In the Thick of Air Pollution: Particles May Contribute to Atherosclerosis

Fine particles damage more than just the lungs—the effects reach all the way to the inner lining of the arteries, report researchers from the University of Southern California’s Keck School of Medicine, led by Nino Kunzli [EHP 113:201–206]. Their paper provides the first epidemiologic evidence linking atherosclerosis with air pollution. It also adds to the growing body of evidence linking pollutants with cardiovascular damage of many types. 
 
To uncover this new evidence, Kunzli and colleagues studied 798 men and women from the Los Angeles area. These study subjects were already participating in two clinical trials investigating other aspects of atherosclerosis, or thickening and hardening of the arteries. The selected participants were generally healthy people over 40 who showed some signs of increased risk of cardiovascular disease. 
 
To gauge the participants’ exposure to fine particles, the researchers began with year 2000 data from 23 monitoring stations in the Los Angeles basin, then interpolated the data, using a geographic information system and geostatistics, to assign long-term mean ambient particulate matter concentrations to each participant’s address. According to the authors, fine particle concentrations have changed little over the past 5–10 years, and the 2000 data are representative of long-term exposure. This novel approach allowed a more accurate approximation of exposures than simply using readings from the closest monitor. 
 
The potential effects of fine particles were assessed using data gathered previously on the thickness of the two inner layers of the carotid artery, which feeds the head and neck. This technique has become generally accepted as a reliable indicator of atherosclerosis, which happens slowly in all people and is a leading contributor to death in most countries. In their statistical analysis of the data the researchers accounted for other factors such as diet, use of vitamin supplements and hormone replacement drugs, physical activity, blood pressure, education, and income. 
 
Overall, the researchers found that for each increase in fine particles of 10 micrograms per cubic meter (μg/m3), the two inner layers of the carotid artery thickened by about 4%. For the selected monitoring stations, fine particle concentrations ranged from 5.2 to 26.9 μg/m3. That means the most-exposed participants in this study experienced about 8% more artery thickening than the least-exposed participants. 
 
However, not all participants were found to be equally vulnerable. Women over age 60 experienced artery thickening of about 15% for each 10-μg/m3 increase. In general women were much more vulnerable than men, although no link was observed for women taking hormone replacement drugs. Others who were significantly more vulnerable included both male and female nonsmokers as well as men and women taking drugs to reduce cholesterol. 
 
The researchers studied one other pollutant, ozone, and found some correlations but no significant link with atherosclerosis. However, they acknowledge some weaknesses in their ability to detect links, given their methods and available data. For example, previous studies cited by the authors indicate that outdoor ozone measurements are very weakly correlated with actual personal exposures. 
 
Kunzli and colleagues acknowledge other limitations as well, and say much more research needs to be done. For instance, the selection of participants focused on a relatively healthy population and may have excluded those at highest risk. Also, the relatively low numbers of people in each subgroup analyzed add to the possibility that the findings for these groups may not be accurate. 
 
In addition, the projections for exposure to fine particulates measured outdoors have some small margin of error. Using just these data also excludes exposures in vehicles, workplaces, and other enclosed settings, where people generally spend the vast majority of their time. Other studies have found significant increases in indoor and in-vehicle concentrations of many pollutants, including fine particles, compared to measurements taken at nearby outdoor monitoring stations.


Background
The blood-brain barrier (BBB), formed by the capillary endothelial cells surrounded by astrocytes, protects the brain, but it also poses an obstacle for the delivery of therapeutic molecules into the brain. Microvessels supplying brain tumors retain some characteristics of the BBB and form a blood-brain tumor barrier (BTB). While adequate delivery of chemotherapeutic drugs has been achieved in systemic tumors, the BTB limits such delivery to brain metastases. Therefore, understanding the biochemical modulation of BBB and BTB is critical for developing strategies to deliver therapeutic agents into metastatic brain tumors.
During the past decade, various strategies have been used to deliver therapeutic drugs selectively to brain tumors and injured brain, including, biodegradable polymers implanted into the tumor cavity [1], convection-enhanced delivery [2,3], and BBB/BTB disruption [4,5]. Our laboratory has focused on pharmacologic modulations to increase BTB permeability and increase delivery of therapeutic drugs selectively to brain tumors with little or no drug delivery to normal brain tissue [6][7][8][9]. This strategy exploits the function of certain vasomodulators that play a key role in modulation of BBB/BTB permeability. It has been demonstrated that bradykinin [10], leukotriene (LTC 4 ) [11][12][13], nitric oxide (NO) [14], c-GMP [8], and potassium channel agonists [15,16] can selectively increase capillary permeability in primary brain tumors, while leaving normal brain unaffected. These findings have already been translated into clinical studies to increase drug delivery selectively to tumor tissue in brain tumor patients [7,[17][18][19]. Modulation of critical molecules involved in selectively increasing BTB permeability could lead to the development of effective strategy to increase chemotherapy delivery to brain tumors.
Large conductance calcium-activated potassium (K Ca ) channels are a unique class of ion channel coupling intracellular chemical and electrical signaling. These channels give rise to outwardly rectifying potassium currents and respond not only to changes in membrane voltage, but also to changes in intracellular calcium. Recent studies suggest that K Ca channel expression levels correlate positively with the malignancy grade of glioma [20]. K Ca channels are also present in cerebral blood vessels, where they regulate cerebral blood vessel tone [21] and, probably, BBB/BTB permeability [15,22]. Evidence from several studies further indicate that K Ca channels play an important role in vasodilation when it is mediated by bradykinin [23,24], NO donors [25], and cyclic GMP [26]. In response to the binding of bradykinin to its type 2 receptors (B2R), intracellular Ca 2+ is increased either by mobilization of Ca 2+ from internal sites and influx [27] or by NO production from NO synthase activation [14]. The increase in intracellular Ca 2+ level activates K Ca channels and alters the membrane potential of cells [28]. Furthermore, previous studies have also shown that bradykinininduced K Ca channel activation in endothelial cells is potentiated by NS1619, a selective K Ca channel agonist [29], and attenuated by a highly selective inhibitor, iberi-otoxin (IBTX) [29][30][31]. We previously demonstrated that K Ca channels are overexpressed in primary brain tumors and tumor microvessels, and such channels respond to NS1619, which selectively increases BTB permeability. The accelerated formation of pinocytotic vesicles appears to be the cellular mechanism by which K Ca channels mediate increases in BTB permeability [15]. Moreover, in a rat brain tumor model, we showed that the B2R expression level on brain tumors directly correlates with bradykinininduced BTB permeability increases [32]. Co-infusion of carboplatin with either NS1619 or a bradykinin analog, RMP-7, led to enhanced survival in intracranial tumorbearing rats and brain tumor patients [17][18][19]22,33]. These data indicate that K Ca channels serve as a convergence point in the modulation of BTB permeability in primary brain tumors.
Brain metastasis is a frequent complication in patients suffering from lung and breast cancer, and a significant cause of morbidity and mortality. Brain metastases are found in approximately 10% of lung cancer patients at the time of diagnosis, and up to 40% of all patients develop brain metastases during the course of their disease [34]. The prognosis of brain metastases from lung cancer is poor, with median survival of 4~5 month. Lung cancer cells that spread to the brain are generally sensitive to chemotherapeutic drug compared with primary brain tumor cells. The BTB, however, prevents the delivery of non-lipid-permeable chemotherapeutic drugs and monoclonal antibodies in sufficient amounts to achieve a therapeutic benefit [35], especially in early stage of brain metastases. Although metastatic brain tumors have ten times more than the incidence of primary brain tumors in the United States, the role and regulation of K Ca channels in metastatic brain tumors to selectively open BTB have not been elucidated. As new therapeutic agents are developed which effectively treat primary tumors, an efficient delivery of these agents selectively to metastatic brain tumors across the BTB will significantly improve treatment efficacy. Here, we studied the role of K Ca channel activation in BTB permeability in a metastatic brain tumor model.

K Ca channel mediates BTB permeability increase in a CRL-5904 metastatic brain tumor model
To determine whether K Ca channels mediate BTB permeability in a metastatic brain tumor model, intracranial CRL-5904 tumor bearing-rats received intravenous infusion of NS1619 (0~120 μg/kg/min), bradykinin, IBTX, or PBS. The transport constant (Ki) was determined by radiotracer [ 14 C] sucrose uptake in the tumor core, tumor-adjacent brain tissue, and contralateral brain tissue. The data were obtained from 3~6 rats for each group. Intravenous infusion of NS1619 at 30 μg/kg/min resulted in a significant increase of Ki values in the tumor center (11.24 ± 1.99 μl/ g/min) as compared with PBS control (6.58 ± 0.75 μl/g/ min; p < 0.05, NS1619 vs PBS). A higher concentration of NS1619 at 60 μg/kg/min further increased Ki values (12.77 ± 1.99 μl/g/min; p < 0.01, NS1619 vs PBS). While, increasing dose to120 μg/kg/min did not elicit any further Ki increase. Intravenous infusion of bradykinin (120 μg/ kg/min) also significantly increased BTB permeability within the tumor, with average Ki values at 13.31 ± 2.48 μl/g/min(p < 0.01, bradykinin vs PBS). Furthermore, NS1619-and bradykinin-induced BTB permeability increases resulted in enhanced delivery of radiotracer [ 14 C] sucrose to the tumor center without any increase to tumor adjacent brain tissue and contralateral normal brain ( Figure 1A). In a separate experiment, we found that co-administration of a specific K Ca channel inhibitor, IBTX (1 μg/kg/min), blocked the increase of BTB permeability induced by NS1619 (Ki value of NS1619+IBTX is 7.58 ± 1.31 μl/g/min; NS1619+IBTX vs NS1619, p < 0.05). Intravenous infusion of IBTX alone did not show any effect on BTB permeability ( Figure 1B). These data confirm that activation of K Ca channel selectively induces BTB opening in tumor tissue, but not tumor adjacent tissue or contralateral normal brain, in a metastatic brain tumor animal model.

Expression of K Ca channels and B2R in CRL-5904 cells, HBMEC and human tumor tissue of lung cancer brain metastases
To examine whether K Ca channels were present in tumor tissue, immunostaining of paraffin-embedded tissue sections from lung cancer brain metastases patients were performed. The results demonstrated that K Ca channels ( Figure 2A) and B2R ( Figure 2B) expressed extensively in tumor masses and microvessels within the tumor. Negative control experiments of K Ca channels ( Figure 2C) and B2R ( Figure 2D) did not show specific staining on the corresponded specimens. Elevated mRNA level of K Ca channels was also detected in lung cancer brain metastases tissues from patients using real time PCR (data not shown). To further determine whether K Ca channels and B2Rs are present in metastatic brain tumor and endothelial cells, we examined their expression by immunocytochemistry. Fluorescence immunostaining showed robust K Ca channel expression in cultured CRL-5904 cells, which distributed on the cell membrane, cytoplasm and perinuclear components ( Figure 3A~C). K Ca channels were also detected in HBMEC ( Figure 3G~I), but the signal intensities were lower compared with that in CRL-5904 cells. B2R expression was detected in both CRL-5904 cells (Figure 3D~F) and HBMEC ( Figure 3J~L) with a higher level of expression in CRL-5904 cells. These data illustrate the presence of K Ca channels in cultured metastatic brain tumor cells, endothelial cells, and most importantly, in human metastatic brain tumor tissue.

Activity of functional K Ca channels in cultured CRL-5904 cells and HBMEC
Since the above data demonstrated the presence of K Ca channels in metastatic brain tumors and endothelial cells as well as metastatic brain tumor tissue, we determine whether the overexpressed K Ca channels were functional. Therefore we examined membrane potential activity of K Ca channels in cultured tumor and endothelial cells and by using a fluorescent dye-based potentiometric assay that measures changes in membrane potential. To activate K Ca channel, NS1619 (25 μM), a K Ca channel agonist, was added to the cells, and membrane potential changes were monitored for up to 300 seconds. Upon activation a decrease in membrane potential was observed in CRL-5904 cells ( Figure 4A) and HBMEC ( Figure 4C), an effect that lasted more than 300 seconds. Furthermore, we introduced bradykinin (25 μM) to the cultured cells and observed membrane potential changes in CRL-5904 cells ( Figure 4B) and HBMEC ( Figure 4D) that lasted for approximately 100 seconds. Moreover, both NS1619 and bradykinin elicited greater hyperpolarization on CRL-5904 cells compared with HBMEC. IBTX (20 nM), a K Ca channel antagonist, reversed the membrane potential changes on both cells caused by NS1619 and bradykinin. Furthermore, we found that NS1619 and bradykinin could induce a dose-dependent membrane potential change in CRL-5904 and HBMEC (data not shown). These data indicate that K Ca channels are functional on both CRL-5904 cells and HBMEC. The K Ca channels can be activated directly by NS1619 or indirectly through B2R signaling by bradykinin.

Co-culture of metastatic brain tumor and endothelial cells increases K Ca Channel expression
We further investigated whether K Ca channels expression is modulated by the interaction of metastatic brain tumor cells and endothelial cells. CRL-5904 cells were co-cultured with HBMEC, and protein and mRNA levels of K Ca channels were examined subsequently. K Ca channel were overexpressed in CRL-5904/HBMEC co-cultures compared to single cultures of either CRL-5904 cells or HBMEC by western blot assay ( Figure 5A). Image quantification analysis showed an approximately 30% increase of K Ca channel expression in co-culture of CRL-5904/ HBMEC compared to individual cultures by normalized to β-actin as an internal control ( Figure 5B). Also, individual cultures of CRL-5904 tumor cells had higher K Ca channel expression than HBMEC. RT-PCR analysis also showed an increase in K Ca channel mRNA levels in co-culture cells compared to individual cultures ( Figure 5C). These data suggest that co-culture of metastatic tumor and brain endothelial cells results in upregulation of K Ca channel.

Immuno-colocalization of K Ca channel expression in a CRL5904 metastatic brain tumor animal model and human lung cancer brain metastases tissue
Since K Ca channel modulators can selectively increase BTB permeability without affecting normal brain, we wanted to know whether K Ca channels were differentially expressed within the tumor mass compared with normal brain tissue. To address this question, we examined K Ca channel and endothelial cell marker von Willebrand factor (vWF) expression in CRL-5904 tumors and human lung cancer brain metastases tissue. An abundance of K Ca channel expression (green) was detected within the tumor mass with robust microvessel formation (red) within the tumor area of CRL5904 tumor ( Figure 6A). In contrast, there was less K Ca channel expression in normal brain tissue adjacent to the tumor mass as well as contralateral normal brain ( Figure 6B). More importantly, colocalization of K Ca channels with vWF within microvessels was observed only within the tumor mass and not in normal brain. These results demonstrate elevated expression of K Ca channels on endothelial cells of capillaries within the tumor mass. Immunostaining of human tissue from lung cancer brain metastases also revealed that K Ca channel expression was colocalized with vWF expression in Expression of K Ca channels and B2R in the lung cancer metastatic brain tumor tissue from patients Figure 2 Expression of K Ca channels and B2R in the lung cancer metastatic brain tumor tissue from patients. K Ca channels (A) and B2R (B) expression were detected on the tumor cells in the tissue. Negative control experiments of K Ca channels (C) and B2R (D) did not show specific staining on the corresponded specimens by deleting of primary antibodies. Scale bar = 50 μm. tumor capillary endothelia ( Figure 6C). These results strongly suggest that the selective BTB permeability increase induced by modulation of K Ca channel in the metastatic brain tumor model is likely due to the overexpresion of K Ca channels on tumor cells and tumor capillary endothelia.

Discussion
We have studied the presence of K Ca channels and B2R in primary brain tumors, however, their occurrence and function in metastatic brain tumors remained to be investigated. In this study, we detected high level expression of K Ca channels in CRL-5904 tumor and brain endothelial cells, which is consistent with previous studies showing K Ca channels expression in RG2 glioma and endothelial cells [15]. Other investigators have also demonstrated that the expression level of K Ca channels correlates with the malignancy grade of glioma in human [36]. Therefore, these data suggest there is an intimate association between K Ca channel expression and brain tumor development, which remains to be fully investigated. Additionally, we detected the presence of B2R in CRL-5904 tumor and endothelial cells. Liu et al also showed that B2R are expressed in cultured RG2, C6 and 9L glioma cells, more interestingly, the expression levels of B2R in tumor cells was directly correlated with the increase of BTB permeability induced by bradykinin in a rat glioma model [32]. Thus, the presence of K Ca channels or B2R in metastatic brain tumor cells or HBMEC may play a functional role in BTB permeability of metastatic brain tumors.
We examined K Ca channel activity on metastatic tumor cells and capillary endothelial cells using a membrane potential assay, which is well-correlated with the patchclamp method, used to measure changes in membrane potential (FLIPR application note, Molecular devices, CA). We found that NS1619 and bradykinin elicited greater hyperpolarization effects on CRL-5904 than on HBMEC. These findings may reflect a higher level of expression for K Ca channels on metastatic tumor cells as compared to endothelial cells. Importantly, the membrane potential change induced by NS1619 lasted 3 times longer than that induced by bradykinin. These data further support the finding, from cellular level, that NS1619elicited increases in BTB permeability in a glioma model last up to 60 minutes compared to the transient effect of bradykinin, which lasts for about 15~20 minutes [15], partially due to B2R internalization [37]. The current data illustrates that the presence of K Ca channel are functional in metastatic brain tumor and endothelial cells. Similar to our findings, Reiser et al demonstrated that bradykinin can directly activate K Ca channels in rat glioma cells [38]. Other studies have shown that bradykinin can activate K Ca channels through a NO-cGMP signalling pathway [39]. Hence, our present study indicates that bradykinin-activated downstream signals, such as activation of K Ca channels, may be modulated to induce membrane potential changes on brain metastatic tumor and endothelial cells.
The presence of functional K Ca channels in metastatic brain tumor and brain endothelial cells suggests that biochemical modulation of K Ca channels could play an important role in therapeutic BTB opening. We further investigated whether the K Ca channels agonist, NS1619 and bradykinin could selectively enhance BTB permeability in a metastatic brain tumor xenograft model. These results showed that intravenous infusion of NS1619 yielded a two-fold increase the unidirectional transport of a radiotracer into metastatic brain tumors; similar to bradykinin-induced BTB permeability increase in metastatic brain tumor-bearing rats. Our previous studies have demonstrated that higher doses of intravenous bradykinin are required to increase BTB permeability compared to intracarotid infusion of bradykinin, reflecting the influence of the first pass effect with intracarotid delivery [15]. In a glioma model, it has been reported that the effects of bradykinin on BTB permeability mediated by B2R resulted in enhanced drug delivery to glioma [40]; and this effect could be attenuated by coinfusion with IBTX [15]. In this metastatic brain tumor model, we further demonstrate the presence of B2R and confirm that the bradykinin effect on permeability is mediated via K Ca channels. Consistent with previous studies [41], current data confirms the selective increase of BTB permeability in brain metastatic tumors but not normal brain tissue. These results suggest that biochemical modulation of K Ca channel induces a selective BTB opening in metastatic brain tumor.

Measurement of functional K Ca channels activity in CRL-5904 cells and HBMEC
Confocal images showed K Ca channels overexpression in tumor tissue and tumor microvessels as compared with normal brain. More importantly, the tumor capillaries showed co-localization of K Ca channels and vWF in tumor area of CRL-5904 tumor and in human metastatic brain tumor tissue. To further study the interaction between tumor and endothelial cells, we co-cultured CRL-5904 metastatic brain tumors and brain endothelial cells. We show that mRNA expression of K Ca channels is upregulated in co-cultured cells compared to indivdual cultures. These data suggest that increased K Ca channel expression and their activity in tumor endothelial cells maybe due to the tumor micro-environment or cell-to-cell communication between tumor and microvessel endothelial cells. This is consistent with reports that brain tumor cells increase K ATP channel expression in endothelial cells [16].
These findings support a pivotal role for K Ca channels in BTB permeability regulation. Recently clinical study showed that trastuzumab, anti-HER2 antibody, while effective in treating tumors outside the brain, fails to treat brain metastases due to its inability to cross the blood brain tumor barrier [42]. Our research has shown that K Ca channel-mediated BTB permeability modulation could be a useful strategy to increase therapeutic agents, such as antibody-based therapies, delivery into metastatic brain tumors.

Conclusion
We present evidence that activation of K Ca channels by a channel-specific agonist can selectively enhance BTB permeability in a metastatic brain tumor rat model. We show K Ca channel and B2R are highly expressed in brain metastatic tumor cells, endothelial cells and lung cancer brain metastatic tissue. The expression level is correlated with K Ca channel activity in these cells. In a metastatic brain tumor model, we demonstrate that NS1619 and bradykinin can selectively open BTB and significantly enhance the radiotracer delivery specifically to metastatic brain tumors. It is also demonstrated that K Ca channels expression can be upregulated in the co-cultures of tumor cells and endothelial cells, as well as in the microvessel endothelia of brain metastases tissue. K Ca channels may be exploited as specific target for selectively pharamacologic modulation of BTB to increase delivery of chemotherapeutic drugs to brain metastases.

Cell Culture
CRL-5904 cells (human non-small cell lung cancer; metastatic site: brain poorly differentiated carcinoma) and human brain microvessel endothelial cells (HBMEC) were obtained from the American tissue culture collection (ATCC, VA) and maintained in RPMI 1640 with 10% fetal bovine serum. Both cell lines were maintained in the com- mon tissue culture condition. For co-culture of CRL-5904 cells with HBMEC, the same number of CRL-5904 cells and HBMEC were co-cultured in growth medium and allowed to achieve 90% confluence. Then, the co-culture and single cultures of cells were harvested for protein or RNA extraction.

Membrane Potential Assay
The functional activity of K Ca channels in CRL-5904 cells and HBMEC was measured using the FLIPR Membrane Potential Assay Kit on a FLEXstation (Molecular Devices, Sunnyvale, CA) as described previously [15]. This kit provided a fast, simple and consistent mix-and read proce-Localization of K Ca channels in brain tumor tissues from xenogenic rat model and patients of lung cancer brain metastases Figure 6 Localization of K Ca channels in brain tumor tissues from xenogenic rat model and patients of lung cancer brain metastases. K Ca channels (green) were localized in the endothelial cells of capillaries (red) in the tumor mass (A). In the contralateral side of brain, much less of K Ca channel was detected in normal brain. (B). K Ca channels are co-localized with von Willebrand factor, a marker of endothelial cells, in human tumor tissue of lung cancer brain metastases (C). Scale bar = 250 μm.
dure. In brief, the cells were seeded in sterile, clear bottom, black 96-well plates (Corning Inc., MA) at density of 2 × 10 3 cells/well to achieve monolayer within 24 h. The monolayer cells were incubated with the membrane potential assay kits reagents for 30 min before loading the compounds. The anionic potentiometric dye that transverses between cells and extracellular solution in a membrane potential-dependent manner serves as an indicator of vasomodulator-induced voltage changes across the cell membrane. Dose response studies were performed with 0 to 50 μM NS1619 or bradykinin with or without IBTX (20 nM). The FLEXstation was set up using the following parameters: excitation 530 nm, emission 565 nm, and emission cut of 550 nm wavelengths. Observations and recordings were made for 300 seconds after adding the compounds. NS1619, bradykinin and IBTX were obtained from Sigma (St. Louis, MO).

In Vivo BBB/BTB Permeability
All of the animals used were conducted in accordance with the Institutional Animal Care and Usage Committee in force at Cedars-Sinai Medical Center. A metastatic brain tumor xenograft model was established using athymic nude rats (180-200 g; Charles River Laboratories, Inc., MA) for BBB/BTB permeability studies. Athymic nude rats were anesthetized with i.p. ketamine and xylazine, and stereotactically implanted with CRL-5904 cells (2 × 10 5 ) in 4 μl of 1.2% methylcellulose/PBS using a Hamilton syringe into the right striatum. The Coordinates were 3.4 mm lateral to bregma and 5.0 mm deep from dura. Ten days after tumor implantation, the femoral arteries of rats were cannulated to measure blood pressure and collect blood, and the femoral vein was also cannulated to administer the drugs and radiotracer. Body temperature was maintained at 37°C. Arterial blood gases, blood pressure and hematocrit were monitored. Animals with abnormal physiological parameters were eliminated from this study. In regional permeability studies, either intravenous drug or PBS was infused into the femoral vein at a rate of 66.7 μl/min for 15 minutes. Five minutes after the start of the intravenous infusion, 50 μCi/kg of the radiotracer [ 14 C] sucrose was injected as an intravenous bolus. Arterial blood pressure was monitored throughout the experimental period with a blood pressure monitor (DigiMed, KY). The unilateral transport constant Ki (μl/g/ min), which is an initial rate for blood-to-brain transfer of radiotracer, was calculated as described by Ohno et al [43]. The Ki was determined by radiotracer [ 14 C] sucrose in the tumor core, tumor-adjacent brain tissue, and contralateral brain tissue using the quantitative autoradiographic (QAR) method as describe previously [44]. Quantitative analysis of the regional radioactivity was performed using a computer (Power Macintosh 7100) and Image 1.55 software (National Institutes of Health, Bethesda, MD). An optimum dose (120 μg/kg/min, i.v.) of bradykinin estab-lished previously was used for Ki measurements. To establish the optimal and safe dose range that would result in selective increase in BTB permeability without appreciably altering system blood pressure, various doses (0~120 μg/ kg/min, i.v.) of NS1619 were administered in metastatic brain tumor bearing-nude rats. Additional experiments were performed by coinfusing of NS1619 with IBTX (1.0 μg/kg/min) to investigate whether inhibition of K Ca channels by IBTX has any effects on NS1619-induced permeability increase.

Western Blot Analysis
The extracted protein samples were quantified to determine total protein concentrations using a protein assay kit (BioRad, CA). Same amount of each sample was fractionated on 10% SDS-polyacrylamide gel and then transferred to a nitrocellulose membrane. The membrane was probed with primary antibodies anti-MaxiKá (1:200; Santa Cruz Biotechnology, CA) and β-actin (1:5000; Sigma, MO), followed by peroxidase-conjugated secondary antibodies. The signals were detected with an enhanced chemiluminescence kit (Amersham Biosciences Corp., NJ). β-actin served as an internal control.
Publish with Bio Med Central and every scientist can read your work free of charge http://www.molecular-cancer.com/content/6/1/22