Alcohol and the hippocampus
       More than 30 years ago, both Ryback (1970) and Goodwin and colleagues (1969a) speculated that alcohol might impair memory formation by disrupting activity in the hippocampus. This speculation was based on the observation that acute alcohol exposure (in humans) produces a syndrome of memory impairments similar in many ways to the impairments produced by hippocampal damage. Specifically, both acute alcohol exposure and hippocampal damage impair the ability to form new long-term, explicit memories but do not affect short-term memory storage or, in general, the recall of information from long-term storage.

Location of the hippocampus in the human brain
(This image was borrowed from www.morphonix.com, a site offering multimedia software aimed at educating children and adults about the brain)

       Research conducted in the past few decades using animal models supports the hypothesis that alcohol impairs memory formation, at least in part, by disrupting activity in the hippocampus (for a review, see White et al. 2000b). Such research has included behavioral observation, examination of slices of brain tissue, neurons in cell culture, brain activity in anesthetized or freely behaving animals, and a variety of pharmacological techniques.
       As mentioned above, damage limited to the CA1 region of the hippocampus dramatically disrupts the ability to form new explicit memories (Zola-Morgan et al. 1986). In rodents, the actions of CA1 pyramidal cells have striking behavioral correlates. Each cell tends to discharge action potentials (events that result in one cell communicating with other cells) when the rodent is in a distinct location in its environment. The location differs for each cell. For instance, while a rat searches for food on a plus-shaped maze, one pyramidal cell might generate action potentials primarily when the rat is at the far end of the north arm, while another might generate action potentials primarily when the rat is in the middle of the south arm, and so on.  Collectively, the cells that are active in that particular environment form a spatial, or contextual, framework within which memories for events are formed.  Because of the location-specific firing of these cells, they are often referred to as "place-cells," and the regions of the environment in which they fire are referred to as "place-fields" (for reviews, see Best and White 1998; Best et al. 2001). Given that CA1 pyramidal cells are critically important to the formation of memories for facts and events, and the clear behavioral correlates of their activity in rodents, it is possible to assess the impact of alcohol on hippocampal output in an intact, fully functional brain by studying these cells.
       In recent work with awake, freely behaving rats, White and Best (2000) showed that alcohol profoundly suppresses the activity of pyramidal cells in region CA1. The researchers allowed the rats to forage for food for 15 minutes in a symmetric, Y-shaped maze and measured the animals' hippocampal activity using microelectrodes (tiny wires) implanted in their brains. Figure 3 displays the activity of an individual CA1 pyramidal cell. The activity--which corresponds to the middle portion of the lower left arm of the maze--is shown before alcohol administration (A), 45 to 60 minutes after alcohol administration (B), and 7 hours after alcohol administration (C). The dose of alcohol used in the testing session was 1.5 grams per kilogram-enough to produce a peak BAC of about 0.16%, twice the legal driving limit (for humans) in most States. As the figure illustrates, the cell's activity was essentially shut off by alcohol. Neural activity returned to near-normal levels within about 7 hours of alcohol administration.
       White and Best administered several doses of alcohol in this study, ranging from 0.5 g/kg to 1.5 g/kg.(Only one of the experiments is represented in figure 3.)They found that the dose affected the degree of pyramidal cell suppression. Although 0.5 g/kg did not produce a significant change in the firing of hippocampal pyramidal cells, 1.0 and 1.5 g/kg produced significant suppression of firing during a 1-hour testing session following alcohol administration. The dose-dependent suppression of CA1 pyramidal cells is consistent with the dose-dependent effects of alcohol on episodic memory formation.

       Below is another representation of the effects of alcohol on the firing of pyramidal neurons in the hippocampus. This one shows images representing cellular activity during 15 min recording sessions, like the image above, but the slides are shown consecutively so that you can see the amount of cellularly activity shrink after the alcohol is administered and then return as the rat sobers up (see the caption for more detail).

Alcohol disrupts the functioning of the hippocampus
The figures above display the firing rate of a single pyramidal neuron in the hippocampus of a rat running around on a maze. Each box in the image represents a 3cm X 3cm region of the maze that the rat entered, at least one, during the recording period. Colored boxes, and particularly boxes with darker colors, indicate that the cell fired lots of action potentials in a particular 3cm X 3cm region of the maze. These cells are absolutely critical for the ability of the hippocampus to form new memories. As can be seen in the figures, alcohol essentially shuts these cells down. This is more than likely one of the central mechanisms by which alcohol produces memory impairments, including blackouts.

Alcohol and hippocampal long-term potentiation
       In addition to suppressing the output from pyramidal cells, alcohol has several other effects on hippocampal function. For instance, alcohol severely disrupts brain cells' ability to establish long-lasting, heightened responsiveness to signals from other cells (Bliss and Collinridge 1993). This heightened responsiveness is known as long-term potentiation (LTP). Because researchers have theorized that something like LTP occurs naturally in the brain during learning (for a review, see Martin and Morris 2002), many investigators have used LTP as model for studying the neurobiology underlying the effects of drugs, including alcohol, on memory.
      In a typical LTP experiment, two electrodes (A and B; please refer to figure below) are lowered into a slice of hippocampal tissue kept alive by bathing it in oxygenated artificial cerebral spinal fluid (ACSF). A small amount of current is passed through electrode A, causing the neurons in this area to send signals to cells located near electrode B. Electrode B is then used to record how the cells in the area respond to the incoming signals. This response is the baseline response. Next, a specific pattern of stimulation intended to model the pattern of activity that might occur during an actual learning event is delivered through electrode A. When the original stimulus that elicited the baseline response is then delivered again through electrode A, the response recorded at electrode B is larger (i.e., potentiated). In other words, as a result of the patterned input, cells at position B are now more responsive to signals sent from cells at position A. The potentiated response often lasts for an extended period of time, hence the term long-term potentiation.

       Alcohol interferes with the establishment of LTP (Morrisett and Swartzwelder 1993; Givens and McMahon 1995; Pyapali et al. 1999; Schummers and Browning 2001), and this impairment begins at concentrations equivalent to those produced by consuming just one or two standard drinks (e.g., a 12-oz beer, 1.5 oz of liquor in a shot or mixed drink, or a 5-oz glass of wine)(Blitzer et al. 1990). If sufficient alcohol is present in the ACSF bathing the slice of hippocampal tissue when the patterned stimulation is given, the response recorded later at position B will not be larger than it was at baseline (that is, it will not be potentiated). And, just as alcohol tends not to impair recall of memories established before alcohol exposure, alcohol does not disrupt the expression of LTP established before alcohol exposure.
      One of the key requirements for the establishment of LTP in the hippocampus is that a type of signal receptor known as the NMDA receptor becomes activated. Activation of the NMDA receptor allows calcium to enter the cell, which sets off a chain of events leading to long-lasting changes in the cell's structure or function, or both. Alcohol interferes with the activation of the NMDA receptor, thereby preventing the influx of calcium and the changes that follow (Swartzwelder et al. 1995). This is believed to be the primary mechanism underlying the effects of alcohol on LTP, though other transmitter systems are probably also involved (Schummers and Browning 2001).

Please click here for a slide discussing the impact of alcohol on NMDA receptor activation.

Indirect effects of alcohol on hippocampal function
        Like other brain regions, the hippocampus does not operate in isolation. Information processing in the hippocampus depends on coordinated input from a variety of other structures, which gives alcohol and other drugs additional opportunities to disrupt hippocampal functioning. One brain region that is central to hippocampal functioning is the medial septum, an area of the forebrain involved in memory, learning, and emotion (Givens et al. 2000). The medial septum sends rhythmic excitatory and inhibitory signals to the hippocampus, causing rhythmic changes in the activity of hippocampal pyramidal cells. In electroencephalograph recordings, this rhythmic activity, referred to as the theta rhythm, occurs within a frequency of roughly 6 to 9 cycles per second (hertz) in actively behaving rats. The theta rhythm is thought to act as a gatekeeper, increasing or decreasing the likelihood that information entering the hippocampus from cortical structures will be processed (Orr et al. 2001). Information entering the hippocampus when pyramidal cells are slightly depolarized (i.e., slightly excited) has a better chance of influencing hippocampal circuitry than signals that arrive when the cells are slightly hyperpolarized (i.e., slightly suppressed).
       Manipulations that disrupt the theta rhythm also disrupt the ability to perform tasks that depend on the hippocampus (Givens et al. 2000). Alcohol disrupts the theta rhythm in large part by suppressing the output of signals from medial septal neurons to the hippocampus (Steffensen et al. 1993; Givens et al. 2000). Given the powerful influence that the medial septum has on information processing in the hippocampus, the impact of alcohol on cellular activity in the medial septum is likely to play an important role in the effects of alcohol on memory. Indeed, in rats, putting alcohol directly into the medial septum alone produces memory impairments (Givens and McMahon 1997).