Introduction

The relationship between the relative breast cancer (BC) risk and seaweed intake among humans is only now unfolding. A small body of research, both in vivo and in vitro, suggests seaweed may be useful in BC prevention (Funahashi et al. 1999; Teas et al. 1984; Yamamoto et al. 1987). Seaweeds are specifically used to treat tumors in Traditional Chinese Medicine and Japanese folk medicine. On a population level, those people for whom seaweed is a regular part of their diet, most notably in Japan, have dramatically lower rates of BC (Hebert and Rosen 1996; Kodama et al. 1991).

Epidemiologic studies done in Japan in the 1980s, before Westernized diets were common, reported that Japanese women had one out of three the rate of premenopausal BC and one out of nine the rate of postmenopausal BC (Reddy et al. 1980). Even today, BC incidence rates for women in Japan are 20/100,000 compared with the US average of 118/100,000 (US Cancer Statistics Working Group 2007). Although genetic predisposition has been proposed, when rates among migrants from Japan to the USA are compared, BC incidence almost doubles after 10 years of residence in the USA (20/100,000 to 35/100,000) (Shimizu et al. 1991) and continues to increase with each successive generation (LeMarchand et al. 1985). Japanese–American women who develop BC have significantly better survival rates than other American ethnic groups (Kanemori and Prygrocki 2005). On the other hand, Asian–American women over 50 years of age living in Los Angeles, especially Japanese–American women, have one of the most rapidly increasing BC incidence rates (Deapen et al. 2002). These data support the hypothesis that lifestyle changes and possibly gene–nutrient interactions are important in BC susceptibility.

Seaweed is a typical part of East-Asian diets, although consumption varies widely among individuals (Fukuda et al. 2007). Brown seaweeds have no land equivalents in terms of their specific components of fiber (alginate), primary carotenoid (fucoxanthin), sulfated polysaccharide (fucoidan and laminarin), and polyphenol defense compounds, each of which has been reported to have strong anti-cancer activity (Kang et al. 2012; Kotake-Nara et al. 2005; Koyanagi et al. 2003; Miao et al. 1999; Son et al. 2003).

Many in vivo and in vitro studies of dietary seaweed report decreased angiogenesis and increased apoptosis of tumor cells (Konishi et al. 2006; Koyanagi et al. 2003; Sekiya et al. 2005), inhibition of tumor cell adhesion and metastasis (Liu et al. 2005), and enhanced immune responses (Maruyama et al. 2006). Although Nishino et al. (2000) have investigated seaweed activation of the urokinase plasminogen system pathway, urokinase-type plasminogen activator receptor (uPAR) was not specifically studied. Based on the wide range of antitumor effects, we investigated the possibility that seaweed could affect uPAR concentrations in women who consume seaweed. The uPAR (CD87), is the membrane receptor for serine protease and is central to plasminogen activation and pericellular proteolysis. uPAR is an extracellular receptor involved in multiple signaling pathways important in maintaining homeostasis, directly interacting with 42 proteins, affecting the extracellular matrix (ECM), cell adhesion, inflammation, immune function (complement) activation, blood coagulation, and tissue repair (Eden et al. 2011). Increased concentrations have been shown to be associated with the presence of cancer, and increased cell surface shedding of uPAR is characteristic of more rapid cancer progression (Boonstra et al. 2011). We therefore included evaluation of uPAR as a possible biomarker for seaweed activity that might be related to BC prevention.

To further assess whether a dietary seaweed intervention could alter protein expression in urine and serum in a non-seaweed-consuming population of healthy postmenopausal women, we used surface-enhanced laser desorption/ionization–time of flight coupled with mass spectrometry (SELDI-TOF-MS). Proteomic analyses have been used to identify cancer biomarkers with high sensitivity and specificity, including those related to BC (Gast et al. 2009; van Winden et al. 2009). SELDI has also been shown to be sensitive enough to be used to identify changes in serum associated with the addition of a novel food (green tea) (Tsuneki et al. 2004).

Materials and methods

Results

Demographic variables for the study participants are presented (Table 1) and show no significant differences. The health variables in Table 2 show cancer survivors had higher screening FSH when compared with controls, indicating possible hormonal differences. It was also notable that the ER+ had lower BMI than the other two groups.

The iodine content in each 0.5 g capsule (n = 4) averaged 38.5 μg. Urinary iodine excretion standardized for creatinine (in μg dL−1 g−1 of creatinine) increased slightly during seaweed supplementation, from 0.04 (95 % CI, 0.02–0.06) after P1 to 0.05 (95 % CI, 0.03–0.06) on S returning to 0.03 (95 % CI, 0.02–0.04) on P2, indicating adherence to the protocol, but also that this low iodine-containing seaweed had little effect on iodine metabolism. These trends were the same for women taking thyroid hormone medication (n = 5) and those not taking thyroid medications (n = 10).

Urinary creatinine-standardized uPAR concentrations decreased by almost 50 % following seaweed supplementation that was reversed after the second placebo treatment. The uPAR concentrations (in pg mL μg−1 of creatinine) were P1, 1.5 (95 % CI, 0.9–2.1); S, 0.9 (95 % CI, 0.6–1.1); and P2, 1.7 (95 % CI, 1.2–2.2; p = 0.01, ANOVA). These trends were observed in each group, regardless of BC history (Fig. 1).

 

Open in a separate window

Fig. 1

Urinary uPAR concentrations compared by BC history and treatment group (error bars indicate 95 % CI)

The SELDI-MS-TOF data identified a total of 14 proteins that were differentially expressed in serum (Fig. 2). Only one serum protein reversibly changed in all three groups with seaweed supplementation (m/z 8,927). Fewer urinary proteins were identified using SELDI (Fig. 3), and only one (m/z 9,776) showed a reversible response to seaweed, increasing slightly, but significantly.

 

Fig. 2

Comparison of differentially expressed serum proteins (m/z) and their relative intensities by BC history and treatment group

 

Fig. 3

Comparison of differentially expressed urine proteins (m/z) and their relative intensities by BC history and treatment group

Discussion

We first proposed that seaweed intake could be a factor in BC prevention in 1981 (Teas 1981), and although the evidence from hundreds of research studies have supported seaweed as important in a host of biological pathways. Our study provides evidence that 5 g day−1 of brown seaweed (Undaria) stimulated significant changes in urinary uPAR and slight changes in two other proteins. For the relatively small amount of seaweed (5 g day−1) to result in a 50 % reduction in uPAR is remarkable. If supported in other studies, it gives a clear mechanism for explaining the various preventive and treatment-related effects of seaweed supplementation. At the very least, it gives a plausible mechanism for explaining the diverse findings of other studies, which include cell adhesion and detachment from the ECM of cancer cell leading to migration and metastasis (Hildenbrand and Schaaf 2009).

In many ways, it is difficult to imagine how a mere 5 g day−1 could have such a profound effect on systemic concentrations of uPAR and specific protein expression. One possibility is that seaweed, like other dietary fibers, is known to increase the mucus bilayer that protects the colonic mucosa, stimulating mucus secretion and thickness (Brownlee et al. 2005). Other suggested consequences of dietary seaweed fiber include increased stool bulk which would provide protection from dietary toxins and carcinogens and reduction in amylase activity leading to decreased glycemic response.

It is interesting that powdered seaweed (1 %, w/v) but not seaweed extracts were responsible for almost 100 % inhibition of amylase and pepsin digestive enzyme activity in vitro (Bobin-Dubigeon et al. 1997). Small water-soluble molecules which would be lost in the production of seaweed extracts were thought to be responsible for the inhibition of protease activities. Thus, the small amount of dietary seaweed used in our study and consumed in Japan might be expected to have a profound effect in humans.

Our study also provides a possible rationale for understanding why BC in non-seaweed-consuming Western populations increases with age, but remains more or less stable for Japanese women (Reddy et al. 1980). uPAR increases after menopause (Chung et al. 1998), and if seaweed, on a population level in Japan, has a significant impact on uPAR concentrations, it may be the basis for the stable rates.

Our investigation using SELDI was interesting. Several of proteins with high intensity have also been identified as important in BC (Gast et al. 2009). uPAR has an m/z protein size beyond the range we measured, but two proteins we identified changed with seaweed. Serum protein (m/z 8,927) has been identified as complement activation C3a des-arginine and may indicate stimulation of an immune response that has also been detected in subjects with early BC (Becker et al. 2004). In the seaweed literature, enhanced immune function has been reported in vitro (Kim and Joo 2008) and in vivo (Kandasamy et al. 2012; Leonard et al. 2012). Increased expression of m/z 8,928 may be associated with the observed enhanced immune function.

The urinary protein m/z 9,776 has not been identified as important in BC but it has been identified as a hormone-responsive protein important in decidualization of the endometrial stromal matrix preceding embryo implantation (Oh et al. 2008). This progesterone-responsive protein increased with seaweed. In a study of conception rates in heat-stressed cows, seaweed supplementation with a different brown seaweed (Ascophyllum) was associated with more than a 7-fold increase in successful pregnancies (Kellogg et al. 2006). This suggests that seaweed supplementation may influence cell surface signaling receptors for steroid hormones.

As BC in Japan is characterized by higher rates of the more favorable ER+ type, we anticipated that women with a history of ER− BC might have different protein responses to dietary seaweed. There were only minor differences in protein expression by BC history, and possibly the variations might reflect underlying differences in BC susceptibility. Further research should be conducted to define whether or not these minor differences reflect meaningful biological variations.

Seaweeds are consumed daily in Japan and are considered to be important in maintaining optimal health. The actual amount of seaweed consumed is somewhat problematic. Several different kinds of seaweed are eaten daily in a variety of forms and dishes and are consumed dried in sushi wrappers, fresh or rehydrated as a side dish, and used as flavor condiments as well as the basis for broth-based (dashi) flavoring added to other foods. It is therefore difficult to estimate the actual daily intake of seaweed. Our best guess is that 5 g/day may approximate the mean Japanese daily intake. A typical serving of the kind of seaweed used in this study, wakame (Undaria), has been estimated at 1 g by the Japanese Ministry of Health (Ministry of Health of Japan 2004). Thus our study could overestimate the contribution of Undaria to the average Japanese consumer, but as a pilot study, we propose that it combines the mean dose of seaweed and isolates a particular kind of seaweed, so the effects we report can be attributed to one seaweed species.

Our primary reason for using whole seaweed rather than a purified fraction is that people in Japan consume seaweed, and it is this dietary preference that appears to be associated with multiple health benefits (MacArtain et al. 2007). The combination of various bioactive components in whole seaweed may be greater than the sum of the individual parts, were they to be purified and tested individually. Many previous in vitro and in vivo studies have used hot or cold water extracts of Undaria, and have proposed that fucoidan, a water-soluble sulfated polysaccharide primarily found in brown seaweeds, as the most bioactive component (Cumashi et al. 2007). In previous studies we have demonstrated that Undaria fucoidan is absorbed by humans, despite its large molecular size (Irhimeh et al. 2005). Although it would be interesting to have used such an extract in our study, it would have required obtaining approval to use the extract as an Investigational New Drug, which would have taken extensive and costly pretesting evaluation to register with the US Food and Drug Administration. As an alternative, we chose to use high-fucoidan seaweed, using only the sporophyll of Undaria, collected at the peak of the summer harvest season. These sporophylls consistently contain 8 % fucoidan/dry weight and are also low in iodine (approximately 80 μg g−1), and 5 g day−1 would be well under the tolerable upper intake level of 1,100 μg/day (National Research Council 2001). In addition, no adverse effects have been recorded in our previous studies using this 5 g day−1 dose of Undaria sporophylls (Teas et al. 2009; Teas and Irhimeh 2012).

One of our study strengths was that it evaluated the introduction of seaweed to a non-seaweed-consuming population, and despite the women continuing with their normal diet, significant changes in protein expression were observed. The reversibility of the changes in the urinary concentration of uPAR and of the two SELDI-identified proteins strongly suggests the changes were attributable to the inclusion of seaweed. As any small proof of principle study, our small sample size was a limitation. However, finding significant differences bodes well for future studies.

Data from uPAR concentration changes indicate that seaweed affects ECM, which is known to influence cell surface signaling, cell adhesion, and growth factor communication and responsiveness in breast tissue. Additional data from SELDI proteomic analysis further supports the role of seaweed supplementation in modulating proteins known to affect cell adhesion, immune response, and inflammatory responses. These effects may be mediated through the ECM and be important in maintaining healthy structure and function in breast cells. In particular, lower uPAR concentrations could reduce the likelihood of breast tissue transformation and cancer cell proliferation and invasiveness.

The role of dietary seaweed in lowering uPAR may therefore be critical in explaining differences in postmenopausal BC incidence and mortality between Japanese women and women in other developed countries. The possibility that dietary seaweed could decrease uPAR is especially important in that there are no anti-uPAR drugs currently available (Degryse 2011). Finding that seaweed can have a significant effect on uPAR could be helpful both in designing future drugs and dietary interventions for people with cancer.

Acknowledgments

Footnotes

References