L-Arabinofuranosidase treatment. To evaluate the influence of
terminally-linked Ara of arabinans and arabinogalactans on the immu­
nostimulatory activity of Et80 fraction, it was treated with α-L-arabi­
nofuranosidase. This enzymatic treatment allowed to recover 79 % of
the mass of Et80 fraction and 58 % of Ara in Et80_Ara subfraction.
Et80_Ara had 61 % of total sugars and was mainly composed by UA (64
mol%), Gal (14 mol%) and Ara (10 mol%). Linkage analysis revealed
that the α-L-arabinofuranosidase treatment promoted the decrease of the
relative content of terminally-, 2,5-, and the 3,5-linked Ara residues, as
well as 3,6-linked Gal residues. This treatment allowed to observe an
7


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Fig. 4. FTIR spectra of samples Et80 and Et80_Sulf acquired by ATR.

immunostimulatory activity of the sample, but the modification of the
galactan moiety may hidden the structural galactans features also
important for the bioactivity of these polysaccharides, with a net zero
balance.

motives from hot water extracted broccoli pectic polysaccharides are
related with lateral chains rich in branched Ara with UA residues that
can confer acidity to the polymer. These structural features can be ob­
tained by enzymatic modification of the hot water extract by pectin
methylesterase and endo-polygalacturonase, followed by anionexchange chromatography. Moreover, it seems that the poly­
saccharides recovered in Enz_HWA had a combination of polysaccharide
structures that may mask the activity of Enz_HWB, as the pristine
Enz_HW, which contained the same structural features, did not present
the same extent of activity.

3.2.2. Pectin methylesterase and endo-polygalacturonase treatment of
purified HW extract
To isolate the branched regions of pectic polysaccharides, the hot
water extract from broccoli was treated with pronase and α-amylase, to
remove proteins and starch-like polysaccharides, respectively. This was
followed by treatment with pectin methylesterase and endopolygalacturonase to methyl de-esterify and hydrolyse homogalactur­
onan domains from pectic polysaccharides. After these enzymatic
treatments, 32 % of HW material was recovered as high-molecularweight material in Enz_HW fraction, which was composed by 60 % of
sugars (Table 1). This enzymatic treatment removed a large extend of
HW Glc (78 %), in accordance with the presence of starch-like poly­
saccharides. Enz_HW was composed mainly by UA (37 mol%), Ara (33
mol%), and Gal (19 mol%), showing a 84 % UA removal, in accordance
with the degradation of homogalacturonans. The remaining UA in
Enz_HW is expected to belong to small homogalacturonan regions or
regions with acetyl groups (Searle-van Leeuwen et al., 1996), as well as
rhamnogalacturonans and type II arabinogalactans containing GlcA
(Vincken et al., 2003; Yapo, 2011). Ara and Gal were also lost after the
enzymatic treatment (49 % and 47 %, respectively), indicating that
these residues were constituents of small side chains, which were
removed upon dialysis. This inference takes into consideration the

absence of side enzyme effects on other polysaccharides beyond the
homogalacturonan moieties.
To obtain pectic polysaccharides with different charge properties,
the Enz_HW fraction was submitted to an anion-exchange chromatog­
raphy (Table 1 and Fig. 3.b). As the compounds eluted mainly with
0.125 M of NaCl in the buffer, it was inferred that they constituted an
acidic fraction (Enz_HWB, 50 %). The fraction that eluted with buffer
without NaCl (Enz_HWA) was not retained in the column, accounting for
36 % of Enz_HW. This non retained fraction had lower amount of UA (34
mol%) than the retained fraction (40 mol%), allowing to infer that
Enz_HWB had more exposed UA residues than Enz_HWA. In addition,
although the amount of Ara and Gal was similar, Enz_HWB had an Ara:
Gal ratio of 2 whereas Enz_HWA had a ratio of 1. Linkage analysis
showed that Enz_HWB had also a slightly higher proportion of 3,5- and
2,3,5-linked Ara than Enz_HWA, allowing to infer the presence of ara­
binans with higher branching degree (0.33) than Enz_HWA (0.24)
(Table 2). Accordingly, 4- and 3,6-linked Gal were recovered mainly in
the non-retained fraction. Noncarbohydrate compounds from Enz_HW
(14 %) were eluted with 0.250 M of NaCl (Enz_HWC).
The incubation of Enz_HWB with spleen cells induced a stimulation
of 46 % of B lymphocytes (Fig. 2), higher than the activation observed
for the hot water extract (16 %) and even Et80 fraction (26 %),
considering the same amount of stimulus (100 μg/mL). However,
Enz_HW and Enz_HWA stimulated less B lymphocytes (13 % and 10 %,
respectively). These results allow to infer that the active structural

3.3. Detailing of structural features from broccoli pectic polysaccharides
by NMR spectroscopy
To detail the structural features of immunostimulatory poly­
saccharides from broccoli, fraction Et80, the one with higher immu­

nostimulatory activity when strategy 1 (ethanol precipitation) was used
to obtain the polysaccharides, and Enz_HWB, the fraction with higher
immunostimulatory activity derived from strategy 2 (enzymatic treat­
ment and anion exchange chromatography separation), were analysed
by NMR spectroscopy. The 13C NMR, HSQC, and HMBC spectra are
represented in Figs. 5, 6, and 7. According to these 1D and 2D NMR
spectra, methylation analysis (Table 2), and literature about pectic
polysaccharides rich in UA, arabinans, and arabinogalactans (Cardoso
et al., 2002; Dourado et al., 2006; Fernandes et al., 2019; Hamed et al.,
2022; Makarova et al., 2013; Makarova et al., 2016; Schols et al., 1990;
Shakhmatov et al., 2014; Shakhmatov et al., 2017; Shakhmatov et al.,
2019), intense signals of the anomeric carbon in the 13C NMR spectrum
were attributed to C-1 of α-Araf (106.3–109.2 ppm), α-GalpA
(99.2–100.2 ppm), and β-Galp (103.1–104.1 ppm) for both Et80 and
Enz_HWB. Based on literature about α-glucans (Chen et al., 2017; Guo
et al., 2015), the signal at 99.5 ppm of Et80 was attributed to C-1 of
α-Glcp. Low intense signals of C-1 from β-Galp (102.9 ppm), α-Rhap
(99.2 and 100.1 ppm), and β-GlcpA (102.9 ppm) were also observed for
both Et80 and Enz_HWB.
Prominent signals in the region of carboxyl groups were observed in
the 13C NMR at δC range of 170–171 ppm, indicating the existence of UA
with different vicinities. Methyl and acetyl esterification were suggested
by both one-bond 13C–1H correlation of NMR resonances, in the HSQC
spectrum (Fig. 6), and by three-bond 13C–1H correlation of NMR res­
onances, in the HMBC spectrum (Fig. 7). The signal of CH3 of methyl
esterified UA at 52.6/3.69 ppm and the signal of CH3 of acetyl esterified
UA at 20.3/2.04 ppm were observed in the HSQC spectrum. The signals
of the carboxyl carbon correlation with methyl groups of CH3CO- at
173.6/2.06 ppm and 172.9/2.05 ppm (176.0/2.09) ppm were observed
in the HMBC spectrum (Makarova et al., 2013; Shakhmatov et al., 2017,

2019). The methyl esterification was residual in Enz_HWB in contrast
with Et80 intense signal of CH3O-group, which justify different chemical
shifts for GalA anomeric carbon and C-5 (Fig. 6). Some assignments were
not possible for GalA due to overlapping of signals and to its lower de­
gree of freedom when compared to other sugar residues (Dourado et al.,
2006; Schols et al., 1990). Overall, the data indicate the presence of a
8


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Legend:

Ara

GlcA

Gal

GalA

Rha

Fuc

Glc

13

Fig. 5. Representation of the C NMR spectra from a. Et80 and b. Et80_Ara. Illustration of sugar residues using the nomenclature proposed by Neelamegham et al.
(2019). *Identification of negative signals of DEPT-135 NMR spectrum from Et80.

methyl and acetyl esterified galacturonan in Et80 fraction and an acetyl
esterified galacturonan in Enz_HWB (Table 3). Methyl esterification and
acetylation, specifically at O-2 and/or O-3 positions, were previously
observed in pectins from Brassica (Westereng et al., 2006).
In 13C and 1H NMR spectra (Fig. 5), C-6/H-6 of deoxy-sugar residues
Rha were observed at 16.1–16.8/1.11–1.18 ppm (Shakhmatov et al.,
2017). These results indicated the presence of t-Rhap, usually found in
the non-reducing terminal of arabinogalactans, and the presence of
(α1→2)-Rhap and (α1→2,4)-Rhap, typical of in RG-I regions, according
with what was observed in methylation analysis.
The negative carbon signals in the 13C DEPT-135 (Fig. 5) found at δC
69.4, 66.3–66.8, 60.9, 60.7, and 60.5 ppm can be attributed to the -CH2of sugar residues, namely the C-5 of Ara and the C-6 of Gal and Glc.
Other assignments of the proton and carbon signals of Ara, Gal, and Glc
were reported in Table 3. The identification of signals belonging to
(α1→5)-Araf, (α1→3,5)-Araf, (α1→3)-Araf, (α1→2,5)-Araf, (α1→2,3,5)Araf, and t-α-Araf indicated the presence of arabinans (Dourado et al.,
2006; Fernandes et al., 2019; Shakhmatov et al., 2014, 2017, 2019).

Downfield chemical shifts of Ara residues were also found in HSQC
spectrum, indicating linkages between Ara-Gal residues, as usually
found in arabinogalactans (Makarova et al., 2016; Shakhmatov et al.,
2014, 2017), and confirmed by the decreasing of these signals when
Et80 fraction was treated with α-L-arabinofuranosidase (Fig. 5.b), where
terminally linked Ara residues were mainly removed from galactans
(Leboeuf et al., 2004).
The identification of signals belonging to t-β-Galp, (β1→4)-Galp,
(β1→3)-Galp, (β1→6)-Galp, and (β1→3,6)-Galp agreed with the pres­

ence of 4-linked galactans, like the ones found in type I arabinoga­
lactans, and 6-linked galactans ramified at 3 like the ones found in type
II arabinogalactans. Galactans are usually found in ramified regions of
pectic polysaccharides and type II arabinogalactans can be found both as
free polysaccharides and as side chains of rhamnogalacturonans
(Makarova et al., 2016; Shakhmatov et al., 2017, 2019). In this work, no
correlation was observed between these two polysaccharides and RG-I,
which can be due to low intensity of these possible linkages.
A methyl group signal not related with carboxyl group of UA was
9


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Fig. 6. Representation of HSQC spectra of fraction Et80 and Enz_HWB. The notations used are given in Table 3.

10


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Fig. 7. Representation of HMBC spectrum of fraction Et80.
Table 3
1
H and 13C chemical shifts (δ) of main residues found in Et80 and/or Enz_HWB.
Residue

C-1

→4)-α-GalpA-(1→

GalA

→4)-α-GalpA-(1→

GalA Me

α-Rhap

Rha

→5)-α-Araf-(1→

A5

→3,5)-α-Araf-(1→

A35

→3)-α-Araf-(1→

A3

→2,5)-α-Araf-(1→

A25

α-Araf-(1→

At

→2,3,5)-α-Araf-(1→

A235

α-Araf-(1→

At*

β-Galp-(1→

Gt

→6)-β-Galp-(1→

G6

→3,6)-β-Galp-(1→

G36

→3)-β-Galp-(1→

G3

→4)-β-Galp-(1→

G4

4-OMe-α-GlcpA-(1→

GA

α-Glcp

Glc

C-2

C-3

C-4

C-5

C-6

H-1

H-2

H-3

H-4

H-5;5’

H-6;6’

100.2
4.84
100.2
4.84
100.1
4.63
107.3
4.96
107.1
5.00
107.1
5.04
107.0
5.07
107.0
5.02
106.7
5.12
109.2
5.12
103.1
4.36
103.1
4.36
103.1
4.36
104.1

4.51
104.1
4.51
102.9
4.36
99.7
5.28

–
–
–
–
–
–
81.0
4.01
79.1
4.16
81.0
4.01
86.7
4.04
81.7
3.99
84.5
4.18
–
–
70.5
3.4

70.5
3.42
69.5
3.53
70.0
3.56
71.7
3.54
72.9
3.22
–
3.53

–
3.97
–
–
–
–
76.6
3.88
83.9
3.95
83.7
3.83
79.4
4.09
76.4
3.83
80.9

4.22
–
–
72.5
3.53
72.5
3.53
79.6
3.55
80.0
3.61
73.2
3.56
73.9
–
–
–

79.0
4.33
–
–
–
–
82.3
4.08
81.1
3.89
81.3
3.99

81.5
4.17
83.9
3.92
81.9
3.99
–
–
–
3.84
–
3.80
–
3.97
–
4.04
77.5
4.05
82.1
3.23
–
–

72.5
4.5
70.5
4.98
–
–
66.5

3.67;3.75
66.8
3.67;3.75
60.9
3.60;3.69
66.8
3.67;3.75
60.9
3.60;3.69
66.5
3.67;3.75
–
–
–
3.53
–
3.84
–
3.84
–
3.53
–
3.54
–
4.17
–
–

170.3

found in the HSQC spectrum at 60.0/3.35 ppm and the chemical shift of
C-4/H-4 at 82.1/3.23 ppm confirmed the existence of 4-O-Me-GlcA. A
cross-peak at 82.1/3.35 ppm in the HMBC spectrum confirmed an Omethyl substituent at C-4 of β-D-GlcpA. According to literature, these
residues can occur as terminals of side chains of (β1→6)-Galp chains
from type II arabinogalactans (Shakhmatov et al., 2014; Shakhmatov

–
16.1–16.8
1.04–1.11

60.4
3.60;3.69
69.2
3.89;3.80
69.2
3.89;3.80
60.4
3.60;3.69
60.4
3.60;3.69
–
–
–
–

CH3O-

52.6
3.67

CH3COO
20.3
2.04

60
3.35

et al., 2017).
NMR spectra interpretation corroborated glycosidic linkage data
obtained by methylation analysis. Furthermore, data revealed that Et80
and Enz_HWB had acetyl-esterified pectic polysaccharides and type II
arabinogalactans with 4-O-Me-GlcpA residues. NMR spectra also
allowed to identify the main differences between fractions Et80 and
11


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Enz_HWB as α-glucans and highly methyl-esterified pectic poly­
saccharides in Et80 fraction. These polysaccharides seem to hinder the
immunostimulatory activity of slightly charged pectic polysaccharides
composed with galactans motifs and with branched Ara residues.

2020, UIDP/50011/2020 & LA/P/0006/2020), and CITAB research
Unit (FCT UID/AGR/04033/2019), for the financial support through
national funds and, where applicable, co-financed by the FEDER, within
the PT2020 Partnership Agreement. SSF thank FCT for the individual
doctoral grant (SFRH/BD/103003/2014). The authors also thank

Monliz SA, Alpiarỗa, Portugal, for providing the broccoli by-products.

4. Concluding remarks

References

The results herein collected showed that broccoli by-products are a
source of pectic polysaccharides with in vitro stimulatory activity on B
lymphocytes. Immunostimulatory pectic polysaccharides can be ob­
tained by preparation of an AIR, which can be stabilized by solvent
drying, and extraction with hot water. The highest immunostimulatory
activity was observed for the acidic fraction of pectic polysaccharides
purified with pronase and α-amylase, methyl de-esterified with pectin
methylesterase, and partially degraded with endo-polygalacturonase
(Enz_HWB). After these treatments to remove protein, starch, and
partially degrade the polygalacturonan backbone, Enz_HWB was a
fraction composed by galacturonans (40 mol%) residually methylesterified but acetyl-esterified, with ramified regions of 4-linked gal­
actans (0.5 mol%), arabinans (35 mol%, with a branching degree of
0.33), and type II arabinogalactans (15 mol%, with a branching degree
of 0.32) with 4-O-Me-GlcA residues.
Pectic polysaccharides with immunostimulatory activity, although
not as high as that observed for Enz_HWB, can also be obtained by a
simple 80 % ethanol precipitation (Et80) of the hot water extract. The
lower immunostimulatory activity of Et80 can be due to the presence of
starch derived α-glucans (7.6 mol%) and highly methyl esterified uronic
acids. The activity of the compounds in these fractions seems not to be
dependent of their molecular weight (from 12 to 400 kDa), neither of the
lower content of terminally linked single or small chain Ara residues (14
mol% of arabinans, with a branching degree of 0.27). Et80 also con­
tained ramified regions of 4-linked galactans (1.0 mol%) and highly

ramified type II arabinogalactans (5.2 mol%, with a branching degree of
0.82).
The presence of pectic polysaccharides with potential immunosti­
mulatory activity in broccoli by-products showed that these by-products
can be valorised as functional food ingredients that improve immune
function and promote health.

Bastos, R., Coelho, E., & Coimbra, M. A. (2015). Modifications of Saccharomyces
pastorianus cell wall polysaccharides with brewing process. Carbohydrate Polymers,
124, 322–330.
Beukema, M., Faas, M. M., & de Vos, P. (2020). The effects of different dietary fiber
pectin structures on the gastrointestinal immune barrier: impact via gut microbiota
and direct effects on immune cells. Experimental & Molecular Medicine, 52,
1364–1376.
Blakeney, A. B., Harris, P. J., Henry, R. J., & Stone, B. A. (1983). A simple and rapid
preparation of alditol acetates for monosaccharide analysis. Carbohydrate Research,
113, 291–299.
Blumenkrantz, N., & Asboe-Hansen, G. (1973). New method for quantitative
determination of uronic acid. Analytical Biochemistry, 54, 484–489.
Busato, B., de Almeida Abreu, E. C., de Oliveira Petkowicz, C. L., Martinez, G. R., &
Noleto, G. R. (2020). Pectin from Brassica oleracea var. italica triggers
immunomodulating effects in vivo. International Journal of Biological Macromolecules,
161, 431–440.
Cardoso, S. M., Silva, A. M. S., & Coimbra, M. A. (2002). Structural characterisation of
the olive pomace pectic polysaccharide arabinan side chains. Carbohydrate Research,
337, 917–924.
Castro, R., Piazzon, M. C., Zarra, I., Leiro, J., Noya, M., & Lamas, J. (2006). Stimulation of
turbot phagocytes by Ulva rigida C. Agardh polysaccharides. Aquaculture, 254, 9–20.
Chen, Z.-E., Wufuer, R., Ji, J.-H., Li, J.-F., Cheng, Y.-F., Dong, C.-X., & Taoerdahong, H.
(2017). Structural characterization and immunostimulatory activity of

polysaccharides from Brassica rapa L. Journal of Agricultural and Food Chemistry, 65,
9685–9692.
Ciucanu, I., & Kerek, F. (1984). A simple and rapid method for the permethylation of
carbohydrates. Carbohydrate Research, 131, 209–217.
Coimbra, M. A., Delgadillo, I., Waldron, K. W., & Selvendran, R. R. (1996). Isolation and
analysis of cell wall polymers from olive pulp. In Plant cell wall analysis (pp. 19–44).
Berlin, Heidelberg: Springer.
de Santana-Filho, A. P., Noleto, G. R., Gorin, P. A. J., de Souza, L. M., Iacomini, M., &
Sassaki, G. L. (2012). GC–MS detection and quantification of lipopolysaccharides in
polysaccharides through 3-O-acetyl fatty acid methyl esters. Carbohydrate Polymers,
87, 27302734.
Diallo, D., Paulsen, B. S., Liljebă
ack, T. H. A., & Michaelsen, T. E. (2001). Polysaccharides
from the roots of Entada africana Guill. et Perr., Mimosaceae, with complement
fixing activity. Journal of Ethnopharmacology, 74, 159171.
Diallo, D., Paulsen, B. S., Liljebă
ack, T. H. A., & Michaelsen, T. E. (2003). The malian
medicinal plant Trichilia emetica; studies on polysaccharides with complement fixing
ability. Journal of Ethnopharmacology, 84, 279–287.
Dourado, F., Cardoso, S. M., Silva, A. M. S., Gama, F. M., & Coimbra, M. A. (2006). NMR
structural elucidation of the arabinan from Prunus dulcis immunobiological active
pectic polysaccharides. Carbohydrate Polymers, 66, 27–33.
Dourado, F., Madureira, P., Carvalho, V., Coelho, R., Coimbra, M. A., Vilanova, M., …
Gama, F. M. (2004). Purification, structure and immunobiological activity of an
arabinan-rich pectic polysaccharide from the cell walls of Prunus dulcis seeds.
Carbohydrate Research, 339, 2555–2566.
Du, X.-j., Zhang, J.-s., Yang, Y., Tang, Q.-j., Jia, W., & Pan, Y.-j. (2010). Purification,
chemical modification and immunostimulating activity of polysaccharides from
Tremella aurantialba fruit bodies. Journal of Zhejiang University SCIENCE B, 11,
437–442.

Duan, J., Dong, Q., Ding, K., & Fang, J. (2010). Characterization of a pectic
polysaccharide from the leaves of Diospyros kaki and its modulating activity on
lymphocyte proliferation. Biopolymers, 93, 649–656.
Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., & Smith, F. (1956). Colorimetric
method for determination of sugars and related substances. Analytical Chemistry, 28,
350–356.
Farache, J., Zigmond, E., Shakhar, G., & Jung, S. (2013). Contributions of dendritic cells
and macrophages to intestinal homeostasis and immune defense. Immunology and cell
biology, 91, 232–239.
Femenia, A., Bestard, M., Sanjuan, N., Rossell´
o, C., & Mulet, A. (2000). Effect of
rehydration temperature on the cell wall components of broccoli (Brassica oleracea L.
Var. italica) plant tissues. Journal of Food Engineering, 46, 157–163.
Fernandes, P. A. R., Le Bourvellec, C., Renard, C. M. G. C., Wessel, D. F., Cardoso, S. M., &
Coimbra, M. A. (2020). Interactions of arabinan-rich pectic polysaccharides with
polyphenols. Carbohydrate Polymers, 230, Article 115644.
Fernandes, P. A. R., Silva, A. M. S., Evtuguin, D. V., Nunes, F. M., Wessel, D. F.,
Cardoso, S. M., & Coimbra, M. A. (2019). The hydrophobic polysaccharides of apple
pomace. Carbohydrate Polymers, 223, Article 115132.
Ferreira, S. S., Monteiro, F., Passos, C. P., Silva, A. M. S., Wessel, D. F., Coimbra, M. A., &
Cardoso, S. M. (2020). Blanching impact on pigments, glucosinolates, and phenolics
of dehydrated broccoli by-products. Food Research International, 132.

CRediT authorship contribution statement
´ nia S. Ferreira: Conceptualization, Investigation, Methodology,
So
Data curation, Writing – original draft. Alexandra Correia: Investiga­
tion, Methodology, Data curation, Writing – review & editing. Artur M.
S. Silva: Methodology, Data curation, Writing – review & editing.
Dulcineia Ferreira Wessel: Supervision, Writing – review & editing.

Susana M. Cardoso: Supervision, Writing – review & editing. Manuel
Vilanova: Supervision, Writing – review & editing. Manuel A. Coim­
bra: Conceptualization, Data curation, Supervision, Writing – review &
editing.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Data availability
Data will be made available on request.
Acknowledgements
Thanks are due to University of Aveiro, Polytechnic Institute of
Viseu, FCT/MCT, LAQV-REQUIMTE (UIDB/50006/2020 & UIDP/
50006/2020), and CICECO-Aveiro Institute of Materials (UIDB/50011/
12


S.S. Ferreira et al.

Carbohydrate Polymers 303 (2023) 120432

Ferreira, S. S., Passos, C. P., Cardoso, S. M., Wessel, D. F., & Coimbra, M. A. (2018).
Microwave assisted dehydration of broccoli by-products and simultaneous extraction
of bioactive compounds. Food Chemistry, 246, 386–393.
Ferreira, S. S., Passos, C. P., Cepeda, M. R., Lopes, G. R., Teixeira-Coelho, M.,
Madureira, P., … Coimbra, M. A. (2018). Structural polymeric features that
contribute to in vitro immunostimulatory activity of instant coffee. Food Chemistry,
242, 548–554.
Ferreira, S. S., Passos, C. P., Madureira, P., Vilanova, M., & Coimbra, M. A. (2015).
Structure–function relationships of immunostimulatory polysaccharides: A review.

Carbohydrate Polymers, 132, 378396.
Gonỗalves, F. J., Fernandes, P. A. R., Wessel, D. F., Cardoso, S. M., Rocha, S. M., &
Coimbra, M. A. (2018). Interaction of wine mannoproteins and arabinogalactans
with anthocyanins. Food Chemistry, 243, 1–10.
Guo, Q., Cui, S. W., Kang, J., Ding, H., Wang, Q., & Wang, C. (2015). Non-starch
polysaccharides from American ginseng: physicochemical investigation and
structural characterization. Food Hydrocolloids, 44, 320–327.
Hamed, M., Coelho, E., Bastos, R., Evtuguin, D. V., Ferreira, S. S., Lima, T., …
Bougatef, A. (2022). Isolation and identification of an arabinogalactan extracted
from pistachio external hull: Assessment of immunostimulatory activity. Food
Chemistry, 373, Article 131416.
Harris, P. J., Henry, R. J., Blakeney, A. B., & Stone, B. A. (1984). An improved procedure
for the methylation analysis of oligosaccharides and polysaccharides. Carbohydrate
Research, 127, 59–73.
Huang, X., Nie, S., & Xie, M. (2017). Interaction between gut immunity and
polysaccharides. Critical Reviews in Food Science and Nutrition, 57, 2943–2955.
Inngjerdingen, K. T., Kiyohara, H., Matsumoto, T., Petersen, D., Michaelsen, T. E.,
Diallo, D., … Paulsen, B. S. (2007). An immunomodulating pectic polymer from
Glinus oppositifolius. Phytochemistry, 68, 1046–1058.
Isogai, A., Ishizu, A., Nakano, J., Eda, S., & Kat¯
o, K. (1985). A new facile methylation
method for cell-wall polysaccharides. Carbohydrate research, 138, 99–108.
Jin, M.-Y., Li, M.-Y., Huang, R.-M., Wu, X.-Y., Sun, Y.-M., & Xu, Z.-L. (2021). Structural
features and anti-inflammatory properties of pectic polysaccharides: A review.
Trends in Food Science & Technology, 107, 284–298.
Kiyohara, H., Uchida, T., Takakiwa, M., Matsuzaki, T., Hada, N., Takeda, T., Shibata, T.,
& Yamada, H. (2010). Different contributions of side-chains in β-d-(1→3,6)-galactans
on intestinal Peyer’s patch-immunomodulation by polysaccharides from Astragalus
mongholics Bunge. Phytochemistry, 71, 280–293.
Knee, M. (1973). Polysaccharides and glycoproteins of apple fruit cell walls.

Phytochemistry, 12, 637653.
Komban, R. J., Stră
omberg, A., Biram, A., Cervin, J., Lebrero-Fern´
andez, C., Mabbott, N.,
… Lycke, N. (2019). Activated Peyer′ s patch B cells sample antigen directly from M
cells in the subepithelial dome. Nature Communications, 10, 2423.
Leboeuf, E., Thoiron, S., & Lahaye, M. (2004). Physico-chemical characteristics of cell
walls from Arabidopsis thaliana microcalli showing different adhesion strengths.
Journal of Experimental Botany, 55, 20872097.
Lindberg, B., & Lă
onngren, J. (1978). Methylation analysis of complex carbohydrates:
General procedure and application for sequence analysis. Methods in Enzymology, 50,
3–33.
Makarova, E. N., Patova, O. A., Shakhmatov, E. G., Kuznetsov, S. P., & Ovodov, Y. S.
(2013). Structural studies of the pectic polysaccharide from Siberian fir (Abies sibirica
Ledeb.). Carbohydrate Polymers, 92, 1817–1826.
Makarova, E. N., Shakhmatov, E. G., & Belyy, V. A. (2016). Structural characteristics of
oxalate-soluble polysaccharides of Sosnowsky’s hogweed (Heracleum sosnowskyi
Manden). Carbohydrate Polymers, 153, 66–77.
Martins, V. M. R., Sim˜
oes, J., Ferreira, I., Cruz, M. T., Domingues, M. R., &
Coimbra, M. A. (2017). In vitro macrophage nitric oxide production by Pterospartum
tridentatum (L.) Willk. inflorescence polysaccharides. Carbohydrate Polymers, 157,
176–184.
Neelamegham, S., Aoki-Kinoshita, K., Bolton, E., Frank, M., Lisacek, F., Lütteke, T., …
Woods, R. J. (2019). Updates to the symbol nomenclature for glycans guidelines.
Glycobiology, 29, 620–624.
Nergard, C. S., Matsumoto, T., Inngjerdingen, M., Inngjerdingen, K., Hokputsa, S.,
Harding, S. E., … Yamada, H. (2005). Structural and immunological studies of a
pectin and a pectic arabinogalactan from Vernonia kotschyana Sch. Bip. ex Walp.

(Asteraceae). Carbohydrate Research, 340, 115–130.
Nunes, C., Silva, L., Fernandes, A. P., Guin´e, R. P. F., Domingues, M. R. M., &
Coimbra, M. A. (2012). Occurrence of cellobiose residues directly linked to
galacturonic acid in pectic polysaccharides. Carbohydrate Polymers, 87, 620–626.
Passos, C. P., & Coimbra, M. A. (2013). Microwave superheated water extraction of
polysaccharides from spent coffee grounds. Carbohydrate Polymers, 94, 626–633.
Passos, C. P., Costa, R. M., Ferreira, S. S., Lopes, G. R., Cruz, M. T., & Coimbra, M. A.
(2021). Role of coffee caffeine and chlorogenic acids adsorption to polysaccharides
with impact on brew immunomodulation effects. Foods, 10, 378.
Peng, Q., Liu, H., Lei, H., & Wang, X. (2016). Relationship between structure and
immunological activity of an arabinogalactan from Lycium ruthenicum. Food
Chemistry, 194, 595–600.

Petkowicz, C. L., & Williams, P. A. (2020). Pectins from food waste: Characterization and
functional properties of a pectin extracted from broccoli stalk. Food Hydrocolloids,
107, Article 105930.
Popov, S. V., & Ovodov, Y. S. (2013). Polypotency of the immunomodulatory effect of
pectins. Biochemistry (Moscow), 78, 823–835.
Ramberg, J. E., Nelson, E. D., & Sinnott, R. A. (2010). Immunomodulatory dietary
polysaccharides: A systematic review of the literature. Nutrition Journal, 9, 54.
Schă
afer, J., Stanojlovic, L., Trierweiler, B., & Bunzel, M. (2017). Storage related changes
of cell wall based dietary fiber components of broccoli (Brassica oleracea var. italica)
stems. Food Research International, 93, 43–51.
Schols, H. A., Posthumus, M. A., & Voragen, A. G. J. (1990). Structural features of hairy
regions of pectins isolated from apple juice produced by the liquefaction process.
Carbohydrate Research, 206, 117–129.
Searle-van Leeuwen, M. J. F., Vincken, J.-P., Schipper, D., Voragen, A. G. J., & Beldman,
G. (1996). Acetyl esterases of Aspergillus niger: Purification and mode of action on
pectins. In J. Visser & A. G. J. B. T.-P. in B. Voragen (Eds.), Pectins and pectinases,

pp. 793–798. Elsevier.
Selvendran, R. R., March, J. F., & Ring, S. G. (1979). Determination of aldoses anduronic
acid content of vegetable fiber. Analytical Biochemistry, 96, 282–292.
Shakhmatov, E. G., Belyy, V. A., & Makarova, E. N. (2017). Structural characteristics of
water-soluble polysaccharides from Norway spruce (Picea abies). Carbohydrate
Polymers, 175, 699–711.
Shakhmatov, E. G., Makarova, E. N., & Belyy, V. A. (2019). Structural studies of
biologically active pectin-containing polysaccharides of pomegranate Punica
granatum. International Journal of Biological Macromolecules, 122, 29–36.
Shakhmatov, E. G., Toukach, P. V., Michailowa, E. A., & Makarova, E. N. (2014).
Structural studies of arabinan-rich pectic polysaccharides from Abies sibirica L.
Biological activity of pectins of A. sibirica. Carbohydrate Polymers, 113, 515–524.
Stevens, B. J. H., & Selvendran, R. R. (1980). Structural investigation of an arabinan from
cabbage (Brassica oleracea var. capitata). Phytochemistry, 19, 559–561.
Taylor, E. J., Smith, N. L., Turkenburg, J. P., D’Souza, S., Gilbert, H. J., & Davies, G. J.
(2006). Structural insight into the ligand specificity of a thermostable family 51
arabinofuranosidase, Araf51, from Clostridium thermocellum. Biochemical Journal,
395, 31–37.
Togola, A., Inngjerdingen, M., Diallo, D., Barsett, H., Rolstad, B., Michaelsen, T. E., &
Paulsen, B. S. (2008). Polysaccharides with complement fixing and macrophage
stimulation activity from Opilia celtidifolia, isolation and partial characterisation.
Journal of ethnopharmacology, 115, 423–431.
Vincken, J.-P., Schols, H. A., Oomen, R. J. F. J., Beldman, G., Visser, R. G. F., & Voragen,
A. G. J. (2003). Pectin - The hairy thing. In F. Voragen, H. Schols, & R. Visser (Eds.),
Advances in pectin and pectinase research, 47–59. Dordrecht: Springer Netherlands.
Waldron, K. W., & Faulds, C. B. (2007). Cell wall polysaccharides: Composition and
structure. In Comprehensive Glycoscience (pp. 181–201). Elsevier.
Wang, X. S., Liu, L., & Fang, J. N. (2005). Immunological activities and structure of
pectin from Centella asiatica. Carbohydrate Polymers, 60, 95–101.
Wang, Z., Cai, T., & He, X. (2018). Characterization, sulfated modification and

bioactivity of a novel polysaccharide from Millettia dielsiana. International Journal of
Biological Macromolecules, 117, 108–115.
Westereng, B., Coenen, G. J., Michaelsen, T. E., Voragen, A. G. J., Samuelsen, A. B.,
Schols, H. A., & Knutsen, S. H. (2009). Release and characterization of single side
chains of white cabbage pectin and their complement-fixing activity. Molecular
Nutrition & Food Research, 53, 780–789.
Westereng, B., Michaelsen, T. E., Samuelsen, A. B., & Knutsen, S. H. (2008). Effects of
extraction conditions on the chemical structure and biological activity of white
cabbage pectin. Carbohydrate Polymers, 72, 32–42.
Westereng, B., Yousif, O., Michaelsen, T. E., Knutsen, S. H., & Samuelsen, A. B. (2006).
Pectin isolated from white cabbage–structure and complement-fixing activity.
Molecular Nutrition & Food Research, 50, 746–755.
Xu, L., Cao, J., & Chen, W. (2015). Structural characterization of a broccoli
polysaccharide and evaluation of anti-cancer cell proliferation effects. Carbohydrate
polymers, 126, 179–184.
Yamada, H., & Kiyohara, H. (2007). 4.34 – Immunomodulating activity of plant
polysaccharide structures. In Comprehensive Glycoscience (pp. 663–694). Elsevier.
Yapo, B. M. (2011). Rhamnogalacturonan-I: A structurally puzzling and functionally
versatile polysaccharide from plant cell walls and mucilages. Polymer Reviews, 51,
391–413.
Yin, M., Zhang, Y., & Li, H. (2019). Advances in research on immunoregulation of
macrophages by plant polysaccharides. Frontiers in immunology, 10, 145.
Zhang, Y., Jiang, Z., Wang, L., & Xu, L. (2017). Extraction optimization, antioxidant, and
hypoglycemic activities in vitro of polysaccharides from broccoli byproducts. Journal
of Food Biochemistry, 41, Article e12387.
Zou, Y.-F., Fu, Y.-P., Chen, X.-F., Austarheim, I., Inngjerdingen, K. T., Huang, C., …
Paulsen, B. S. (2017). Polysaccharides with immunomodulating activity from roots
of Gentiana crassicaulis. Carbohydrate Polymers, 172, 306–314.
Zou, Y.-F., Zhang, Y.-Y., Fu, Y.-P., Inngjerdingen, K. T., Paulsen, B. S., Feng, B., … Yin, Z.Q. (2019). A Polysaccharide Isolated from Codonopsis pilosula with
Immunomodulation Effects Both In Vitro and In Vivo. Molecules, 24, 3632.

13