2.2.4. Gut Flora Results Study 2

A permutational multivariate analysis of variance was used to determine if there were any significant differences in gut microbiota between the two treatment regimens across the three time points, resulting in comparisons between six groups. There were no significant differences between the six groups (*p* > 0.05). A correlation plot was then used to assess the movement of gut microbiota per regime group and time point (Figure 6), on which each data point represented a time point per regime group. After visual inspection of the correlation plot, it was apparent that the two regime groups at baseline varied greatly, with movement occurring at each time point. A SIMPER analysis was conducted to determine whether there were any consistent increases or decreases in genera common to both the active treatments and not the placebo treatment. There were three genera identified that consistently changed in both regimes after the SXRG84 treatment with no effect or an opposing effect in the placebo groups—these were an increase in both *Fusicatenibacter* and *Parabacteroides* and a decrease in *Clostridium.*

There were a number of genera at baseline that significantly correlated with baseline measures of weight, BMI, waist circumference, lipids, glucose, CRP, and cytokines, which are summarized in Table 8. There were also significant correlations between the change of cytokines that were observed in this study (IFNγ, IL-1β, TNF-α, and IL-10) and the change in specific gut genera (below, Table 9).

**Figure 6.** Correlation plot of gut microbiome movement per treatment group. Green triangles are regime AB (starting on placebo then crossing to SXRG84 treatment), blue triangles are regime BA (starting on SXRG84 treatment then crossing to placebo). T1 is baseline measure, T2 is measure from six weeks, and T3 is measure from 12 weeks.

**Table 9.** Cytokines that significantly changed in Study 2 correlate with a change in specific gut microbiota genera.


#### **3. Discussion**

This research presents two approaches to clinical studies in humans following the ingestion of a unique ulvan polysaccharide, SXRG-84. The studies explored the intervention effects on metabolic health outcomes wherein the links between lipid markers, inflammation status, and gut microbiota composition were recognized.

The primary outcome measure of Study 1 was plasma lipids, for which a significant 10% reduction in non-HDL cholesterol was observed in overweight participants. Study 2 was then powered to show the reduction in non-HDL cholesterol seen in Study 1, although this was not confirmed with potentially less metabolically challenged participants at baseline.

There was strong agreement between the two studies that dietary SXRG84 effectively reduced inflammatory markers. In the first study, the marker CRP was significantly reduced (−27%) in the 4 g/day dose group. In Study 2, a wider range of proinflammatory cytokines were reduced: IFN-γ (3.4 vs. 7.3 pg/mL), IL-1β (16.2 vs. 23.2 pg/mL), and TNF-α (9.3 vs. 12.6 pg/mL), as well as the anti-inflammatory cytokine IL-10 (1.6 vs. 2.1 pg/mL) (*p* < 0.05). These marked findings indicate a positive effect on metabolic health over a relatively short period of time.

Each study looked at gut microbiomes. There was no consistent effect on the microbiome seen between the two studies, although Study 1 demonstrated a significant change in overall composition and abundance of microbiota in SXRG-84 treatments versus placebo, and some key biota that were important in this shift were identified.

In more detail, Study 1 subgroups (based on BMI) presented changes in lipids, inflammation, and insulin levels and shifts in gut flora. The reduction in non-HDL cholesterol in overweight participants on the 2 g dose also showed a trend to reduction in the atherogenic index (log(triglycerides/HDL-cholesterol)). Although the changes observed in non-HDL cholesterol were much lower than those observed in statin trials [26], this group was not necessarily hypercholesterolemic to begin with. Non-HDL cholesterol is a clinically relevant target as it has strong correlations with atherogenic lipoproteins and is suggested to be a better predictor of CVD events than LDL-cholesterol [6].

Although Study 2 was adequately powered to show a reduction in non-HDL cholesterol, Study 1 findings were not confirmed. One reason for this null finding may be that Study 2 used a different population from Study 1. Whilst the two study populations did not differ in total cholesterol, non-HDL cholesterol, and LDL-cholesterol levels at baseline (Table 10), they differed in regard to CRP, fasting glucose, and HDL-cholesterol at baseline, with participants from Study 1 having significantly higher levels. The higher glucose and CRP indicate a slightly more metabolically challenged group in Study 1, which may have elicited a stronger treatment effect.

**Table 10.** Comparison of baseline data from Study 1 to Study 2.


Data presented as absolute values and (percentage) for female gender; mean ± standard deviation for normally distributed variables or median (25th, 75th percentile) for non-normally distributed variables. BMI—body mass index, OGTT—oral glucose tolerance test. Two groups were compared using an independent samples t-test on normally distributed data, log-transformed data (§), square root-transformed data (‡), or the Wilcoxon signed rank test (¥).

The inconsistency of lipid-lowering effects highlights the need for further research with this extract, with recruitment restricted to hypercholesterolemic participants. Although Study 2 failed to show an improvement in lipid levels, the major finding was the improvements in proinflammatory cytokines after SXRG84 treatment, including IFN-γ, IL-1β, TNF-α, and the generally anti-inflammatory IL-10 [27].

Previous work from animal trials has shown that certain seaweed glycan extracts reduce plasma lipid levels through the action of bile acid sequestering [21,28,29], supported by the increase in bile acids in the feces of these animals following seaweed glycan supplementation [21]. Mechanistically, plasma lipids are lowered as they are required for the synthesis of bile acids; thus, the removal of lipids from the circulation is upregulated [21]. It was of interest that the genus Bilophila was one of the microbiotas that decreased during SXRG-84 treatment in support of this hypothesis [30,31]. Another potential reason for the lipid-lowering effects is via short-chain fatty acid propionate produced by *Bacteroides* and

*Akkermansia* [32], which both increased in Study 1. Propionate production has been shown to increase up to five times more from L-rhamnose than from other sugars [33] (SXRG84 is rhamnose rich).

In Study 1, there were significant reductions in CRP in overweight participants on 4 g of SXRG84 and a trend towards a reduction in CRP in obese participants. Although CRP is a nonspecific marker of inflammation, it is predictive of coronary heart disease and gastrointestinal diseases [34]. There was a trend towards a reduction in the two-hour insulin response to the OGTT, with no observed changes to glucose levels, suggesting an improvement in insulin sensitivity.

In Study 2, the observed reductions in a suite of inflammatory cytokines (IFN-γ, IL-1β, and TNF-α) may be beneficial as elevated levels are implicated in metabolic and cardiovascular conditions. IFN-γ has been implicated in the development of cardiovascular disease [35]. Additionally, IL-1β and TNF-α levels are chronically raised in metabolic disease [36]. Reducing these proinflammatory cytokines may benefit overweight participants who are otherwise generally healthy but may be at increased risk. IL-10 is generally regarded as an anti-inflammatory cytokine, and it is possible that the reduction indicates a reduced inflammatory pressure [27]. In contrast, probiotic supplementation in ulcerative colitis generally decreases proinflammatory cytokines and raises IL-10 levels.

In Study 1, the change in microbiome species composition over time (six weeks) was significantly different and more variable for participants on SXRG84 versus placebo. The impact of dietary SXRG84 on the gut flora can be summarized as an increasing shift of up to 15 taxa. In Study 1, of the 15 taxa responsible for significant differences in the active groups vs. placebo groups, 25% of the significant shift was explained by a quartet of assumed beneficial or probiotic microbiota, including *Pseudobutyrivibrio*, *Bifidobacteria*, *Akkermansia*, and *Clostridium*. These bacteria are known to respond positively to soluble dietary glycans in the distal colon. *Akkermansia* and *Bifidobacterium*, which are thought to be important for a broad range of health-related processes [37], are regarded as target organisms by researchers in the field of metabolic disease and gut health-related disorders. *Bifidobacterium* has specifically been shown to increase in response to larger molecular polysaccharides, more so than for the recognized fructo-oligosaccharides, including those with rhamnose [38]. There has been a lot of recent work focusing on the beneficial effects of probiotic *Akkermansia* intervention, and it is suggested that there is a strong synergistic relationship between the host and the bacterium in defending the gut lining and reducing leaky gut-triggered inflammation in exchange for increased mucilage production for food [39]. Achieving healthy levels of *Akkermansia* has been identified as a potential probiotic target to decrease inflammation, reduce obesity, and improve insulin sensitivity [39].

*Akkermansia* seems to have a high specificity, growing only on specific polysaccharides, including amine sugars, in the presence of proteins [40]. This makes sense in the case of this study, which includes amine-polysaccharides. The mechanism for protection by *Akkermansia* is still not fully understood but it is thought to relate to endocannabinoids that modulate glucose metabolism and protect against pathogenic bacteria [37,39]. *Bacteroides* was shown to decrease in all treatment groups at the largest magnitude. It is unclear as to why this occurred in all groups; however, Bacteroides is a dominant genus in the human gut, and has been shown to reflect a more western diet that is high in animal fat and protein [41]. Therefore, a reduction across the study population may suggest improved dietary habits in the participants. This, however, is only supported in the 4 g treatment group in our dietary analysis.

In Study 2, we did not observe a significant difference in gut microbiome composition between the two regimes at each timepoint. We *did* observe a consistent change in certain genera, including an increase in *Fusicatenibacter* and *Parabacteroides* and a decrease in *Clostridium* while on the SXRG84 treatment and not while on the placebo treatment for both regimes. *Parabacteroides* have increased in rats after laminarin supplementation [42], as has the species *Parabacteroides distasonis* [43], which is also a common species in rats fed with alginate. *Parabacteroides distasonis* has been identified as a laminarin fermenter [43]. It is possible, as our work has suggested, that *Parabacteroides* also responds to seaweed polysaccharides from Ulva Sp., which may infer benefits to the host. This is because *Parabacteroides* has been shown to modulate immunity [44] by suppressing the increase in inflammatory cytokines (IFN-γ, IL-12, IL-17, and IL-6) from gut tissue and increasing serum antibodies in a murine model of intestinal inflammation [44]. In Study 1, there was an increase in beneficial species of *clostridium*. In contrast, Study 2 showed a decrease in *Clostridium. Clostridium* as a genus, and species of clostridium have had different responses to seaweed extracts. In a murine model, *Clostridium cluster* XIVb and XI decreased in prevalence after laminarin supplementation [42]. Alternatively, both *Clostridium histolyticum* and *Clostridium coccoides* did not respond to ten different low-molecular-weight polysaccharides from either alginate or agar seaweeds when they were inoculated with human feces [45]. In Study 2, *Clostridium* at baseline was negatively correlated with the baseline value of the cytokines that were reduced (IFN-γ, IL-1β, TNF-α, and IL-10), suggesting an overall anti-inflammatory role of *Clostridium*. Further work is needed to determine the effects of SXRG84 extract on *Clostridium* levels across species as the current evidence is contradictory.
