Why SIBO Keeps Coming Back: MMP-9, EMT, and the Collapse of Host Capacity

One part of my Host Capacity Model that I have been researching is what happens when chronic intestinal inflammation stops being only an immune problem and begins changing the actual identity and metabolism of the intestinal epithelial cell. This matters because the intestinal lining is not simply a passive wall separating bacteria from the bloodstream. Epithelial cells are active metabolic cells. They maintain tight junctions, absorb nutrients, produce enzymes, communicate with immune cells, respond to microbial metabolites, regulate the chemical environment immediately surrounding microorganisms, and use mitochondrial metabolism to maintain normal tissue function. If the cellular program that creates a healthy epithelial cell begins to change, the microbial environment can change with it.

One pathway I have been looking closely at involves MMP-9, E-cadherin, beta-catenin, and epithelial-to-mesenchymal transition, or EMT. But MMP-9 does not simply appear by itself. Chronic microbial and inflammatory signaling can drive its expression. Gram-negative bacterial lipopolysaccharide, or LPS, can activate TLR4 and downstream signaling through pathways such as MyD88, TAK1, IKK, and NF-kB. Cytokines such as TNF-alpha and IL-1 beta can also activate NF-kB and MAPK pathways, including p38 and JNK, while oxidative stress and other inflammatory signals can increase AP-1 activity. NF-kB and AP-1 are important transcriptional regulators of inflammatory genes, including MMP-9. In a chronically inflamed intestinal environment, the sequence can therefore begin with microbial products and immune activation, leading to persistent NF-kB and MAPK signaling and increased MMP-9 activity.

In simplified terms, the pathway I am investigating looks like this: microbial products such as LPS increase TLR4 signaling, inflammatory cytokines such as TNF-alpha and IL-1 beta increase NF-kB and MAPK signaling, oxidative stress reinforces these pathways, and MMP-9 expression and activity can rise. Once MMP-9 remains elevated, the consequences extend beyond inflammation because MMP-9 can degrade components of the extracellular matrix and disrupt epithelial junctional architecture.

One important target in this model is E-cadherin. E-cadherin is part of the adherens-junction system that physically connects neighboring epithelial cells. Beta-catenin is normally associated with E-cadherin at these junctions. When E-cadherin is disrupted, beta-catenin can be released from its normal membrane-associated position into the cytoplasm.

Normally this free beta-catenin should not simply accumulate. The cell has a regulatory destruction complex involving APC, Axin, GSK-3 beta, CK1, and other proteins. Beta-catenin is phosphorylated, recognized by the ubiquitin system, and degraded by the proteasome. This is one of the mechanisms preventing uncontrolled beta-catenin signaling.

But chronic inflammatory signaling can interfere with this regulation. If beta-catenin becomes sufficiently stabilized, it can accumulate in the cytoplasm, enter the nucleus, and interact with TCF and LEF transcription factors.

At this point the biology becomes much more important than simple barrier damage. We are no longer talking only about a junction being temporarily opened. We are talking about the possibility that the cell begins changing its transcriptional identity.

Nuclear beta-catenin signaling can interact with programs involving Snail, Slug, and Twist, major regulators associated with EMT. These transcription factors can suppress epithelial differentiation programs and reduce expression of proteins required for maintaining a polarized epithelial phenotype. E-cadherin can decline further, creating another feedback loop. Tight-junction proteins such as ZO-1 and occludin can also decline as the epithelial architecture becomes progressively remodeled.

This is epithelial-to-mesenchymal transition. I think the easiest way to understand EMT is not simply that the cell becomes “damaged.” The cell begins changing what type of cell it is trying to be.

A normal intestinal epithelial cell has a very specialized program. It knows which side faces the intestinal lumen and which side faces the underlying tissue. It builds tight junctions between neighboring cells. It expresses digestive enzymes and transporters. It maintains microvilli. It controls nutrient absorption. It communicates with immune cells. It metabolizes microbial products. It generates ATP through mitochondrial metabolism and continuously participates in maintaining the ecological conditions surrounding the microbiome.

During significant EMT-like remodeling, parts of this differentiated epithelial program can be suppressed. This is where several components of my Host Capacity Model come together.

Before even discussing MCT1, consider what happens when the epithelial cell begins losing proteins such as intestinal alkaline phosphatase, ZO-1, occludin, and E-cadherin.

Intestinal alkaline phosphatase, or IAP, is not simply another digestive enzyme. It is part of the intestinal defense system. IAP can dephosphorylate and detoxify inflammatory microbial molecules such as LPS and helps participate in maintaining mucosal homeostasis. If the differentiated epithelial program responsible for expressing IAP is weakened, the host may become less capable of neutralizing microbial inflammatory pressure at the luminal surface.

That creates an important potential feedback mechanism. More biologically active LPS can mean greater TLR4 stimulation, greater NF-kB activation, more inflammatory cytokine signaling, and potentially more MMP-9 expression. The pathway can therefore begin feeding back into itself.

At the same time, losing ZO-1 and occludin compromises the tight-junction architecture that regulates movement between epithelial cells. Losing E-cadherin weakens adherens junctions and epithelial polarity. So now the tissue is not merely inflamed. Several of the systems required for maintaining its normal physical identity are being compromised simultaneously.

The model can therefore begin to look like this: LPS and inflammatory cytokines increase NF-kB and MAPK signaling, MMP-9 rises, E-cadherin is disrupted, beta-catenin signaling changes, Snail, Slug, and Twist reinforce EMT-like remodeling, and normal epithelial programs involving E-cadherin, ZO-1, occludin, IAP, and metabolic transport systems become progressively impaired.

Then we reach another important part of the model: MCT1 and butyrate utilization.

The research material I have been examining proposes that as epithelial identity is lost, expression of transport systems such as MCT1 and SMCT1 may also fall alongside other epithelial functions. I think this is particularly important because it changes how we should think about microbial metabolites.

We frequently hear that butyrate is beneficial, so the natural assumption becomes that if we increase butyrate-producing bacteria or give butyrate directly, the epithelial cell should recover. But there is an important missing step.

The presence of a metabolite is not the same thing as the host’s ability to use that metabolite.

Butyrate sitting in the intestinal lumen still has to reach the epithelial cell, cross its membrane, enter intracellular metabolism, and be oxidized by mitochondria before many of its metabolic effects can occur.

MCT1, encoded by SLC16A1, is one of the transport systems capable of moving monocarboxylates such as short-chain fatty acids across cellular membranes. SMCT1, encoded by SLC5A8, provides another mechanism for short-chain fatty-acid uptake. If the epithelial cell reduces expression of these transporters as it loses its differentiated phenotype, the bottleneck is no longer necessarily how much butyrate the microbiome produces. The bottleneck may become whether the host epithelial cell can transport it.

This creates an important distinction in the Host Capacity Model: microbial production does not equal host utilization.

A person could theoretically have adequate luminal butyrate while the epithelial tissue has impaired transport capacity. Adding even more butyrate does not automatically repair a transporter that is poorly expressed. And even successful transport does not guarantee that the mitochondria are capable of oxidizing the substrate normally.

Once butyrate enters a metabolically competent epithelial cell, it can be converted through beta-oxidative pathways into acetyl-CoA. Acetyl-CoA enters the TCA cycle. The TCA cycle generates NADH and FADH2. These reducing equivalents deliver electrons to the mitochondrial electron transport chain. Electron flow drives proton pumping across the inner mitochondrial membrane, establishes the electrochemical gradient used by ATP synthase, produces ATP, and ultimately requires oxygen as the terminal electron acceptor at Complex IV.

So there is an entire host pathway downstream of microbial butyrate production: butyrate must be produced, transported through MCT1 or other transport systems, taken into the cell, metabolized to acetyl-CoA, oxidized through the TCA cycle, coupled to electron transport, and converted into useful cellular energy.

Any one of those steps can become a bottleneck.

If MCT1 and SMCT1 expression declines, substrate entry can decline. If mitochondrial enzymes are damaged, transport alone will not fix the problem. If the NADH/NAD+ redox system becomes disrupted, oxidative flux can slow. If iron-sulfur centers within mitochondrial proteins are damaged by oxidative or nitrosative stress, electron transport can become impaired. If Complex I, II, III, or IV function is compromised, the cell may have substrate available and still fail to generate normal respiratory flux.

Inflammation can make this considerably worse.

NF-kB activation can induce iNOS, increasing nitric oxide production. In an environment where mitochondrial dysfunction and inflammatory signaling also increase superoxide production, nitric oxide can react with superoxide to generate peroxynitrite. Peroxynitrite is particularly damaging because it can modify proteins, oxidize lipids, disrupt iron-sulfur centers, interfere with mitochondrial respiratory enzymes, and create additional metabolic stress.

The epithelial cell can therefore become trapped from both directions. Its transcriptional identity is changing, reducing expression of proteins and transport systems that characterize a healthy epithelial phenotype, while inflammatory oxidative and nitrosative stress damages the mitochondrial machinery required to use the substrates that still reach the cell.

This is where the microbial ecology becomes important.

A functioning epithelial cell does not simply absorb nutrients. Its metabolism helps shape the biochemical environment surrounding microbes. Mitochondrial oxygen consumption, epithelial redox state, inflammatory signaling, mucus biology, antimicrobial molecules, nutrient transport, and barrier integrity all contribute to microbial selection pressure.

If epithelial respiration declines, oxygen handling near the mucosa changes. If inflammation simultaneously increases iNOS activity, nitrate and other alternative electron acceptors can become more available. Facultative organisms such as E. coli and Klebsiella possess respiratory flexibility that allows them to exploit environmental conditions that many strict anaerobic organisms cannot exploit as efficiently.

This means inflammation can potentially change microbial competition itself. Instead of thinking only about whether bacteria are being killed, we have to think about what metabolic environment is selecting which bacteria are able to expand.

Now the entire feedback loop becomes clearer.

Microbial expansion and LPS increase TLR4 signaling. TLR4, TNF-alpha, IL-1 beta, oxidative stress, NF-kB, AP-1, p38, and related inflammatory pathways increase MMP-9 pressure. MMP-9 contributes to disruption of E-cadherin and extracellular architecture. Beta-catenin becomes dysregulated and can enter the nucleus. TCF and LEF signaling interacts with Snail, Slug, and Twist. The epithelial cell moves toward an EMT-like remodeling program. E-cadherin, ZO-1, occludin, IAP, and potentially transport systems associated with the differentiated epithelial phenotype become impaired. Barrier integrity worsens. LPS detoxification capacity may fall. MCT1 and SMCT1-mediated substrate handling can decline. Butyrate utilization becomes less efficient. Mitochondrial oxidative metabolism weakens. ATP production and epithelial oxygen consumption change. Inflammatory iNOS activity and reactive nitrogen species create further mitochondrial injury. The local ecological environment becomes increasingly favorable to inflammatory facultative pathobionts. Their expansion generates additional microbial products and immune activation, and the cycle begins again.

This is why I think some chronic SIBO cases become extraordinarily difficult to resolve.

The bacteria may have started the inflammatory pressure, or they may have expanded because host physiology was already impaired. But after enough time, cause and consequence can begin reinforcing each other.

At that point, reducing bacterial biomass is only one part of the problem.

Antibiotics can kill susceptible organisms. Herbal antimicrobials can reduce microbial abundance. An elemental diet can dramatically reduce nutrient substrate available for fermentation. These approaches may be valuable and sometimes necessary.

But none of them automatically tells the epithelial cell to start expressing E-cadherin again. They do not automatically restore ZO-1 or occludin. They do not necessarily restore intestinal alkaline phosphatase. They do not automatically increase MCT1. They do not repair a damaged mitochondrial electron transport chain. They do not automatically reverse chronic NF-kB signaling or restore the differentiated metabolic identity of the epithelium.

This is also why I am cautious about reducing the solution to simply “take butyrate.”

If the problem is insufficient microbial butyrate production, providing more substrate may be useful. But if the epithelial cell has reduced MCT1 or SMCT1 expression, the bottleneck may be transport. If transport is intact but mitochondrial oxidative capacity is impaired, the bottleneck is downstream metabolism. If mitochondrial metabolism is functional but inflammatory signaling continues driving EMT-like remodeling, then the cell may continue losing the differentiated program required to maintain normal intestinal function.

The question therefore becomes much more precise. Where is the bottleneck?

Is butyrate not being produced? Is it not being transported? Is it not being oxidized? Is mitochondrial electron transport impaired? Is inflammation continuously suppressing epithelial differentiation? Is NF-kB remaining chronically active because LPS exposure has never been controlled? Has reduced IAP created another reason LPS signaling persists? Are tight junctions still compromised? Is motility allowing microbial biomass and inflammatory products to remain in the small intestine for too long?

This is the central idea behind my Host Capacity Model. When someone with chronic SIBO repeatedly improves after antimicrobials and then relapses, I do not only want to know which organism returned. I want to understand what part of the host never regained the capacity to control that organism.

Was it motility? Gastric acid? Bile? Mucosal immunity? Secretory IgA? IAP? Barrier architecture? E-cadherin? ZO-1 and occludin? MCT1? Mitochondrial metabolism? Oxygen and redox control? Or persistent inflammatory signaling through LPS, TLR4, TNF-alpha, NF-kB, and MMP-9 that continues pushing epithelial cells away from their normal differentiated state?

That is why I increasingly view chronic SIBO as more than a question of what bacteria are present. In some cases, the deeper question may be what the intestinal host has lost the capacity to do.

Killing the bacteria may reduce the microbial burden. But if the biological environment that selected those organisms remains unchanged, the intestine may simply recreate the same ecosystem again.

And that may be one reason some people keep treating SIBO without ever truly changing the conditions that allow it to return.One part of my Host Capacity Model that I have been researching is what happens when chronic intestinal inflammation stops being only an immune problem and begins changing the actual identity and metabolism of the intestinal epithelial cell. This matters because the intestinal lining is not simply a passive wall separating bacteria from the bloodstream. These epithelial cells transport nutrients, maintain tight junctions, communicate with immune cells, regulate antimicrobial defenses, control the chemical environment around microbes, and continuously use mitochondrial metabolism to maintain their structure and function. If that cellular program begins to change, the microbial environment can change with it. One pathway I have been looking closely at involves MMP-9, E-cadherin, beta-catenin, and a process known as epithelial-to-mesenchymal transition, or EMT. Under normal conditions, neighboring intestinal cells are held together by junctional proteins that include E-cadherin. Beta-catenin is associated with these structures and is normally tightly controlled inside the cell. During persistent inflammation, however, MMP-9 activity can increase. MMP-9 is a matrix metalloproteinase capable of degrading extracellular and junctional proteins, and disruption of E-cadherin can release beta-catenin from its normal membrane-associated position. Free beta-catenin is normally controlled by a destruction complex involving APC, Axin, and GSK-3 beta, but when this regulation becomes impaired, beta-catenin can accumulate in the cytoplasm and move into the nucleus.

Once beta-catenin reaches the nucleus, it can interact with transcription factors called TCF and LEF. This is an important transition because the problem is no longer only that the cell is inflamed. The cell can begin changing which genes it expresses. Beta-catenin signaling can participate in activation of transcriptional programs involving Snail, Slug, and Twist, which are involved in epithelial remodeling and EMT-like changes. These programs can suppress normal epithelial characteristics, including proteins such as E-cadherin, ZO-1, and occludin. The cell begins moving away from its normal highly polarized, metabolically specialized intestinal phenotype and toward a more remodeled mesenchymal-like state. This is important because a healthy epithelial cell has a very specific identity. The side facing the intestinal lumen is different from the side facing the bloodstream. It has specialized transporters, tight junctions, nutrient-handling machinery, mitochondrial programs, and signaling systems that allow it to behave like an intestinal epithelial cell. If that identity begins to erode, the cell can also lose some of the machinery that normally helps it regulate its surrounding microbial environment.

This is where the pathway connects with another major part of my Host Capacity Model, which is butyrate utilization. Butyrate is often discussed as if its presence in the intestine automatically produces a benefit, but production and utilization are not the same thing. A metabolite can be present in the lumen while the host cell is poorly equipped to transport or metabolize it. Butyrate first has to cross the epithelial membrane through transport systems that include MCT1 and SMCT1. Once inside the cell, it can be converted into acetyl-CoA and enter mitochondrial oxidative metabolism. Acetyl-CoA feeds into the TCA cycle, which produces reducing equivalents that deliver electrons to the electron transport chain. The mitochondria then use that electron flow to produce ATP while consuming oxygen. So the biologically relevant sequence is not simply bacteria produce butyrate and the gut becomes healthy. It is butyrate production, transport through MCT1 and SMCT1, intracellular availability, conversion into acetyl-CoA, TCA-cycle metabolism, electron transport, ATP generation, and oxygen consumption by the epithelial cell. If the epithelial cell begins losing expression of MCT1 and SMCT1 during abnormal remodeling, then butyrate may still be present next to the cell while the cell becomes progressively less capable of using it. The material I have been studying specifically describes loss of these transport systems as part of this epithelial identity shift.

That distinction has major implications. If MCT1 and SMCT1 expression falls, less butyrate may enter the epithelial cell. If intracellular substrate availability falls, mitochondrial oxidation can decline. Reduced oxidative metabolism can mean less TCA-cycle flux, less electron transport, less ATP generation, and less oxygen consumption by the epithelial tissue. This matters because epithelial metabolism is part of microbial ecology. The epithelial cell is not simply keeping bacteria physically separated from the bloodstream. Its metabolism influences oxygen tension, redox conditions, nutrient availability, barrier integrity, inflammatory signaling, and the ecological pressures that determine which microorganisms gain a competitive advantage. If mitochondrial respiration falls, the tissue consumes less oxygen and the local redox environment can shift. The material I have been examining connects loss of epithelial transporter expression and metabolic function with reduced oxygen consumption and an environment that may increasingly favor organisms such as E. coli and Klebsiella. Inflammation can compound this problem. Increased iNOS activity can raise nitric oxide production, while nitric oxide reacting with superoxide can generate peroxynitrite. These reactive nitrogen species can damage mitochondrial proteins, iron-sulfur centers, respiratory enzymes, and membranes, further reducing oxidative capacity. At that point, the host can become trapped in a self-reinforcing loop in which microbial overgrowth promotes inflammation, inflammation increases MMP-9 activity, MMP-9 contributes to E-cadherin disruption, beta-catenin signaling changes, Snail, Slug, and Twist drive epithelial remodeling, MCT1 and SMCT1 expression falls, butyrate utilization becomes less efficient, mitochondrial respiration weakens, oxygen handling changes, barrier proteins decline, and the local environment increasingly favors organisms adapted to inflammatory and metabolically disturbed conditions. Those organisms then generate more microbial products and inflammatory pressure, feeding the cycle again.

This is why I do not think chronic SIBO can always be understood as simply having too many bacteria that need to be killed. Antibiotics, herbal antimicrobials, or elemental diets may reduce microbial biomass and may improve symptoms, sometimes significantly, but they do not automatically restore epithelial differentiation, MCT1 expression, butyrate transport, mitochondrial oxidation, E-cadherin, ZO-1, occludin, mucosal immune function, or the metabolic state of the tissue. This is also why simply giving more butyrate may not always address the real bottleneck. If the problem is reduced microbial production, increasing butyrate availability may help. But if the limiting factor is transport, then more butyrate in the lumen does not necessarily restore MCT1 or SMCT1. If the limiting factor is mitochondrial oxidation, then getting more butyrate into the cell may still not restore normal metabolism. And if chronic inflammation has shifted the transcriptional identity of the epithelial cell itself, the deeper problem may be restoring the cellular program that allows the tissue to transport, metabolize, and respond to these substrates normally.

That is one of the central ideas behind my Host Capacity Model. When someone repeatedly relapses after antimicrobial treatment, I do not only want to know which bacteria came back. I want to know what part of the host never recovered. Was intestinal clearance still impaired? Was mucosal immunity weak? Was bile signaling abnormal? Was barrier integrity still compromised? Was mitochondrial metabolism dysfunctional? Was transporter expression reduced? Could the epithelial cell still properly utilize metabolites such as butyrate? Was chronic inflammation continuing to push the tissue toward an abnormal remodeling state? Chronic SIBO may therefore be more than a problem of what is growing in the intestine. In some cases, the deeper problem may be what the host has lost the capacity to do. Until that lost capacity is identified and restored, killing bacteria may reduce the consequence while leaving the biological environment that keeps recreating the problem largely unchanged.