1. Human Exposure and Bioavailability

1.1. Exposure

According to ^[1], lignans over-all intake in European countries was estimated to 1.23 (0.89 – 1.73) mg/day. Among these substances are the enterolignan precursors. However, as already mentioned, and detailed below, not all consumers are able to produce high levels of the estrogenic enterolignans from their non-estrogenic parent compounds and there is a large variability in the conversion efficiency in enterolignans producers. In enterolactone producers, a great variability also exists in the conversion rate. Therefore, the enterolignans bioavailability can only be studied in enterolignans producers. Nevertheless, even if the lignan exposure is considered to be low, in Western studies, where isoflavones and lignans are measured in urine samples, enterolignans tend to be more concentrated than isoflavones ^[2].

1.2. Gut Flora Involvement

In humans, the enterolignans enterolactone and enterodiol are produced by specific bacteria clusters that are not present in all consumers. However, in ^[3] the Tmax for enterolactone was found to be observed at 24-36 hours post-ingestion while it is 12 to 24 hours post-ingestion for enterodiol. Such a large Tmax is explained by the gut bacterial origin of both compounds and allows both accumulation in blood reaching a steady state level during chronic ingestion periods. As a consequence, the level of enterolactone in human consumers’ serum and urine can vary widely, being potentially in the same range and even higher than isoflavones concentration i.e. between 0.03 to 1 µM in plasma. For instance, in ^[4] the median serum enterolactone concentration was 13.8 nM (range: 0–0.0956 µM) and 16.6 nM (range: 0–0.182.6 µM) in men and women, respectively. The conversion which can be seen in Figure 1 is the metabolism of seccoisolariciresinol diglucoside into enterolactone. It is the one most studied so far. This conversion requires a deglycosylation into seccoisolariciresinol and then a demethylation. The demethylation step can be undertaken by the strain Clostridium sp. SDGMt85-3Db which showed the highest initial rate of deglycosylation in ^[5]. Bacteroides distasonis and B. fragilis were found to be predominant in enterolactone producers even if the strain Clostridium sp. SDG-Mt85-3Db was also sometimes detected. Bacteria involved in the demethylation step of seccoisolariciresinol diglucoside also demethylated other lignans. In particular, Peptostreptococcus productus demethylated the lignans pinoresinol, lariciresinol and matairesinol, and Eggerthella lenta catalysed the reduction of pinoresinol and lariciresinol into secoisolariciresinol.

Figure 1. Metabolic transformation of lignans into estrogenic enterolignans by gut bacteria clusters.

Nowadays, investigations on these bacterial metabolisms aim at finding ways to influence the gut ecosystem to promote the production of these compounds of health interest ^[6].

1.3. Blood Concentrations

Enterolactone levels in human blood are in the nM range. According to ^[7] and based on measurements performed on 69 Northern Mexican women, the mean enterodiol and enterolactone levels in plasma were 0.2 ± 0.6 nM and 3.8 ± 4.1 nM, respectively. Besides, based on serum enterolignans measurements from 55 Swedish menopausal women, the median plasma enterolactone level was 16.7 nM (range: 0.3 - 176.9 nM) ^[8]. Again, due to their hydroxyl groups both enterolactone and enterodiol exist in plasma as glucuronide and sulfate conjugates ^[9]. Therefore, their plasma concentrations in active forms can only be 5% of their total concentrations i.e. <10 nM. If higher concentrations are tested in vitro, they could stimulate other cell signaling pathways that may not be relevant in vivo.

2. Beneficial Effects

2.1. Hormonal Effects

2.1.1. Menopausal symptoms

Among all phytoestrogens health effects that were studied so far in humans, menopausal symptoms are those most documented. Many studies were undertaken using dietary supplements containing various classes of phytoestrogens to check for their efficiency. Currently, the market of food supplements offers preparations based on soy, clover, alfalfa, kudzu, linseed and hop. Other plants are used for menopausal symptoms including black cohosh, chasteberry or yam but their action modes are not strictly estrogenic. Although effects are reported and confirmed by meta-analyses, the effects of phytoestrogens on menopausal symptoms are still debated. The reasons are: a great interindividual variability, different effects according to the peri- or post-menopausal status, strong placebo effects and studies based on self-declarations.

The effects of linseed and lignans have not been extensively studied on menopausal symptoms. In their majority, the studies are not RCT compared to placebo and therefore the results are not as strong as they should be expected. However, in ^[10], 87 women were enrolled in a RCT compared to placebo and soy. The treatments were: ground flaxseed muffins (n=28); soy flour muffins (n=31); and wheat flour muffins (n=28) as control. Hot flashes were less severe with flaxseed (P = 0.001) compared to placebo but no effect was really recorded on the other parameters measured. Similarly, a double-blind, placebo-controlled RCT ^[11] compared two groups of postmenopausal women: the first one consumed two slices of bread with 25 g of flaxseed (46 mg lignans) and the second group consumed wheat bran (< 1 mg lignans) as control. The treatment lasted for 12 weeks and similar hot flashes frequency reduction and Kupperman Index were reported in both groups. Likewise, in the study by Pruthi et al. ^[12], an effect was recorded on perimenopausal women showing initially at least 14 hot flashes per week and using 40g of flaxseed flour daily for 6 weeks. A decrease was observed in hot flashes scores after flaxseed therapy (mean 57%; median 62%) from 7.3 hot flashes to 3.6 per day. However, 50% of the volunteers experienced mild or moderate abdominal distention, 29% experienced mild diarrhea, 4% experienced flatulences, and 21% withdrew because of toxicities. Indeed, 40g of crushed linseed is quite a large dose that is able to affect gut transit. There was no placebo group in this study and therefore the effect could not be fully characterized per se. Three major reasons can explain such uncertainty in the efficacy of vasomotor symptoms treatment with lignans. (1) The enterolignans production can be highly variable from one person to another. If these enterolignans are expected to be the active compounds, a large variability in the efficacy of the treatment should be expected. (2) Lignans are commonly present in the Western food and sometimes the dosages used were not high enough to induce a significant difference with the casual diet intake. (3) Menopausal symptoms are known to be sensitive to placebo effect and according to several authors ^[13], a 50% reduction of vasomotor symptoms can be attributed to placebo effect alone. Such an effect is increased since hot flashes measurements are based on self-declarations by the volunteers.

2.1.2. Bone health

Only few studies dealt with the association of lignans and enterolignans to bone health ^[14][15]. They were quite insufficient to prove any effect. The study by Dodin and coworkers [14] estimated the dietary intake of lignans and the median intake of total phytoestrogens at 876 μg/day, using questionnaires. Such figure is most probably underestimated since enterolignan-precursors are significantly present in the Western diet. Nevertheless, according to the authors, enterolignans estimated from food intake regardless of gut flora efficiency, were positively associated with bone density in postmenopausal women. However, this association became non-significant when dietary Ca²⁺ was added to the model. The same conclusion arose from the study by Kuhnle et al. ^[15]. This study estimated the phytoestrogens intake in more than 2580 women and 4930 men. It confirmed a positive association of BMD with methoxylated and hydroxylated isoflavones when Ca²⁺ was present and also showed a positive association of enterolignans precursors with BMD. However, the latter association was no longer significant when Ca²⁺ was taken into account. Enterolignans plasma levels should have been determined in these studies to really help in determining their effects on the health issue examined.

2.1.3. Estrogen responsive tissues

Few robust data are available so far linking the estrogenic enterolignans and estrogen-dependent diseases. This can be explained by : (1) all sources of lignans are not yet characterized in food ^[16], (2) not all humans can efficiently transform lignans precursors into estrogenic enterolignans ^[17], and (3) enterolignans are not measured usually in the fluids of volunteers taking part in clinical investigations.

Lignans were first correlated to reduction of prostate cancer risk in the 80’s ^[18]. However, the meta-analysis by Saarinen and co-workers ^[19] found no correlations at population level. A few studies involving flaxseed intake in men with a diagnosed prostate cancer did show a decrease of both serum total PSA and proliferation rate of benign epithelium. The authors also showed a significant decrease in total testosterone and free androgen indices ^[20]. According to them, enterolignans may have a protective effect on prostate cancer but the normal dietary intake is insufficient to induce such protection. More recently, it was shown that prostate cancer pre-operating treatment with flaxseed supplement induced the decrease of several tumour biomarkers ^[21]. Namely, NFkB, Ki67 and VEGF were significantly decreased, and it was concluded that flaxseed supplementation inhibits cancer cell growth and potentially reduces tumour angiogenesis in patients with prostate cancer.

Considering breast cancer, lignans and enterolignans are admitted to be protective in the global opinion [16]. Such a view is based on population studies showing that fruits and vegetables consumption globally reduces all risks of cancers including breast cancers ^[22]. Although enterolignans have estrogenic properties they can be defined as SERMs (Selected Estradiol Receptor Modulators) since they act essentially through the AF-2 transactivation domain of ERa which activates cell differentiation ^[23]. Conversely, and as mentioned earlier, it is the AF-1 transactivation domain which is most likely to be involved in proliferation phenomena. This mechanism was further confirmed in vivo on nude mice transplanted with MFC-7 cells ^[24]. In this study, dietary enterolignans were able to prevent the MCF-7 cell proliferation induced by the dietary isoflavone genistein. Regarding an interaction with GPER, no data were found so far. In addition, studies showed that a higher intake of lignans was associated with lower risk of developing cancer. On several occasions, the cancers receptor status was mentioned ^[25][26]. Finally, as for isoflavones, a preventive effect at the initial phase of breast cancer progression could be retained. Indeed, in ovariectomized rat, enterolactone was shown to decrease the occurrence and size of DMBA (dimethyl benz(a)anthracen)-induced mammary tumours at a dose of 10 mg/kg/day ^[27]. Because DMBA was able to transform healthy mammary cells in cancer cells, the enterolactone effect would target the initial phase of cancer progression.

2.2. Metabolic Beneficial Effects

2.2.1. Effect on cholesterol

Several intervention studies gathering low numbers of subjects tended to show a lowering effect of lignan consumption on LDL or triglyceride levels. These studies monitored the enterolactone and/or enterodiol presence either in urine or both blood and urine. Because of the low numbers of subjects and the modest effects, these studies were not always convincing ^[28][29]. Besides, other population studies were performed giving rise to contrasting results. In ^[30], 1492 male and female subjects of the NHANES cohort were examined for several cardiovascular risk factors and the urinary levels of enterolactone were measured. The subjects were split into 3 tertiles based on their urinary enterolactone levels. The study showed that enterolactone was significantly positively and negatively correlated with HDL-cholesterol and triglycerides, respectively. Besides, LDL and TC were not significantly associated. In parallel, in the study by Frankenfeld ^[31], performed on 2260 subjects from the NHANES cohort, the effects of the two enterolignans were not equivalent. In this study, the population was monitored according to the fibres intake and the urinary levels of enterodiol and enterolactone. Four quartiles were defined for fibres intake as well as two levels for urinary enterolignans (high and low). The study showed that enterodiol was not associated with any of the criteria analysed. However, enterolactone urinary levels were associated with reduced obesity, reduced C-Reactive Protein, reduced blood pressure, reduced triglycerides, higher serum HDL-cholesterol and lower occurrence of metabolic syndrome. In this study, the fibres intake was only associated with lower blood pressure and higher HDL-cholesterol. These data would sustain that enterolactone may have an effect per se. However, the differences between the two criteria considered (fibre intake and enterolactone urinary levels) may also be related to a more accurate determination of enterolignans levels than fibres intake.

2.2.2. Effect on metabolic syndrome

Flax-seed supplemented diets were tested several times on metabolic syndrome or some of its components, like abdominal adiposity, waist circumference, lipid and glucose plasma levels, insulin resistance among others. The studies published showed all an improvement of the criteria associated with metabolic syndrome and notably in randomized clinical trials ^{[32][33][34][35]}. However, lignans are associated to fibres and fruits & vegetables intakes, which are known to reduce the risk of metabolic disorders, cardiovascular diseases and diabetes. In addition, the ability of the gut flora to form enterolignans and especially the estrogenic enterolactone is quite variable among humans and flaxseed contains seccoisolariciresinol di-glucoside which can both be metabolised into enterolignans or act as such. Therefore, it is somehow uneasy to confer the effect of flaxseed to estrogenic enterolignans themselves. The study by Frankenfeld ^[31] as an observational study related enterolactone plasma levels to several criteria associated with metabolic syndrome. This study was performed on 2260 adults over 20 years old, extracted from the 2003–2010 NHANES study. The subjects were divided in two groups according to their urinary enterolactone or enterodiol levels. Only the 226 subjects with the highest and the lowest enterolignans urinary levels were considered for comparisons. Finally, the high urinary enterodiol concentrations were not associated with obesity or cardiometabolic risk factors. However, the high urinary enterolactone concentrations were inversely associated with obesity, abdominal obesity, high serum C-reactive protein, high serum triglycerides, low serum HDL cholesterol, and reduced risk of metabolic syndrome. Such associations are not definitive proofs. To sustain these reservations, the randomized clinical controlled trial from Eriksen and co-workers ^[36] did not show significant effect of seccoisolariciresinol supplementation (280 mg/day) in a whole grain rye diet compared to a whole grain wheat diet without supplementation. However, the monitoring of enterolactone levels in plasma and urine did show that the supplementation that was maintained for 4 weeks, dramatically increased the plasma enterolignan levels. Therefore, if an effect is to be expected it should occur after 4 weeks. Anyway, these data, that were not fully conclusive, appeal for more well designed clinical trials.

2.2.3. Effects on diabetes

Several clinical studies were performed to correlate estrogenic enterolignans with health benefits and among them T2DM incidence. The link with enterolactone can usually be made through enterolignans measurements in plasma or urine. However, such correlation cannot constitute a causal link since many other substances can be found in flaxseed or other sources of lignans like vegetables and fruits. For instance, the cross-over randomized control trial by Pan et al. ^[37] was performed in 68 T2DM patients with mild hypercholesterolemia. They received a placebo or 360 mg flaxseed lignans for 12 weeks before an 8 weeks wash-out and another 12 weeks alternative treatment. The lignan supplement significantly improved HbA1c compared to placebo, but no significant changes were observed on fasting glucose and insulin concentrations, insulin resistance and blood lipid profiles. Therefore, it can be said that the effect is modest. To sustain this finding the study by Talaei et al. ^[38] did not show any effect of urinary phytoestrogens: isoflavones on the one hand and enterolignans on the other hand, on T2DM risk in Singapore subjects. Briefly, the subjects were enrolled in the frame of the Singapore Chinese Health Study cohort. Initially none of the volunteers had a T2DM profile and they were followed for 6 years from 1999-2004 to 2006-2010. The subjects were shared into 4 quartiles according to their initial phytoestrogens measurements in urine and odd ratios (OR) correlating diabetes biomarkers and phytoestrogens quartiles were calculated. Whatever the phytoestrogens i.e. genistein, daidzein, glycitein, equol or enterodiol and enterolactone, none of the OR was significant when 95% confidence intervals were taken into account. However, lignan intake is known to be lower in Asia than in Western countries while isoflavone intake is higher. In this context, this study shows that, in China, dietary exposure to lignans is unlikely to have an effect in consumers. Conversely, in the USA ^[39], where the lignan intake is higher and results in higher urine concentrations compared to the Singapore study previously cited (2.1 nmol/mg creatinine in the urine of USA control subjects vs 0.95 nmol/mg creatinine in the urine of Chinese control subjects) enterolactone was found to be negatively correlated with T2DM incidence. This study is a nested cases-control study based on two cohorts: the NHS and the NHSII. Enterolactone and enterodiol were measured in urine at inclusion (1995-2001) and followed until the end of 2008. The subjects were divided into 4 quartiles based on their enterolignans urine levels and the difference between the two extreme quartiles was correlated to T2DM incidence in the two populations. Although enterolactone was significantly associated with a reduced risk of declaring a T2DM, enterodiol was only marginally significantly efficient. Finally, the study by Eriksen et al. ^[40] was a case-control study included in a much wider cohort, i.e. the Danish Diet, Cancer and Health cohort. The study examined the link between the enterolactone plasma levels at pre-diagnosis and diverse mortality causes of the subjects including diabetes. Subjects were recruited between 1993 and 1999 and they were followed until the end of 2009. During that period, a group of 640 subjects deceased from different causes i.e. diabetes (n = 48), cardiovascular (n = 141), cancer (n = 243), respiratory affections (n = 63) or other causes (n = 102). When plasma enterolactone was considered, 4 quartiles could be determined, each containing 295 or 296 people. The Hazard Ratios calculation relating initial enterolactone plasma levels and death causes showed that the risk of death was significantly decreased in the highest quartile for diabetes and cancer. This study examined for the first time the risk of death associated with enterolactone exposure. It indicated that providing that the intake is sufficient and the follow up duration long enough, enterolactone – not enterodiol- is likely to sign a preventive effect of diet on the incidence of T2DM. These results did not establish a definitive causal relation between enterolignans and disease occurrence.

3. Adverse Effects

A reference doses that is probably underestimated has been drawn out for enterolactone based on the ingestion of 7-hydroxymatairesinol in rat. Indeed, as seen previously many lignans can be considered as precursors of enterolactone and enterodiol and such a NOAEL does not reflect the dietary situation in humans. Nevertheless, in ^[41] the NOAEL for 7-hydroxymatairesinol established in rat was 460 – 740 mg/kg/day depending on the physiological endpoint examined. According to the classical decision rules applied in toxicology, the reference dose in human should be 4.6 – 7.4 mg/kg bw/day while it was found to be 1.64 – 18.2 µg/kg bw/day of matairesinol in France according to ^[42].

3.1. Hormonal Based Effects

3.2.1. Pituitary interactions

Estrogens are known to regulate pituitary reproductive hormones, namely Follicle-Stimulating Hormone (FSH) and Luteinising Hormone (LH). Such an effect is due to the regulation of the hypothalamic Gonadotrophin Releasing Hormone (GnRH) ^[43]. Depending on the cycle period, estradiol can either stimulate or repress pituitary hormone release. Using this property, contraceptive drugs have been developed essentially based on the synthetic ethynyl-estradiol which pharmacokinetic is longer than that of estradiol conferring it a higher potency. Therefore, expecting an effect of phytoestrogens on pituitary hormones release seems sensible. If such an effect is recorded it should induce menstrual cycle impairment and steroid synthesis modifications. Doing that phytoestrogens can act as endocrine disruptors and affect male and female fertility. As will be seen below, such effects are sometimes recorded but other studies failed to identify any endocrine disruption. This may be due to low dosages of phytoestrogens, to short treatments or too few tested subjects.

As far as researchers know, there is no published data on the effect of lignans and enterolignans on human fertility, women menstrual cycles, and men or women hypothalamus and pituitary hormones.

3.2.2. Estrogen based toxic effects

Animal studies were performed to assess the reproductive toxicity of the enterolignans. In ^[41] the enterolignan precursor 7-hydroxymatairesinol as potassium acetate complex was administered for days 0–21 of gestation at three different doses 140–180, 460–740, and 1190–2930 mg/kg bodyweight/day. The highest dose was not well tolerated and induced a loss of appetite in 50% of the gestational dams. However, the study showed no effects on reproductive performances of any treatment after external, visceral, and skeletal examination of the foetuses. Based on weight alteration, the NOAEL for maternal effects was fixed at 460–740 mg/kg/ bodyweight/day whereas the NOEL for foetal development was fixed at 1190–2930 mg/kg bodyweight/day. In addition, still in rats, Colins and co-workers ^[44], who investigated the effect of flaxseed (20 or 40%) or defatted flaxseed meal (13 or 26%) added to AIN-93 diet on gestation only, or on gestation and maturation in a lifetime study, showed that flaxseed did not affect foetal development but did affect indices of postnatal development such as the oestrous cycles in females. Although several studies related urinary lignans and steroid levels in men and women, the only study which was found dealing with a reproductive physiology effect is ^[45]. In this study, the authors tested a flaxseed powder inducing high concentrations of urinary lignans, on the menstrual cycle of 18 normally cycling women. They proceeded in a balanced randomized cross-over study where each subject had their usual diet for 3 cycles as control and a supplementation with flaxseed for another 3 cycles. The comparison was performed on second and third cycles of each diet. Apart anovulatory cycles which occurred only in the control group, flax cycles were associated with longer luteal phase (12.6 ± 0.4 vs. 11.4 ± 0.4 days; p=0.002). There were no significant differences on either estradiol, estrone or DHEA-S, prolactin, or sex hormone-binding globulin concentrations. Progesterone concentrations was not modified but the luteal phase progesterone/estradiol ratios were higher during the flax cycles. Mid-follicular phase testosterone concentrations were slightly higher during flax cycles. These data were not confirmed by subsequent studies.

3.2.3. Thyroid based toxic effects

Few data are available on the influence of enterolignans on thyroid hormone levels in humans. However, a recent study ^[46] positively correlated higher urinary enterolactone levels with elevated TSH levels in girls who were 12-19-year-of-age (β=0.196, 95% CI: 0.081, 0.311). In the 12-19-year-of-age male group, enterodiol was significantly positively correlated with TSH and TT3 (TT3: β=3.444, 95% CI: 0.150, 6.737; TSH: β=0.104, 95% CI: 0.005, 0.203). Such an association would suggest an impairment of the thyroid function in humans. However, the data are too scarce to firmly conclude.

4. Conclusions

Enterolignans exhibit moderate estrogenic activities at relevant dietary intakes. As such, they are suspected to exert beneficial effects in cases of estrogen deficiency. Acting as SERM no deleterious effects on estrogen dependent cancers were observed. Conversely, a protective effect was reported in vitro and in vivo in animal models and in human populations exposed to enterolignans precursors. The status of enterolignans is still unclear. They cannot be found in conventional diet and are produced by the gut flora. They act as SERMs and may reach active concentrations in biological fluids and may exhibit slight estrogenic beneficial effects. They can potentially prevent T2DM and metabolic syndrome.