γ-Glutamyl-Se-methylselenocysteine (GGMSC) has recently been identified as the major Se compound in natural garlic and selenized garlic. Our working hypothesis is that GGMSC serves primarily as a carrier of Se-methylselenocysteine (MSC), which has been demonstrated in past research to be a potent cancer chemopreventive agent in animal carcinogenesis bioassays. The present study was designed to examine the in vivo responses to GGMSC or MSC using a variety of biochemical and biological end points, including (a) urinary Se excretion as a function of bolus dose; (b) tissue Se accumulation profile; (c) anticancer efficacy; and (d) gene expression changes as determined by cDNA array analysis. Our results showed that like MSC, GGMSC was well absorbed p.o., with urinary excretion as the major route for eliminating excess Se. When fed chronically, the profile of Se accumulation in various tissues was very comparable after treatment with either GGMSC or MSC. In rats that had been challenged with a carcinogen, supplementation with either GGMSC or MSC resulted in a lower prevalence of premalignant lesions in the mammary gland, and fewer mammary carcinomas when these early lesions were allowed to progress. More importantly, we found that a short term GGMSC/MSC treatment schedule of 4 weeks immediately after carcinogen dosing was sufficient to provide significant cancer protection, even in the absence of a sustained exposure past the initial 4-week period. With the use of the Clontech Atlas Rat cDNA Array, we further discovered that the gene expression changes induced in mammary epithelial cells of rats that were given either GGMSC or MSC showed a high degree of concordance. On the basis of the collective biology, biochemistry, and molecular biology data, we conclude that GGMSC is an effective anticancer agent with a mechanism of action very similar to that of MSC.

Past research showed that garlic cultivated with Se fertilization has unique attributes as a cancer preventative (1, 2). The enrichment of Se in garlic is dependent on the intensity of Se fertilization. With the use of a rat mammary carcinogenesis model, it has been reported (3) that the anticancer potency of garlic with a moderate level of Se enrichment (100–300 ppm Se dry weight) was very similar to that of garlic with a high level of Se enrichment (>1000 ppm Se). In these experiments, the freeze-dried garlic powder was added to the animal diet to reach a final concentration of 2–3 ppm Se. Recent analytical studies, however, provided evidence that the major form of Se is different in these two types of selenized garlic. Characterization by high-performance liquid chromatography with inductively coupled plasma mass spectrometry showed that MSC3 and GGMSC are the predominant forms of organic Se in the highly enriched and moderately enriched garlic, respectively (2). The structures of these two compounds are illustrated in Fig. 1. GGMSC is also found in natural garlic, accounting for ∼31% of total Se in garlic itself and ∼53% in commercial garlic powder (4). γ-Glutamyl-S-alk(en)ylcysteines have been well documented as sulfur storage compounds in garlic (5, 6, 7). Because Se is known to follow the sulfur assimilatory pathway in plants (8), it is not surprising that GGMSC is produced when garlic is fertilized with inorganic Se salts.

MSC has been investigated extensively as a chemopreventive agent in animal carcinogenesis bioassays (1, 9, 10, 11). Little information, however, is available on the activity of GGMSC. After oral administration, a dipeptide such as GGMSC is probably hydrolyzed by a specific peptidase to produce MSC. Thus, on both the systemic and the cellular levels, GGMSC is expected to behave very similarly to MSC. The present study was designed to compare the in vivo responses to GGMSC and MSC using a variety of biochemical and biological end points, including: (a) tissue Se accumulation profile; (b) urinary Se excretion as a function of bolus dose; (c) anticancer efficacy; and (d) gene expression changes as determined by cDNA array analysis.

Se Compounds.

MSC was obtained from Selenium Technologies, Inc. (Lubbock, TX). GGMSC was supplied by one of the authors (E. B.). The synthesis of this compound and its chemical confirmation were described previously in detail (12).

Tissue and Urine Se Analysis.

Pathogen-free female Sprague Dawley rats were purchased from Charles River Breeding Laboratories (Wil mington, NC). They were fed the standard AIN-76A diet (13) for 5 days to acclimatize them to the powdered ration. The AIN-76 mineral mix provides 0.1 ppm Se (as sodium selenite) to the diet. At 50 days of age, the animals were divided equally into three groups (n = 6 rats/group) and fed one of the following diets: (a) basal diet containing 0.1 ppm Se (or 0.1 mg Se/kg of diet); (b) basal diet + 3 ppm Se as GGMSC; and (c) basal diet + 3 ppm Se as MSC. All of the rats were sacrificed after 1 month of feeding. Liver, kidney, mammary gland, skeletal muscle, and plasma were collected, frozen immediately in liquid nitrogen, and stored at −80°C until ready for analysis. Se concentrations were determined by the fluorometric method of Olson et al.(14).

Prior to the start of the GGMSC or MSC diet, a separate group of rats were fasted overnight for the urinary excretion experiment. In the morning, they were given by gavage a single dose of GGMSC or MSC (dissolved in 1 ml of water) before being placed individually in glass metabolic cages for the collection of urine and feces (n = 6/group). The dose of each compound was equal to the amount consumed in 1 day by an animal fed a diet containing 3 ppm Se (identical to the level used in the above experiment). For a 50-day old female rat, the average food intake is about 13 g per day. Thus the daily ingestion of Se from a diet containing 3 ppm Se is calculated to be 40 μg. Urine and feces were collected at the end of 24 and 48 h as described previously (15). During this time, animals were given free access to water and food (basal diet containing 0.1 ppm Se). Se concentrations in the urine and feces samples were determined by the fluorometric method (14). The background value from control animals (not given GGMSC or MSC) was subtracted from the experimental measurements. The results are expressed as percentage of the dose excreted in each 24-h period.

Quantification of Premalignant Lesions in the Mammary Gland.

Within a few weeks after animals were injected with a mammary gland carcinogen, the appearance of premalignant lesions in histological sections could be detected and quantified (16). These early transformed colonies, which are known as IDPs, are the precursors for the eventual formation of palpable carcinomas (16). In this experiment, 50-day-old rats were injected i.p. with MNU at a dose of 50 mg/kg body weight. They were then divided into three dietary groups (n = 6 rats/group): control (basal diet), GGMSC (at 3 ppm Se), or MSC (also at 3 ppm Se). Animals were killed at 4 weeks after Se supplementation. The procedure of IDP quantification was described previously in detail (10).

Briefly, at necropsy, the abdominal-inguinal mammary gland chain on both sides was excised in one piece, fixed in methacarn, and processed in a Tissue-Tek Vacuum Infiltration processor. Each mammary gland whole mount was then divided into six segments and embedded into paraffin blocks. Ribbons of 5-μm thickness were cut from each block and placed on slides that had been treated with 3-aminopropyltriethoxysilane. Every tenth section was heat immobilized, deparaffinized in xylene, rehydrated in descending grades of ethanol, and stained with H&E. These H&E-stained slides were examined under the microscope for the appearance of IDPs using the criteria described by Russo et al.(16). Once a section showing the pathology of an IDP lesion was found, the in-between slides were similarly stained to confirm the histology. The size of each IDP lesion could thus be estimated operationally by the number of serial sections showing the same pathology. The total IDP count data were analyzed by the χ2 test using a Poisson regression model (17).

Mammary Cancer Chemoprevention Bioassay.

The protocols of MNU administration and GGMSC/MSC treatment were identical to that of the IDP experiment, with the exception that more animals were used for the chemoprevention bioassay (n = 30/group). Because of the limited amount of GGMSC available, rats were fed either the GGMSC or MSC diet for only 4 weeks after MNU injection before returning to the basal diet for the remaining duration of the experiment. Previous work with selenized garlic showed that a short-term treatment schedule of 4 weeks immediately after carcinogen dosing was sufficient to provide significant cancer protection, even in the absence of a sustained exposure past the initial 4-week period (18). Because GGMSC and MSC are both constituents of selenized garlic, the chemopreventive efficacy of the 4-week supplementation protocol was repeated in the present design. Rats were palpated weekly to determine the appearance, size, and location of mammary tumors. The experiment was terminated at 24 weeks after MNU administration. At necropsy, all of the tumors were excised and fixed for histological examination. Only confirmed carcinomas are reported in the results. Tumor incidences at the final time point were compared by χ2 analysis, and the total tumor yield was compared by frequency distribution analysis as described previously (19).

Mammary Epithelial Organoid Isolation.

Rats used in these experiments were injected with MNU at 50 days of age and fed a control, a GGMSC, or a MSC diet for 1 month before sacrifice (n = 6/group). At necropsy, the abdominal-inguinal mammary gland chain was excised in one piece and was then subjected to mammary epithelial organoid isolation according to the method described by Hahm and Ip (20), with slight modifications. Minced mammary glands from each group were pooled and digested with class 3 collagenase (Worthington Biochemical, Freehold, NJ) and grade II dispase (Roche, Indianapolis, IN), each present at 1.5% (w/v). The digestion was carried out at 37°C in a gyroshaker with a speed of 70 rpm for 1.5 h, the digest was then filtered through a series of Nitex filters. To obtain an enriched mammary epithelial organoid population, the stromal contaminants were removed by adherence to plastic during a 2-h incubation in DMEM/F12 medium containing 10% fetal bovine serum. Total RNA was isolated from recovered mammary epithelial organoid using TRIzol reagent (Life Technologies, Inc., Grand Island, NY) at a concentration of 1 ml per 107 cells according to the procedure supplied by the manufacturer.

cDNA Microarray Analysis.

32P-cDNA probes were prepared using the Atlas Pure Total RNA Labeling system (Clontech, Palo Alto, CA). After DNase I treatment, 50 μg of total RNA isolated from each mammary epithelial organoid preparation was mixed with biotinylated oligo (dT), which binds the mRNA fraction. Streptavidin magnetic beads were added to remove the biotinylated oligo (dT)/mRNA complex, which was then collected using a magnetic particle separator. cDNA probe synthesis was performed directly on the beads, using the captured mRNA as the template, with a gene-specific primer mix provided by the manufacturer. This mix contains only primers for genes that are represented on the array. The use of such a labeling method results in much higher sensitivity and a concomitant reduction in nonspecific background signals compared with the use of conventional oligo (dT) or random primers. The probes synthesized from control and GGMSC-treated samples were hybridized side-by-side to two identical Atlas rat cDNA Expression Arrays (Clontech, Palo Alto, CA). After overnight hybridization and a high-stringency wash (according to the procedure described in the Atlas Array User Manual), the arrays were scanned by a Storm Phosphorimager (Molecular Dynamics, Sunnyvale, CA) with a 4-day exposure. The two arrays were then stripped and reused for a second experiment involving control- and MSC-treated samples.

The two array images were analyzed and compared using the Clontech AtlasImage 1.5 software. First, the phosphorimage of each array was separately aligned and fine-tuned to the AtlasImage Grid Template. After normalizing the signal values of all of the genes on the arrays (global normalization), the two aligned arrays were compared with each other using the control group as the reference to generate a color schematic diagram of the changes in the treatment group and a tabular report of the data.

In an attempt to determine the enteral absorption of GGMSC versus MSC, rats were given a single oral dose of 40 μg Se of either compound. As noted in the “Materials and Methods” section, this dose is equivalent to the amount ingested per day from a diet containing 3 ppm Se. No Se was found in the feces during day 1 or day 2, which suggests that both compounds were well absorbed from the intestinal tract. Table 1 shows the proportion of the Se dose recovered in urine over a 48-h period. Approximately 40% of the dose was excreted in urine during the 1st day after a bolus gavage of either GGMSC or MSC. The rate of urinary elimination decreased markedly during the 2nd day, dropping off to about 6–7% for both compounds. There was, however, no statistical difference in the total amount of urinary Se recovered between the two groups.

Rats fed a diet containing 3 ppm Se as either GGMSC or MSC showed significant increases in tissue total Se after 1 month of treatment (Table 2). Interestingly, the magnitude of the elevation was quite similar with the two compounds. Different tissues are known to exhibit a hierarchy of Se accumulation. Consistent with our previous observations, liver and kidney were able to retain more Se (∼3-fold increase) when compared with mammary gland, muscle, and plasma (∼2-fold increase or less). The data in Table, 2 therefore, represent the Se status of the animals at the time when their mammary glands were excised for the quantification of the premalignant IDP lesions.

Table 3 shows the IDP results from MNU-treated rats in the three dietary groups. As noted in the “Materials and Methods” section, the size of each lesion was determined operationally by the number of serial sections showing the same pathology. Because of the very time-consuming nature of the analysis, we could only process six animals per group. There were a total of 42 IDPs found in the control group (last column of Table 3). Treatment with GGMSC or MSC reduced the number of IDPs by 48 and 57%, respectively (P < 0.05). All of the lesions were categorized into five classes with each containing ≤10, 11–20, 21–30, 31–40, or >40 serial sections. To analyze the size distribution data, a repeated measures option was added to the Poisson regression because most animals presented lesions in more than one size class. No significant differences were found by treatment within a given size class, probably because of the small sample number in each category when the data were segregated; this reduced statistical power to detect significant differences because of treatment.

Table 4 shows the results of the mammary cancer chemoprevention experiment in rats given GGMSC or MSC for only 4 weeks after MNU administration. It can be seen that both treatments were equally effective in inhibiting mammary tumorigenesis. The total number of carcinomas were reduced by 45–50% despite limiting the length of intervention with GGMSC or MSC to a short time during the early stage of carcinogenesis.

In view of the similarity of the cancer chemopreventive efficacy of GGMSC and MSC, our next approach was to use the technology of cDNA microarray analysis in an attempt to determine whether the biological activity of these two compounds could be distinguished at the molecular level. Clontech markets a 588-gene Atlas rat cDNA Expression Array that contains a handful of cell cycle and apoptosis regulatory genes that are of interest to us. As noted in the “Materials and Methods” section, the molecular analysis was done using mammary epithelial organoids obtained from MNU-treated rats fed a control, a GGMSC, or a MSC diet for 1 month. Therefore this was a sample consisting predominantly of normal ductal and alveolar cells mixed with a very small population of IDP cells. It is noteworthy to point out that we did the experiment with RNA from mammary epithelial cells, and not RNA from the whole mammary gland of which adipocytes and stromal cells constitute the major components. The integrity of the mammary organoid RNA was confirmed by gel electrophoresis.

Fig. 2 shows the actual hybridization pattern. Across all three of the groups, ∼8% of the genes on the array were expressed above the detection limit. This seemingly muted level of activity might simply reflect the quiescent nature of the mature mammary epithelium. We have unpublished data showing that, if whole mammary gland RNA were used for the analysis, as much as 38% of the genes on the array would give rise to a quantifiable signal.4 Suffice it to note that the whole mammary gland consists predominantly of adipocytes, fibroblasts, macrophages, and other immune cells, which may be metabolically and functionally more active than the epithelial cells. The implication of the array data generated from isolated mammary epithelial organoids versus that from whole mammary gland will be addressed in a separate report.

Fig. 3 shows the color schematic comparison (produced by the Clontech AtlasImage 1.5 software) of the array data between the control and treatment with either GGMSC or MSC. The figure legend provides an explanation of how to interpret the colored grid. Table 5 summarizes the gene expression profile changes induced in mammary epithelial cells of rats given either GGMSC or MSC. A total of 21 genes were found to be affected by one or the other compound. Of these 21 genes, 15 were shared with both compounds. This trend represents a concordance rate of 71%. The significance of some of the genes will be elaborated on in the “Discussion.” Table 5 also reports the magnitude of gene expression changes. Although the values were not necessarily identical between the two groups, the general consistency of the data suggests that the molecular effects of GGMSC and MSC are quite similar.

Prior to its detection in selenized garlic, GGMSC has been reported to be present in Astragulus bisulcatus(21) and Melitotus indica L. (22). The former is a Se accumulator weed, and the latter is a Se-tolerant grassland legume. Neither is a marketable food product commonly consumed by humans. The discovery of GGMSC as a naturally occurring phytochemical in garlic, coupled with the early work by Ip et al.(23, 24) that selenized garlic has much higher anticancer activity than regular unenriched garlic, provided the impetus to the present investigation of the bioactivity of GGMSC. Structurally, there is one distinctive feature about this dipeptide. Most peptide bonds involving glutamic acid are of the α variety. With GGMSC, the selenoamino acid is linked to the γ-carboxyl group of glutamic acid as shown in Fig. 1. Thus, GGMSC is unlikely to be acted on by aminopeptidases, which are known to liberate an amino acid via scission of the peptide bond adjacent to the free amino group.

We found that GGMSC, when given as an acute bolus dose, is quantitatively absorbed from the gastrointestinal tract like MSC. Urinary excretion is apparently a major route for eliminating the excess Se from GGMSC. If biliary excretion is to play an important role in the metabolic disposition of GGMSC, we would have seen a marked increase in fecal Se on the 2nd day. This is clearly not the case. In fact, the amount of Se recovered in two consecutive 24-h urine samples of rats given GGMSC was strikingly similar to that of rats given MSC.

We also found that in rats fed either a GGMSC diet or a MSC diet for 1 month, the tissue Se accumulation profile was comparable between the two groups. In rats that had been challenged with a carcinogen, treatment with either reagent resulted in a lower prevalence of premalignant lesions in the mammary gland, and fewer mammary carcinomas when these early lesions were allowed to progress to malignancy. These two pathological end points responded equally well to intervention by GGMSC or MSC. Thus for all practical purposes, GGMSC is serving mainly as a carrier of MSC. After ingestion as a dietary constituent, the bulk (not necessarily all) of GGMSC is likely to be hydrolyzed by γ-glutamyl transpeptidase in the gastrointestinal tract (25), releasing MSC for absorption and systemic delivery to other tissues. Further confirmation of this hypothesis will require rigorous chemical speciation of individual Se metabolites. The methodology for this kind of sophisticated analysis still needs to be refined, especially with respect to its application to biological samples.

It is noteworthy that treatment with GGMSC or MSC for just 1 month was able to provide a lasting protection against subsequent cancer development. For cells that have sustained irreparable DNA damage, apoptosis is a means for their elimination. The appearance of a defined premalignant lesion (e.g., an IDP) is the net result of cell proliferation minus cell death. Thus a down-sizing in the population of premalignant lesions can in effect be achieved by enhancing cell death either in the absence of or in addition to decreasing cell proliferation. We have preliminary data indicating that MSC is able to induce apoptosis in IDP cells in vivo(26). We expect GGMSC to be just as effective as MSC in increasing apoptosis, although this remains to be confirmed.

cDNA microarray analysis is a powerful tool for gene expression studies because of its ability to simultaneously monitor the expression pattern of a wide spectrum of genes in a single experiment. Our plan, therefore, was to use this technology to elucidate the gene expression profile changes that occurred in the mammary gland after intervention with either GGMSC or MSC. If the molecular effects of GGMSC and MSC are closely related to each other, we should predict a high degree of concordance of the results obtained from the two compounds between the two analyses. The data summarized in Table 5 indeed confirm our expectation. It should be kept in mind that the microarray analysis was carried out at a particular point in time. Depending on whether GGMSC or MSC is used as the exogenous source of Se, the pharmacokinetics and pharmacodynamics of Se metabolism in the mammary epithelial cells may not exactly duplicate one another with these two treatments. That could be one of the underlying reasons why we failed to obtain identical mirror images from the GGMSC array data versus the MSC array data. We did not attempt to verify the alterations in transcript level by Northern blot analysis, because at this stage of our research, we are more concerned about the overall pattern of changes in gene expression induced in the target tissue by two metabolically related Se compounds. It would be imprudent to dismiss the possibility that the gene expression snapshot taken of the mammary organoid might have been distorted because of the rather vigorous procedure involved in isolating the target cells. These artifactual interferences, whatever they may be, should be universal to the control, the GGMSC, and the MSC samples. One needs to keep in mind that the array analysis was used here simply as a molecular thermometer to gauge the degree of similarity (or dissimilarity) of the responses to GGMSC or MSC. We have no intention to go beyond the boundary of this interpretation.

Although the Clontech Atlas rat cDNA Expression Array seemingly contains a “potpourri” of genes, we have learned a few valuable lessons concerning the application of the technology to the study of chemoprevention at the molecular level. For the purpose of the present discussion, we will highlight only two observations and their inferences in particular. Observation 1: we could not detect any changes in the expression of proliferating cell nuclear antigen, cyclins, cyclin-dependent kinase, cdk inhibitors, JNKs, or ERKs in either the MSC or GGMSC treatment group. Neither could we detect changes in the apoptosis regulatory genes including bcl-2, bax, Gadds, caspases, Fas-L, TNF-α and TNF receptor. These negative gene expression results are generally consistent with the negative immunohistochemistry results of cell cycle and apoptosis biomarkers from mammary gland sections of normal ductal and alveolar cells, as reported recently (10). As much as it is comforting to know that these two very different approaches are providing congruent information about the effect of Se, our intention is to explore new ways of making use of the microarray technology in chemoprevention research, and especially in a positive sense. What are the clues that can be gleaned from looking at hundreds of genes?

Observation 2: of an assortment of 588 genes in the Array, only 15 genes were found to be modulated by both GGMSC and MSC. Some of them are likely to be worthy of further investigation. The ones that pique our curiosity the most include superoxide dismutase (oxidant stress), c-jun (regulation of transcription), a G-protein (signal transduction), and adipocyte fatty acid binding protein (homologous to, or possibly the same as, mammary-derived growth inhibitor). Regardless of the opportunities open for more in-depth mechanism studies, an important message that can be distilled from the data is that it is possible to identify molecular targets as potential biomarkers of MSC intervention efficacy even if we are limited to working with normal differentiated cells.

In summary, the present study conclusively demonstrates that GGMSC is bioavailable. On the basis of the collective biology, biochemistry and molecular biology data, we believe that GGMSC is an effective anticancer agent with a mechanism of action similar to that of MSC. By extrapolation, it is reasonable to propose that plants producing other γ-glutamyl selenoamino acids, such as γ-glutamyl selenomethionine (4, 27, 28), are likely to be good sources of bioactive Se. This paper exemplifies a paradigm of collaborative research in natural product chemistry and biomedical science in developing new strategies for intervention of cancer.

Fig. 1.

Structures of GGMSC and MSC.

Fig. 1.

Structures of GGMSC and MSC.

Close modal
Fig. 2.

Phosphorimage of hybridization pattern of Clontech Atlas rat Expression Array of mammary epithelial cell samples from control versus GGMSC-treated rats (A) or MSC-treated rats (B).

Fig. 2.

Phosphorimage of hybridization pattern of Clontech Atlas rat Expression Array of mammary epithelial cell samples from control versus GGMSC-treated rats (A) or MSC-treated rats (B).

Close modal
Fig. 3.

Schematic comparison of the array data between control versus treatment with GGMSC (A) or MSC (B). Each box, representing a gene on the array, is divided in half horizontally. The top half, the ratio of the signal between the two arrays; bottom half, the difference of the signal between the two arrays. To be considered significantly different, the ratio signal and the difference signal must exceed their respective thresholds, which are set as follows: up-regulation ratio threshold, >2; up-regulation difference threshold, >10; down-regulation ratio threshold, <0.5; down-regulation difference threshold, <−10.

Fig. 3.

Schematic comparison of the array data between control versus treatment with GGMSC (A) or MSC (B). Each box, representing a gene on the array, is divided in half horizontally. The top half, the ratio of the signal between the two arrays; bottom half, the difference of the signal between the two arrays. To be considered significantly different, the ratio signal and the difference signal must exceed their respective thresholds, which are set as follows: up-regulation ratio threshold, >2; up-regulation difference threshold, >10; down-regulation ratio threshold, <0.5; down-regulation difference threshold, <−10.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported by National Cancer Institute Grants CA45164 and CA27706 and Roswell Park Cancer Institute Core Grant CA16056.

3

The abbreviations used: MSC, Se-methylselenocysteine; GGMSC, γ-glutamyl-MSC; MNU, methylnitrosourea; IDP, intraductal proliferation.

4

Y. Dong and C. Ip, unpublished data.

Table 1

Total Se excretion in urine of rats given a single dose of either GGMSC or MSCa

Treatment% of dose excretedb
0–24 h24–48 h
GGMSC 43.1 ± 3.6 6.7 ± 0.5 
MSC 37.4 ± 3.3 6.2 ± 0.6 
Treatment% of dose excretedb
0–24 h24–48 h
GGMSC 43.1 ± 3.6 6.7 ± 0.5 
MSC 37.4 ± 3.3 6.2 ± 0.6 
a

Rats were gavaged with a single dose of 40 μg of Se as either GGMSC or MSC (see “Materials and Methods” section for dose justification).

b

Results are expressed as mean ± SE (n = 6).

Table 2

Tissue Se levels in rats fed either GGMSC or MSCa

TreatmentSe concentration (μg/g or ml)b
LiverKidneyMammaryMusclePlasma
None 3.3 ± 0.1 4.0 ± 0.1 0.12 ± 0.01 0.72 ± 0.03 0.41 ± 0.02 
GGMSC 9.1 ± 0.6c 11.8 ± 0.8c 0.25 ± 0.02c 1.5 ± 0.05c 0.67 ± 0.04c 
MSC 10.5 ± 0.7c 13.4 ± 1.2c 0.29 ± 0.02c 1.6 ± 0.06c 0.70 ± 0.05c 
TreatmentSe concentration (μg/g or ml)b
LiverKidneyMammaryMusclePlasma
None 3.3 ± 0.1 4.0 ± 0.1 0.12 ± 0.01 0.72 ± 0.03 0.41 ± 0.02 
GGMSC 9.1 ± 0.6c 11.8 ± 0.8c 0.25 ± 0.02c 1.5 ± 0.05c 0.67 ± 0.04c 
MSC 10.5 ± 0.7c 13.4 ± 1.2c 0.29 ± 0.02c 1.6 ± 0.06c 0.70 ± 0.05c 
a

Rats were fed either GGMSC or MSC at a level of 3 ppm Se in the diet for 1 month.

b

Results are expressed as mean ± SE (n = 6).

c

P < 0.05, compared with the corresponding control value.

Table 3

Reduction in the number of IDP lesions by GGMSC or MSC in the mammary gland of rats given MNUa

TreatmentSize distribution of IDP lesionsTotal no.
≤10 sections11–20 sections21–30 sections31–40 sections>41 sections
None 11 10 42 
GGMSC 22b 
MSC 18b 
TreatmentSize distribution of IDP lesionsTotal no.
≤10 sections11–20 sections21–30 sections31–40 sections>41 sections
None 11 10 42 
GGMSC 22b 
MSC 18b 
a

The data represent all the lesions from a total of six rats/group.

a

P < 0.05, compared with the control group.

Table 4

Mammary cancer prevention by GGMSC or MSC given for 1 month after MNU administrationa

TreatmentTumor incidenceTotal no. of tumors
None 25/30 63 
GGMSC 12/30b 36b 
MSC 10/30b 32b 
TreatmentTumor incidenceTotal no. of tumors
None 25/30 63 
GGMSC 12/30b 36b 
MSC 10/30b 32b 
a

Rats were given either GGMSC or MSC at 3 ppm Se for 1 month after MNU administration, and then switched back to a basal diet containing 0.1 ppm Se. The basal diet was continued for 5 more months until the end of the experiment.

b

P < 0.05, compared with the corresponding control value.

Table 5

Summary of gene expression profile changes induced by either GGMSC or MSC in mammary epithelial cells of rats given NMU

The values represent the fold of change compared with the control.a
 GGMSC MSC 
Genes up-regulated   
Prothymosin-α 8.0×  
Copper-zinc superoxide dismutase 7.0× 3.5× 
DNA-binding protein inhibitor ID1 2.5×  
Macrophage inflammatory protein-2 precursor  2.0× 
Genes down-regulated   
c-jun 4.3× 2.1× 
Glutathione S-transferase P subunit 2.2× ndb 
Clusterin, apoliprotein J,sulfated glycoprotein 2 2.3× 2.3× 
Nur 77 early response protein, NGF-1 7.0× 7.0× 
Guanine nucleotide-binding protein G(I)α2 subunit 2.5× nd 
40S ribosomal protein S11 2.5× 2.3× 
40S ribosomal protein S12  2.6× 
40S ribosomal protein S17 2.1× 2.6× 
CD 87 2.6× 3.0× 
Na+/K+ ATPase α1 subunit 2.3× 7.0× 
Na+/K+ ATPase β3 subunit  5.0× 
Tissue inhibitor of metalloproteinase 2 2.3× 3.0× 
Tissue inhibitor of metalloproteinase 3 5.5× 5.5× 
Proteasome subunit RC6-1 2.7× 2.7× 
Proteasome activator rPA28 subunit α 3.0× 4.0× 
Mast cell protease 1 precursor (RMCP-1) 2.0× 6.0× 
Adipocyte fatty acid-binding protein  3.8× 
The values represent the fold of change compared with the control.a
 GGMSC MSC 
Genes up-regulated   
Prothymosin-α 8.0×  
Copper-zinc superoxide dismutase 7.0× 3.5× 
DNA-binding protein inhibitor ID1 2.5×  
Macrophage inflammatory protein-2 precursor  2.0× 
Genes down-regulated   
c-jun 4.3× 2.1× 
Glutathione S-transferase P subunit 2.2× ndb 
Clusterin, apoliprotein J,sulfated glycoprotein 2 2.3× 2.3× 
Nur 77 early response protein, NGF-1 7.0× 7.0× 
Guanine nucleotide-binding protein G(I)α2 subunit 2.5× nd 
40S ribosomal protein S11 2.5× 2.3× 
40S ribosomal protein S12  2.6× 
40S ribosomal protein S17 2.1× 2.6× 
CD 87 2.6× 3.0× 
Na+/K+ ATPase α1 subunit 2.3× 7.0× 
Na+/K+ ATPase β3 subunit  5.0× 
Tissue inhibitor of metalloproteinase 2 2.3× 3.0× 
Tissue inhibitor of metalloproteinase 3 5.5× 5.5× 
Proteasome subunit RC6-1 2.7× 2.7× 
Proteasome activator rPA28 subunit α 3.0× 4.0× 
Mast cell protease 1 precursor (RMCP-1) 2.0× 6.0× 
Adipocyte fatty acid-binding protein  3.8× 
a

Calculated based on the criteria set for signal threshold, ratio threshold, and difference threshold according to the instructions contained in the manual of the AtlasImage software program.

b

nd, not detectable.

We thank Todd Parsons and Cassandra Hayes for their technical assistance.

1
Ip C. Lessons from basic research in selenium and cancer prevention.
J. Nutr.
,
128
:
1845
-1854,  
1998
.
2
Ip C., Birringer M., Block E., Kotrebai M., Tyson J. F., Uden P. C., Lisk D. J. Chemical speciation influences comparative activity of selenium-enriched garlic and yeast in mammary cancer prevention.
J. Agr. Food Chem.
,
48
:
2062
-2070,  
2000
.
3
Ip C., Lisk D. J. Efficacy of cancer prevention by high selenium-garlic is primarily dependent on the action of selenium.
Carcinogenesis (Lond.)
,
16
:
2649
-2652,  
1995
.
4
Kotrebai M., Birringer M., Tyson J. F., Block E., Uden P. C. Characterization of selenium compounds in selenium-enriched and natural sample extracts using perfluorinated carboxylic acids as HPLC ion-pairing agents with ICP- and electro spray ionization-mass spectrometric detection.
Analyst
,
125
:
71
-78,  
2000
.
5
Virtanen A. I. Studies on organic sulfur compounds and other labile substances in plants.
Phytochemistry
,
4
:
207
-228,  
1965
.
6
Koch H. P. Lawson L. D. eds. .
Garlic. The Science and Therapeutic Applications of Allium sativum L. Related Species
,
: Williams & Wilkens Baltimore, MD  
1996
.
7
Block E. The organosulfur chemistry of the genus Allium–Implications for organic sulfur chemistry.
Angew. Chem. Int. Ed. Engl.
,
31
:
1135
-1178,  
1992
.
8
Shrift A. Metabolism of selenium by plants and microorganisms Klayman D. K. Gunther W. W. H. eds. .
Organic Selenium Compounds: Their Chemistry and Biology
,
:
763
-814, Wiley-Interscience New York  
1973
.
9
Ip C., Zhu Z., Thompson H. J., Lisk D., Ganther H. E. Chemoprevention of mammary cancer with selenium-allylselenocysteine and other selenoamino acids in the rat.
Anticancer Res.
,
19
:
2875
-2880,  
1999
.
10
Ip C., Thompson H. J., Ganther H. E. Selenium modulation of cell proliferation and cell cycle biomarkers in normal and premalignant cells of the rat mammary gland.
Cancer Epidemiol. Biomark. Prev.
,
9
:
49
-54,  
2000
.
11
Ip C., Thompson H. J., Zhu Z., Ganther H. E. In vitro and in vivo studies of methylseleninic acid: evidence that a monomethylated selenium metabolite is critical for cancer chemoprevention.
Cancer Res.
,
60
:
2882
-2886,  
2000
.
12
Block E., Birringer M., Jiang W., Nakahodo T., Thompson H. J., Toscano P. J., Uzar H., Zhang X., Zhu Z. Allium chemistry: synthesis, natural occurrence, biological activity, and chemistry of Se-alk(en)ylselenocysteines and their γ-glutamyl derivatives and oxidation products.
J. Agr. Food Chem.
,
49
:
458
-470,  
2001
.
13
Ip C., Hayes C., Budnick R. M., Ganther H. E. Chemical form of selenium, critical metabolites, and cancer prevention.
Cancer Res.
,
51
:
595
-600,  
1991
.
14
Olson O. E., Palmer I. S., Carey E. E. Modification of the official fluorometric method for selenium in plants.
J. Assoc. Offic. Anal. Chem.
,
58
:
117
-121,  
1975
.
15
Ip C., Lisk D. J., Ganther H., Thompson H. J. Triphenylselenonium and diphenylselenide in cancer chemoprevention: comparative studies of anticarcinogenic efficacy, tissue selenium levels, and excretion profile.
Anticancer Res.
,
17
:
3195
-3200,  
1997
.
16
Russo J., Tay L. K., Russo I. H. Differentiation of the mammary gland and susceptibility to carcinogenesis.
Breast Cancer Res. Treat.
,
2
:
5
-73,  
1982
.
17
McCullagh P. Nelder J. A. eds. .
Generalized Linear Models
,
: Chapman and Hall, Ltd. London  
1989
.
18
Ip C., Lisk D. J., Thompson H. Selenium-enriched garlic inhibits the early stage but not the late stage of mammary carcinogenesis.
Carcinogenesis (Lond.)
,
17
:
1979
-1982,  
1996
.
19
Horvath P. M., Ip C. Synergistic effect of vitamin E and selenium in chemoprevention of mammary carcinogenesis in rats.
Cancer Res.
,
43
:
5335
-5341,  
1983
.
20
Hahm H. A., Ip M. M. Primary culture of normal rat mammary epithelial cells within a basement membrane matrix. I. Regulation of proliferation by hormones and growth factors.
In Vitro Cell. Dev. Biol.
,
26
:
791
-802,  
1990
.
21
Nigan S. N., McConnell W. B. Seleno amino compounds from Astragalus bisculcatus. Isolation and identification of γ-l-glutamyl-Se-methylseleno-l-cysteine and selenium-methylseleno-l-cysteine.
Biochim. Biophys. Acta.
,
192
:
185
-190,  
1969
.
22
Guo X., Wu L. Distribution of free seleno-amino acids in plant tissue of Melilotus indica L. grown in selenium-laden soils.
Ecotoxicol. Environ. Saf.
,
39
:
207
-214,  
1998
.
23
Ip C., Lisk D. J., Stoewsand G. S. Mammary cancer prevention by regular garlic and selenium enriched garlic.
Nutr. Cancer
,
17
:
279
-286,  
1992
.
24
Ip C., Lisk D. J. Enrichment of selenium in allium vegetables for cancer prevention.
Carcinogenesis (Lond.)
,
15
:
1881
-1885,  
1994
.
25
Taniguchi N., Ikeda Y. γ-Glutamyl transpeptidase: catalytic mechanism and gene expression.
Adv. Enzymol. Relat. Areas Mol. Biol.
,
72
:
239
-278,  
1998
.
26
Ip, C., Dong, Y. Methylselenocysteine modulates proliferation and apoptosis biomarkers in premalignant lesions of the rat mammary gland. Anticancer Res., in press, 2001
27
Kotrebai M., Birringer M., Tyson J. F., Block E., Uden P. C. Identification of the principal selenium compounds in selenium-enriched natural sample extracts by ion-pair liquid chromatography with inductively coupled plasma- and electrospray ionization-mass-spectrometric detection.
Anal. Commun.
,
36
:
249
-252,  
1999
.
28
Kotrebai M., Tyson J. F., Block E., Uden P. C. HPLC of selenium compounds utilizing perfluorinated carboxylic acid ion-pairing agents and inductively coupled plasma- and electrospray ionization-mass spectrometric detection.
J. Chromatogr.
,
866
:
51
-63,  
2000
.