*2.1. Carrageenans*

Polysaccharides were extracted from red seaweed, namely *Chondrus armatus* and *Tichocarpus crinitus*, with water at 90 ◦C. The obtained extracts were purified of low-molecular-weight impurities by column filtration, and polysaccharides were precipitated from solutions by precipitation in alcohol. The yield of polysaccharides from *C. armatus* was 50% (designated as unfractionated or total polysaccharide ( Σ)), and the yield from *T. crinitus* was 30%. The polysaccharides were separated using 4% KCl into KCl-insoluble and KCl-soluble fractions. The structures of the polysaccharides were studied by 13C-NMR and FTIR spectroscopy. The obtained spectra were compared with the spectra of polysaccharides isolated by us earlier from these species of algae [37,38]. The identity of the spectra indicated that the KCl-insoluble fraction from *C. armatus* was κ-CRG (G4S-DA-carrabiose) with traces of ι-CRG (G4S-DA2S-carrabiose), which were randomly distributed along the polysaccharide chain as a single insertion [37]. The KCl-soluble fraction from *C. armatus* was represented mainly by λ-CRG. According to data obtained by 13C-NMR and FTIR spectroscopy, the KCl-insoluble polysaccharides from *T. crinitus* had hybrid structures. On the basis of the analysis of spectra compared with data obtained previously, these polysaccharides were identified as κ/β-CRG [38]. Therefore, CRGs are differentiated by the presence of 3,6-anhydro-D-galactose in κ- and κ/β-CRG as well as by the different numbers and positions of ester sulfate groups. λ-CRG has a higher degree of sulfatation.

The chemical structures of the disaccharide repeating units of the CRGs and viscometric molecular weights of CRGs [39], calculated by the Mark–Kuhn–Houwink equation, are listed in Table 1.



#### *2.2. Atomic Force Microscopy of LPS and LPS–GRG Mixtures*

The influence of CRGs on the morphology of LPS was investigated by atomic force microscopy (AFM). According to the AFM data, κ-CRG at 0.1 mg/mL (Figure 1a) is a densely branched network structure formed by fibers with a height of about 1.0–1.5 nm and a lateral size of 51 ± 15 nm (root mean square (σRMS) = 0.726 nm). The network structure was also observed for κ/β-CRG at 0.1 mg/mL (Figure 1b). The network is also formed by fibers but they are grouped into bundles and as a result the network looks less dense than the network of κ-CRG. These fibers have a height of about 0.9–1.5 nm, with a thickness of 36 ± 11 nm (σRMS = 0.705 nm). λ-CRG differs from the other investigated carrageenan types with a high sulfation degree and the absence of 3,6- anhydrogalactose. λ-CRG is presented in the form of a random coil (Figure 1c). According to the AFM data, λ-CRG at 0.1 mg/mL forms an unordered structure consisting of particles with an irregular spherical segmen<sup>t</sup> (σRMS = 0.951 nm). The density of the particles is about 1.1 × 1010 cm<sup>−</sup>1, they are distributed almost uniformly over the surface, but there are a number of particle agglomerations. The height and lateral size of the particles are about 1.8–2.0 nm and 38 ± 9 nm, respectively.

The AFM image of *E. coli* LPS shows micelle-like particles typical for amphiphilic polymers (σRMS = 25.859 nm). The large particles easily seen in Figure 1d have a density of 2.6 × 108 cm<sup>−</sup><sup>2</sup> and they are nonuniformly distributed over the sample surface. The average large particle height is about 50.9 nm. Most particles have an elongated irregular shape with a large lateral size of 651 ± 452 nm and a small lateral size of 334 ± 115 nm; the aspect ratio is about 2. A detailed analysis of the large particles revealed that they consist of small particles, of which the lateral size can only be roughly estimated to be about 40 nm because of a lack of resolution.

**Figure 1.** AFM topography images: (**a**) κ-CRG, 0.1 mg/mL; (**b**) κ/β-CRG, 0.1 mg/mL; (**c**) lambda-CRG, 0.1 mg/mL; and (**d**) LPS, 0.05 mg/mL.

The image of the *E. coli* LPS with CRG mixtures at the 2:1 *w*/*w* ratio between components (Figure 2) showed that the macromolecular structures of the mixtures are quite different from the structures of the initial components at the same concentrations. In the case of κ-CRG and κ/β-CRG, the number of separate fibers is significantly decreased. In the complex κ-CRG + LPS, one can see fragments of the network with many spherical particles as can be seen in clean *E. coli* LPS (Figure 2a), and some worm-like structures (Figure 2a, inset). The lateral size and height of the spherical particles are 35 ± 3 and 1.65 ± 0.60 nm, respectively, while the width and height of the fibers in the worm-like structures are 60 ± 18 and 7.0 ± 1.1 nm, respectively. The fibers in the κ/β-CRG + LPS mixture are organized in a continuous network with round-shaped holes (Figure 2b), while the worm-like structures are either incorporated into the network or just lie over it. The width and height of the fibers in the worm-like structures are 49 ± 11 and 6.6 ± 1.4 nm, respectively, which are slightly smaller than the corresponding values in the κ-CRG + LPS mixture. The value of σRMS for both mixtures (Figure 2a,b) increased to more than twice that for the initial CRGs: 2.594 nm for κ-CRG + LPS and 1.989 nm for κ/β-CRG + LPS.

The morphology of λ-CRG in the complex with *E. coli* LPS also changed (Figure 2c). Instead of separated particles, one can see short fibers that are uniformly distributed over the surface rather than arranged in a network. The width and height of the fibers are 27 ± 3 and 0.74 ± 0.15 nm, respectively. *E. coli* LPS in this mixture also forms the worm-like structures that are incorporated into a layer of uniformly distributed fibers of λ-CRG. The width and height of the fibers in the worm-like structures are 62 ± 13 and 8.6 ± 2.0 nm, respectively.

**Figure 2.** AFM topography images of mixtures in ratio 2:1 *w*/*w*: (**a**) κ-CRG + LPS; (**b**) κ/β-CRG + LPS; and (**c**) λ-CRG + LPS.

#### *2.3. E*ff*ect of LPS and CRGs on the Production of Cytokines by Human Cells*

Experiments were performed to determine the effect of CRGs on the activation of the synthesis of LPS-induced IL-10 and TNF-α in human blood cells.

As shown in Figure 3a, CRGs induced the synthesis of TNF in cells only at high polysaccharide concentrations. Σ-CRG at C = 1 μg/mL showed a high activity with respect to the synthesis of the pro-inflammatory cytokine, similar to the effect of LPS at C = 10 μg/mL. At the same time, CRGs induced the synthesis of IL-10 in a dose-dependent manner similar to LPS. κ/β-CRG can induce IL-10 synthesis and is weakly dependent on the polysaccharide concentration.

**Figure 3.** TNF-α (**a**) and IL-10 (**b**) levels stimulated by *E. coli* LPS and carrageenans. Concentrations of samples: 10 μg/mL (white column), 1 μg/mL (black), 100 ng/mL (gray), 10 ng/mL (light gray), and 1 ng/mL (black-dot). Mean (±SD) contents of cytokine in serum are presented. Whole blood was obtained from five healthy subjects and incubated with the samples at different concentrations. The level of cytokines in serum of normal donors was considered as a negative control used for statistical significance calculation. \* Differences between the samples and the control were significant, *p* < 0.05.

The preliminary incubation of cells with LPS (C = 10 μg/mL) followed by treatment with λ-CRG did not have an influence on IL-10 production, whereas treatment with κ- and κ/β-CRG (Figure 4) increased the production of IL-10 and slightly reduced the synthesis of TNF in cells.

**Figure 4.** TNF-α (**a**) and IL-10 (**b**) levels stimulated by preliminary incubation cells with *E. coli* LPS (10 min), and then CRGs. Concentrations of samples: 10 μg/mL (white column), 1 μg/mL (black); 100 ng/mL (gray); 10 ng/mL (light gray), and 1 ng/mL (black-dot). Mean (±SD) contents of cytokine in serum are presented. The level of cytokines in serum stimulated by *E. coli* LPS (10 μg/mL) was considered as a control used for statistical significance calculation. \* Differences between samples and the control were significant, *p* < 0.05.

#### *2.4. E*ff*ect of CRGs on Stress Reaction of Mice Induced with E. coli LPS*

When considering the effect of *E. coli* LPS injection for five days, it was found that the physiological status of animals exhibits considerable changes. The stress response to LPS- intoxication in mice is manifested by thymus involution, thyroid atrophy, and adrenal gland hypertrophy, at the same time as an increase of the serum corticosterone level, a decrease of the concentrations of adenosine triphosphate (ATP) and glycogen in liver, and lactate acidosis (Table 2).

The results of this study show that administration of κ-CRG, κ/β–CRG, and --CRG led to the minimization of physiological disorders and metabolic homeostasis in mice exposed to LPS- intoxication. The relative masses of the thymus, thyroid, and adrenal glands of animals in the LPS + --CRG group differed from the norm only by 22%, 10%, and 12%, respectively, whereas, in the mice of LPS group, the deviations were 36%, 21%, and 30%, respectively. In the LPS + κ/β-CRG group, the relative masses of the thymus, thyroid and adrenals were less than in the control group by 16%, 7% and 5%, respectively (Table 2). After the administration of the κ- and κ/β-CRGs, the serum corticosterone level in mice intoxicated with *E. coli* LPS were reduced by 32% relative to the LPS group. The administration of --CRG and κ/β-CRG led to 22% and 17% lowered lactate storage in mice liver relative to the LPS group. The use of κ-CRG, κ/β-CRG, and --CRG contributed to more e ffective conservation of the energetic substrate ATP and glycogen in liver. The concentration of ATP and glycogen in mice liver were on average 18% higher than in the LPS group.

In this study, κ-CRG significantly minimized adrenal gland hypertrophy, and the depletion of ATP and glycogen in mice liver, whereas λ-CRG did not have an e ffect on the stress response of mice exposed to *E. coli* LPS.

**Table 2.** The e ffect of di fferent types of CRG on some physiological parameters in mice intoxicated with *E. coli* LPS.


BM, body mass; ATP, adenosine triphosphate; mean ± SD (n = 8 observations); a *p* < 0.05 compared with the control group; b *p* < 0.05 compared with the LPS group used Student's *t*-test.

#### *2.5. The E*ff*ects of Food Supplement Carrageenan-FE on the Immune System and Hemostasis Parameters in Patients with Food Borne Toxicoinfection*

The medico-biological study was carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) and the biologically active food supplement Carrageenan-FE based on the -- CRG, was used.

Investigation of the parameters of the hemostasis system showed that 18 patients with food borne toxicoinfection, caused by *Salmonella enterica*, showed signs of a hypercoagulation (first phase of disseminated intravascular coagulation) and 24 patients showed symptoms of a hypocoagulation (second phase syndrome). For patients with hypercoagulation, there was a reduction of platelet aggregation by an average of 65%, and, in patients with hypocoagulation, the degree of aggregation increased on average by 22%. In patients on the third day of treatment with CRG, there was a decrease of leukocytosis, decreasing to normal amounts of activated immune cells, and an increase in relation to the total population of T-lymphocytes and the absolute number of lymphocytes in peripheral blood and their immunoregulatory subpopulations (Table 3).


**Table 3.** The average values of immunological parameters in patients with acute enteric infections in the course of the disease with treatment Carrageenan-FE.

mean ± SD (n, see Table 3); \* *p* < 0.05 compared with the control group; † *p* < 0.05 compared with the without CRG-FE group; ‡ *p* < 0.05 compared with the on admission to hospital group used Student's *t*-test.
