| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Bench |
Division of Chemistry for Materials,1 Faculty of Engineering, Graduate School of Mie University; Division of Therapeutic Blood Purification,2 Mie University School of Medicine, Ochanomizu University,3 Tsu, Mie, Japan
Correspondence to: T. Horiuchi, Division of Chemistry for Materials, Faculty of Engineering, Graduate School of Mie University, 1577 Kurima-Machiyacho, Tsu, Mie, 514-8507 Japan. horiuchi{at}chem.mie-u.ac.jp
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONDISCLOSUREREFERENCES |
|---|
Background: Reactive oxygen species (ROS) have been
speculated as possible inducers of structural or functional changes that lead
to a hyperpermeable state in patients on long-term peritoneal dialysis. This
study aimed to compare localization of tight junction-associated proteins
(TJPs), which relate to solute permeability characteristics, between human
peritoneal mesothelial cell (HPMC) monolayers and human umbilical vein
endothelial cell (HUVEC) monolayers under oxidative stress. Methods: HPMCs and HUVECs were cultured on a polymer
mesh until transepithelial electrical resistance reached a plateau. Solute
permeation tests were conducted using FITC-labeled dextrans. Localization of
TJPs was observed under a confocal laser scanning microscope. These
experiments were carried out with/without 0.1 mmol/L
H2O2. In addition, ROS production as well as the amounts
of intracellular reductive glutathione (GSH) and oxidative glutathione were
measured. Results: When the monolayers were exposed to 0.1 mmol/L
H2O2/medium for 2 hours, the HPMC monolayer revealed a
significant reduction in transepithelial electrical resistance (from 32.5
± 3.4 to 17.4 ± 4.9 
·cm2) with
delocalization of TJPs, particularly occludins. The HUVEC monolayer remained
stable and exhibited an unremarkable change in TJP organization. Compared to
the HUVEC monolayer, the HPMC monolayer exhibited two- to threefold higher
2',7'-dichlorofluorescein intensities that increased in a
dose-dependent manner. HUVECs contained approximately 2.5-times more GSH than
HPMCs. This supported the lesser production of ROS when exposed to 0.1 mmol/L
H2O2 for 24 hours. HUVECs used 8.03 nmol/mg GSH protein
to maintain TJP localization, while only 3.75 nmol/mg GSH protein was
available for the HPMCs. Conclusion: The HUVEC monolayer, which was less
permeable to middle-to-high molecular weight solutes, was more tolerant
against ROS stress than the HPMC monolayer. Availability of intracellular GSH
is an important issue in maintaining the integrity of the mesothelium.
KEY WORDS: Human peritoneal mesothelial cells; human umbilical vein endothelial cells; solute permeability; occludin; zonula occludens-1 (ZO-1); transepithelial electrical resistance; oxidative stress; 2',7'-dichlorofluorescein (DCF); glutathione.
Solute and water transport via the peritoneum, which is composed of the mesothelium and the interstitium embedding a network of capillaries, is a major principle of continuous ambulatory peritoneal dialysis therapy. This complex structure, however, hinders our efforts to find the most critical region in the peritoneum when a patient has an unsatisfactory outcome, such as insufficient water and solute removal.
To assess the functional integrity of the peritoneum, the three-pore model developed by Rippe and Stelin has been clinically utilized as the most informative analytical tool (1). In this model, solute permeation data are fitted to theoretical equations on the basis of three hypothetical pores. Although an ultrasmall pore with estimated diameter of 4 – 6 Å has been hypothesized as the water transport channel (i.e., aquaporins), the morphologically corresponding pathways for the other two pores (i.e., small pores and large pores) have not yet been satisfactorily determined.
In the peritoneal equilibration test (PET) commonly used clinically, water soluble substances with low molecular weight, such as creatinine and urea, are monitored and evaluated as dialysate-to-plasma (D/P) ratios (2). Although change in effective peritoneal surface area is one of the crucial factors influencing the D/P ratio, the more intrinsic property, namely solute permeability, should be discussed. Which pathway in the three-pore model would correspond to D/P ratio in a PET? Small pores would be responsible for this pathway because their estimated pore size is around 40 Å. Moreover, the intercellular pathway (lateral) could be specified because these substances do not have a specific transport channel on the plasma membrane.
In the cellular physiology of epithelium and endothelium, an intracellular junction apparatus has been noted as a solute transport barrier, in particular the tight junction (TJ) located at the most apical position (3–5). The role of the TJ as a solute transport barrier differs in different types of cells. For example, higher solute transport barrier is noted in the vascular endothelial cells of the brain aortic capillary and the cornea (6,7), and in the epithelial cells of the bowel and the cornea (8,9). Our previous study, wherein we constructed a human peritoneal mesothelial cell (HPMC) monolayer on a polymer mesh, suggested the existence of a solute transport barrier that was also controlled by tight junction-associated proteins (TJPs) (10). These TJP organizations relate to epithelial-to-mesenchymal transition in which transforming growth factor-beta1 (TGF-β1) downregulates E-cadherin through expression of transcriptional factor Snail (11). It has been reported that Snail modulates TJPs of Madin–Darby canine kidney cells, resulting in increased ion permeability (12). As described elsewhere, TGF-β1 is not the only factor, other factors such as reactive oxygen species (ROS) and advanced glycated end products may be involved in this cellular event (13,14).
Based on this evidence, we hypothesized that TJ characteristics, which relate to solute permeability and epithelial-to-mesenchymal transition, differ between peritoneal resident cells (e.g., the mesothelium and the endothelium). In order to understand the mechanism of solute transport via the peritoneum and its correlation to TJP organization, it is worthwhile to characterize both the HPMC monolayer and the human umbilical vein endothelial cell (HUVEC) monolayer in the same experimental setting. The specific purposes of the present study were to compare (1) solute permeability, (2) susceptibility of TJP organization to oxidative stress, (3) ROS productivity, and (4) antioxidative capacity between HPMCs and HUVECs.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONDISCLOSUREREFERENCES |
|---|
CONSTRUCTION OF HPMC AND HUVEC MONOLAYERS ON A MEMBRANE SUPPORT
To construct an in vitro model of the HPMC and HUVEC monolayers, a
cell suspension (5 x 104 cells/cm2) was seeded and
cultured on a polyester mesh (Transwell, 0.4 µm pore size, 12-well type;
Costar, MA, USA) using 10% fetal bovine serum (FBS)/medium. The inner and
outer chambers were filled with 0.5 mL and 1.5 mL culture medium respectively;
the culture medium was replaced every 3 days. In all the experiments described
below, HPMCs were used within three passages.
MEASUREMENT OF TRANSEPITHELIAL ELECTRICAL RESISTANCE
Transepithelial electrical resistance (TER) was measured daily or every 2
days using an EVOM volt ohm meter with STX-2 electrodes (World Precision
Instruments, Sarasota, FL, USA). The electrodes were inserted into both ends
of the mesh. Alternating current of less than ±20 µA was applied
between the electrodes at a frequency of 12.5 Hz. Prior to measurement, the
culture medium was replaced with fresh medium (0.1% FBS/medium) and maintained
at 37°C. The resistance of each monolayer was multiplied by the effective
surface area to obtain the electrical resistance of that monolayer (in
·cm2). To calculate the normalized TER of each
monolayer, background TER of a blank polyester mesh was subtracted from the
TER of the respective cell monolayer. Any study using the monolayer was
conducted 2 days, but not more than 3 days, after steady state was
achieved.
SOLUTE PERMEABILITY TEST (SPT)
Once stable TERs were obtained, the permeabilities of the HPMC and HUVEC
monolayers were determined by measuring changes in the concentrations of
molecular markers. Fluorescein isothiocyanate (FITC)-labeled dextrans
(molecular weights 4, 10, and 70 kDa) were used as fluorescent molecular
markers. Each FITC-labeled dextran was added to the inner chamber of the
chamber at a final concentration of 50 µg/mL. An equimolar amount of
unlabeled dextran was added to the outer chamber of the polyester mesh system
to maintain an isotonic condition. At 2 hours and 4 hours after the addition
of a molecular marker, a 10-µL volume of each sample was collected from
both sides of the chamber. Each sample was assessed using a fluorescence
spectrophotometer (F-2000; Hitachi, Tokyo, Japan) at an excitation wavelength
of 490 nm and an emission wavelength of 520 nm. The solute permeability
coefficient (SPC), which is the flux of FITC-labeled dextran across the
monolayer, was calculated using the following equation:
 |
t is the time interval (s), and C
is the concentration of the molecular marker
(Ca0 = concentration at the apical side at time
0; Cb4 = concentration at the basal side at 4
hours; Cb2 = concentration at the basal side at
2 hours).
IMMUNOHISTOCHEMICAL STAINING OF OCCLUDINS AND ZONULA OCCLUDENS-1 (ZO-1)
Occludins and ZO-1 were immunohistochemically stained in accordance with a
previously described protocol
(17). Briefly, HPMCs and
HUVECs were cultured on an eight-chambered slide (Nunc, NY, USA) coated with
types I and III collagen (Cellmatrix Type 1-A; Nitta Gelatin, Osaka, Japan).
Following confluence, HPMCs were washed twice with phosphate-buffered saline
(PBS). The medium was replaced with fresh medium (0.1% FBS/M199 for HPMCs and
0.1% FBS/EGM-2 for HUVECs) and the cells were incubated for another 24 hours.
Subsequently, the cells were washed twice with PBS and then cultured for 30
minutes in the test media. The cells were then fixed in cold acetone/methanol
at –20°C for 5 minutes, rehydrated in PBS, and blocked for 1 hour in
PBS containing 20% Block Ace (Dainippon Seiyaku, Tokyo, Japan). This was
followed by overnight incubation with a primary antibody at 4°C. A rabbit
anti-ZO-1 antibody (Zymed Laboratories, San Francisco, CA, USA) and a rabbit
anti-occludin antibody (Zymed) were used as primary antibodies at a dilution
of 1:100 and 1:50 respectively. After incubation with the primary antibody,
the cells were washed 5 times with PBS. Subsequently, an appropriate secondary
antibody was applied for 1 hour at room temperature, following which the cells
were again washed 5 times with PBS. Polyclonal swine anti-rabbit FITC-labeled
immunoglobulins (Dako, Glostrup, Denmark) were used at a dilution of 1:100 and
incubated at room temperature for 1 hour. Stained specimens were examined
under a confocal laser scanning microscope (Fluoview FV1000; Olympus, Tokyo,
Japan).
EXOGENOUS H2O2 SUPPLEMENTATION INTO THE HPMC AND HUVEC MONOLAYERS
To examine the effect of H2O2 on solute permeation
and TJP localization, the HMPC and HUVEC monolayers were supplemented with
exogenous H2O2. Prior to the SPT, the effect of
H2O2 on enzymatic activities of mitochondria was
assessed in a dose-dependent manner using the
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay,
which is an indicator of the activity of mitochondrial dehydrogenases. The
H2O2 was applied at final concentrations of 0.05, 0.1,
0.5, and 1.0 mmol/L and exposure time was set at 2 hours.
For the SPT, 0.1 and 1.0 mmol/L were selected as H2O2 concentrations based on the results of the MTT assay. After 30-minutes' exposure of the HPMC and HUVEC monolayers to 0.1 mmol/L and 1.0 mmol/L H2O2/medium, SPTs were performed according to the protocol described above.
Immunohistochemical examinations were also carried out to assess the effect of H2O2 on TJP delocalization.
OBSERVATION OF INTRACELLULAR ROS PRODUCTION USING 2',7'-DICHLOROFLUORESCEIN DIACETATE (DCFH-DA)
To quantify intracellular ROS production, DCFH-DA (Sigma, Japan), a
fluorescence probe that permeates the cell membrane, was added to the HPMC and
HUVEC monolayers at a final concentration of 10 µmol/L after 15 minutes'
incubation with 0.01, 0.1, and 1.0 mmol/L H2O2/medium.
DCFH-DA was allowed to incorporate into the cells for 15 minutes at 37°C;
DCF produced as a result of ROS-induced DCFH oxidation was detected under a
confocal laser scanning microscope at an excitation wavelength of 480 nm and
an emission wavelength of 530 nm
(18).
MEASUREMENTS OF INTRACELLULAR REDUCTIVE GLUTATHIONE (GSH) AND OXIDATIVE GLUTATHIONE (GLUTATHIONE DISULFIDE; GSSG)
The amounts of intracellular GSH and GSSG were measured using an enzymatic
recycling method (19).
Briefly, after discarding the culture medium, the monolayer cultured on a
35-mm diameter culture flask was washed twice with PBS. The scraped cells were
resuspended in 80 µL 10 mmol/L HCl and then frozen at –80°C for
15 minutes. Thawing and freezing of the cell suspension was repeated a couple
of times following deproteinization with 5% sulfosalicylic acid. Supernatant
(20 µL) was added to a mixture of 30 µL glutathione reductase and 20
µL of 1 mmol/L NADPH. After incubation for 10 minutes at 37°C, 20 µL
10 mmol/L DTNB was added and absorbance was measured at 405 nm at a certain
time interval.
To measure the amount of GSSG, 2-vinylpyridine was added prior to the above-mentioned GSH measurement because 2-vinylpyridine inhibits the reduction of GSSG to GSH by binding to GSSG. Therefore, the amount of GSSG could be calculated by subtracting the GSH amount from the total glutathione amount.
DATA ANALYSIS
All data are expressed as mean ± SD. Statistical significance of
TER, MTT, and solute permeabilities were calculated using Student's t-test.
The level of significance was set at p < 0.05. For the analysis of
covariance, we used SAS software for statistical analysis (SAS/STAT User's
Guide Release, 6.03 Edition; Tokyo, Japan).
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONDISCLOSUREREFERENCES |
|---|
·cm2 on the eighth day (n = 7)
for the endothelial monolayer, and at 32.3 ± 2.8
·cm2 on the fifth day (n = 4) for the
mesothelial monolayer (10).
This was accompanied by confluence of cells, as previously mentioned.
|
As distinctly seen in Figure
2, solute permeability of the HUVEC monolayer on the polymer mesh
was lower than that of the HPMC monolayer, indicating that the HUVEC monolayer
is more resistant to permeation by middle and high molecular weight solutes. A
real SPC (Kreal) was calculated by subtracting the blank SPC from
the measured SPC in an inverse fashion:
 |
|
|
|
·cm2) during the
2-hour exposure to 0.1 mmol/L H2O2/M199 (n =
7). On the other hand, there was no significant reduction in the TER of the
HUVEC monolayer during the 2-hour exposure to 0.1 mmol/L
H2O2/EGM-2 (n = 9), as shown in
Figure 5.
|
Figure 6 shows the SPCs of the polymer mesh (pore size 0.4 µm) for each molecular marker with and without the HPMC and HUVEC monolayers. As distinctly seen in the figure, the solute permeability of the HUVEC monolayer on the polymer mesh was lower than that of the HPMC monolayer, indicating that the HUVEC monolayer is more resistant to the permeation of middle and high molecular weight solutes.
|
INTERCELLULAR LOCALIZATION OF OCCLUDINS AND ZO-1 UNDER OXIDATIVE CONDITIONS
Immunohistochemical staining of occludins and ZO-1 clearly showed the
intercellular localization of TJPs. In the control, TJPs, occludins, and ZO-1
were localized along the intercellular spaces, as shown in the left panels of
Figure 7. On the other hand,
the localized occludins diminished from the intercellular space when the HPMC
monolayer was exposed to 0.1 mmol/L H2O2, while they
were maintained in the HUVEC monolayer. Disappearance of localized ZO-1 was
also observed; however, ZO-1 disappeared to a lesser extent than the
occludins.
|
|
AMOUNT OF INTRACELLULAR GSH AND GSSG WITH/WITHOUT 24-HOUR EXPOSURE TO 0.1 MMOL/L H2O2
The amount of total glutathione per milligram proteins was significantly
greater in HUVECs than in HPMCs (23.4 ± 11.7 vs 7.19 ± 3.74
nmol/mg protein) (Figure 9).
Due to the 24-hour exposure to 0.1 mmol/L H2O2, the
amount of GSH decreased significantly, from 12.2 ± 8.67 to 4.17
± 0.98 nmol/mg protein in the HUVECs and from 4.83 ± 3.10 to
1.08 ± 0.51 nmol/mg protein in the HPMCs. There was no significant
change in the amount of total glutathione in either HPMCs or HUVECs.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONDISCLOSUREREFERENCES |
|---|
For the endothelial monolayer, the SPC of D-70 (3.8 x 10–6 cm/s) was almost similar to the values of albumin (4.5 – 5.6 x 10–6 cm/s) and dextran (3 – 8 x 10–6 cm/s) reported previously (10). As can be seen in Figure 3, the real SPC of D-70 for the HPMC monolayer was approximately eight times higher than that for the HUVEC monolayer, whereas only two times higher SPC of D-4 was noted. Therefore, these SPC and TER results indicate that, compared to the HPMC monolayer, the HUVEC monolayer is less permeable to middle-to-high molecular weight solutes but their permeabilities to smaller molecules are similar.
Clinical data usually obtained from PETs reflect overall mass transfer area coefficients, the inverse parameter of which is represented as overall mass transport resistance: Roverall = 1/KoverallAoverall = Rmeso + Rendo = 1/KmesoAmeso + 1/KendoAendo (20). Since each resistance is represented as an inverse of the product of surface area and SPC, the contribution of Rmeso to Roverall may increase when vascular surface area (Aendo) increases. For example, a two-times increase in vascular surface area (Aendo) gives a solute transport resistance of the mesothelium (Rmeso) equal to a 4-kD solute. On the other hand, solute transport resistance (Rmeso) for 10 kDa would be less influenced by the same increment. If SPTs using at least two marker solutes with widely varying molecular weights are available, the differences in solute transport profiles between HPMCs and HUVECs might enable an educated guess as to which layer might be deteriorated.
SOLUTE TRANSPORT AND INTERCELLULAR LOCALIZATION OF TJPs UNDER OXIDATIVE CONDITIONS
It has been suggested that repeated exposure to fresh peritoneal dialysate
induces an oxidative stress to peritoneal resident cells, particularly in
long-term peritoneal dialysis. In the present study, we employed exogenous
supplementation of H2O2 to assess the effect of ROS
production and its effects on TJP organization. According to Makino et
al.'s recent study, exogenous H2O2 is reduced in a
stepwise manner mainly by glutathione when present in low concentrations, and
by catalases when present in high concentrations
(21). The crossover point is
assumed to be around 0.1 mmol/L. In the present study, we employed
supplementation of exogenous H2O2 at concentrations
ranging from 0.01 to 1.0 mmol/L for the DCFH-DA study, and at a concentration
of 0.1 mmol/L for the other experiments (TER, SPT, TJP staining, and GSH).
Based on a previous report
(22), it can be supposed that
the actual intracellular concentrations are probably between one-seventh and
one-tenth of the supplemented concentrations.
It has been widely reported that TER and/or solute permeability reflect a well-formed intercellular junction in various types of cells (3–9). The intercellular space between epithelial and endothelial cells is bridged by a set of specialized structures, namely, TJs or ZO, zonula adherens, desmosomes, and gap junctions. Tight junctions, the most apical structures of the junctional complex, act as a diffusion barrier by controlling the passage of ions, water, and other molecules as well as by maintaining the polarity of cells. Therefore, localization of TJPs is a key factor in comprehending solute transport through a cell layer. Loss of occludin localization was more prominent in HPMCs (Figure 7); this may be evidence enough to speculate that occludin plays a role in the barrier function for electrolyte mobility. A unique extracellular loop of occludin, which contains approximately 60% tyrosine and glycine residues, may partially participate in charge selectivity. Claudins, which are also TJP members, may contribute to the diversity of permeability across different cell types because there are 20 types of claudins expressing differently on each cell type (23). It is speculated that both number and type of TJPs differ between HPMCs and HUVECs, and their expressions along the lateral region might be attributed to the integrity of the extracellular matrices.
INTRACELLULAR PRODUCTION OF ROS
Oxidative stress is usually balanced by the oxidant and antioxidant
capabilities of a cell. Scavenging of various types of oxidants occurs largely
through H2O2 generation, followed by reduction of
H2O2 to H2O by an antioxidative enzyme,
various peroxides, and catalases, resulting in an intracellular
H2O2 concentration of <1 µmol/L (approximately 1
– 700 nmol/L) under physiological status
(22).
DCFH-DA was used in our experiment to detect the intracellular production
of ROS because DCFH-DA incorporated into the cells is immediately
de-acetylated to DCFH by esterase, a membrane enzyme. This form of DCFH can
remain inside the cell. When ROS exist in proximity to this compound, DCFH is
oxidized to DCF, which is a fluorescein form. Almost all ROS can oxidize DCFH
despite the different reactivities among the ROS, which are as follows
(24):
 |
In our experimental conditions, both ·OH and H2O2 were the major ROS detected by fluorescein intensity and resulted from the imbalance between the intracellular antioxidative activity and the excess amount of exogenous H2O2. While some exogenous H2O2 that passes through the membrane is reduced to H2O by GSH and/or catalase, the remaining H2O2 would contribute to ·OH generation by the gain of electrons from reduced-type transient metal ions such as Fe2+ and Cu+. The latter reaction, known as the Fenton reaction, is the first-order reaction until adequate amounts of reduced-type transient metal ions exist. As seen in Figure 8, there were significant dose-dependent increases in DCF intensity in both HPMC and HUVEC monolayers. In particular, the HPMC monolayer exhibited a stronger intensity than the HUVEC monolayer when exposed to a concentration of up to 1.0 mmol/L H2O2. The difference in the intensity between the HPMC and HUVEC monolayers reveals that the antioxidative capability of HUVECs is much stronger than that of HPMCs. This finding may be important in terms of the primary host defense capacity of HPMCs.
AMOUNT OF INTRACELLULAR GSH AND GSSG
These results help us understand the reasons for the differences in the
amounts of antioxidants between the two monolayers. Glutathione (a
tri-peptide: L-
-glutamyl-L-cysteinyl-glycine) is the most abundant
intracellular antioxidant responsible for regulating the
oxidation–reduction balance.
The GSSG reductase–DTNB recycling method, which was employed in this study, is generally considered a specific, sensitive, rapid, and reliable method (19). Under oxidative stress, GSH is oxidized to GSSG, which is subsequently reduced to GSH by glutathione reductase to maintain a constant intracellular GSH concentration. Under an excess load of oxidative stress, such as 24-hour exposure to 0.1 mmol/L H2O2, which was applied in our experiment, the balance between GSH and GSSG shifted toward the oxidation state, as shown in Figure 9. Our results clearly indicate that the amount of GSH per milligram protein as well as total glutathione amount (GSH + GSSG) are significantly higher in HUVECs than in HPMCs.
Maintaining the integrity of the peritoneal mesothelial layer would contribute to the preservation of natural host defense, which plays an important role in susceptibility to foreign bodies (25). In their in vitro study, Shostak et al. demonstrated D-glucose-induced H2O2 production in rat peritoneal mesothelial cells (26). Lee et al. (18) have reported that a high concentration of glucose (100 mmol/L) causes DCFH-sensitive ROS generation in HPMCs; the intensity of the ROS corresponded to that of the supplemented 0.1 mmol/L H2O2.
Our basic in vitro results reasonably support the results of the In vivo preclinical study with respect to the use of antioxidative agents. Breborowicz et al. and Tager et al. demonstrated the effect of L-2-oxothiazolidine carboxylic acid, a glutathione precursor, on glucose-induced dysfunction of mesothelial cells (27,28). Based on these results, antioxidative treatment of the mesothelium to maintain peritoneal integrity may be important to prevent not only a hyperpermeable state but also epithelial-to-mesenchymal transition.
|
CONCLUSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONDISCLOSUREREFERENCES |
|---|
|
DISCLOSURE |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONDISCLOSUREREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
We thank Prof. Kusunoki and Dr. Tonouchi (Mie University School of Medicine, Japan) for providing the omentum samples.
Received 2 October 2007; accepted 26 June 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONDISCLOSUREREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
