Report LIFE 98

Technical description of methodology

Since 1992 several scientists have been suggesting that a number of reproductive disorders in men and female are, as in wildlife, caused by anthropogenic industrial compounds that have the capacity to interfere with the body's endocrine (hormone) system. Testicular cancer, urethral abnormalities and the decrease in sperm count and quality in the male and breast cancer and endometriosis in the female are disorders of the reproductive system which have increased in many countries over a short period of time. Therefore they rather seem to reflect changes in environmental factors or "life style" than genetic factors.

The project aimed to tackle the endocrine issue in a multidisciplinary way:

• Firstly, existing screening assays for endocrine disrupters (xenoestrogens) have been validated and standardised. In addition, the development of new tools to study more closely the effects of estrogens on the different compartments of the hypothalamo-pituitary-gonadal axis was initiated. This should allow 1) a more accurate detection of these compounds in the environment, 2) a more fundamental knowledge on the role of estrogens in male fertility.
• Secondly, the project has investigated the possibility to lower human exposure to xeno-estrogenic substances through 1) the use of an enhanced water filtration system and 2) the ingestion of "health food" types of product.

In vitro tests and gas chromatography mass spectrometry

Most research to date in the field of endocrine disruption has focused on estrogenic chemicals. Although the role of xenoestrogens in the alleged decrease of sperm quality is one of the most controversial scientific subjects of the past years, the assays that are available to date to measure estrogenicity all suffer from a lack of relevance for male fertility. The basic mechanism behind estrogen action at the cellular level starts with an estrogenic compound (ligand) entering the cell and the nucleus and binding to one of the possible estrogen receptor isoforms (ERalpha or ERbeta) via their respective "Ligand Binding Domain" (LBD). Subsequently, the receptor dimerizes and binds via its "DNA Binding Domain" (DBD) to specific sequences on the DNA called "Estrogen Receptor Elements" (ERE). Parallel to the binding of the ER to the ERE, a number of co-activators and/or co-repressors bind to the ligand occupied ER, enhancing or repressing its transcriptional activity. The interplay between the ligand, the receptor, and the relative presence of the co-activators/repressors /integrators determines the cell specificity of response. This cell type specificity of estrogen response is the experimental paradigm behind the booming research into selective estrogen modulators (SERMs) used for the treatment of osteoporosis, breast cancer and the post-menopausal syndrome (Figure 1).

Figure 1: General estrogen receptor mechanism. The pleiotropic response could be the endogenous production of a protein (e.g. inhibin B in the primary Sertoli cells or LH in the pituitary cell line), growth (Sertoli cell line), or the production of an enzyme like -galactosidase or luciferase (yeast assay and Sertoli cell assay).

Male fertility is regulated through the hypothalamic-pituitary-gonadal axis via a series of complex feedback mechanisms. Gonadotropin releasing hormone (GnRH) is produced by the hypothalamus and secreted in the hypothalamo-pituitary portal vein in a pulsatile manner. These pulses of GnRH entrail the production of Luteinizing Hormone (LH) and Follicle Stimulating Hormone (FSH) by gonadotropic cells present in the anterior lobe of the pituitary. The latter two peptide hormones affect two different cell types in the testes: the Leydig cells that react on LH with the production of the steroid testosterone and the Sertoli cells that respond to FSH with the production of a number of secretory proteins, the most important being androgen binding protein and inhibin B.

Testosterone, after peripheral transformation to 17-estradiol, and inhibin B exert on their turn a negative feedback on the pituitary. Estradiol has also been described as affecting the GnRH pulse generator in the hypothalamus (Figure 2).

Figure 2: Hypothalamic-Pituitary-Gonadal axis

Most in vitro tests nowadays are artificially designed. In many cases the cell system does not endogenously express the estrogen receptor, but is transcribed from a plasmid that is brought into the cell by specific molecular techniques. The response elicited is often the result of an interaction of the ER with a consensus ERE, integrated in a sensitive promotor upstream of the genetic code of a luciferase enzyme. In most of the developed assays, the cell system (with its cofactors and promotors) is not relevant for male fertility. This part of the project aimed at the development of a complementary set of test methods to study the impact of hormone disrupting substances on the different levels of the hypothalamic-pituitary-gonadal axis:

• We have investigated the role of estrogens in the male gonads using primary cultures of mouse Sertoli cells, cells that play a crucial role in spermatogenesis.
• We have characterized an immortile Sertoli cell line and evaluated the effects of several estrogens on their growth characteristics. In addition we evaluated the functionality of the estrogen receptors in this cell line using transfection studies. This should ultimately lead to a sensitive and reproducible test system for the evaluation of the cell specific response of several types of xenoestrogens.
• Thirdly, we have studied the effect of estrogenic substances on the release of gonadotropin releasing hormone from guinea pig hypothalamus explants. These experiments were however unsuccessfull due to non reproducibility.
• We have subsequently explored the use of an immortalized murine pituitary cell line. We characterized this cell line extensively and set the stimulation conditions for estrogenic compounds.
• We have revealed the characteristic features of the guinea pig GnRH peptide both in vitro and in vivo. This is a crucial step for the development of an in vivo guinea pig model that should allow the corroboration of the above described in vitro test results with the LT2.
• We have implemented, optimized and validated a yeast assay for estrogenic chemicals. This assay does not display cell specific agonist or antagonist characteristics (see above) and can thus be used as a control for the effects seen in the other assays. In addition it can be used for the screening of environmental samples.
• Finally, we have developed and validated an LC/MS identification method for the determination of estrogenic compounds in water.

Through the use of these bioassays we will be able to study the effects of endocrine disruptive compounds, which may lead to the identification of previously undetected types of hormone disrupters. At this moment, we are screening Flemish surface waters, together with effluents from sewage treatment works and industrial discharges for the presence of estrogenic compounds using the yeast assay together with the LC/MS method and a previously developed estrogen receptor binding assay.

Lowering human exposure to endocrine disrupting substances: "Water Filter" and "Health Food"

Although thanks to a number of regulations, the use of several endocrine disrupters will be reduced in the coming years, humans and wildlife are still exposed to them, because of their previous release in the environment (e.g. PCB’s, some organochlorine insecticides) and their low biodegradability and important bioaccumulation in the body, and because of their widespread and poorly regulated use in some (developing) countries (e.g. DDT). This makes the development of systems preventing ingestion of those compounds more than necessary. Food and drinking water are generally considered as the major sources by which humans (and wildlife) are exposed to hormone disrupting substances. In this Life98 project we proposed to approach this problem from two angles of incidence.

Firstly, we investigated the feasibility of using the cloned estrogen receptor as a tool to extract estrogenic compounds from water. At the time of the project proposal, a number of water filtration systems for household use existed on the market. These were mostly build around a cartridge filled with activated carbon, a material with a proven high adsorptive capacity for lipophilic compounds. However it is known that the adsorptive performance of activated carbon decreases with the polarity of the compounds. Exemplary is the fact that every year in Flanders, several pesticides are detected in drinking water above their maximally allowed concentrations, although drinking water companies apply an expensive activated carbon filtration step. Therefore we initiated the development of a filter system in which the core would be a biological compartment that should eliminate xenoestrogens via the binding mechanisms that play an important role in the hormone signal transduction pathway in the human body. We originally conceived the complete filter as a modular device in which the different parts could be regenerated separately. It would consist of 1) a prefilter necessary to eliminate particulate matter from the water 2) an antibacterial filter placed behind the prefilter to trap microorganisms 3) a chemical filter (octadecyl (C₁₈) siloxane or PSDVB disks) to retain the bulk of the environmental contaminants 4) an affinity filter for oestrogenic substances, based on the use of human estrogen receptors 5) and finally a second antibacterial filter placed at the outlet, in order to eliminate any bacterial contamination that may have occurred during passage through the filter (Figure 3).

Figure 3: Original concept of the filter using C₁₈ disks.

During our research we focused on the development of the biological compartment of the filter, as this part was considered to be the most challenging one. In addition we evaluated the possibility of using classical filters for particulate matter, microbial filtration disks and the C₁₈ and PSDVB membranes for the prefiltration module of the filter.

We successfully cloned the estrogen receptor in yeast as well as in human K562 cancer cells. This yielded six receptors of which two exhibited excellent 17-estradiol binding characteristics, comparable to the values found for endogenous human estrogen receptors alpha in international literature. The production of the these two receptors was scaled-up and attempts were undertaken to purify them. However, during these experiments, it became clear that the estrogen receptors were not stable (desintegration) and thus quickly lost their specific binding characteristics. For these reasons, we concluded that the cloned recepors would probably not permit the development of the biological filter cartridge.

In a last attempt to develop a biologic filter system that would enable the removal of trace amounts of endocrine disruptive substances, we started from a completely new experimental setup investigating the feasibility of using manganese oxide particles for the oxidative breakdown of estrogenic compounds. Manganese oxide is a well-known solid phase oxidant, and its surface redox reactions with xenobiotic organic chemicals have been extensively studied. The degradation of substances such as humic substances (humic acids and fulvic acids), atrazine, hydroxylamines, ascorbate and different types of catechols, are known to happen faster in the presence of MnO₂. In natural waters, the main manganese source is Mn(II). It has been proposed that the oxidation of Mn(II) in humic rich environments is a possible mechanism whereby bacteria can utilize the large biologically recalcitrant pools of carbon contained in humic substances. After being oxidized, the manganese precipitates around cells or accumulates on slime layers or sheaths. This precipitated manganese then can abiotically oxidize humic and fulvic acids releasing low molecular organic compounds such as pyruvate, acetone, formaldehyde, and acetaldehyde. The latter are easily bioavailable for the Mn-oxidising organisms (Figure 4). In addition, manganese oxide has been shown to adsorb metals and organic compounds, further enhancing its possible use as a removal substrate for noxious environmental agents in water.

Figure 4: Proposed ethinylestradiol degradation in which the MnO₂ operates as a catalytic surface.

The above described characteristics of manganese and Mn-oxidising bacteria were integrated into an experimental paradigm, starting with manganese as a reusable catalyst at the interphase of surface chemistry and biological degradation (Figure 4) In this reconversion cycle, organic compounds are oxidatively degraded into bioavailable compounds. The latter form a simple carbon source for the bacteria that convert the reduced Mn back to Mn(IV). Manganese loss in the treated water is thus prevented, making manganese treatment possibly a cheap step in advanced sewage water treatment. As a first step in the development of such a biocatalytical treatment filter, we investigated the catalytic properties of MnO₂ for the removal of EE2.

Lactobacillus based "Health Food"

The gastrointestinal microflora is a complex ecological system, normally characterized by a flexible equilibrium. The most important role of the microflora, from the point of view of the host, is the prevention of invasion and colonization by other pathogenic microorganisms through physical displacement, competition for nutrients and lactic acid and other "short chain fatty" acid production resulting in an acidic pH of the epithelial surface which is toxic to most pathogens. In addition to the complex barrier functions that indigenous bacteria provide, they also function to synthesize nutrients in the colon such as vitamin B12, biotin and vitamin K. They produce the enzyme lactase that converts the milk sugar Lactose to which a number of humans are intolerant, into glucose and gaLactose. Finally, they are described of having an immunostimulatory property. Bifidobacteria and Lactobacilli are Gram-positive lactic acid-producing bacteria constituting a major part of the intestinal microflora in humans and other mammals.

As an adjunct to the diet, foods or supplements that contain microorganisms that salutarily colonize the human mucosal surfaces have been referred to as probiotics. This good microflora can provide a protective effect only when a proper balance is maintained among all the different bacteria that normally reside in the intestine. Therefore, the most frequently used probiotic genera are Lactobacilli and Bifidobacteria. Health benefits of probiotics are being vigorously investigated today at various medical and research centers around the world. Results from clinical studies are showing that specific probiotic bacteria can alleviate or prevent diverse intestinal disorders and/or diseases. Another interest is to find alternatives to classical antibiotic treatments because of rapid development of antibiotic resistance, as well as to a multitude of negative side effects and allergic reactions. Oral ingestion is an important exposure route by which endocrine disruptive chemicals can enter the human body after intestinal absorption. The aim of this research chapter was to direct the scientific research towards the development of a "health food" that can be applied to reduce the bioavailability of these compounds in the gastrointestinal tract.

In a first experimental phase, we evaluated our experimental paradigm and compared three commercially available probiotics for their adsorptive capabilities towards atrazine and 2,4-D in relation to the effect of supplemented activated carbon (AC). In a second phase, we evaluated an in-house Lactobacillus strain for its survival capacity in the stomach and the bile using a Simulator of the Human Microbial Ecosystem (SHIME). This is a dynamic in vitro method, developed previously in our laboratories. It consists of 6 compartments of which the first 3 represent the stomach, the duodenum/jejunum and the ileum and the last 3 are filled with well defined microbial suspension of the different parts of the large intestine (Figure 5).

Figure 5: Scheme of the Simulator of the Human Intestinal Microbial Ecosystem (SHIME)-reactor.

Finally, we investigated the influence of estrogens towards the functioning of the microbial community in the large intestine using the SHIME. We looked at the sorptive characteristics of colon bacteria on estrogens spiked to the intestinal suspension with and without supplementation of a probiotic strain that was put forward in the previous part.