Australasian Science Magazine

The liver’s unusual cellular architecture makes blood flow slower than in other organs. Perforations of the lining of the blood vessels within the liver allow a unique contact between the circulating blood cells (red) and the liver cells (green), and explains why activated T cells (blue) are efficiently retained in the liver. Credit: Frederic Sierro

By Patrick Bertolino and David Bowen

Recent discoveries about the unusual behaviour of immune cells in the liver could open the way to new strategies for transplantation and the treatment of viral hepatitis.

The relationship between the liver and the immune system is unique. During pregnancy, the foetal liver produces the baby’s first blood cells. After birth, the liver hands over this role to the bone marrow but continues to maintain an unusual relationship with the immune system.

In fact, the liver is the only place outside of the lymph nodes where white blood cells, or T cells that have not seen foreign material before, can be retained and activated. Strangely, though, T cells activated in the liver do not go out and kill foreign invaders of the body – they are actually tolerant of them. To ensure this, the liver destroys most of the T cells it has activated.

We have recently found how it goes about doing this. Not only does the process explain several puzzling observations of the behaviour of the liver, it could also lead to new treatments for a variety of conditions, including viral hepatitis.

The liver is the largest internal organ of the body and has always been considered exceptional. The ancient Greeks seemed to have a feel for its properties: in Greek mythology Prometheus, a Titan who irritated Zeus by stealing fire and giving it to humans, was punished by being chained to a rock where his liver was eaten away each day by an eagle, only for it to regenerate to its original size every night.

This ancient story beautifully illustrates a real scientific observation. A liver that has had two-thirds of its mass removed can restore the missing tissue in less than 2 weeks.

In addition to its remarkable regenerative capacity, the liver is the central factory of the body. It performs a large range of the biochemical functions that keep the body ticking over, including detoxification, the production of lymph and plasma proteins, and the synthesis of hormones and coagulation factors.

Put simply, we cannot live without our livers.

Although these regenerative and operational properties of the liver are well documented, its relationship with the immune system has been neglected. This is surprising as several scientific observations indicate that it is unusual. In particular, the liver is able to calm the immune system down and induce a state of unresponsiveness to foreign bodies.This is known by immunologists as “tolerance”.

Why the liver has developed this capacity is still a matter of speculation. As it receives and filters blood directly from the intestine, one suggestion is that the liver is strategically located to process ingested proteins from the gut and induce tolerance to proteins in food. This hypothesis has been supported by experiments showing an immune response against food when blood originating from the gut was surgically diverted to bypass the liver.

But liver-induced tolerance has also been observed in different settings, one of the most important being in transplantation. The first evidence suggesting that the liver influences the immune system in transplantation comes from work by Sir Roy Calne, who reported in 1969 that liver transplants were accepted in outbred pigs without the need for immuno­suppressive drugs. This observation, confirmed in several other species including mice and rats, is unusual.

Most transplanted organs require ongoing administration of immunosuppressive drugs to prevent rejection, but liver grafts seem to be an exception to this rule, and are more likely to be accepted spontaneously. Consistent with the observation in animals, human liver transplants also require less immunosuppression than other organs, and about 15% of patients can successfully be weaned off immuno­suppressive therapy – a higher percentage than for any other transplant.

A decade after the pioneering observation by Calne it was shown that recipients of liver grafts also accepted skin or heart transplants from the same donor yet rejected them from others. Liver transplants were also able to reverse severe ongoing rejection of earlier organ transplants from the same donor, including heart, pancreas and skin transplants. These experiments re­inforced the view that liver grafts were not ignored by the immune system, but rather induced a donor-specific tolerance.

Several liver-specific viruses, such as hepatitis C virus, cause persistent infections. Despite producing detectable virus-specific T cells, 80% of patients infected with hepatitis C virus fail to eradicate them and become chronically infected. Thus it is tempting to speculate that the hepatitis C virus uses the properties of the liver to trick the immune response and persist in the host.

Recent evidence suggests that the liver, unlike other organs in the body, is not a passive player in the immune response. The liver tissue’s unusual architecture makes blood flow slower than in other organs, and the lining of the blood vessels within the liver contains perforations that are not present in any other organs. This allows a unique contact between the circulating blood cells and the liver cells, and explains why activated T cells are efficiently retained in the liver.

The slow and intermittent blood flow in the liver also alters the circulation of T cells that have not yet been activated. Naïve T cells normally circulate via the blood and lymph nodes, but do not produce the special molecules required for adhering to and penetrating blood vessel walls, so they cannot enter the tissue of a solid organ. In the lymph nodes, naïve T cells scan the surface of specialised cells known as dendritic cells, which present fragments of foreign proteins that they have captured from tissues.

Any T cell that recognises and can link to a foreign fragment becomes activated and programmed to undergo massive proliferation. At the same time, these activated T cells develop the ability to either kill cells carrying the foreign fragment, or to secrete several molecules that assist the immune response.

During activation in the lymph nodes, T cells are also programmed to make the molecules needed for them to enter organ tissues. This is the main reason why tissue access is normally restricted to cells that have undergone activation in lymphoid tissues.

About 10 years ago, our research group demonstrated that the liver is an exception to this. Naïve T cells circulating via the blood were retained and activated within the liver by hepatic cells.

The activation of naïve T cells outside the lymph tissues is likely to be due to the unusual structure and blood flow in the liver, which allows T cells to adhere to the blood vessel walls despite not having specialised adhesion molecules. However, the T cells activated in the liver receive different programming than those activated in lymph tissues. While CD8 T cells activated in the lymph nodes expand and become efficient killers, T cells activated in the liver are poor effectors and do not accumulate. In contrast to T cells activated in lymph nodes, the immune response initiated in the liver is associated with tolerance.

The liver cells responsible for programming naïve T cells do not provide the molecules required to induce a strong and persisting T cell response. Although the liver does contain some resident dendritic cells, they are largely immature, and unable to prime such an immune response.

Not only do T cells activated in the liver fail to become efficient killer cells, our studies have also demonstrated in mice that after 2 days they show high levels of Bim, a molecule that promotes programmed cell suicide. As a result of Bim activity, naïve T cells activated in the liver prematurely die and are therefore unable to accumulate and contribute to an effective immune response.

Although Bim-dependent T cell death certainly contributes to liver-induced tolerance, it does not explain all the available data. We have always observed an unexplained 80–90% drop in the total number of liver-induced T cells during the first 24 hours after their activation. This cell loss occurs before Bim can be made by T cells, and is even observed if the T cells are altered so they cannot make Bim – suggesting that the early cell loss is Bim-independent. In fact, most T cells disappear in the first 8 hours, and for this reason have been missed in previous studies.

In a recent article published in the Proceedings of the National Academy of Sciences we explored why this loss after activation occurs. Our studies led us to the unexpected discovery that T cells commit mass suicide. They actively invade the main liver cells and are destroyed within lysosomes, membrane-bounded sacs that degrade molecules, bacteria and cellular debris (Fig. 1). This mechanism is rapid, efficient and able to kill off very large numbers of T cells.

Figure 1. Schematic diagram of a liver sinusoid illustrating the fate of T cell in hepatocytes. Naïve CD8 T cells activated in the liver cross the vascular barrier and invade hepatocytes. The endocytic vesicle containing the T cell fuses with lysosomes, leading to the rapid degradation of the T cell.

We believe that this novel mechanism is used by the liver to remove unwanted T cells very early in the immune response, before they start to proliferate and at a time when they have not yet acquired the ability to kill cells, thus limiting any tissue damage following their activation. For reasons that remain unclear, 10–20% of T cells survive this degradation and begin to divide. Although these T cells expand, they do not become efficient killer cells. They appear to be the liver-activated T cells that die in a Bim-dependent manner after a few days.

The liver thus seems to have evolved two different and effective pathways to dispose of potentially harmful T cells after their activation: a predominant rapid mechanism in which liver cells cannibalise CD8 T cells, and a backup mechanism that mops up the remaining cells via programmed cell death (Fig. 2).

Figure 2. Model illustrating the different fates of naïve CD8 T cell activated in liver and lymph nodes. (B) Activation of naïve CD8 T cells in the liver leads to tolerance as a result of degradation in hepatocytes (suicidal emperipolesis) and Bim-dependent apoptosis of remaining cells. (A) But in the absence of cell cannibalism in lymph nodes, T cells activated by dendritic cells (DC) expand and acquire a different program leading to their survival and development into cell killers. Cytotoxic T cells (CTLs) generated in lymph nodes are thus programmed to kill rather than invade and die within antigen-expressing hepatocytes.

This intriguing mechanism of cannibalism of T cells probably explains many of the unusual immunological properties of liver tissue, and opens up new avenues for therapies against infections by liver viruses and to assist transplantation.

If this process holds the key to the liver’s ability to induce tolerance, we predict that drugs that block it will promote the normal functioning of T cells in the liver. So, in patients infected with the hepatitis C virus, these drugs would promote the survival of virus-specific T cells that would normally be eliminated, and the liver would no longer be a sanctuary for the virus. In transplantation, on the other hand, it would be beneficial to treat organ recipients with drugs that promote this process and eliminate any T cells responsible for rejection.

Liver immunology is a relatively new discipline. The emerging role of the liver in the immune system suggests that we are just at the beginning of a new era that will further unravel the mysteries of this fascinating organ.