Copyright © 2002 by The Eaton T. Fores Research Center

So much has been written about the benzodiazepines – probably the most widely taken family of psychotropic drugs in history – that it would be exceedingly hard to write a better or more complete review of this field than the great many that can already be found in the literature. Nevertheless, I am going to try to present an exhaustive overview of these drugs, from the serendipitous events leading to their discovery, to the explosion of their use around the world, to their pharmacokinetics and pharmacodynamics, and to their uses, both licit and illicit.  A serious discussion of drugs and pharmacology cannot dispense with organic chemistry, physiology, and biology, and I make no attempt to avoid these subjects.   However, I do approach them in what is intended to be an accessible way.  It is my hope that seeing the chemical structures, as well as learning about the pharmacological activities, will result in a much deeper understanding of the subject.   Since this is not a textbook, I cannot afford to spend time explaining general chemistry, or very much organic chemical theory, but new words, whether they are conceptual or simply chemical names, are italicized the first time they are used.  I make the assumption that readers know what atoms and molecules are.  I truly hope that this is a valid assumption.  Animal models used in behavioral pharmacology laboratories are explained in a readily understandable way.  Finally, some effort is made to demystify biochemical concepts like receptors, ionophores, "binding" (the quotes are there because the "binding" of a drug to its receptor is not "binding" at all in any chemical sense), enzymes, re-uptake pumps, and so on.   This is not as difficult a task as it might seem, since all of these things are more-or-less the same thing.

If the material proves to require too much background to be approachable, we recommend that you backwards chain to the next lower level you need to understand.  Questions can also always be sent to comments@etfrc.com.  We hope that you both enjoy what follows and deepen your understanding through it.

Prior to the invention of the benzodiazepines, the most commonly used drugs for sedation and sleep were the barbiturates, that is, the 5,5-disubstituted derivatives of barbituric acid (which is itself without any pharmacological activity) which had been invented at the dawn of the 20^th century. While effective, the barbiturates were exceedingly toxic and highly addictive – abrupt withdrawal could cause death – and barbiturate poisoning accounted for a great number of deaths every year. Many sedative-hypnotics not derived from barbituric acid have been synthesized and marketed, but all eventually proved to have the drawbacks of the barbiturates (although a few were just jaw-droppingly euphoric). The therapeutic index – the difference between an effective dose and a poisonous dose – of these drugs was very low. The drugs were called primary and continuous central nervous system (CNS) depressants, reflecting the fact that, depending on the dose, all levels of CNS depression from mild sedation to surgical anesthesia to coma and death could be produced. Medicine clearly needed a safer, less addictive kind of sedating drug: a sleeping pill that one stood a reasonable chance of waking up again after taking.

Methyprylon and methaqualone.

In the mid-1950s, the Hoffman-LaRoche company, with chemist Leo E. Sternbach and pharmacologist Lowell Randall heading the effort, began a systematic search for such a "minor tranquilizer." Chlordiazepoxide (RO 5-0690) was the result of a mistaken synthesis on the part of one of Sternbach's bench chemists of the last of a series of analogs of a certain dye compound (which they thought were heptodiazines, but which turned out to be quinazolone-3-oxides) they had been making. One could say that the decision to test RO 5-0690 when it was dug up in a dusty corner of the lab, where it had been placed some years back, was similar to the "peculiar presentiment" that led Dr. Albert Hoffman to re-evaluate the 25th compound in his ergot-derived series, which compound eventually restructured Western civilization such that my kids are incredibly spoiled, but that's another story for another time.

Randall's pharmacologists wondered why RO 5-0690 alone showed all of these behavioral properties (most notably, "taming" at vastly sub-ataxic doses), when the prior 39 analogs were all inactive. The answer turned out to be that it wasn't an analog after all – it was a brand new compound, the first benzodiazepine, which was initially called methaminodiazepoxide and then chlordiazepoxide (CDZ), which remains its USAN name. Chlordiazepoxide goes by the tradename Librium®. Once again, psychopharmacology, showing considerable ataxia itself, had stumbled forward. This is all, of course, if I'm remembering my psychopharmacochemical history correctly, which is questionable, given the effects of benzodiazepines on memory. But the gist of my account is probably related in some way to what really happened, even if the type of relationship is one of diametrical opposition. Subsequently, the biochemical geniuses over at Hoffman-LaRoche figured out that the epoxide moiety on CDZ not only had nothing to do with anything, but it actually reduced activity. Analysis of the activity of chlordiazepoxide's primary metabolite, demoxepam, led to the introduction of diazepam about four years later. Everyone liked diazepam much better, and pretty soon, no one could remember why they had thought chlordiazepoxide was such a big deal, once again demonstrating the effects of these compounds on memory. By the mid-1970s, over 8000 tons (a lot of 5 mg tablets) of benzodiazepines were sold every year. The benzodiazepines proceeded to do to the propanediol carbamates, notably meprobamate, and the barbiturates pretty much what the DeathStar did to Alderaan, what Bill Gates did to Netscape, what Bush has done to the United States, etc.  Sales of barbiturates and related drugs became increasingly rare until they were unheard of.

Ten years later, the series was perfected with the introduction by Upjohn of the triazolo analogs, and the displacement of Valium as the Number One most prescribed drug by Xanax (alprazolam) [-lam is the suffix for 1,4-triazolo-benzodiazepines; -zam for 1,5-benzodiazepines; and –pam for the typical 1,4-benzodiazepines]. Xanax was remarkably potent (down to the microgram range, like LSD), and doctors were actually trusting enough to buy the company's rap that it didn't cause dependence. Of course, it causes far more intense dependence than diazepam or any other previously marketed benzodiazepine. About 10% of people who are prescribed Xanax become addicted to it (probably more), which adds up to millions of people. As the labeling says, "Certain clinical events, some of them fatal, are a direct consequence of physiological dependence to Xanax ... some patients will prove resistant to all tapering regimens." Nothing significant has happened in the benzodiazepine world since then, although it’s worth knowing that over 2000 benzodiazepines have been synthesized and the relationship between their molecular structure and their effects (which pharmacologists call the SAR, or Structure-Activity Relationship) has been worked out in great detail (more on that subject later).

Pharmacology is a combination of physiology and organic chemistry, the branch of chemistry that deals with the enormous number of ways that carbon atoms can be arranged. That is why I’ve been careful to make note of proper oganic chemical terminology and to explain some rudimentary facts about molecular geometry in this article. When proper organic chemistry terminology is used, it is possible to visualize the molecule simply from its name. This is true of the word "benzodiazepine" as well. We will build a typical anxiolytic benzodiazepine up from its component parts, learning a bit of organic chemistry nomenclature in the process.

Six carbon atoms arranged in a ring make a compound called cyclohexane, which is drawn as shown above. At each bend in the ring, there is an implicit carbon (C) atom. Carbon always has four bonds attached to it. Where fewer are shown, there are implicit hydrogen (H) atoms connected by single bonds.

Cyclohexane is called an aliphatic molecule, because the electrons in it are stable and associated with particular carbon atoms. But there is another way these same six carbon atoms can be arranged, to form an aromatic (resonant) molecule called benzene. You can picture the double bonds in benzene rapidly alternating with the single bonds; or you can forget the bonds altogether and imagine the shared electrons smeared out across the entire molecule.

A seven-membered ring of carbon atoms with one nitrogen in it is called an azepine ring. If it has two nitrogen atoms, it is a diazepine ring.  The relative positions of the nitrogen atoms are recorded using a standard numbering.  In this picture, the topmost nitrogen is designated position one.  So, the complete name of the above molecule is 1,4-diazepine. Now we’ve drawn benzene and diazepine. How far can we be from benzodiazepine?

And, indeed, the above molecule is 1,4-benzodiazepine. From a chemical perspective, this is all that is required for a molecule to be a benzodiazepine. However, from a pharmacological perspective, there are several additional features required before any drug activity will be seen.

Almost all active benzodiazepines, except those possessing a fused heterocyclic ring or a thionyl group, have a carbonyl group at position 2. Carbonyl, or ketone, is often abbreviated "one" (pronounced "own," not like the number one). If this is clear, then you already knew it.

Two other features are required before we will have the basic pharmacological benzodiazepine skeleton. The first of these is another benzene ring, separated from the heterocyclic benzodiazepine ring system by a single bond. A benzene ring is called a phenyl group when it is part of a larger molecule. There is one final requirement, without which our beautiful molecule will have no drug effect. There must be an electron-attracting substituent at position 7. The halogens – chlorine, flourine, bromine, and iodine – are nice attractors of electrons, which is to say, they are quite electronegative atoms (flourine, in fact, is the most electronegative of all elements).

 If we take care of these two details, and add an N-methyl group to the nitrogen at position 1, we will have arrived at a prototype benzodiazepine: diazepam, tradenamed Valium®. This drug exhibits all three of the basic benzodiazepine effects: (1) skeletal muscle relaxation, (2) anticonvulsant activity, and (3) anti-anxiety effects at doses far lower than those that cause ataxia (loss of balance).

What’s "metabolism?" It’s one of those words that everyone thinks they know the meaning of, until they try to define it. In Biochemistry and Bioenergetics, as well as in medicine, "metabolism" has a very specific meaning. Metabolism can be broken down into anabolism and catabolism. They are the sequence of chemical reactions, generally enzyme-catalyzed, which synthesize what the organism needs, and destroy what is toxic to the organism, respectfully. In pharmacokinetics, the focus is generally on catabolism, for this is the process by which the organism gets rid of xenobiotics, substances which come from outside the body and are recognized as foreign. Drugs are xenobiotics. The organism mobilizes a wide range of chemical strategies to rid itself of xenobiotics.

In a few cases, anabolism plays a role in the "absorption-fate-elimination" cycle that pharmacokinetics documents. These are the cases in which the drug taken is actually a prodrug, something which must undergo further metabolism before it becomes active. Several benzodiazepines are, in fact, prodrugs whose metabolism produces nordazepam, the long-lived primary metabolite of diazepam and final common pathway for the metabolism of many benzodiazepines.

For the most part, benzodiazepines are not water soluble and cannot be readily prepared for injection. Virtually all benzodiazepine use outside of hospitals is oral. Benzodiazpines therefore must traverse the same "obstacle course" that all oral drugs run before they can so much as be absorbed: first, the drug must go into water solution simply to cross the gastric mucosa. Even doses prepared as solutions may precipitate as they move between the low pH environment of the stomach and the high pH environment of the gut. In the GI tract, there will be a host of digestive enzymes, each waiting for its chance to destroy the drug. Should the drug survive, it must then pass through the gastric mucosa into the venous side of the GI tract to become systemically available. Once in the circulation, the drug will encounter plasma proteins – random albumins which will nonselectively bind up to 95% of the drug. And finally, if the drug is a psychotropic one, it will have to get across the blood-brain barrier.

Phase I and Phase II Reactions

Oil and water don’t mix. Similarly, things that dissolve in water don’t dissolve in oil. The difference between water-soluble and oil-soluble chemicals turns out to be extremely important in pharmacology. Drugs which dissolve in water are called hydrophilic, which means "water loving," while drugs that dissolve in oils are called "lipophilic," which means "fat loving." Just to make things more confusing, pharmacologists often use the converse of these words: things that are hydrophillic are said to be lipophobic (afraid of fats), while things that are lipophilic are said to be hydrophobic (afraid of water).

In the body, lipophilic drugs have a number of great advantages. They cross all sorts of barriers more easily, since cell membranes are lipids. In particular, for a psychotropic drug to be lipophilic allows it to penetrate the blood-brain barrier and gain access to the CNS. The lipophilicity, or degree to which a drug is soluble in fat, can be quantified and expressed in many ways, including the n-octanol:water partition coefficient, which won’t be discussed here. In addition to its advantages, lipophilicity also has some drawbacks: in particular, lipophilic compounds are so at-home in the body that, unless mechanisms existed for making them polar and thus hydrophilic, they might roam the body forever, at no point having their effect terminated. Fortunately, the body turns nonpolar (lipophilic) compounds into polar (hydrophilic) ones via a two-step process (hence the title of this subsection). During a Phase I reaction, also called oxidation or functionalization, a polar substituient, such as a hydroxyl group (–OH), is added to the lipophilic molecule. The addition of this functional group allows the molecule to become a target for other kinds of reactions. Phase II reactions, also called conjugation reactions, cause the formation of a covalent bond between the functional group added by the Phase I reaction (or, occassionally, some other functional group) and a bulky, highly polar molecule derived from within the body. For our purposes, glucuronic acid is the molecule most likely to be attached by a Phase II reaction, and the resulting molecule is called a glucuronide. The highly polar glucuronide is usually pharmacologically inactive, and is so constructed that it can be rapidly removed from the circulation by the kidneys. The conjugate is then excreted into the urine. This process is often referred to as though it were a single process, and it is called oxidation-conjugation.

What does it mean for a molecule to be polar or non-polar? Consider that reality, on the chemical scale, consists entirely of molecules, energy, and the occasional atom. What attributes do molecules have? That is, which aspects of them is it coherent to ask questions about? Does it make sense to ask what color a molecule is? Or what its political views are? No, of course not. Color and beliefs are not attributes of molecules, so such questions would obviously be absurd. Science understands reality as having a fundamentally third-person ontology. That means that everything there is to be said about reality can be described from an outside point of view. On this view, there is, for example, nothing it is like for me to accidentally put my hand on a lit burner on my stove, because anything "inward" or "intrinsic" or "first-person" is, according to scientism, not part of reality. A molecule has just three attributes from a chemical perspective: size, shape, and charge distribution.   Polar molecules are polar because their charge is unevenly distributed.  They are more positively charged in one place, and more negatively charged in another.   That is, they have poles.  Nonpolar molecules are usually devoid of the electron-attracting groups that distort the charge distribution of the molecule.   They have no poles.  They are often long chains of carbon atoms, whose electrons are pretty evenly distributed.

Much of what we have been learning is not specific to benzodiazepines at all, but is simply organic chemistry and general pharmacology. Should you investigate other classes of drugs, you will find that those drugs too are small, organic molecules; that their lipophilicity determines many things about them; that they exhibit patterns of absorption, fate, and elimination; and that they undergo Phase I and Phase II reactions just as benzodiazepines do. To that extent, we’re studying a mixture of chemistry, pharmacology, and biochemistry, and using a certain class of drugs, the benzodiazepines, as our example. Later we shall see how psychology, philosophy, and phenomenology also bear on the questions raised by psychotropic drugs (it should be noted that acquiring phenomenological knowledge of such matters is currently illegal in most of the world).

Diazepam has many routes of metabolism. For the moment, we are going to focus on a minor one, the para-hydroxylation of the aromatic ring at position 5, which is an example of a Phase I oxidative reaction, followed by the etheric conjugation of this product with glucuronide, which is a Phase II conjugation reaction. For the sake of simplicity, we won’t discuss the bioenergetics of these enzyme-catalyzed reactions; the curious reader can find this information in any textbook of biochemistry and probably all over the Web as well. Following this, we will detail the general scheme for all benzodiazepine metabolism. We will see that, although there are many benzodiazepines, most of them are converted to one of a handful of active metabolites almost immediately, and it is this metabolite or metabolites, not the parent drug, that accounts for all of the effects of an administered dose. So, let’s begin by clarifying in physical terms what we mean by "oxidation" and "conjugation" reactions.

 

Above is shown one possible (although rare in reality) oxidation-conjugation pathyway by which diazepam can be eliminated from the body.  In the first picture of the diazepam molecule, I have explicitly numbered the carbon atoms, in case anyone was having trouble with the numbers in the organic chemistry naming convention. Note that the numbers on the 5-aryl (5-phenyl) ring are followed by the "prime" symbol, e.g., 2’, 3’, 4’. This is to differentiate those positions from the numbered positions on the heterocyclic benzodiazepine ring system to which it is attached. Also, notice that the numbers on the 5-phenyl ring stop at 4’. Since this ring is connected to the rest of the molecule by a single bond, and since molecular substituients can rotate freely around single bonds, there would be no meaning to numbering the rest of the carbon atoms: 6’ is the same thing as 2’, since the ring can simply flip over, and likewise, 5’ would mean the same thing that 3’ does. Chemists have other names for these special positions on the benzene ring, too: 2’ is called ortho, 3’ is called meta, and 4’ is called para.

The attachment of a hydroxyl group (–OH) to a phenolic ring or other aromatic structure is called aromatic oxidation. This reaction is catalyzed by the cytochrome P-450 monooxygenases, a group of related enzymes in the liver, and also requires molecular dioxygen – O2 – which is supplied by an NADPH-dependent reductase, something that would require several more documents to explain. Suffice it to say that you’re always breathing in dioxygen, and the metabolic processes of your body are constantly producing energy (ATP) from your respiration. These processes are also the ones that donate the oxygen atoms used in aromatic oxidation.  Following the reaction, one oxygen atom has been incorporated into the metabolized drug, and the other has been incorporated into a water molecule.

Before the reaction can begin, a susceptible site for oxidation must be found. There are two main factors that make a given ring, or a given position on a ring, suitable or unsuitable. The first is charge distribution. Phenolic rings with electron attracting groups attached to them are not favored (when we say that something is "not favored" in chemistry, it doesn’t mean that the other molecules look down upon it, it just means that the environment would require too much energy to carry out the reaction in). Stearic factors are the other principal determinants of ring suitability. Stearic factors simply have to do with how large and bulky various parts of a molecule are. If a potential site of oxidation lies hidden behind two gigantic ring systems, it can be forgotten about. The energy cost of moving the rings aside is just too high. Remember, the enzyme doesn’t have any "desire" to oxidize things. The enzyme doesn’t "want" anything. It simply helps things move to more stable (lower) energy states.

The para-position (4’) on the 5-phenyl ring of diazepam is a good site for oxidative attack. The ring has no substituients drawing electrons away from it; nor is it crowded in by other bulky parts of the molecule. The P-450 monooxygenase moves in, and before you can say "5-phenyl-7-chloro-1-methyl-1,4 benzodiazepin-2-one," the deed is done and the diazepam molecule has become 4’-hydroxydiazepam – a compound which, by the way, is virtually inactive. But now it’s more polar. It likes water more. And it has a nice, juicy oxygen atom sticking out of it just waiting to react with something.

Now the drug is set up for conjugation (in this case, glucuronidation) to occur. Here, unfortunately, it's difficult even make much of an attempt to explain what's going on.  Phase II reactions involve things like endogenous substrates becoming activated in the form of coenzymes, and the connection of the activated endogenous substrate to the oxidized part of the drug by glucuronyltransferases. It’s not possible to get much of an idea of how this works without a lot of study (which we encourage, of course, but this isn’t a biochemistry textbook). We can, however, say what it does: Phase II reactions essentially "tag" the molecule for removal by the kidneys. At that point, the drug has been eliminated from the body and the pharmacokinetic journey that began when someone swallowed it is over.

The illustration below details the major pathways, through functionalization, oxidation, and conjugation, of representatives from several structural classes of benzodiazepines.

There are a great many benzodiazepines. Now, we’re going to very briefly examine how their chemical structures affect their pharmacological activity.

First of all, what exactly is the pharmacological activity of this family of drugs? We’ll talk about the molecular basis of their activity when we discuss pharmacodynamics, but for now, let’s just state what they do, since that’s what we’ll be correlating their structures with. The benzodiazepines produce a wide range of effects, and are used for all sorts of things. In medicine, they’re usually used for anxiety and insomnia, though they’re occassionally used in musculoskeletal injuries, alcohol detoxification, and several other things. Three of their effects are considered hallmarks: (1) an anticonvulsant effect, (2) a muscle relaxant effect, and (3) an anti-anxiety effect.

Behavioral Pharmacology

The study of the effects of psychotropic drugs on animal behavior is called behavioral pharmacology.  This kind of work has lately come to be called "psychopharmacology," but this is a misnomer, since "psycho," from the Greek psyche, always denotes something about experience, not behavior.  We can't do psychopharmacology on rats, because the rats can't tell us what the effect of the drug on their consciousness -- their thoughts, feelings, moods, and so on -- was.  Behavioral pharmacology is sometimes the attempt to infer these things from behavior; other times, simply the effort to rapidly screen drugs for marketability.  The former scenario occurs mostly in academia; the latter, in industry.  Doing behavioral pharmacology with the intention of discovering uses of a new compound to relieve human psychic distress requires the development of animal models of various human troubles.  There is a real problem here, because most human psychospiritual troubles are the result of higher brain and corresponding mental functions which rodents, for example, simply do not have.

There are animal models for all of the major benzodiazepine effects.  The models in use during the development of these drugs differed from lab to lab, but certain features of them were the same.  Anticonvlusant activity was generally measured by the ability of the drugs to inhibit seizures induced by the convlusant drugs picrotoxin or pentylenetetrazol (which, bizarrely, was sold by prescription during the 1960s as a "tonic" for the elderly under the tradename Metrazol®), or else by strychnine or maximal electroshock induced seizures.  Convlusions due to the latter can be inhibited only by large drug doses, and these doses are strongly ataxic. 

Myorelaxation (muscle relaxant effect) was often measured by dragging rodents by the tail across a screen, and measuring the force with which they were able to resist being pulled.  Benzodiazepines caused the animals to lose their grip on the screen as a result of far less force than animals who had received vehicle (placebo).   Ataxia can be measured in rodents by placing them on a rotating rod running the length of their bodies.  The longer an animal could stay on the rotating rod without falling off, the less ataxic it was judged to be.  Certain benzodiazepines, such as clonazepam, the 7-nitro analog of lorazepam, shown below, could produce significant loss of muscle tone (hypotonia) without significant ataxia.  Tolerance, however, almost always develops to the myorelaxant, anticonvulsant, and ataxic effects of these drugs.

Animal Models of Psychic Distress

The area of producing animal models of human psychic distress really brought out the creativity in psychopharmacologists.  Models of pain (for testing new analgesics), depression, anxiety, and psychosis could reach almost Mengleian levels of inventiveness.  Removing newborns from their mothers, inflaming rats' paws with subcutaneous yeast injections and then putting them into a press, forcing lab animals to swim through freezing cold water with weights on their bodies, long periods of isolation, starvation, and inflicting totally random and unpredictable punishments with electrical shocks were just a few of the tricks in the psychopharmacologist's book.  Among the more bizarre procedures was muricide, which required "muricidal" rats -- rats which would spontaneously kill mice placed in the cage with them -- and a decrease in the number of mice killed by the rat ostensibly indicated antidepressant activity.   The reasoning, or what passed for it, behind this kind of assay is anyone's guess.   And there is methamphetamine aggregate toxicity, in which mice are given large doses of methamphetamine (speed) and packed together in a single cage to maximize stress, until they die.  Decreases in the death rate caused by a test drug was taken as evidence that the drug had "antipsychotic" activity.  When such methods "worked," that is, predicted drug action in human beings, so many things were generally wrong with the model that it seemed likely that the correlations were pure coincidence.  Of course, there are similarities in the way all mammals behave in response to torture, but does this really say anything about the human psyche, let alone anything about putative "chemical imbalances" and other biopsychiatric ideas that are so far removed from their speculative roots that the connection is essentially imaginary?

Animal models of anxiety were, all things considered, relatively short on sadism.  Large numbers of mice were not crowded together in a cage having been given huge overdoses of methamphetamine in order to drive them nuts.  There was no need to stress them to exhaustion to produce "behavioral despair," since antidepressant effects weren't being sought.  Instead, animals were generally placed in cages where they could walk into an adjoining chamber to obtain food or water.   After this behavior was learned, the animals began to be shocked for stepping into this room.  A visual or audible cue always preceeded the start of the punishment situation.  The animals quickly learned not to try to eat or drink when they detected this cue.  Benzodiazepines, sometimes in very low doses, caused the animals to disregard the cue, and seek out their food or water without regard to punishment, where previously they would have waited until the cue was over before venturing out.  For some reason, this was interpreted as evidence for a selective "anti-anxiety" effect.  It seems that a more reasonable conjecture would have been that the animals were simply stoned, but a review of the literature shows that this interpretation apparently did not occur to any of the scientists doing this work.  Of course, people always tell their bosses whatever their bosses want to hear, and these scientists' bosses wanted to hear that the scientists had found a selective anti-anxiety drug.

In the illustration above, I’ve shown the structures of six benzodiazepines. I’ve picked these six from the thousands available because they’re well-known, and each makes an important point about the relation of structure to activity in these drugs.

Following is an extremely short synopsis of the structure-activity relationship of benzodiazepines:

1. There must be an electron attracting substituient at position 7.

2. The first benzodiazepine, chlordiazepoxide (above), was a 2-amino benzodiazepine. However, research showed that this compound was quickly metabolized into several benzodiazepin-2-ones, and subsequent work focused on those compounds, as they were more potent. Eventually, triazolobenzodiazepines (see alprazolam and triazolam, above) were found to be much more potent the 2-carbonyl drugs, and work began to be done with triazolo and other fused ring compounds. For some reason, the triazolo compounds do not require any substitution at position 7, contrary to what is indicated above.

3. Positions 6, 8, and 9 should be left unsubstituted.

4. The presence of a phenyl at the 5-position increases activity, and in practice, all benzodiazepines in use have this substituient. 5-phenyl rings with electron-attracting (generally halogen atoms) groups at the 2’ position show greatly increased activity, and produce greater amnesia (see lorazepam, above, and compare alprazolam and triazolam, above). 4’ (para) substitution decreases or abolishes activity.

5. The 4,5 double bond should not be moved or saturated.

6. The N-substituient at position 1 should be small for higher intrinsic activity; however, drugs with large substituients at this position have been prepared and marketed (e.g., flurazepam, prazepam, halazepam, not shown). They owe their activity to metabolic dealkylation, often to nordazepam.

With that out of the way, let's get back to the topic at hand.  The discovery of the "opiate receptor" in 1974 set off a firestorm of "grind 'n' bind" mania, with neuropharmacologists grinding up rat's brains like mad, trying to make radioactive drugs stick to parts of them better than the mirror-image, inactive forms of those drug molecules did.  It wasn't long until the discovery of "the benzodiazepine receptor" was announced in 1977 in a paper called "Benzodiazepine Receptor: Demonstration in the Central Nervous System," a title that sounds an awful lot like the title of Pert and Snyder's original paper announcing the demonstration of the opiate receptor, but of course, scientists aren't the most creative people on Earth.  It seemed that it was possible to make radioactive benzodiazepines stick to brain membranes, too.  Today it is recognized that there are at least three distinct binding sites for the benzodiazepines, now designated the "omega" receptor.  Furthermore, these receptors aren't isolated, but rather are parts of a macromolecular complex that includes (1) the GABA-A receptor, (2) the benzodiazepine receptor, (3) the binding site for the barbiturates, and (4) the chloride ionophore, the channel that allows negatively charged chloride ions to enter the neuron and drive it to further polarization, making it less likely to fire.  What's more, all of these things interact: the chloride ionophore is what's called a ligand-gated ion channel (a ligand is simply something that binds to something).  When the inhibitory neurotransmitter gamma-aminobutyric acid (GABA) binds to its binding site, the result is that the chloride channel opens and allows Cl- ions in, hyperpolarizing the cell and making it less excitable.  Now, the benzodiazepine site, when bound, allosterically alters the shape of the GABA site so that it fits GABA better: it has a greater affinity for the GABA molecule.  The barbiturate binding site also modifies the effects of GABA, but is capable of opening the ionophore by itself -- something that benzodiazepines can't do, and almost certainly what makes them so much safer than barbiturates.  A general drawing of this complex is shown below.  It's important to remember that the placement of the various receptors on the complex are arbitrary.  In fact, receptors that interact allosterically are more likely to be close to one another than this diagram shows.  This is because allosteric interaction involves the twisting of some atoms and bonds in one part of the peptide, with the result that atoms in bonds in other parts change their positions (and thus, perhaps, their affinity for various molecules) too.

A schematic drawing of the GABA/benzodiazepine/barbiturate/chloride ionophore macromolecular complex.