Henri R. Ford, MD: Nitric Oxide and Gut Barrier Failure

But, at any rate, we'll try to talk to you a little bit about a slightly different perspective with regards to the role of nitric oxide in diseases.

[ Slide 01 ]   NO and Intestinal Physiology

You've heard earlier from Tim how NO can be protective. Well, NO is indeed a molecule of diverse biologic properties. It may play a paradoxical role in intestinal physiology. On the one hand, it may be protective whereas on the other hand it may also be detrimental.

First let's consider some of the protective effects. NOS-2 protein, the inducible isoform of the enzyme, is actually found throughout the GI tract and we know that NO may be important for intestinal permeability as well as intestinal motility. It has very potent antimicrobial properties and furthermore it is up regulated following endotoxin challenge, which may be very important for microbial elimination.

[ Slide 02 ]   untitled

Well, we also know that the sustained up regulation of the inducible isoform of the enzyme may also be detrimental. It may lead to decreased endothelial cell viability. We know that NOS-2 is up regulated in inflammatory bowel disease, and furthermore peroxynitrite, which is a potent oxidant formed by the reaction of superoxide and nitric oxide can induce experimental colitis.

[ Slide 03 ]   NO in NEC

Well several years ago - about seven years ago - we asked the question, "Does NO play a role in the pathophysiology of necrotizing enterocolitis?" And we looked at the resected specimens from 14 babies, premature infants that had undergone resection for perforated NEC and what we found was that NOS-2 was expressed in the intestinal epithelium in virtually all the cases. And furthermore, 3-nitrotyrosine residues, which is the footprint of peroxynitrite, since peroxynitrite by itself cannot be measured directly in the tissues, we had to find another mechanism. But we know that peroxynitrite will target specific tyrosine residues in proteins and lead to their nitration and you can then detect such nitrated protein by looking at immunoreactivity to 3-nitrotyrosine. So we found that these were present and furthermore co-localized with areas of epithelial injury. So not only is NOS-2 present in the epithelium we have evidence that NO must be generated and peroxynitrite was also there and co-localized with enterocyte apoptosis.

[ Slide 04 ]   Translocation

Well to begin to address the mechanism of this process we had to use a reductionist approach since we couldn't really experiment in 600g infants. So we used the well-established model of endotoxin induced bacterial translocation from the gut lumen to the lymph node, liver, spleen and blood. What we wanted to know was whether or not nitric oxide plays a role in the process of translocation and if so, does a nitric oxide scavenger such as NOX, a-dithiocarbamate or aminoguanidine, the competitive inhibitor of the enzyme, could modulate such bacterial translocation.

[ Slide 05 ]   NOX, a-dithiocarbamate

This is NOX, a-dithiocarbamate, which actually chelates iron and this complex in turn will scavenge NO. This is the molecule that we relied mostly for the data I'm going to show you.

[ Slide 06 ]   Materials and Methods

The method that we used in our approach was to place a subcutaneous pump, an Alzet pump, in the back of a Sprague-Dawley rat, allow him to recover and then challenge him with endotoxin and then look for bacterial translocation in his organs. We also look for nitrite-nitrate in the plasma, and then subsequently do morphologic studies using confocal and transmission electron microscopy. We also did some ex-vivo studies using the Ussing chamber.

[ Slide 07 ]   Bacterial Translocation

Okay, this is some of our data. We showed that indeed LPS, this is a saline control, in the LPS treated animals there was significant incidence of translocation but NOX clearly suppressed that activity.

[ Slide 08 ]   Incidence of Bacteremia

Furthermore if you look at the incidence of bacteremia it was completely abrogated by administration of this NO scavenger.

[ Slide 09 ]   Transmucosal Passage of Beads

When we looked at the intestinal segment ex-vivo within the Ussing chamber, again this is over a 180 minute observation period, we had fluoresceinated beads within the chamber and we found that NOX essentially abrogated the passage of the beads. And this is in the saline controls.

[ Slide 10 ]   Effect of NOX on Bacterial Translocation

Well does it matter when you give this nitric oxide scavenger? Well we did some time delay experiments, administering the scavenger at 4, 8, up to 18 hours after endotoxin challenges. We can see a gradual increase in the incidence of translocation. But even up to 12 hours after the administration of endotoxin, NOX still had a protective effect in reducing the incidence of translocation which at least in the trauma setting may have very, very important clinical applications.

[ Slide 11 ]   Epithelial Breaks

When we used confocal microscopy we see that there was evidence of epithelial break within the intestine and this is presumably some of the sites where the bacteria, these are fluoresceinated bacteria, presumably have traversed the epithelium to enter the lamina propria. And that's our saline control.

[ Slide 12 ]   NOX in Epithelial Breaks

And when we treated again with the NOX compound we can see that the epithelium is well preserved and the bacteria are not really penetrating.

[ Slide 13 ]   3-Nitrotrosine Residues

Furthermore we looked for evidence of 3-nitrotyrosine residues and again these bright marks here, the arrows, are pointing to evidence of 3-nitrotyrosine residues in the lamina propria in the saline control.

[ Slide 14 ]   NOX and 3-Nitrotrosine Residues

And these were markedly reduced in the NOX treated animals.

[ Slide 15 ]   NO Effect on Expression of NOS Message

This is just an in situ hybridization to show that indeed expression of iNOS message was not actually affected and this was regardless of the treatment.

[ Slide 16 ]   Aminoguanidine as Competitive Inhibitor

And furthermore, using aminoguanidine, which is a competitive inhibitor of the enzyme, we were able to show the same types of protective effect. These are the animals that received saline and on that side, animals receiving aminoguanidine, and we can see that there is evidence of epithelial break near the villus apex as well as some sloughing of the mucosa. This is a slightly higher magnification of what we see in panel A. And by electron microscopy we can see that the microvillus border is markedly destroyed whereas animals that received the competitive inhibitor of NO had preservation of the intestinal epithelial architecture. There is a little bit of villus core separation and some mucosal edema, but we do not see the sloughing of the apex. And similarly when you look by TEM, the microvillus border is essentially intact compared to the saline controls.

[ Slide 17 ]   Evidence of Apoptosis

Next we looked for evidence of enterocyte apoptosis. This is a TUNEL assay, a double staining technique. On this side here we look at the Hoechst stain, which stains all the nuclei, and here this is our FITC stain which stains all the apoptotic enterocyts.

[missing audio]

Epithelial injury was occurring and we can that this is where we have epithelial injury in the saline controls.

[ Slide 18 ]   NOX in Apoptosis

Such changes were really very sparse when you treat the animals with NOX and the integrity of the epithelium was well perserved.

[ Slide 19 ]   untitled

Immaturity of the neonatal gut barrier may predispose to bacterial invasion leading to LPS release. Subsequently we have NOS-2 gene transcription and with consequent formation of NO and peroxynitrite within the enterocyte. And we know that excess production of peroxynitrite, while it may have some protective effect in terms of bacterial killing, it may also lead to DNA strand break, enterocyte apoptosis, further tissue disruption, further bacterial invasion, and systemic sepsis.

[ Slide 20 ]   Global Hypothesis

Well, this led us to this global hypothesis that at least some of these damages may indeed be caused by peroxynitrite. And this is true for NEC, this is true for inflammatory bowel disease and just any form of endotoxemia.

[ Slide 21 ]   Proposed Mechanism of NO-Induced Enterocyte Apoptosis

So we needed to take it further. We wanted to begin to at least address some of the mechanisms.

And what are some of the potential mechanisms that we need to look at? Well again, the peroxynitrite that is generated intracellularly may affect the mitochondria, leading to mitochondrial energy depletion, which in turn may activate the caspaces, and that cascade ultimately may lead to activation of effector caspases, that could lead subsequently to DNA fragmentation and cell death.

[ Slide 22 ]   Experimental Design

We had to use yet another reductionist approach to begin to address these mechanistic questions. And we resorted to the IEC6 cells, which is a rat epithelial cell line, of crypt origin. We examined whether or not that particular cell line would be responsive to peroxynitrite, and what would happen if we exposed it to peroxynitrite.

[ Slide 23 ]   % Apoptosis in IEC-6 Cells Exposed to OONO

Well, what we found is that if we exposed these cells to 12.5 micromolar peroxynitrite, which is a dose that is actually fairly attainable within pathological conditions in-vivo, there was a gradual increase in the incidence of apoptosis so that by 24 hours nearly 63 percent of the cells had undergone apoptosis.

[ Slide 24 ]   The Underlying Mechanism

Well next we wanted to address what is the underlying mechanism for this observation? We used this DiOC6 dye which is a dye that incorporates within the membrane of healthy mitochondria whereas in the sick mitochondria if there is any alteration in the transmembrane potential there is not going to be any uptake.

[ Slide 25 ]   DiOC6 Flow Cytometry

And you can look by flow cytometry at the intensity. So that if the die is incorporated, there is going to be much higher intensity, and vice versa - if it is not, there will be a lower intensity.

[ Slide 26 ]   IEC-6 Cell

Well what we found was that PBS causes a fairly high intensity. This is the actual readout here. This dose of peroxynitrite decreased it nearly 40 to 50 percent. When we used an electron chain uncoupler, this was clearly the most potent, and the combination of peroxynitrite and this uncoupler really pretty much destroyed all of the intensity. So here we see that indeed peroxynitrite is able to alter the transmembrane potential of the IEC6 cell line.

[ Slide 27 ]   IEC-6 Cell Procaspase-3 Protein Cleavage

So we wanted to subsequently determine whether or not there was cleavage of procaspase-3, which many believe to be the most important effector of the caspase cascade, leading to cleavage of ICAD, which is the inhibitor of the caspase-associated deoxyribonucleases which are believed to be the terminal effectors of nucleosomal fragmentation or apoptosis.

The procaspase-3 enzyme is about 30 kD and this is the Jurkat cell which is our positive control, and in this case we know the pro-enzyme can be cleaved to a 20 kD fragment. There can also be alternate cleavage to molecules of down to 17 kD. And within the peroxynitrite treated IEC6 cells what we see that there is a 17 kD fragment that we detect within 6 and 18 hours, thus indeed showing that this is the potential mechanism by which peroxynitrite is mediating its effects in the IEC6 cells.

[ Slide 28 ]   Peroxynitrite Effects on PARP Cleavage

We also wanted to know whether or not PARP was cleaved, this is the poly (ADP-ribosome) polymerase which is one of those DNA repair enzymes that can initiate a fairly futile DNA repair process that would lead to bioenergetic deprivation. And again, this is our PBS control, our negative control. TNF is known to pretty much cause PARP activation cleaving the 116 kD molecule to the 85 and 31 fractions. We can see that this is what occurred here. The peroxynitrite-treated cells begin to show cleavage as early as 4 hours and by 12 hours we can see that there is pretty extensive cleavage and that corresponds to areas of apoptosis.

[ Slide 29 ]   Summary

Well, just to synthesize it all, it appears that IEC6 cells exposed to OONO or peroxynitrite do indeed undergo apoptosis and that's in a time-dependent fashion. And that I believe procaspase-3 and PARP may play a role in this process and it's associated with alteration in the mitochondrial transmembrane potential.

[ Slide 30 ]   Hypothesis

The current working hypothesis in our lab is that the gut barrier failure that is mediated by nitric oxide or perhaps by peroxynitrite is really the result of an imbalance between enterocyte proliferation and enterocyte apoptosis.

[ Slide 31 ]   Src Kinase Signaling Pathway - 1

And that pretty much has led us into the area of signal transduction, which we're not going to talk about extensively, but there is evidence at least for us within the gut that there may be, that the SRC kinase family of tyrosine kinases may be playing a critical role in that process.

[ Slide 32 ]   Src Kinase Signaling Pathway - 2

And we know that if you activate via the Ras pathway, you're going to get proliferation. You can decrease elimination if you go through a PI3 kinase route because it decreases apoptosis, predominantly by preventing bad phosphorylation which would hetero-dimerise with BCL2, or you can also again promote cell growth by activating focal adhesion kinase which will immobilize the cell and allow it to undergo this process. And our initial studies, again we're not going to go into those details, clearly show that PI3 kinase importantly is significantly suppressed in this system. And that could potentially at least explain another way of enhancing apoptosis within the intestinal epithelium. And the increase in Src kinase that we observe may be secondary, some epiphenomenona, of the PI3 kinase blockade. But that will be for another time. Well I really want to thank you for the privilege of the floor.

Thank you.