We have been analyzing the effects of Kunitz-type protease inhibitor bikunin, also known as urinary trypsin inhibitor (UTI) derived from human amniotic fluid or urine, in detail, and have confirmed that it is useful for the treatment of mouse and human cancer metastasis (1). Bikunin suppresses upregulation of urokinase-type plasminogen activator (uPA) and its specific receptor uPAR expression, activation of mitogen-activated protein (MAP) kinase, cancer cell invasion in vitro, and peritoneal-disseminated metastasis in vivo. Treatment of cancer patients with bikunin may be beneficial in the adjuvant setting to delay the onset of metastasis development and/or in combination with cytotoxic agents to improve treatment efficacy in patients with advanced ovarian cancer (1). Bikunin also exhibits anti-inflammatory activity in protection against inflammation (2). Previously, we showed that bikunin significantly inhibits lipopolysaccharide (LPS)-induced TNF-α production, suggesting a mechanism of anti-inflammation by bikunin through control of cytokine induction during inflammation (2). It becomes difficult to collect materials (human urine) for purification of bikunin. Therefore, we have been investigating another Kunitz-type trypsin inhibitor from plants.
Proteins of soybeans are widely used in animal and human nutrition (3). In addition to the bulk of the seed storage proteins, approximately 6% of soybean proteins are classified as inhibitors of trypsin and chymotrypsin. The two major classes of inhibitors are the Kunitz trypsin inhibitor (KTI), which inhibits trypsin, and the Bowman-Birk inhibitor (BBI), which inhibits trypsin and chymotrypsin (4). Recently, we tested the effect of KTI and BBI on signal transduction involved in the expression of uPA and tumor invasion (5), showing that uPA upregulation observed in ovarian cancer cells was inhibited by preincubation of the cells with KTI, whereas BBI failed to repress uPA upregulation, and that cell invasiveness was specifically inhibited by treatment of the cells with KTI (5). More recently, we have analyzed the effects of KTI in detail and have confirmed that KTI is useful for the treatment of cancer metastasis in vivo: dietary supplementation of KTI significantly prevents spontaneous, experimental, and peritoneal disseminated metastasis in a mouse model (6). On the other hand, BBI has been extensively studied for its ability to prevent carcinogenesis in many different model systems (7).
In addition to its protease inhibitory effects, bikunin and KTI play a role in inhibiting the MAP kinase signaling cascade (4, 5), which results in suppressing expression of several target molecules, including the TNF-α responsible for inflammation. Suppression of signaling activation may account for prevention by KTI against inflammation. Research from various directions has clarified many aspects of the biological function of KTI, and provides a new vista on its role in inflammation and cancer (5, 6). Thus, a growing number of evidence demonstrated that KTI is not only a protease inhibitor but is also an anti-inflammatory agent. However, such mechanisms do not readily explain the anti-inflammatory function of KTI in vivo, which is largely unaddressed at present.
The pivotal role of TNF-α in inflammation (8) and the potent anti-inflammatory activity of KTI raises the question of whether the induction of TNF-α during inflammation serves as a target of anti-inflammation by KTI. To explore critical role of exogenous KTI and BBI, we used an LPS-induced death model in vivo. A new approach is to use KTI and BBI, preferably as an oral therapy, suitable for chronic administration and as a continual restraint of inflammation. This will be possible experimentally with KTI and BBI, which is attributed to its greater solubility. In the present study, i.p. and daily oral administration of KTI and BBI were examined to determine whether they would inhibit LPS-induced death. Furthermore, we investigated the effects of KTI and BBI treatment on cytokine expression and phosphorylation of MAP kinase in peritoneal macrophages. Our findings provide new insights into a mechanism of protection against inflammatory diseases by KTI.
MATERIALS AND METHODS
Highly purified preparations of KTI and BBI were kindly supplied by Fuji Oil Co. (Tokyo, Japan) (5, 6). Anti-ERK1/2, anti-phospho-ERK1/2, anti-p38 (all Santa Cruz Biotechnology, Santa Cruz, CA), anti-JNK, anti-phospho-JNK, and anti-phospho-p38 antibodies (all Cell Signaling Technology, Beverly, MA) as primary antibodies, and horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin (Ig) G (Sigma-Aldrich, Tokyo, Japan) as a secondary antibody. Unless otherwise specified, reagents for cell culture and all other chemicals including protease inhibitors were of the highest purity and were obtained from Sigma-Aldrich.
All protocols for animal studies were reviewed and approved by the Animal Research and Ethics Board of Hamamatsu University. The care and use of the animals were in accordance with the Institution's guidelines. Specific pathogen-free C57BL/6 mice were purchased from SLC (Hamamatsu, Ichinocho, Japan). Mice matched for gender (female) and age (10-15 weeks) were used in the following LPS (Escherichia coli serotype O111:B4; Sigma; 1 mg of LPS corresponds to ∼1,000,000 endotoxin units) shock experiments. The animals were maintained under the following standard conditions: 22°C ± 2°C, 45 ± 10% relative humidity, and 12-h light/12-h dark cycles each day. Powdered CE-2 (CLEA Japan, Tokyo, Japan) was used as a basal diet. Experimental diets mixed with KTI (5, 15, or 50 g/kg) or BBI (5, 15, or 50 g/kg) were freshly prepared each week, and a feeder was placed in each cage. The experimental diets and water were offered ad libitum. They were randomized into groups so that mean body weights in the group were approximately equal.
Preparation of macrophages and cell culture
Elicited peritoneal macrophage cells were prepared from female mice 4 days after intraperitoneal inoculation of 1 mL of 10% thioglycollate broth (9). Cells were seeded at 2.0 × 106 in 6-well plates. After incubation for 1 h, nonadherent cells were removed by washing the wells with RPMI-HEPES twice, and remnant cells were cultured and stimulated for different periods of time in RPMI 1640 medium containing 10% fetal calf serum (FCS). Cells were counted using a hemocytometer, and viability was assessed by trypan blue staining. Cells were then treated with several reagents, including SBTIs (KTI and BBI; 0.2, 1, or 5 μM) in the presence or absence of test drugs in serum-free medium. The conditioned medium was produced by incubating the cells (∼90% confluent) for 9 h in a serum-free medium. All samples were stored at −20°C until use. In parallel, cells treated in the same condition in different dishes were harvested and counted using a hemocytometer. Protein was measured by the Bradford assay (Bio-Rad, Hercules, CA) using bovine serum albumin as the standard.
Cell activation was performed in RPMI containing serum and LPS in the absence or presence of KTI or BBI at the concentrations indicated in the text for a different period of time at 37°C in 5% CO2.
Assay of plasma levels of TNF-α, IL-1β, and IL-6
Mice pretreated with or without KTI or BBI were injected intraperitoneally with LPS at a dose of 30 mg/kg. Blood was taken from the heart into heparinized syringes under mild anesthesia with diethyl ether at the indicated times after LPS injection. Blood was collected 1 h later for the assay of TNF-α, and 5 h later for the assay of IL-1β or IL-6. Plasma was separated by centrifugation at 5000g for 10 min. Concentrations of TNF-α, IL-1β, and IL-6 in plasma were determined with immunoassay kits (TFB, Tokyo, Japan).
Western blot analysis
For the analysis of total ERK1/2 and its phosphorylation, Western blot was performed as described (5, 6).
The nonparametric Mann-Whitney U test was used to determine differences between two groups, and statistical significance was accepted when P < 0.05. Statistical differences in survival curves among the groups of mice were analyzed by log-rank test. The Instat software package (GraphPad Software, San Diego, CA) was used for statistical analyses.
Intraperitoneal administration of KTI protects against bacterial LPS-induced lethality
The mice without LPS injection in each group did not show any significant difference in body weight (data not shown). We were interested in investigating whether i.p. injection or daily oral administration of soybean trypsin inhibitor can protect LPS-induced lethality in mice in vivo. In an initial experiment, we examined the suppressing effects of i.p. injection of KTI (50 mg/kg) and BBI (50 mg/kg) on LPS-induced lethality after i.p. injection of LPS. The results of a representative experiment are shown in Figur In control mice, LPS (30 mg/kg) induced 90% lethality by 6 h after challenge, whereas mice pretreated with i.p. injection of KTI all survived. Twelve hours after LPS challenge (30 mg/kg), 10/10 control mice died after LPS challenge in contrast to 3/10 KTI-pretreated mice. At 24 h after LPS injection, the survival rate was 0% for control mice compared with 70% for KTI mice. The KTI mice were found to exhibit significantly lower mortality than controls after i.p. injection of LPS at a dose of 30 mg/kg (P = 0.045). At a dose of 30 mg/kg LPS, the survival rate was higher in mice pretreated with the i.p. injection of 10 mg/kg KTI than in controls, but was not statistically significant (P = 0.493, not shown). At 10 mg/kg, LPS lethality was 20% for KTI mice, in contrast to 80% for control mice after 24 h of treatment (P = 0.219; Fig. 1B). Even at the dose of 5 mg/kg, the survival rate was higher in KTI mice than in controls, but was not statistically significant (P = 0.613; data not shown). At all of the LPS concentrations tested, a lower level of death was induced in KTI mice compared with controls. In contrast, BBI failed to improve LPS-induced lethality.
Daily oral administration of KTI protects against bacterial LPS-induced lethality
We next examined whether daily oral administration of KTI can protect LPS-induced lethality in mice in vivo (Fig. 2). Intraperitoneal injection of 30 mg/kg LPS into mice resulted in death of 80% (8/10) of the mice within 24 h after injection (Fig. 2A). In contrast, 9/10 mice were rescued by the daily dietary supplementation of KTI (50 g/kg; P = 0.041), respectively. Intraperitoneal injection of 10 mg/kg LPS resulted in death of 50% (5/10) of the mice within 24 h after injection. A lower level of death was induced in KTI mice compared with controls, but was not statistically significant (P = 0.375). These significant differences in mortality demonstrate that KTI protects against LPS-induced death, irrespective of whether KTI was administered intraperitoneally or orally.
Suppressing effects of KTI on TNF-α, IL-1β, and IL-6 levels in plasma after challenge with LPS
To determine the effect of daily oral administration of KTI or BBI on LPS-induced proinflammatory cytokine levels, mice were pretreated with daily dietary supplementation of KTI or BBI and were injected with LPS, and the plasma cytokine levels were determined 1 h for TNF-α (Fig. 3A) and 5 h for IL-1β (Fig. 3B) and IL-6 (Fig. 3C) after LPS injection. Daily oral administration of KTI significantly reduced cytokine expression in a dose-dependent manner. Daily oral administration of KTI together with 30 mg/kg LPS largely inhibited the LPS-induced maximum levels of TNF-α, IL-1β, and IL-6 for 45%, 47%, and 38%, respectively. The proinflammatory cytokines TNF-α and IL-1 may play an important role in the fatal outcome of sepsis (10). Therefore, inhibition of the LPS-induced plasma levels of these cytokines by concomitant daily oral administration of KTI is expected to render rodents less susceptible to a lethal dose of LPS. In contrast, daily oral administration of BBI did not reduce LPS-induced upregulation of cytokine expression.
Suppressing effect of KTI on LPS-induced upregulation of cytokine expression in peritoneal macrophage cells isolated from mice
We investigated whether LPS-induced upregulation of TNF-α expression was affected by KTI or BBI in peritoneal macrophage cells isolated from mice in an in vitro experiment. We used peritoneal macrophages to investigate reasons for the lower levels of TNF-α in plasma of KTI mice after LPS administration. We determined the cytokine levels in medium from macrophage cells after LPS challenge. LPS stimulated TNF-α synthesis in control macrophages in a dose- and time-dependent manner (data not shown). Macrophages pretreated with KTI significantly reduced TNF-α levels when stimulated with LPS at 100 ng/mL for 9 h (Fig. 4A). Thus, LPS-induced upregulation of TNF-α expression was significantly suppressed by KTI (5 μM). In contrast, BBI did not affect the LPS-induced upregulation of TNF-α expression. These results strongly suggest a role of KTI in direct suppression of LPS-induced upregulation of TNF-α expression.
In a parallel experiment, we determined the direct expression of IL-1β and IL-6 in macrophage cells pretreated with different concentrations of KTI or BBI after LPS challenge. As shown in Figure 4, B and C, KTI decreased LPS-stimulated IL-1β and IL-6 synthesis in macrophage cells in a dose-dependent manner. BBI has no ability to suppress LPS-induced upregulation of cytokine expression. Macrophage cells treated with KTI and BBI showed no evidence of cytotoxicity as measured by the trypan blue exclusion test (data not shown).
KTI specifically suppresses phosphorylation of MAP kinase proteins in LPS-stimulated macrophage cells
LPS stimulation of monocytes activates several intracellular signaling pathways that include three MAP kinase pathways: ERK1/2, JNK, and p38. These signaling pathways, in turn, activate a variety of transcription factors that include NF-κB (p50/p65) and AP-1 (c-Fos/c-Jun), which coordinate the induction of many genes encoding inflammatory mediators (11). Macrophages were stimulated with LPS (100 ng/mL), and the lysates were immunoblotted with antiphosphoprotein antibodies. This resulted in the rapid (within 5 min) phosphorylation of ERK1/2, JNK, and p38, which peaked at 15 min, and was still apparent after 30 min (H. Kobayashi, R. Yoshida, Y. Kanada, Y. Fukuda, and T. Yagyu, unpublished data).
To investigate whether LPS-mediated signaling is regulated by KTI or BBI, we examined the effect of these compounds on LPS-stimulated signaling activation. The cells pretreated with KTI (0.2, 1, or 5 μM) or BBI (5 μM) for 1 h were treated with LPS (100 ng/mL) for 15 min, and then the phosphorylated bands were analyzed by Western blot (Fig. 5). The ERK1/2 phosphorylation was down-regulated by KTI in a dose-dependent manner (Fig. 5A). In a parallel experiment, KTI also significantly suppressed LPS-induced upregulation of phosphorylation of JNK (Fig. 5B) and p38 (Fig. 5C) in a dose-dependent manner. KTI (1 μM) was sufficient to suppress activation of MAP kinases, demonstrating that KTI may play an important role in the suppression of LPS-induced activation of MAP kinase signaling pathway. These data speculate that KTI may inhibit TNF-α expression, possibly through suppression of the MAP kinase-dependent signaling cascade. As expected, BBI did not inhibit LPS-induced phosphorylation of ERK1/2.
Intense efforts have been made in our laboratory to understand the role of Kunitz-type protease inhibitors in cancer and inflammation. Similar to human urine-derived bikunin, a soybean KTI has been assigned a classical protease inhibitory role. We have been investigating the mechanism leading to the inhibitor-dependent suppression of MAP kinase signalings (2, 5, 6, 12-16). In the present study, we used i.p. injection or daily oral administration of KTI, which exhibited significantly longer survival against lethal doses of bacterial LPS, suggesting a potential involvement of the KTI in the progression and pathogenesis of endotoxic shock. Our interesting findings presented here showed that dietary supplementation of KTI plays a role in suppressing endotoxic shock in vivo. Furthermore, the anti-inflammatory effects of KTI were demonstrated in the absence of side effects such as weight loss (data not shown). We also showed that KTI significantly reduced TNF-α, IL-1β, and IL-6 expression in response to LPS in vitro and in vivo. Additionally, the in vitro experiments suggest that KTI down-regulates cytokines expression through suppression of LPS-induced activation of three MAP kinase pathways. Thus, the results in the current study provide the first evidence that dietary KTI reduces LPS-induced lethality, suggesting that KTI would be the more likely candidate for use in future clinical trials.
In the animal model, we applied these compounds as a mixture with a basal diet (up to 50 g of inhibitor in 1 kg powdered CE-2). Our preliminary data demonstrated that approximately 10 mg of purified KTI/mouse/day is needed to reduce mortality and cytokine levels, when the compound was to be given by lavage (H. Kobayashi, R. Yoshida, Y. Kanada, Y. Fukuda, and T. Yagyu, unpublished data). We estimate that at least 30 to 40 g of purified inhibitor would be needed to treat a patient with about 70 kg of body weight. We cannot simply recommend the same amount of KTI because of species specificity. Further studies are required for clinical trials.
On the basis of our previous data, we had shown that KTI binds to certain cells via specific binding sites (5). Membrane-associated KTI binding sites are believed to represent the rate-limiting step for KTI-dependent signal transduction or cellular uptake of the KTI. Although the inhibitory effects of bikunin and KTI imply a similar role for MAP kinase-dependent signaling events, our previous study (5) showed that KTI does not inhibit bikunin binding to the cells. The inhibitory effect of KTI may be seen at different sites rather than we have previously observed for bikunin (5). Thus, both inhibitors may act by different mechanisms to inhibit MAP kinase. These data allow us to speculate that more than one binding site for Kunitz-type inhibitors are expressed in some cells and that KTI and bikunin may act at separate sites.
Recent data indicate that bikunin and KTI may have other functions that are unrelated to protease inhibition (5, 6). At present, we have no data as to whether the KTI-dependent protease inhibitory activity may be crucial in the reduction of LPS-induced upregulation of cytokine expression or LPS-induced lethality in animals. KTI inhibited the MAP kinase signaling cascade, whereas BBI, another type of soybean trypsin inhibitor, was inactive. Therefore, it is unlikely that trypsin inhibitory activity alone is specific for suppression of signaling cascades. The specific molecular structure of Kunitz-type inhibitor may be important for its action. In support of this idea, the protease inhibitor domain of bikunin is inactive to prevent signal transduction (17). The domain of bikunin can be divided into two regions: N-terminal Kunitz domain I region (amino acids 1-77) and a C-terminal Kunitz domain II region (amino acids 78-143) with the active site of trypsin inhibitor. Domain I is involved in the cell-associated binding site and thus, the protease inhibitor domain does not contribute to cell-binding specificity. Furthermore, domain II alone was inactive in preventing signal transduction (17). This may rule out the trypsin inhibitory domain as the site of action.
It has recently been documented that activated proteases from the pancreas are released from the gut during shock conditions, leading to secondary activation of inflammatory peptide fragments and increased mortality (18). Intestinal pancreatic protease inhibition significantly attenuates intestinal ischemia-induced shock by reducing the systemic inflammatory response syndrome and gut injury (18). Therefore, another possibility is that the effect is not directly upon the innate immune cell but rather a nonspecific trypsin inhibitor effect within the gut, preventing the generation of secondary mediators released by the damaged gut into the circulation. At this time, we have no data as to whether this inhibitor is absorbed intact from the gut such that there is functional in vivo inhibition intravascularly, or whether there is antitrypsin activity in the gut after oral administration. In contrast to this idea, however, our results demonstrated that LPS-mediated signaling is directly abrogated by KTI in an in vitro assay system using peritoneal macrophages. Therefore, KTI's function may not be a nonspecific antitrypsin effect within the gut. We finally confirmed that KTI did not directly affect LPS activation when cells were challenged with LPS preincubated with KTI for 1 h (H. Kobayashi, R. Yoshida, Y. Kanada, Y. Fukuda, and T. Yagyu, unpublished data).
It has been established that cells treated with LPS demonstrate significant stress fiber polymerization with pseudopodia formation (19). ERK1/2 activation is frequently dependent upon an intact cytoskeletal microenvironment within the macrophage (19). We must note that stress fiber formation requires de novo protein synthesis, including MAP kinases (19). Therefore, we speculate that KTI inhibits MAP kinase activation, which results in suppression of LPS-induced stress fiber formation. This may abrogate LPS-induced signaling pathways such as cytokine production. Macrophages treated with KTI alone did not show the morphological change when cells were not stimulated with LPS. Therefore, we speculate that this inhibitor does not directly affect the cytoskeleton of the inflammatory cells.
We believe that the observed effects of KTI are due to inhibition of these MAP kinase pathways. There is some evidence (20) that inhibitors of MAP kinase are effective in animal models of shock: the acute respiratory distress syndrome (ARDS) and shock are major causes of morbidity after injury. Alveolar macrophage chemokine and cytokine release after hemorrhage and sepsis is regulated by NF-κB and MAP kinase. Inhibition of NF-κB or the upstream MAP kinase significantly decreased LPS-stimulated macrophage activation. Because enhanced release of inflammatory mediators by macrophages may contribute to ARDS after severe trauma, inhibition of intracellular signaling pathways represents a target to attenuate organ injury under those conditions (20).
Humans are consuming some active trypsin inhibitor in their daily lives (21). However, the effect of trypsin inhibitor in soybean on the prevention of inflammation has not been reported. Soy intakes has been associated with reduced rates of breast cancer (22). Furthermore, BBI can effectively inhibit experimental tumorigenesis in animals (23). Naturally occurring protease inhibitors are also occurring in other plants, such as in potato tubers. However, there have been no reports on suppression of inflammation by protease inhibitors derived from other plants.
In conclusion, our findings presented here suggest for the first time that i.p. and oral administration of soybean-derived KTI plays a role in suppressing LPS-induced lethality or endotoxic shock. We postulate that in severe gram-negative bacterial infection, KTI may form a defense mechanism against the development of sepsis. Finally, the present study should be of great value for research on the development of cytokine production-blocking agents as therapeutic drugs for septic shock or severe inflammation. Treatment might still be very beneficial for patients with inflammation because KTI might be provided in a large scale and at a low price from soybeans.
1. Kobayashi H, Suzuki M, Hirashima Y, Terao T: The protease inhibitor bikunin, a novel anti-metastatic agent. Biol Chem
2. Matsuzaki H, Kobayashi H, Yagyu T, Wakahara K, Kondo T, Kurita N, Sekino H, Inagaki K, Suzuki M, Kanayama N, Terao T: Bikunin inhibits lipopolysaccharide
-induced tumor necrosis factor α induction in macrophages. Clin Diagn Lab Immunol
3. Friedman M, Brandon DL: Nutritional and health benefits of soy proteins. J Agric Food Chem
4. Birk Y: Protein proteinase inhibitors in legume seeds: overview. Arch Latinoam Nutr
5. Kobayashi H, Suzuki M, Kanayama N, Terao T: A soybean Kunitz trypsin inhibitor
suppresses ovarian cancer cell invasion by blocking urokinase upregulation. Clin Exp Metast
6. Kobayashi H, Fukuda Y, Yoshida R, Kanada Y, Nishiyama S, Suzuki M, Kanayama N, Terao T: Suppressing effects of dietary supplementation of soybean
trypsin inhibitor on spontaneous, experimental and peritoneal disseminated metastasis in mouse model. Int J Cancer
7. Kennedy AR: Chemopreventive agents: protease inhibitors. Pharmacol Ther
8. Hanada T, Yoshimura A: Regulation of cytokine signaling and inflammation
. Cytokine Growth Factor Rev
9. Seledtsova GV, Seledtsov VI, Taraban VY, Samarin DM, Kozlov VA: A role for interferon-γ and transforming growth factor-β in erythroid cell-mediated regulation of nitric oxide production in macrophages. Immunology
10. Imahara SD, O'Keefe GE: Genetic determinants of the inflammatory response. Curr Opin Crit Care
11. Guha M, Mackman N: LPS induction of gene expression in human monocytes. Cell Signal
12. Kobayashi H, Yagyu T, Inagaki K, Kondo T, Suzuki M, Kanayama N, Terao T: Bikunin plus paclitaxel markedly reduces tumor burden and ascites in mouse model of ovarian cancer. Int J Cancer
13. Kobayashi H, Yagyu T, Inagaki K, Kondo T, Suzuki M, Kanayama N, Terao T: Therapeutic efficacy of once-daily oral administration of a Kunitz-type protease inhibitor, bikunin, in a mouse model and in human cancer. Cancer
14. Tanaka Y, Kobayashi H, Suzuki M, Kanayama N, Suzuki M, Terao T: Upregulation of bikunin in tumor-infiltrating macrophages as a factor of favorable prognosis in ovarian cancer. Gynecol Oncol
15. Suzuki M, Kobayashi H, Tanaka Y, Kanayama N, Terao T: Reproductive failure in mice lacking inter-α-trypsin inhibitor (ITI): ITI target genes in mouse ovary identified by microarray analysis. J Endocrinol
16. Yagyu T, Kobayashi H, Wakahara Y, Matsuzaki H, Kondo T, Kurita N, Sekino H, Inagaki H, Suzuki M, Kanayama N, Terao T: A Kunitz-type protease inhibitor bikunin disrupts ligand-induced oligomerization of receptors for transforming growth factor (TGF)-β and subsequently suppresses TGF-β signalings. FEBS Lett
17. Suzuki M, Kobayashi H, Tanaka Y, Hirashima Y, Terao T: Structure and function analysis of urinary trypsin inhibitor (UTI): identification of binding domains and signaling property of UTI by analysis of truncated proteins. Biochim Biophys Acta
18. Fitzal F, DeLano FA, Young C, Rosario HS, Schmid-Schonbein GW: Pancreatic protease inhibition during shock attenuates cell activation and peripheral inflammation
. J Vasc Res
19. Cuschieri J, Gourlay D, Garcia I, Jelacic S, Maier RV: Hypertonic preconditioning inhibits macrophage responsiveness to endotoxin. J Immunol
20. Jarrar D, Kuebler JF, Rue LW 3rd, Matalon S, Wang P, Bland KI, Chaudry IH: Alveolar macrophage activation after trauma-hemorrhage and sepsis is dependent on NF-κB and MAPK/ERK mechanisms. Am J Physiol Lung Cell Mol Physiol
21. Miyagi Y, Shinjo S, Nishida R, Miyagi C, Takamatsu K, Yamamoto T, Yamamoto S: Trypsin inhibitor activity in commercial soybean
products in Japan. J Nutr Sci Vitaminol (Tokyo)
22. Lu LJ, Anderson KE, Grady JJ, Kohen F, Nagamani M: Decreased ovarian hormones during a soya diet: implications for breast cancer prevention. Cancer Res
23. Kennedy AR, Kritchevsky D, Shen WC: Effects of spermine-conjugated Bowman-Birk inhibitor (spermine-BBI) on carcinogenesis and cholesterol biosynthesis in mice. Pharm Res