1%, 36.0%) and severe dysplasia groups (41.5%, 41.8%) were significantly higher than in normal epithelium and mild dysplasia (p < 0.01) ( Fig. 7 and Table 1). In electron microscopic observations, glycogen granules were seen in the cells

of normal epithelium but were not seen in the cells of IU ( Fig. 8). In conclusion, IU surrounding OSCC has various types of dysplasia. In oral mucosa, iodine–glycogen reaction does not occur in dysplastic mucosa due to the lack of glycogen granules in the cytoplasm of those cells. This area has high positive PCNA and p53 cells with malignant potentiality. Therefore, see more from the standpoint of molecular biology, we suggest iodine-staining method as strongly useful tool in defining the region of epithelial dysplasia. On the other hand, we are always unable to see a clear margin with IU. It is known that keratinized www.selleckchem.com/products/BAY-73-4506.html squamous epithelia, such as the gingiva and the hard palate, were less reactive to iodine. Iodine–glycogen reaction does not occur in epithelial

cells with chronic inflammatory change. Furthermore, we sometimes suffer from drug induced reaction such as iodine hypersensitivity. For this reason, we must look for a more convenient and useful tool for the detection of safety margin. More recently, Pierre et al. have introduced an approach using fluorescence visualization (FV) technology, fluorescence imaging device, and VELscope (LED Dental Inc., White Rock, British Colombia, Canada [19])

(Fig. 9). This simple handheld device uses a blue light (400–460 nm) to illuminate a collagen matrix or a flavin adenine dinucleotide (FAD). A selective filter in the eyepiece allows the viewer to directly visualize the pale green autofluorescence that is given off by normal tissue (fluorescence visualization retention, FVR). On the other hand, abnormal tissue shows decreased autofluorescence and appears as a dark brown in comparison with green surrounding normal tissue. FER This dark brown is the so called fluorescence visualization loss (FVL). FVL is caused by absorbing a specific blue light due to the breakdown of the collagen matrix and decrease in FAD. The oxidized form of FAD is important fluorophores that are good indicators of cellular metabolism. It has been shown that fluorescence intensity due to FAD decreases with dysplastic progression. We have suggested this device as a suitable adjunct for margin delineation since 2010. Therefore we investigated to evaluate the detection of surgical margin comparison between FV and vital staining with iodine, and the accuracy of autofluorescence examination in its ability to delineate high-risk oral mucosal lesions. To clarify the usefulness of FV compared to vital staining with iodine, we investigated surgical margin of early OSCC in comparison between FVL and IU. Thirty one cases of T1 and early T2 OSCC patients were examined in this study.

The subjects included 5 males and 15 females aged 54–82 years

(mean age: 67.6 years) (Table 2). All subjects had periodically received maintenance with prosthodontists in excess of 20 years, and their denture conditions were favorable. All measurements were performed by one trained dentist. The subjects had no history of brain disease and had not been diagnosed with Alzheimer’s disease. Informed consent was obtained from each individual according to the method approved by the Ethics Committee of the Tsurumi University School of Dental Medicine (approval number 305: accepted on August 31, 2005). The occlusal contact area and occlusal force were measured with and without dentures for functional assessment. For the subjective assessment of dentures, the Visual Analog Scale (VAS) was used selleckchem to evaluate comfort during chewing and the degree of satisfaction. After gum chewing with and without dentures, electroencephalograms (EEG) were taken for 3 min (Fig. 7). EEG were analyzed employing the diagnosis method of neuronal dysfunction (DIMENSION) for brain function assessment. For objective evaluation of the denture function, the occlusal contact Tyrosine Kinase Inhibitor Library in vitro area and occlusal force were measured between with and without dentures using a Dental Prescale Occluzer® (FPD-705; GC Co., Tokyo, Japan) and Dental Prescale® (50H without wax, GC Co., Tokyo, Japan). The subjects clenched their

teeth for 3 s in centric occlusion in order to measure the occlusal contact area and very occlusal force. Statistical analysis was performed using the Wilcoxon rank sum test (p < 0.05). Changes in the average occlusal contact area and occlusal force between with and without dentures are shown in Fig. 8. The average occlusal contact area and occlusal force significantly increased with compared to without dentures in all subjects (p < 0.05). It has been reported that the occlusal contact area and occlusal force are closely correlated with the masticatory efficiency [43], [44], [45] and [46]. These results suggest that subjects

could safely masticate due to an increase in the occlusal contact area and an adequately distributed occlusal force. After whole measurements, to evaluate the stress of wearing dentures, Comfort on chewing and the degree of satisfaction were evaluated using the Visual Analog Scale (VAS). The subjects were asked two questions (How is the level of comfort while chewing with or without dentures? What is the degree of satisfaction with or without dentures?). VAS between discomfort (left end (0)) and comfort (right end (100)) on a line 100 mm long was filled out by subjects. The subjective assessments of dentures using VAS are shown in Fig. 9. Comfort during chewing with dentures significantly increased in 17 of the 20 subjects (p < 0.05). The degree of satisfaction significantly increased in 13 of the 20 subjects compared to those without dentures (p < 0.05).

We performed a tracheotomy with suturing of the distal stump of Decitabine the trachea diastase to the skin and suturing of the previous tracheotomy breach. During surgery, a careful dissection of the trachea was conducted in its distal portion and the anonymous artery was protected by a muscle flap. At the end of surgical treatment, mechanical ventilation through a tracheal cannula was hindered

by the reduced length of the residual trachea below the tracheotomy (about 2 cm from the tracheal carina). Any cannula or endotracheal tube could not be secured to the trachea using the cuff. The need to guarantee mechanical ventilation to the patient led to the implementation of a cuff securing system in the two main bronchi. Therefore, we selectively intubated the main bronchi under bronchoscopy guidance using two tubes (Portex Tracheal Tube, ID

5.5 mm; OD 7.4 mm), then inflating the cuffs at both main selleck stem bronchi inlets (Fig. 2). A Y-shaped bridge was then added for ventilator connection to allow the same ventilation mode for both bronchial systems. Thus, we achieved an adequate minute volume to the patient, allowing us to correct the blood gas levels. During mechanical ventilation, no significant leaks were reported and ventilatory parameters remained stable. Four days later, a bronchoscopic examination showed no evidence of alterations on both main bronchi. The patient died forty days after surgery of sepsis. Currently, tracheotomy is largely performed on patients presenting with acute respiratory failure requiring prolonged mechanical ventilation so as to facilitate weaning, reduce the effort of breathing and curb complications

due to prolonged intubation.1 Although the optimum timing of tracheotomy in critically ill patients with acute respiratory failure still remains controversial, the current trend is to anticipate the procedure within the first week of mechanical Cepharanthine ventilation. The results of a Cochrane meta-analysis performed on five clinical trials showed a significant reduction of days on mechanical ventilation and of hospitalization in ICU wards with early rather than late tracheotomy,2 while no significant difference was reported in mortality or risk of occurrence of hospital-acquired pneumonia. Complications of tracheotomy can be acute, related to or subsequent to the surgical procedure (haemorrhage, pneumothorax, infections, incidental decannulation), or late stage.3 The most frequent late complication is the formation of granulation tissue, with subsequent tracheal stenosis, which can remain asymptomatic for a long time, but which, in some cases, can lead to severe respiratory failure.4 Other types of late complication are rare but life-threatening events, such as tracheoesophageal fistula and tracheo-innominate artery fistula.5 Therefore, the presence of a cuff with a higher critical pressure may cause ischemia and tracheal mucosal necrosis.

Hydrolysis conditions were modified from the method

described by Aziz, Edwards, Lean, and Crozier (1998). The crude extract (5 mg) was mixed with 2 ml of 1.2 N HCl containing RG7420 nmr 20 mM DETC sodium salt in a hydrolysis vial. The hydrolysis was performed in a heating module (Reacti-Therm Heating/Stirring Module No. 18971; Pierce, Rockford, IL) at 90 °C for 2 h. The hydrolysate was then diluted to 1 mg extract/ml with water containing 20 mM DETC sodium salt prior to chromatographic analysis. All samples were filtered through 0.20-μm PTFE membrane filters prior to chromatographic analysis. Separation of polyphenols was performed on a UHPLC system (Agilent Technologies 1290 Infinity, Waldbronn, Germany) (Kong et al., 2012). The stationary phase consisted of an Agilent Zorbax Eclipse Plus C18 (50 × 2.1 mm, 1.8 μm) column and 5 μl of the sample were injected into the system. A binary mobile phase made up of 0.1% trifluoroacetic acid (TFA) (A) and acetonitrile (B), with the flow rate adjusted to 0.6 ml/min, was employed. Separation of polyphenols was achieved using a linear gradient system: 5–15% B in 6 min; 15–25% B in 3 min; 25–60% B in 3 min; 60–80% B in 0.6 min; 80–100% B in 0.8 min. Absorption spectra were

monitored in the region of 200–500 nm throughout the analysis. The polyphenolic compounds were detected at 280 and 325 nm by the diode array detector. All polyphenolic standards were prepared in 50% methanol containing

20 mM DETC sodium salt. Phenolic acids Cytoskeletal Signaling inhibitor and flavonoids were identified by comparing the retention time (tR) and absorption spectra of the samples with those of authentic standards. External standard was used to plot the calibration curve (0–80 μg/ml). Results were expressed as μg/g dry weight (dw) of sample. The percentage of free and bound polyphenols was calculated. The limit of detection (LOD) and limit of quantification (LOQ) were determined as described by the guidelines from the International Conference on Harmonization (ICH) (1996). Three calibration curves were plotted using three sets of polyphenolic standards, selleck chemical injected at concentrations ranging from 2.5 to 20 μg/ml. The equations of the calibration curves were then derived. Mean of the slopes (S) and standard deviation of the intercepts (σ) were calculated. LOD and LOQ were subsequently estimated according to the following formulae: LOD=3.3×σ/SLOD=3.3×σ/S LOQ=10×σ/SLOQ=10×σ/S The in vitro antioxidant assays were conducted only on the freeze dried samples, as it was shown to be a better drying method for polyphenols compared to the air drying method. The serum oxidation assay was modified from the method of Bem et al. (2008), using serum from healthy volunteers. A 0.4-ml solution of serum (final concentration of 25%) was treated with 70 μl of the aqueous extracts of B.

Chitosan is obtained by the alkaline deacetylation of chitin, one of the most abundant biopolymers in nature, present in the exoskeletons of crustaceans and also the cell Cilengitide in vivo walls of fungi and insects (Kumar, 2000). One of the advantages of chitosan which attracts greatest interest is its versatility. This polymer can be easily

modified by chemical or physical processes to prepare chitosan derivatives. The material can be quickly modified physically and obtained in different forms including powder, nano particles, gel beads, membranes, sponge, honeycomb, fibres or hollow fibres. The presence of a high percentage of reactive amino groups, generally higher than 80%, distributed in its polymeric matrix, allows chemical changes to be carried out. The chemical modification of chitosan may be necessary to prevent the dissolution of the polymer when the reactions

are performed in acidic solutions, and/or to change its properties, such as improving its ability to adsorb metals (Guibal, 2004). This biopolymer has been crosslinked with different substances including glutaraldehyde, 1,1,3,3-tetramethoxypropane, PD0325901 research buy ethyleneglycol diglycidyl ether, epichlorohydrin, glyoxal, carbodiimide, and tripolyphosphate and has been used in many different fields (Osifo et al., Interleukin-2 receptor 2008). Spray drying is a technique for the formation of microparticles which has been used by researchers for different ends. It is employed in a wide variety of processes ranging from the manufacture of food products to pharmaceuticals (Tonon, Brabet, & Hubinger, 2008). This well-established technique has been around for over a century, but it remains an active field of innovation, driven by the ever increasing demand for more sophisticated particles.

It has many advantages over other techniques for the preparation of particles, such as excellent reproducibility and speed in obtaining the microspheres (Vehring, Foss, & Lechuga-Ballesteros, 2007). It is used to produce dry powder from solutions or suspensions in three steps of operation: atomization of the liquid feed, drying of the droplets once they are formed, and motion of the droplets to model the spray drying process (Shabde & Hoo, 2008). The atomization of the biopolymer chitosan by this technique is generally used in pharmacological processes, especially in controlled-drug delivery systems, and produces good results. Recently, the spray drying technique has been employed to obtain microspheres of chitosan crosslinked with 8-hydroxyquinoline-5-sulphonic acid and glutaraldehyde, as a new adsorbent for metallic ions (Vitali, Laranjeira, Gonçalves, & Favere, 2008).

e., one year of Central European sun). Under this condition, the polymer degraded to expose, but not necessarily release, free CNTs. Recently, a study was published which conducted an initial, task-based comparative assessment to determine the potential for release of carbon nanofibers (CNFs) during dry material handling, wet cutting, grinding, and sanding (by machine and hand) of plastic composite material containing CNFs (Methner et al., 2012). Using a combination of direct reading instruments and filter-based air sampling methods for airborne mass and

TEM, concentrations were measured and characterized near sources of particle generation, in the breathing zone of the workers, and in the general work area. Tasks such as surface grinding of composite material and manually transferring dry CNFs produced substantial increases in particle number concentration. GSK-3 cancer Concomitant increases in mass concentration were also associated with most tasks. Over 90%, i.e. 12 out of 13 samples taken during abrasion of CNF composites examined via TEM, indicated that releases of CNFs do occur, mainly as agglomerated CNF, and that the potential for exposure exists, although exposure levels were not quantified. Degradation of the polymer/CNT matrix potentially provides key step(s)

in the release of CNTs in all phases of the life cycle including manufacturing, product or article life/usage and end of life. Several other recent papers have provided useful discussions of polymer nanocomposite degradation, selleck kinase inhibitor including polymer CNT composites (Nguyen et al., 2011, Petersen et al., 2011 and Wohlleben et al., 2011). The potentially important role of abrasion in the release of nanoparticles from polymer matrices has been discussed by Wohlleben and coworkers (Wohlleben et al., 2011). Abrasion increases exposure to polymer-CNT simply by enhancing surface area Tacrolimus (FK506) to mass. In addition to these direct effects, the creation of much smaller particles also enhances dispersion by atmospheric and aquatic routes. Degradation generally decreases the

tensile strength of the polymer matrix thus increasing its susceptibility to abrasion and breakdown to small particles, i.e. referred to as the “chalking” phenomenon in some cases (Wohlleben et al., 2011). Fragmentation to smaller particles can in turn increase exposure to light and hydrolytic and/or microbial breakdown. However, current results have shown that nanoparticles remain associated with the debris that results from sanding of polyoxymethylene and polyamide with embedded inorganic nanoparticles (Wohlleben et al., 2011). So far, one generic release scenario for CNTs in composites has been published (Nowack et al., 2012). These authors have evaluated how different environmental conditions affect the alteration of the composite material, as well as the transformation of the CNTs once they are released from the composite.

2 and Table 2). There was agreement between years in that there was no difference for transplant survival and vitality between grouped and scattered retention trees. Also, the survival of autumn transplants was in both survey years significantly higher than the survival of spring transplants. However, transplant vitality differed significantly between survey years with autumn transplants being significantly more vital in clearcuts in 1996 but showing

no significant difference in 2008 (Table 2 and Table 3). The most important conclusion from our 14-year old transplantation experiment is that transplants of L. pulmonaria survived better on retained aspens on clearcuts than on forest trees, indicating that aspens left at clearcutting represent a suitable Selleck Venetoclax habitat for this species. The positive effect of retention trees was especially high on northern sides of tree stems, and thus microhabitat conditions seem decisive for species survival. Also transplant OTX015 manufacturer vitality was higher on northern sides of tree stems, but this did not differ significantly between retention trees and forest trees, indicating that some factor seriously affects

transplant survival in the forests. One possible explanation might be gastropod grazing which has been increasingly noticed as an ecological driver of epiphytic population occurrences (e.g. Asplund et al., 2010). For L. pulmonaria, a positive correlation has been found between gastropod abundance and grazing damage ( Vatne et al., 2010), and snails in the boreal zone are known to be promoted by aspen since the litter of this tree species has a relatively high pH ( Karlin, 1961). It is likely that the grazing pressure is lower on clearcuts than in forests,

since many snails are sensitive to disturbance and microclimatic changes ( Hylander, 2011). The higher survival on retained trees is unexpected since L. pulmonaria is most common in old-growth forest ( Gärdenfors, 2010), i.e. the response of transplants does not match the actual occurrence pattern. However, large differences have been observed between potential and actual niches in lichen transplant Endonuclease studies. For instance, Sillett et al. (2000) found that transplants of L. pulmonaria were tolerant to open habitat conditions one year after transplantation, and Gauslaa et al. (2006) found L. pulmonaria transplants to have larger biomass growth in clearcuts than in old forests. Gauslaa et al. (2006) describe the long-term persistence of this species as a balance between light availability, where high levels benefit growth, and desiccation risk, since drought can drastically decrease populations. The relatively shady north side of retention trees is intermediate between the sun-exposed south sides of retention trees and the often very dark spots in old forests, and thus seems a favorable environment for L. pulmonaria.

, 2011 and PEN, 2013). Completed syntheses of the PEN data have not yet been published, but preliminary analyses provide results that are consistent with those of earlier NTFP studies (Table 1). There have been many studies investigating ancient forest management practices for indigenous food plants in parts of Latin America (e.g., Levis et al., 2012 and Peters, 2000) and Southeast Asia (e.g., Michon, 2005 and Wiersum, 1997), but relatively few in Africa see more (although see, e.g., Leakey et al., 2004 and Maranz

and Wiesman, 2003). Ancient harvesting, managed regeneration and cultivation have, for example, led to genetic changes in many Amazonian fruit trees and palms (Clement, 1989). These include abiu (Pouteria caimito), Amazon tree grape (Pourouma cecropiifolia), araza (Eugenia stipitata), biriba (Rollinia mucosa), peach

palm (Bactris Selleck PD-1/PD-L1 inhibitor gasipaes) and sapota (Quararibea cordata). In Africa, rarer reports of changes in the characteristics of fruits attributed to ancient domestications include bush mango (Irvingia gabonensis and Irvingia wombolu) and safou (Dacryodes edulis) ( Leakey et al., 2004). Again, areca (Areca catechu), coconut (Cocos nucifera) and date (Phoenix dactylifera) are all palms for which changes in fruit size, in the proportion of useable product, and in the ability to be propagated, are attributed to long-past Dehydratase human selections ( Clement, 1992), while an expanding list of global studies on ancient domestications includes many more food trees ( Clement, 2004). In perhaps the best studied case, in Amazonia, Amerindian populations declined after European colonial contact, which resulted in the erosion of the rich tree crop genetic heritage they had established (Clement, 1999). The effects of pre-Columbian forest management remain, however, including high density aggregations of useful trees close to ancient anthropogenic ‘dark earth’ soils (Clement and Junqueira, 2010) and in interfluvial regions (Levis et al., 2012), with Brazil nut (Bertholletia excelsa) being

the most famous example ( Shepard and Ramirez, 2011). A review of molecular genetic studies ( Clement et al., 2010) suggested that current centres of genetic diversity in fruit and nut trees are generally located in the centre of the Amazon Basin along the major white water rivers where large pre-Colombian human populations developed, while the periphery of the basin has had an important role in domestication origins. This suggests that subtle differences in the focus of management programmes for conservation and genetic improvement may be required in different geographic regions of the Amazon, and indicates the importance of germplasm exchange and dispersal during ancient domestication processes.

Changes of ±10% http://www.selleckchem.com/products/pci-32765.html in the final concentration of master mix and primer pair mix were tolerated by both the PowerPlex® ESI Fast and ESX Fast Systems. An increase of either master mix or primer pair mix to a final concentration of 1.2× had minimal effect on these systems. However, decreasing the concentration of master mix or primer pair mix to 0.8× adversely

affected the signal and balance, particularly for the PowerPlex® ESI Fast Systems (Fig. 1 and Supplemental Fig. 2). Direct amplification is facilitated by the inclusion of AmpSolution™ Reagent in the reaction. Inclusion of AmpSolution™ Reagent in the amplification reaction has no effect on the signal or balance of the profile obtained whether 500 pg of DNA is amplified for 30 cycles or 10 ng for 26 cycles (Supplemental Figs. 3 and 4). No additional amplification artefacts were seen in the presence of AmpSolution™ Reagent over those documented in the technical manuals (data not shown) [14], [15], [16] and [17].

Increasing cycle number from 28 to 30 cycles resulted in the anticipated www.selleckchem.com/products/forskolin.html increase in signal across all loci for the PowerPlex® ESI 17 Fast and ESX Fast 17 Systems. At 32 cycles, the increase in signal was not uniform across all loci, resulting in a locus-to-locus imbalance (Supplemental Fig. 5). Similar results were obtained for the two 16 plexes (data not shown). Increasing cycle number did not result in the appearance of additional artefact peaks in the no-template amplifications reactions (data not shown). We looked at the effect of increasing cycle number from 25 to 27 cycles on blood FTA® cards (Fig. 2) and buccal FTA® cards

(Supplemental Fig. 6), blood on ProteinSaver™ 903® cards (Supplemental Fig. 7), Bode Buccal Collectors (Supplemental Fig. 8), and SwabSolution™ extracts (Supplemental Fig. 9). Signal tended to increase with cycle number for all direct amplification samples. Methane monooxygenase When using two 1.2 mm buccal FTA® punches, full profiles were obtained with all samples at all cycle numbers. Dropout at one locus (in this case one allele at SE33 in one replicate of donor 2) was seen at 25 cycles when using one 1.2 mm buccal FTA® punch (Supplemental Table 3, data not shown for 16 plexes), but not with any of the other direct amplification sample types at any cycle number. The genotypes obtained for a given donor were concordant with each other between cycle numbers and across direct amplification sample types tested. Increasing the annealing temperature to 62 °C from the recommended 60 °C resulted in a significant reduction in signal at amelogenin, D8S1179 and FGA with the PowerPlex® ESI Fast Systems (occasional drop-out at amelogenin and D8S1179 at 62 °C) and dropout occurring at 64 °C along with D2S441 and in some replicates at D2S1338 and D19S433. Overall, 90–100% of alleles were obtainable at 62 °C and 61–76% at 64 °C with the PowerPlex® ESI Fast Systems (Supplemental Fig. 10).

8A, even 5 μl of HA/ml did not stimulate reactivation of HIV-1 in ACH-2 cells, as characterized by western blot analysis AG-014699 research buy of the p24 Ag, while a 48 h treatment led to a comparable increase in expression of p24 Ag in cells stimulated with PMA only as well as with PMA and HA. Stimulation of the cells with 10 U/ml of TNF-α led to an even higher expression of p24 Ag, while 1 U/ml induced a relatively smaller expression of p24 Ag. On the other hand, any concentration of phytohemagglutinin A tested (PHA; 0.5, 2.5; 5 μg/ml) alone or in combination with 1 μM ionomycine did not yield a positive signal of p24 Ag in western blot analysis (Fig. 8A and data not

shown). ELISA analysis of culture supernatants revealed similar changes in levels of the p24 antigen as the western blot analysis (Fig. 8B). However, it is obvious that the overall release of p24 by ACH-2 cells stimulated with PMA for 48 h was stronger than by ACH-2 cells stimulated with PMA and HA for the same time period. This effect is possibly due to the death

of the see more PMA- and HA-stimulated cells or to the inhibitory effects of CO and bilirubin on HIV-1 reactivation as discussed below. The same stimulatory agents were also used for treatment of A2 and H12 cells for 48 h. As shown in Fig. 8C, expression of EGFP was stimulated with HA alone weakly in both cells, very strongly with PMA and even more strongly with PMA and HA. The stimulation with 10 U/ml of TNF-α or 0.5–1 μg/ml PHA was comparable to the effect of PMA, while the stimulation with 1 U/ml TNF-α induced a relatively weaker expression of EGFP. It can be observed that the effect of 1 U/ml TNF-α was comparable to the effect of HA (2.5 μl/ml)

in H12 cells, while it was stronger in A2 cells. The stimulatory effects of individual agents on the expression of EGFP were also studied using flow cytometry (Fig. 8D, Supplementary data Table S3). Again, these results reveal similar tendencies as western blot analysis, but as mentioned above, H12 cells reveal a higher background expression of EGFP in untreated cells than A2 cells, and in general respond with a smaller fold-increase than A2 cells. Based on various criteria used in this analysis, it can be concluded that A2 cells are more responsive to TNF-α than H12 Vasopressin Receptor cells. When analyzing the cell viability, neither PMA nor TNF-α alone or in combination with HA were found to decrease it. On the other hand, PHA reduced cell viability relatively strongly. In addition to the previous studies, we have explored the ability of T-cells to get activated by PMA in the presence of HA. The A3.01 cells were stimulated with PMA and expression of CD69 on the cell surface was determined. In these assays, HA revealed no negative effects on the T-cell activation characterized by this activation marker at any concentration of PMA tested (1 and 10 ng/ml; data not shown), especially not even at the lowest concentration used throughout the experiments (0.5 ng/ml; Fig.