User:Scain1990/sandbox

Thank you for coming today for my first doctoral exam. I am interested in the role of neuroinflammation in contributing to the memory loss that is seen following alcohol consumption.

Alcohol has been rated as the most harmful drug of abuse by the NIAAA, surpassing every other drug, including heroin and cocaine, in terms of both harm to self and harm to others.

Alcohol affects the body in many ways, even in low doses. At low and moderate levels of consumption, the immune system is weakened, leading to greater risk of infection and illness. Effects on the circulatory system are well documented, ranging from damage to the heart, increased blood pressure, and increased risk of stroke. The NIAAA has also documented that drinkers are at increased risk for cancers of the mouth, esophagus, stomach and liver. In long-term heavy drinkers, liver damage can result.

More specifically, we are interested in alcohol on the central nervous system. Alcohol exhibits a strong effect on the CNS, and the high vasculature to the brain leads to high concentrations of alcohol within the CNS as compared to other tissues such as muscle (Eckhardt et al., 1998). It exhibits an effect on many neurotransmitter systems including the serotonergic, dopaminergic, glutamatergic and GABA-ergic systems, which result in its anxiolytic, sedative, and addictive properties. Altered activity of serotonin receptors and dopamine receptors leads to sedative and addictive properties commonly associated with alcohol. Addiction research has been conducted through the use of operant conditioning and conditioned place preference tasks, and has shown that animals which initially find alcohol to be aversive, can be conditioned to freely consume alcohol and sucrose and will even consume solutions with greater concentrations of ethanol over time. Within memory processes, alcohol has been shown to act as an antagonist to the NMDA receptor, which is crucial in the formation of new memories through its role in mediating synaptic plasticity. Studies on the GABA-ergic receptors show that alcohol leads to increased release of GABA and greater activation of the GABAa receptor. This receptor is a gated ion channel for chloride and leads to hyperpolarization and decreased excitatory post-synaptic potentials in neurons of the hippocampus.

Memory deficits have been seen in human participants. In a study by Jones, subjects received either alcohol in orange juice or orange juice with a thin film of alcohol on top of it. In this way, they were unaware of whether they were part of the alcohol group or the control group based on taste. They were then asked to memorize words which were flashed on the screen and report them at a later time, either immediately or slightly later. This allowed the experimenter to separate immediate and short term memory effects of alcohol. As the figure shows, subjects displayed similar word recall at the baseline, prior to receiving alcohol, but both immediate recall and short-term memory recall were impaired in the alcohol group. A test of long-term memory conducted 24 hours later showed that subjects had no difficulty recalling words learned before alcohol was administered but had impaired recollection of words following exposure to ethanol. This seems to suggest that alcohol impairs the acquisition of new memory but not the consolidation of previously learned material. The general premise that arises from human work with ethanol suggests that alcohol does not affect the pathway from sensory input into short-term memory but selectively effects the encoding of new memories from short-term into long-term memory.

In animal subjects, memory can be tested in a variety of different ways. Some key examples are the Morris Water Maze and fear conditioning, both of which are used by us, as well as the radial arm maze and Barne maze, which are not.

The Morris Water Maze task is a commonly used test for hippocampal, spatial memory. In the Morris Water Maze task, a 1.8 meter diameter pool is filled with water mixed with white paint and a platform is submerged within it so that the water is 1cm above the platform. The animal is then trained from different start locations over the course of a week, swimming within the pool until either they find the platform or 45 seconds have passed, at which point they are moved to the platform. In naïve animals, they generally swim around the outside of the pool but as they go through more sessions, their performance improves and they are able to find the platform quicker based on spatial cues placed around the pool. The platform location stays constant in between sessions and the location is unchanged for the entire test period. In vehicle/control animals, they can navigate from any start location to rapidly find the location of the platform within the pool, showing evidence of learning between sessions. Ethanol-treated animals do not show improvement however. With further training, they still show difficulty finding the platform within 45 seconds, even after a week of sessions.

As this figure, taken from earlier work conducted by our lab, shows, animals treated with saline show a great deal of improvement over the course of the test days while the ethanol treated animals show significantly slower swim speeds and do not show any improvement. This gives evidence of either a motor dysfunction or a loss of memory so a further test is required in order to confirm that memory is affected.

On the final day of the Morris Water Maze task, the platform is removed from the pool and the animals again have 45 seconds to swim in the pool. What we see is that the saline treated animals spend a vast majority of their time within the quadrant where the platform was originally located compared to the other three quadrants of the pool. The ethanol treated animals spend far less time in the target quadrant and spend close to even time in all of the quadrants. This provides evidence that the impairment is in memory as it shows clearly that the ethanol treated animals spend less time in the target area and more time in the other three zones of the pool.

Even more evidence for the detriment to memory can be seen by recording the number of times the animal passed through the location where the platform was previously placed. Vehicle animals spend more time in the target quadrant and also pass through the platform location significantly more often than animals that are treated with alcohol. This further indicates that alcohol compromises memory within the Morris Water Maze task.

In order to determine whether the deficit to memory is a global memory deficit or if it is localized to a specific area of the brain, we use a fear conditioning paradigm. In this task, animals are acquisitioned to a chamber on day zero. On the training day, they receive a dose of alcohol prior to being placed in the chamber. After a period of time, a tone is played within the chamber which co-terminates with a 1mA footshock. On the last day of the procedure, the animal is placed again within the chamber in one of two scenarios for a test. The amount of time that the subject spends frozen is considered an accurate measure for fear. In the contextual task, the chamber remains completely unchanged from the training day, and the animal is simply left in there for a period of time. A cued task is also done in which the chamber’s appearance is changed. The animal then experiences the tone again and freezing is recorded. The key detail here is that the contextual task is dependent on the hippocampus while the cued task is dependent on the amygdala. With this task, we have the potential to determine whether memory in general is impaired or whether the impairment is only seen in one type of memory process.

Preliminary results from an experiment conducted on post-pubescent animals treated with either alcohol or deionized water show contextual data and two cued data types: the no-tone data is used to show that the animals exhibit no increased fear to the chamber following the change in context. As you can see, this is valid. Freezing is not significantly different between the alcohol-treated and vehicle-treated animals and both groups exhibit very low levels of freezing when exposed to the new chamber. The contextual data indicates that the vehicle animals freeze when re-introduced to the previous context but freezing is impaired in the alcohol-treated animals. This indicates that alcohol leads to an impairment in the acquisition of new hippocampal memory, which was also seen in the Morris Water Maze data I previously showed. When the tone was presented to both groups of animals, no significant difference in freezing was seen and both groups displayed a high level of freezing. This indicates that fear memory is conserved in these animals, suggesting that the amygdala is unaffected by alcohol treatment. With the information provided by the Morris Water Maze and the information provided by fear conditioning, we conclude that hippocampus dependent memory is negatively affected while amygdalar memory is preserved. Furthermore, this pair of experiments eliminates the possibility that a global memory deficit is seen following treatment with ethanol.

The general theory for how alcohol is working in the hippocampus is through some pattern of decreased excitation and increased inhibition. The hippocampus is networked with excitatory glutamatergic neurons and previous work has shown that alcohol acts in an atagonistic way and blocks the excitation of neurons within the hippocampus. Likewise, prior research has suggested that alcohol increases the release of GABA from inhibitory interneurons, leading to greater chloride influx and reduced excitatory postsynaptic potentials. This comes together to create a system that is less able to generate the activation which leads to the formation of new memories and thus leads to a loss in memory formation within the hippocampus.

One possible mechanism, and what I am interested in exploring, is neuroinflammation. Basically, neuroinflammation is used to protect neurons from harm presented by infection, or damaging stimuli. It is activated via the immune system within the central nervous system, made up of astrocytes, microglia and mast cells. Research looking into neuroinflammation has shown that microglial cells produce reactive oxygen species designed to cause cell damage to potentially harmful, foreign substances but also contribute to damage caused to neural tissue (Need to reread Skaper et al.). The obvious issue with immunity between the outside system and the central nervous system is the blood-brain barrier. This layer prevents most outside materials from entering into the central nervous system in an attempt to prevent infection. This includes cells that make up the immune system. However, Shubayev and Myers used an anterograde tracer for tumor necrosis factor alpha, a key inflammatory molecule, to identify positions within the nervous system where TNF-alpha appeared following treatment to the periphery with lipopolysaccharide, an infectious agent and piece of the cell membrane of bacterial cells. They determined that TNF-alpha had the ability to travel along peripheral neurons in a post- to pre-synaptic pattern, and most importantly, could be identified in neurons of the central nervous system. This seems to indicate that TNF-alpha at the very least has the potential to circumvent the blood-brain barrier and enter the central nervous system through peripheral neural pathways. Alongside that, other research has found that stimulation of the peripheral immune system can lead to neuroinflammation. In this study, animals treated with lipopolysaccharides were sacrificed and immune factors found within the brain showed a significant increase over baseline values. This suggests a link between the peripheral immune system and the immune system of the CNS. The exact mechanism of this type of activation has not been determined at this point although some research has shown promising information pinpointing certain cytokines as being able to pass through the blood-brain barrier.

Cytokines are a key part of the overall immune response. These are small proteins which act to trigger signaling cascades, leading to changes in immune cells throughout the body. They come in two distinct types: pro-inflammatory which encourage inflammation and anti-inflammatory, which work to bring an inflammatory response to an end within the system. Proinflammatory cytokines work by increasing blood flow and the number of immune cells such as T cells within an area of infection, and by producing biochemical products such as reactive oxygen species which further increase the proinflammatory response. Anti-inflammatory cytokines work to reverse these changes when the body is no longer at risk by blocking the generation of new cytokines, eliminating products generated through cytokinergic activation, and signaling deactivation of cells in the immune system.

In a healthy individual, these two types of cytokines exist in equilibrium with each other. For our purposes, we are primarily interested currently in IL-1beta, TNF-alpha, and IL-6 on the pro-inflammatory side and IL-4 and IL-10 on the anti-inflammatory side.

Along with that differentiation, it is also possible to separate cytokines into distinct families. In most cases, such as with the interleukin 1 family, the function of the members is generally pro-inflammatory. These cytokines work by activating the IRAK family of proteins, leading to increased synthesis and creation of additional pro-inflammatory cytokines. Along with increasing the concentration of cytokines, it also increases the density of receptors for other pro-inflammatory cytokines.

The hematopoietin family is kind of a “grab-bag” of cytokines, ranging from anti-inflammatory IL-4 to pro-inflammatory IL-6. In this case, the family is defined by the conserved appearance of the receptor family that responds to these cytokines. Primarily, these cytokines work by altering the function of T helper cells in the immune system. Basically, IL-6 plays a crucial role in activating these cells, leading to a heightened immune response in affected tissue. IL-4 on the other hand does the opposite, and leads to the deactivation of T helper cells and a general decrease in the level of inflammation.

The final important family of cytokines we are interested in are the tumor necrosis factor family. These are in charge of the cell death/life signaling cascades that determine whether apoptosis should occur. When foreign materials enter the body, these cytokines are activated through their own signaling pathway, the JAK/STAT pathway and are secreted from the macrophages and other cells within the immune system. Cells that go through TNF-induced apoptosis are lysed into small vesicles and then digested by the macrophages in order to be completely broken down. As indicated in the name, these cytokines are also important in the destruction of tumors and cancerous growths by increasing the blood supply and immune response to areas where tumor growth have been detected.

So overall, what you see in the event of inflammation is something like this. The pro-inflammatory cytokines are up-regulated and lead to increased inflammation to targeted tissue and the anti-inflammatory cytokines are down-regulated. This in itself is a natural process and exists in a series of checks and balances. An increased pro-inflammatory cascade generally also leads to an increase in anti-inflammatory cytokines. As the number of anti-inflammatory cytokines increases, it can begin to rein in the inflammatory response and eventually bring it to an end. The problem is when the inflammation caused by cytokines lasts for prolonged periods of time. In cases like this, it can become increasingly systemic and eventually may even affect the brain and central nervous system.

That being said, cytokines do occur naturally within the central nervous system and, in fact, some are even produced as part of general processes such as the generation of long-term potentiation. Kipnis et al., looked at cognition in animals lacking T cells within the hippocampus and found that their absence led to decreased cognition and altered homeostatic conditions. They used the Morris Water Maze task akin to what we have used here and saw a similar trend in performance. As with us, vehicle animals were able to learn rapidly to escape the pool to the platform while the T cell deprived animals showed no evidence of learning. IL-4 in particular also effects other pathways which have been implicated in synaptic plasticity and learning/cognition, including Nerve Growth Factor and Brain Derived Neurotrophic Factor. IL-1 family cytokines are also naturally occurring and are actually produced naturally in long-term potentiation within the hippocampus. What is really interesting here is that deficiencies and over-production of IL-1 both led to memory deficits. They examined memory versus controls in animals that under-expressed IL-1 cytokines and animals that over-expressed them and found similar deficits to memory in the contextual part of a fear conditioning task. This and the work on IL-4 indicate that the important part for these proteins is that homeostasis is conserved. Over and under activation can both be detrimental to overall brain function.

In general, we are interested in a few mechanisms of potential activation by cytokines within the brain.

Long-term potentiation is generated as a piece of the learning process. In brain sections, stimulation of neurons with rapid bursts of stimulation will lead to greater excitatory postsynaptic potential and greater ease of generating action potentials. This increased activity is associated with memory and the generation of patterns of activation. Cytokines play a significant role in modulating LTP. Increased IL-1beta has been found to decrease the propagation of LTP within the hippocampus. With IL-1beta, a key receptor pairing: MyD88 and IL-1RI occur in tandem with the NMDA glutamate receptor. This receptor is crucial for the formation short and long, long-term potentiation by increasing the release of neurotransmitters in short-lasting LTP and by causing changes in cellular mechanisms and synaptic plasticity in longer-lasting LTP. Bellinger et al conducted an experiment that showed that IL-6, as with IL-1beta, leads to an inhibition in LTP generation. Both also can lead to decreased duration of LTP generation, potentially suggesting another mechanism underlying memory deficits.

Changes in the glutamate system have also been seen following alterations to the cytokine milieu. TNF-alpha has been found to lead to consistent overactivation of glutamate receptors. This can lead to neurotoxicity as it overworks the cells and alters the electric properties of their cell membrane. This neurotoxicity can be increased when reactive oxygen species are added to the equation. In this experiment by Zou and Crews, brain sections prepared from animal subjects were exposed to a variety of substances. They found that ethanol in tandem with glutamate, hydrogen peroxide and TNF-alpha, led to the highest degree of cell death. The microscopy picture indicates that this combination caused the greatest degree of cell injury and death and the statistics show a significant degree of cell death compared to control samples.

The third primary process of damage to neural tissue resulting from neuroinflammation is through the production of reactive oxygen species. As stated before, these have been shown to lead to damage and reactive stress on afflicted cells. IL-1beta in particular leads to the creation of ROS by activating the enzyme superoxide dismutase. In a study by Kelly et al, they wanted to determine what role IL-10 played in remediating the effects triggered by ROS. They first found that injection of IL-1beta and IL-10 led to control levels of EPSP production. They continued this experiment to determine what exactly IL-10 was doing to prevent changes in EPSP generation. To test whether ROS was being affected by the IL-10, they injected animals with hydrogen peroxide. Hydrogen peroxide in the body will naturally produce ROS. What they initially found was that animals treated with H2O2 showed lower EPSPs over time. Animals treated with IL-10 and H2O2 however did not show diminished EPSP. This seems to suggest a role for IL-10 in eliminating memory deficits that are brought on by exposure to ROS.

Basically, these phenomena paint a very broad picture of the ways in which cytokines negatively affect memory processes within the hippocampus. Unfortunately, not much research has been conducted so far to determine whether cytokine activation brought on by alcohol use is to blame for alcohol induced memory deficits. What we do have is research that has implicated cytokine activation in other disease processes which leads to memory deficits.

Fonken et al were interested in the effect that fine particulate matter such as that found in air pollution has on memory processes. Particulate matter has been previously found to increase levels of pro-inflammatory cytokines within the brain. They used a Barnes Maze in which animals are placed in a brightly lit box and are required to locate a hole which leads to a dark, “safety box”. Other holes are present throughout the maze, but only the one leads to the escape target. Following five days of training trials, a probe trial was conducted and the number of times the subject went back to the location where the safety zone had been previously was recorded. What they found was that latency to locate the safety box was stunted following exposure to the particulate matter. They also exhibited a greater number of errors prior to correctly locating the safety box and show poor performance within a probe trial. Biochemically, animals that had been exposed to the particulate matter also showed greater levels of TNF-alpha and IL-1beta within the hippocampus. This indicates that exposure to airborne particulate matter led to an increase in cytokine activation and led to hippocampal inflammation.

Ischemic stroke represents another pathway through which cytokines are activated. In victims of stroke, blood loss to areas of the brain can set off a chain of events leading to excitotoxicity and cell death as neurons are starved of oxygen. This can result in memory loss and changes in cognition which can last for a prolonged period of time. Zhai et al triggered ischemic stroke by blocking the blood flow to one hemisphere of the brain in anesthetized animals. This allowed them to potentially compare the ischemic and non-ischemic hemispheres of the brain within the same animal. They then used RT-PCR to examine the expression of mRNA on both sides of the brain following this procedure and continued to analyze samples at regular time points following alleviation of ischemia. What they found was a localized increase in TNF-alpha and IL-1beta expressing mRNA starting at a very early time point in the experiment and building up to about 6 hours following stroke. At the six hour mark, they also indicate an increase in mRNA expression for IL-10. They suggest that this indicates that an upsurge in pro-inflammatory cytokine activation is followed by an increase in anti-inflammatory activation, signaling the end of neuroinflammation and a return to equilibrium. This research together shows that increased pro-inflammatory factors are associated with the beginning and continued damage following ischemic stroke and that an upswing in expression of anti-inflammatory factors is associated with the return to normal.

Alzheimer’s Disease is also characterized by cytokine activation and severe deficits in memory formation and retrieval. The characteristic feature of this disorder is the production of Amyloid Beta plaques. These are found within the brain of people suffering from Alzheimer’s extracellularly and are associated with memory deficits in these people. Eikelenboom et al found that stimulating the amyloid beta plaques with immune factors led to the production of cytokines. These included IL-1beta, TNF-alpha and IL-6, suggesting that these cytokines may play a role in the memory deficits that have been seen in this disease. Importantly, all three of these cytokines have been seen to have increased activity following exposure to ethanol, further indicating that they may be critical in the memory deficits seen following its consumption.

The last disorder to look at is delirium. This is similar to Alzheimer’s but differs in the rapidity of its onset. Unlike with Alzheimer’s, where detriments can slowly amount over the course of years, delirium is characterized by a rapid detriment in memory and cognitive processes. It is believed to result from systemic inflammation and has been widely explored for its post-operative significance seen as delirium following surgical procedures. In the experiment conducted by Murray et al, they injected animals with either ME7, a prion responsible for causing early stage neurodegeneration and causing priming of microglial cells in the hippocampus. Behaviorally, they found that animals which had received injections of ME7 and LPS performed worse in a T-maze task, suggesting that neuroimmune response led to memory deficits related to ME7. Supporting this, they found that animals that had received ME7 and saline showed no significant difference versus animals that had received saline or LPS injections along with normal brain homogenate. Examining cytokine levels, they found that IL-1beta and TNF-alpha were both increased from baseline values in animals treated with LPS although animals treated with ME7 actually showed lower levels of these two factors at one hour following injection.

So now the question that needs to be answered is how this relates to alcohol. I have indicated mechanisms which underlie cytokine based inflammatory effects within the brain and gave a number of examples of diseases which have an increase in cytokine activation connected to them. What research in the last few decades has shown is that cytokine activation and immune activity are altered based on even low-level exposure to alcohol. It is well established that alcohol use and especially abuse results in liver damage and cirrhosis of the liver. Inflammation has also been documented in the heart and lungs following exposure to alcohol. Long-term alcoholics are even at a higher risk for pneumonia due to alterations in the immune system that result from chronic binge drinking. As I have stated previously, IL-1beta, IL-6 and TNF-alpha appear to be major pro-inflammatory cytokines responsible for the inflammation that is seen throughout the system following consumption of alcohol. IL-4 and IL-10 act to eliminate the effects of inflammation from the system following alcohol exposure but are reduced after consumption.

So what we know is that alcohol leads to inflammation within the body, which is responsible for many of the health problems associated with it. Increased inflammation leads to increased cytokines and inflammation within the brain, and following alcohol consumption that is what we see. Neuroinflammation has the potential to cause long-term changes in brain structure and function, potentially resulting in the long-term deficits that can be seen following binge alcohol consumption. The long-term inflammation is further supported by the continued high concentration of cytokines that we see even long after the last consumption of alcohol. These most likely result from the poor filtration out of the central nervous system and the lack of disposal cells designed to rein in inflammatory cytokines.

To conclude, I have shown that alcohol leads to memory deficits following its use which are limited to the hippocampus. Cytokines of the immune system are up-regulated following consumption of alcohol and increased levels within the body have been shown to lead to a number of disease processes such as stroke, pneumonia, various cancers, and a multitude of liver conditions. A peripheral rise in cytokine concentrations has been found to lead to an increase in concentrations within the central nervous system, through a mechanism that is not yet fully understood. Within other disease processes, such as Alzheimer’s, delirium, stroke and following exposure to air pollution, cytokine levels are increased in the hippocampus, resulting in memory deficits to spatial and contextual memory processes. Similar increases are seen in the brain following consumption of alcohol, but whether this is the specific cause of memory loss has not been determined at this time. So we conclude that neuroinflammation may result in memory loss following ethanol exposure but the exact role is not fully understood at this time.

I’d like to thank City College and the Graduate Center, my mentor, Ratna Sircar, my lab mates, many of whom are in the audience today and of course my committee for meeting with me today for my examination. I will open up the floor to any questions.