Debra Ann Fadool. Encyclopedia of Life Sciences: Supplementary Set. Volume 24, Wiley, 2007.

Pheromones are a vehicle for chemical communication in vertebrates and serve to relay information important for reproduction, mate selection, species and gender identification and social status. Activation of cellular signaling cascades in the vomeronasal organ combined with hormonal changes mediating reproduction are driven by pheromones that must be transduced from an external chemical message into an electrical signal read by the central nervous system.

Introduction

It is amazing to think that our perception of the world is only as refined as our biological sensors! Sensory systems capture information from the environment and convey these signals to higher cortical centres in the brain using the language of the central nervous system, the action potential. Whether the sensory system captures a photon of light, a mechanical vibration, or a chemical on the tongue, these diverse signals must be ‘transduced’ into electrical signals so that we may process the information to render an internal depiction of our external world. Electrical signals passed around by the thinking brain are generated by a single type of protein, the ‘ion channel’. Some of the most important signals for a mammal, are not visual in nature, but form the basis of chemical communication. Mammals use a specialized portion of the olfactory system, called the ‘vomeronasal organ’ (VNO), to transduce chemical information central to mate choice, species recognition and reproductive readiness; all of which are triggered through a series of cellular events that activate an ion channel to carry the electrical signal to the brain.

Chemical Communication

Most higher vertebrates rely upon the detection of body chemicals that are received at a specialized portion of the olfactory nervous system, the VNO. The chemical constituents encoded by the VNO allow the animal to extract important information about the provider concerning social hierarchy, reproductive status, gender or species identification or prey/predator recognition. Since these often innate signal properties provided by the chemicals represent a special form of information passage, scientists refer to this process as ‘chemical communication’.

Anatomy of the olfactory system

In most vertebrates, olfactory sensory perception is mediated by nerve cells called sensory neurons at two anatomically distinct locations. In the first region, the sensory neurons are called ‘olfactory sensory neurons’ (OSNs) because they are contained in the ‘main olfactory epithelium’ (MOE), or in simple terms, the nose. In the second region the sensory neurons (called ‘vomeronasal sensory neurons’ (VSNs)) are contained in the VNO, also referred to as the ‘Jacobson’s organ’, named after the Danish scientist who carefully discovered this hidden organ nearly two centuries ago. Owing to the thoroughness of his anatomical observations, Ludvig L. Jacobson discovered the VNO as a rich innervated organ, hidden in the nasal septum of mammals, and one which had escaped the attention of contemporary anatomists of the nineteenth century. According to his anatomical report to the Royal Veterinary Society in 1811, ‘the organ was carefully concealed in the foremost part of the nasal cavity, in close contact with the nasal septum on the palatal elongations of the intermaxillary bone.’ Previously, only a secretory duct had been discovered, so Jacobson deduced that the organ’s function was secretory in nature, but also conjectured that it was a sensory organ. His discovery was renamed organon vomeronasale (Jacobsoni) in 1895 and today is well accepted as the site for transduction of body chemicals into electrical signals; the basis of chemical communication influencing reproductive and social behaviours in mammals.

The sensory neurons of both the VNO and the MOE are derived from a common embryological origin, and are arranged in a highly ordered array or neuroepithelium, and contain neurons with similar lifespans. Unlike almost all other nerve cells of the central nervous system, neurons in both organs are atypically regenerative. They are replaced from a precursor population every 2-3 months and contain a phylogenetically conserved protein called ‘olfactory marker protein’ (OMP) upon reaching maturity. Scientists are trying to unlock the molecular mechanism for such regenerative powers of the olfactory sensory neurons so that a tool for recovering nerve growth following injury or disease might be invented. In both VNO and MOE, the sensory neurons are bipolar in structure or cellular morphology. In the MOE, the neuronal process that carries the electrical signal to the cell body of the sensory neuron, called the dendrite, terminates in a tuft of cilia, which projects into the mucus layer of the nose and binds odorant molecules. It is the specialized structure of the cilia that contains the olfactory transduction machinery for odorant reception and discrimination. In contrast, VSNs have dendritic processes that completely lack cilia but possess finer hair-like filaments called microvilli. It has been shown that the microvilli of VSN contain the transduction machinery to encode body chemical information. If one introduces a vital dye that is excitable under ultraviolet light into the VNO cavity of a living vertebrate, and mixes this dye with a small quantity of detergent, the detergent will shear the microvilli tips of the VSN dendritic processes, and the dye will enter the cell, permitting great resolution of the neuronal architecture of the VNO. The dye is coupled to a sugar molecule and can migrate into the brain of the animal, to visualize the target or central projection of the VNO, the accessory olfactory bulb.

A division of structure and function

Albeit similar neuronal morphology and origin, the two olfactory systems diverge in function. Olfactory information is transmitted from each olfactory system to different brain regions. In the MOE, processes carrying electrical information away from the sensory cell body, called axons, project to the main ‘olfactory bulb’ (OB), which sends most of its fibres to the olfactory cortex and other higher sensory centres. Axons from the VSNs of the VNO project alternatively to an accessory olfactory bulb, which then bypasses higher cognitive centres by projecting to a discreet locus within the amygdala and then relaying to the hypothalamus, which may regulate the hormonal or endocrine state of the animal. The chemical nature of the odour ligand received by each olfactory system is also different. Volatile odorants are detected in MOE; odour molecules that are vaporizable and delivered in the air to the nasal passages. The VNO, however, mostly detects specialized nonvolatile odorants that have to be transported to the sensory neurons through the vomeronasal duct and usually transferred by physical contact. Although the latter type of chemosignal or pheromone is poorly described for many vertebrates, it is known that the vertebrate VNO detects steroid or fatty acid compounds, histocompatibility complex antigens in urine, skin and vaginal secretions, collagen-like compounds, and generally high-molecular-weight compounds as recognition molecules derived from body chemicals. The detection of these specialized chemosignals thus allows the VNO a different primary functional role in sensory perception; the detection of molecules emitted by conspecifics (same species). Hence, the VNO plays a strong role in the execution of species-typical behaviours, innate social or sexual behaviours and the initiation of specific neuroendocrine changes.

What is a pheromone?

The term ‘pheromone’ was first coined by Karlson and Luscher in 1959 to describe the physiology and behaviour of insects following exposure to a released chemical for the purpose of communication. The word pheromone comes from the assembly of two Greek roots, ‘pherein’, meaning to transfer, and ‘hormone’, meaning to excite. Explicitly, a pheromone was defined as ‘a substance that is secreted to the outside by an individual and received by a second individual of the same species, in which it releases a specific reaction, for example, a definite behaviour or a developmental process.’ In the most restricted sense, pheromone communication is thought to be driven by natural selection, whereby both the sender and receiver increase their evolutionary fitness by the mutually beneficial communication. Pheromone communication has since been classified into two types: priming pheromones and signalling pheromones. ‘Priming pheromones’ produce a long range of change in the physiological state of the receiver, generally by eliciting a neuroendocrine secretory change that ‘primes’ the receiver for subsequent pheromone responses. Examples of mammalian priming pheromones include those that accelerate puberty, suppress or synchronize oestrus, or suppress oestrus in subordinate females. In contrast, ‘signalling pheromones’ produce immediate stereotypical behaviours that may be modified by experience. The best known example in mammals for a signalling pheromone would be hamster male-mounting behaviour that is precipitated following the physical sampling of pheromone contained in vaginal fluid.

Pheromone-mediated Behaviours

While it is recognized that most pheromones implicated in the control of behaviour or physiological processes are transduced in the VNO, a few are not. Other chemicals involved in prey detection, but not pheromonal, can be transduced in the VNO. For example, the pig pheromone androstenone is a volatile molecule found in the saliva of sexually aroused boars that is received in the MOE and not in the VNO. Snakes, for example, are known to detect earthworm secretions (prey chemosignal not pheromonal) through the VNO. This complicates the science of chemical communication when the term pheromone is used too liberally, but, yet extremely interesting, discreet behaviours have been described and attributed to the VNO that tell us much about the nature of pheromones.

General VNO-related behaviours

Owing to the nonvolatile nature of the chemosignal being received at the VNO, vertebrate animals have been observed to perform patterned behaviours that may be involved in either investigation of the chemosignal or promoting access of the chemosignal to the organ. In mammals, the tongue is used by bulls, dogs and opossums, for example, to direct pheromone chemicals to the VNO. Teeth chattering, nuzzling and head-bobbing, may all act to help deliver or discriminate the chemical most effectively. Ungulates (hoofed animals), elephants and cats have the most marked VNO-related behaviour in mammals referred to as ‘flehmen behaviour’. These animals have a distinct display during nasal-oral contact with pheromones contained in urine or vaginal secretions. The animals lift their head, lift the upper lip or contract the nose, and suppress breathing momentarily. Incorporation of radioactive tracer dyes into the pheromone fluid secretions was used in studies of goat flehmen behaviour to demonstrate the positive transport of the pheromone into the VNO through the oral cavity or mouth.

Pre-mating behaviours: female receptivity and deterrence of young males

The receptivity of the female in mammals begins with the pheromonal control of ovulation. In most mammals, ovulation does not operate on a cyclic basis, rather oestrus induction depends upon the pheromone stimulation by the male, a process that is referred to as reflex ovulation. In prairie voles, for example, a drop of male urine placed on the nose of a female vole, initiates rapid uterine weight gain as well as a complex series of serum rises in several hormones including ‘luteinizing hormone’ (LH), ‘luteinizing hormone releasing hormone’ (LHRH) and ‘prolactin’ (PRL). In the Asian elephant, a pre-ovulatory urinary pheromone called Z7-12:Ac serves as a powerful cue of female sexual readiness whereby its detection by male elephants evokes erections and other sexual behaviours. The concentration of Z7-12:Ac increases 100 times from early ovarian follicular stages to just prior to ovulation, and as it increases, the frequency of male stereotypical flehmen behaviour increases in direct relation. Also, in the Asian elephant, another premating pheromone is secreted from the temporal gland just below the eye. Although the chemical composition of this pheromone secretion is not yet known, socially immature males are distinguished and deterred by older males who avoid contact with the subordinate animals by increasing the malodorous content of the secretion. These premating pheromones are secreted in a period called musth, which is an annual interval of heightened sexual activity and intensified aggression for the wild male elephant. Such chemical communication is thought to be advantageous for the conspecifics that may recognize both the degree of aggression and sexual maturity prior to initiating physical interactions.

VNO impairment or removal

The first experiments that tied reproductive behaviours to the proper functioning of the mammalian VNO were performed in male guinea-pigs, which failed to mount or demonstrate typical male copulatory behaviours (intromission, mounting and thrusting) when the VNO was impaired. Similarly, impaired females did not demonstrate lordosis, a female mating position assumed where the head is up, tail is down, and male intromissions are facilitated by a laying down posture. If innervation from the VNO is severed in sexually naive mice or hamsters, then mating behaviours are more severely disrupted than when sexually experienced animals lose their use of the VNO. In hamsters, a pheromone called aphrodisin is released in vaginal discharge that increases copulatory behaviour in males and is associated with increased hormonal secretion of ‘gonadotrophin releasing hormone’ (GnRH) also in the male. If GnRH is injected into VNO-removed male hamsters or if pheromone-containing vaginal fluid is sampled by the male animal during copulation prior to removal of the VNO, mating behaviours are less affected. Thus prior sexual experience or hormone release induced by VNO activation may produce changes in neuronal circuits in the brain so that mating behaviours can occur in spite of the lack of VNO input.

The Bruce effect

One of the more striking physiological effects that occurs in female rodents and governed by pheromone communication combined with endocrine changes is called the ‘Bruce effect’ named after Dr. Hilda Bruce, the scientist who first described this phenomenon in 1959. Shortly after copulation, if a strange male is introduced inside the cage of a female mouse, the fertilized eggs will fail to implant in the uterine wall and the pregnancy is said to be ‘blocked’.

What is the purpose for such a physiological effect, one might ask? It represents low evolutionary fitness for the new male to raise a pup litter that is not his own, thus he will typically destroy a foreign litter upon birth. It is presumed to be energetically more advantageous for the female not to commit metabolic resources to a potentially destroyed set of offspring, thus the Bruce effect assures spontaneous abortion. The memory of the pheromone chemosignal of the first male is thought to be stored in the accessory olfactory bulb, and since the pheromone of the second (strange) male is not identical, it activates messengers in the female endocrine axis of the brain to deceive physiologically the normal pregnancy implantation processes and the fetus is reabsorbed. Experiments suggest that the pheromonal memories can last up to 30 to 50 days, after which the second (strange) male odour lacks the capacity to prevent egg implantation. This is a very strong, lasting memory, when the mean length of gestation for rodents is considered to be only 22 days.

Acceleration of puberty

Any dog owner will admit that the behaviour of scent marking by urine is a prominent behaviour in mammals to indicate the reproductive receptiveness, the social status and the territory or identity of the donator. Male hamsters and gerbils scent mark near the openings of female burrows or territories to demonstrate their ability for mating. Female mice prefer urine derived from dominant, unfamiliar male mice over that derived from subordinate, prepubertal or sibling mice. In male mice, small molecular weight odour molecules are associated with proteins called ‘major urinary proteins’ (MUPs). Although the precise nature of these odour molecules is not fully understood in most mammals, high circulating concentration of male reproductive hormones called androgens, can stimulate the excretion of MUPs in the male urine. Thus MUP is thought of as a pheromone-binding protein, and molecules that can associate or co-purify with MUP attract female mice and repel other male mice by escalating male-aggressive behaviours. Interestingly, during prepubertal stages, female mice are more attracted to female urinary signals than those contained in male urine, but have preference for male urinary proteins at the time of puberty. In fact, just the presence of MUP or even a small peptide sequence of the amino (N) terminus of MUP may be sufficient to increase the uterine mass in prepubertal females almost 2-fold, a general index of puberty acceleration. Formally, the acceleration of puberty induced by male urinary pheromones is referred to as the ‘Vandenbergh effect’. Group housing female mice in small inclosures can actually elicit the opposite response to produce a delay in puberty, governed by the increase in a small molecular-weight molecule in female urine. Although mice undergo spontaneous oestrus, every 4-6 days, this spontaneity can be blocked, called anoestrus, by group housing, and the blocked cycling can be restored by molecules bound to MUP in male mouse urine. This latter interesting phenomenon is referred to as the ‘Whitten effect’, named after the scientist who discovered this behaviour in 1959.

Vomeronasal Organ Signal Transduction Machinery

Although Ludvig Jacobson made the conjecture two centuries ago that the VNO was a sensory organ, to date we do not know the complete cellular details of how pheromone molecules or chemosignals are received at the vomeronasal sensory neurons and transduced into electrical signals. Cell signalling molecules are often organized in a ‘cascade’, whereby the action of an upstream component, may negatively or positively impose a response in an adjacent or downstream protein or molecule. Since cells use a limited array of known signalling molecules assembled into operational cascades, this narrows the search for potential transduction molecules or machinery that would be available to encode pheromone information. Regardless of the intermediary steps, the end result is an electrical signal that provides information to the brain. Scientists are intrigued to investigate species-specific electrical responses or the neural basis for reproductive or innate behaviours.

Many ligands initiate a cascade of physiological effects by binding to the surface of the cell membrane at specialized proteins called receptors. Protein receptors that capture the binding of pheromone ligands in VNO belong to a large family of receptors known as the ‘G-protein coupled receptors’ (GPCRs). Binding of GPCRs is linked to the activation of ‘GTP-binding proteins’ (G proteins), so named because they bind guanylate triphosphate (GTP), during the process of activation. Activated G proteins cause a conformational change in an effector enzyme that produces one or more additional molecules to increase signalling. These latter molecules are called second messengers, because they act secondary to amplify the first signal. In the VNO, the downstream effector enzyme is ‘phospholipase C’ (PLC), which regulates the production of two second messenger molecules, ‘diacylglycerol’ (DAG) and ‘inositol 1,4,5-trisphosphate’ (IP₃).

Pheromone receptors

The putative pheromone-binding receptors (also called ‘vomeronasal receptors’ (VRs)) have seven transmembrane spanning domains, a structural feature noted in all GPCRs, however, they do not share any of the conserved sequence motifs exhibited by members of the previously characterized olfactory receptors. The two families of VRs are classified as V1R and V2R; the V1R family members share common sequences with taste receptor proteins that transduce bitter tastants, while the V2Rs have a large hydrophobic N-terminal extracellular domain that is similar to the metabotrophic glutamate receptors. The third to fifth transmembrane domain is highly variable in amino acid sequence in members of the V1R family, and thus this site is thought to have the capacity to accommodate different reactive groups of diverse pheromone molecules to provide selectivity of pheromone binding similar to a lock and key. Members of the V2R family, lack this hypervariable region, yet selectivity could be achieved by the pheromone molecule binding to the large N-terminal domain instead. In rodents, VSN that express V2R genes are located in the basal portion of the VNO epithelium, and send their axon projections to the posterior portion of the accessory olfactory bulb. In contrast, VSN that express V1R genes are located in the apical portion of the VNO epithelium and send their axon projections to the anterior portion of the accessory olfactory bulb.

The GTP-binding proteins

When it was discovered that the specialized olfactory G protein called G_olf was not expressed in the VNO, the search for other G-protein subtypes was initiated. Keeping in mind the spatial segregation of VR expression and respective neural projections to the accessory bulb, it was surmised that there can be two populations of neurons in the rodent VNO, sorted by apical and basal position, and that each can have different transduction machinery. This was indeed what was discovered; two different G proteins were expressed in the VNO and were segregated anatomically. The V1R class of pheromone receptors is coexpressed with the G_αi2 G protein in the apical layer of the VNO epithelium, whereas the G_αo G protein is coexpressed with the V2R class of pheromone receptors in the basal layer. It has been postulated that this topographical segregation of transduction cascades in the rodent VNO serves to sort sex-specific versus interstrain recognition behaviours by selective activation of each of the pathways by different classes of pheromones. In other vertebrates, the segregation of VRs and G-protein subtypes cannot be found. In turtle, goat, horse, dog and monkey, for example, only a single type of G protein, G_αi2 G protein, is expressed. There has been a paucity of molecular work performed in these species, hence the class or potential co-localization of VRs in these species is unknown. In animals not belonging to Rodentia, there may be other means for encoding functional specificity. For example, in turtles, there is a sexually dimorphic pattern of G-protein expression where G_αi2 G-protein expression is higher in female VNO than that of male VNO and the expression of several transduction proteins is elevated or cellularly redistributed in coincidence with reproductive maturity.

Calcium ion channels

There are three different ion channels that conduct calcium and which are highly expressed in VNO, namely the ‘transient receptor potential channel’ (TRPC2), the ‘inositol 1,4,5-trisphosphate receptor’ (IP3R), and a ‘calcium-activated nonselective cation channel’ (CaCN). Scientists have used patch-clamp electrophysiology of single vertebrate VSNs stimulated with pheromones to measure the produced electrical signals governed by activation of these channels and receptors. Researchers, jointly with engineers, have constructed multiple assemblies of recording devices, called multielectrode arrays, which offer the advantage of recording populations of VSNs and the electrical interaction between neighbouring VSNs. Similarly, pheromone responses across large populations of VSNs have been visualized through the use of dyes that are excited during the release of calcium, another indication of the function of these channels in the VNO. Despite the available technology, the precise nature of the cascade events between production of the messengers and movement of calcium ion through one or more of these channels is currently not understood but is under active investigation. Complicating the identification and regulation of the cascade, is the recent finding that ion channels and receptors do not often act in isolation, but are a component of multiple interacting proteins that regulate each other through a scaffold of assembled proteins in close proximity or through tight interactions. For example, the IP₃ receptor has been found to be linked to the TRPC2 channel in the rat VNO through a ‘protein-protein interaction’, which may use an accessory protein Homer as the means for forming the complex. This leads to further questions, such as, is the flow of calcium through the TRPC2 regulated by formation of the complex or does the flow of calcium through the TRPC2 channel activate a secondary channel (CaCN) to trigger the action potential? The regulation of the gating or opening of the TRPC2 channel in the presence of interacting proteins and regulatory ions may clarify the pause in understanding of which protein components present in the VNO produce the primary electrical event, and which components modulate the event to encode information contained in the pheromone molecule.

One manner in which to tease apart the elements of the transduction cascade is to delete the expression of the protein in mice. These transgenic animals are referred to as ‘knock-out mice’ for the manipulated gene of investigation. Although there can be built in compensatory mechanisms for the loss of a gene product, which can confound interpretation, these knock-out mice can identify the degree of involvement for a VNO transduction protein through loss of function. For example, V1R knock-out mice (a subset of this family was knocked-out) do not electrophysiologically respond to pheromone ligands that mediate puberty acceleration or delay, but still respond to those that mediate intermale aggression and the timing/ induction of oestrus. Female V1R knock-out mice exhibit a reduced level of maternal aggression to strange males and male V1R knock-out mice emit ultrasonic vocalizations to females quite normally but do not attempt to mount them as frequently as the control or wild-type mice. TRPC2 knock-out mice appear to produce a different deficit in VNO function that does not precisely match the effects of VNO surgical removal but signifies the importance of this VNO ion channel. Although electrophysiological responses to pheromones can be elicited from the VSNs of these transgenic mice, responses are only induced at extremely high and unphysiological concentrations of pheromones. Responses cannot be elicited through application of the second-messenger DAG as is observed for control or wild-type mice. Male TRPC2 knock-out mice fail to exhibit male aggression towards strange (nonsibling) males and they participate in courtship displays and sexual mounting behaviours with both male and females – they can no longer discriminate gender.

Do Humans Detect Pheromones?

The answer to this question depends upon one’s accepted definition of pheromone, the more common location of detection, and is always sure to spark heated debates! Most anatomists judge that both the accessory olfactory bulb and the VNO are vestigial after the third month of human gestation. In human autopsies that reveal a pit-like VNO structure in some individuals, neuroscientists cannot find axons that might lead sensory information to higher brain centres and the cells contained in the pit do not express OMP. Geneticists have discovered the presence of human genes that code for VR and TRPC2; however the genetic sequences indicate that the DNA could not be transcribed to functional receptor or channel proteins; in genetic terms these human genes are nonfunctional pseudogenes. There is one VR gene that appears to be intact but is expressed in the MOE not VNO, and it is not clear if it has function. Despite these negative predictions, there are interesting reports of pheromone-based behaviours in humans. Odourless armpit compounds taken from women in the later stages of their menstrual cycle, can accelerate both the spiking of ‘luteinizing hormone’ (LH) responsible for ovulation and shorten the menstrual cycle of recipient women. A human pheromone contained in body secretions thus may be a mechanism for synchronized menstrual cycles observed in women who workout or live together. A compound called androstadienone may represent another human pheromone and has been demonstrated to subconsciously alter brain glucose metabolism in areas of the brain not typically associated with olfaction, but affecting psychological state. Another class of compounds, called vomeropherins, have been described to alter respiratory frequency and heart rate. Perhaps human intuitive impressions may be a mere result of pheromones used to communicate the genetic, hormonal, metabolic, dietary and social cues serving to shape reproductive fitness, but whose primacy has been lost through lack of evolutionary pressure.