12

  TRANSGENERATIONAL  DEVELOPMENTAL  PLASTICITY – AN EPITOME  OF  EVOLUTIONARY  CHANGE

 

The epigenetic information for transgenerational developmental plasticity originates in the CNS.

 

 

Metazoans sometimes respond to specific environmental stimuli by inducing adaptive phenotypic changes in their offspring. Such cases of transgenerational developmental plasticity (TDP) are, essentially, not different from evolutionary changes, hence the study of the TDP is of paramount importance in understanding the nature and mechanisms of the evolutionary change. Available evidence shows that induction of new phenotypic characters in the offspring during TDP involves no changes in genes (it affects the whole population exposed to the stimulus, takes place within one/a few generations). Based on the empirical evidence on the transgenerational developmental plasticity, now we are able to reconstruct the basic mechanism of the sudden induction of new phenotypic characters in the offspring. In a generalized form, that mechanism looks as follows: the specific environmental stimulus is received by sensory neurons, converted into electrical spike trains that are transmitted for processing in specific neural circuits. The processing, which is a computational non-genetic process, generates epigenetic information in the form of a chemical output (neurotransmitter, neuromodulator, etc.) that is released on specific secretory neurons. This stimulates the neuron to secrete a neurohormone that starts the signal cascade for deposition in the gamete(s) of a parental factor that determines development of the new phenotypic character in the offspring.

 

 

Examples of developmental plasticity considered in the previous chapter represent discrete adaptive changes taking place within the lifetime of metazoan organisms in response to specific or stressful environmental stimuli, such as changes in the physical and social environment, changes in the food resources, presence of predators, etc. These phenotypic changes do not affect the offspring in the absence of the inducing factors that triggered their appearance in parents. In the classical biological terminology all of them represent “acquired characters”. Conventionally, the inability  to transmit acquired characters in the offspring has been explained with the fact that no changes in genes and genetic information encoding these characters have occurred. Hence, all of them are evolutionarily irrelevant.

However, an ever-increasing number of biologists believe that, in many cases, the developmental plasticity is a component of evolutionary process (Thompson, 1991) or a stage in the process of the evolution of morphological innovations and novelties (West-Eberhard, 2003).

The fact that the evolution of animal morphology, by human lifespan standards takes very long periods of time, impedes attempts to directly observe changes in the developmental mechanisms enabling evolution of morphological innovations in metazoans. One empirical way to overcome this difficulty is to look for possible short-term biological phenomena where similar mechanisms may be working. Fortunately, it seems that such a phenomenon exists and its study may contribute to the understanding of the nature, mechanisms, and origins of the evolutionary events (Cabej, 2004g). This is the phenomenon of transgenerational developmental plasticity where offspring displays and transmits to the next generation new characters that its parent(s) or close predecessors lacked.

Both the evolutionary change and  transgenerational developmental plasticity consist in appearance in the offspring, and transmission to the future generations, of phenotypic (behavioral, morphological, physiological and life history) characters that the the parents did not inherit but may (or may not) have acquired during their lifetime. The similarity of results (heritable transmission of new characters to the offspring) may suggest that mechanisms and the source of information for evolutionary change and transgenerational developmental plasticity are similar. For metazoan evolution is inherently too parsimonious to afford for developing two separate mechanisms for achieving the same result.

Being experimentally accessible and transgenerational, this form of developmental plasticity certainly transcends the field of alternative phenotypes to account for the still little known mechanisms of evolutionary change.

In view of the fact that any heritable change in a phenotypic character requires investment of new information, and in the case of transgenerational plasticity no changes in genes are involved, the question arises: Where does the new information for transgenerational phenotypic changes come from?

Let’s briefly review the experimental and observational evidence showing that metazoan organisms, in response to specific stressful environmental stimuli can not only adaptively change their phenotype (behavior, morphology, physiology, and life history) but also transmit phenotypic changes to the offspring.

 

 

Transgenerational Developmental Plasticity - Inherited Changes without Changes in Genes

 

Transgenerational Developmental Plasticity in Nature

 

The first report of an experimentally induced transgenerational change in morphology was described in 1934 by Richard Woltereck. By cultivating in Italy, a native Denmark strain of Daphnia cucullata, he obtained a Dauermodifikation (Germ. long-lasting modification): within a number of generations, the crustacean changed its morphology and transmitted the morphological change, in the absence of the inducing factor, to the offspring for 40 generations before reverting to its ancestral form (Woltereck, 1934).

When Daphnia cucullata, olfactorily perceives its predator’s kairomones in the environment, it doubles the size of its own and of its offspring’s (F₁) helmet, which makes it less vulnerable to the attacks of the predator (Agrawal et al., 1999). Under laboratory conditions, on visually perceiving small-scale turbulence, the crustacean also develops extremely large helmets, more elongated than those induced by the predator kairomone ( Laforsch and Tollrian, 2004). In J. Maynard Smith’s interpretation:

 

The change in morphology is adaptive; it occurs in response to an environmental stimulus; and once it has occurred, it is transmitted through the egg. Almost certainly, it is caused by changes in activation and not by changes in the base sequence” (Maynard Smith, 1999).

 

It is also reported that, in response to stressful external stimuli, such as fish smells and food shortage, Daphnia reduces the body size of its own as well as that of the offspring (Hanazato et al., 2001).

On detecting the presence of the predator midge (Chaoborus flavicans) larvae, Daphnia pulex develops a neck spine containing several teeth, delays its reproductive maturity and increases the body size of the offspring. In the presence of another of its predators, the water bug, Notonecta glauca, it produces offspring of smaller body size that reach reproductive maturity at a younger age (Lüning, 1992). In both cases, the inherited change is adaptive and, obviously, involves no changes in genes.

Under unfavorable environmental conditions, Daphnia magna, an all-female asexually reproducing species, gives birth to a sexually reproducing generation (male + female individuals). The latter produces freezing- and desiccation-resistant eggs, which only hatch into male individuals. The transgenerational change is also experimentally induced by administration of insecticidal juvenile hormone analogs, pyriproxyfen and methoprene (Olmstead and LeBlanc, 2003), and by exposing maternal organisms with maturing oocytes in the ovaries to methyl farnesoate (Rider et al., 2005; figure 12.1).

In response to crowding, amictic female rotifers of the cyclically parthenogenetic species, Brachionus angularis, give birth to mictic daughters (sexually reproducing generation), which produce haploid eggs developing into males or, when fertilized, developing into diapausing eggs. Amictic females hatching from diapausing eggs respond very slowly to crowding, to reach the normal sexual reproduction response only after several generations (Schröder and Gilbert, 2004; Stelzer and Snell, 2003; Gilbert and Schröder, 2004). The crowding chemicals, density-dependent kairomones released by B. angularis, like other kairomones, act neurally, and may target a maternal tissue, possibly the nervous system, which then stimulates some oocytes to differentiate into mictic females (Gilbert and Schröder, 2004).

 

 

Figure 12.1. Diagrammatic representation of the putative juvenoid signaling pathway (From Rider et al., 2005).

 

Another small rotifer, Brachionus calycifloris, when detects the presence in the environment of its predator, Asplanchna brightwelli, produces offspring with an additional pair of spines, which protects them from being eaten by the predator. The character is transmitted for many generations (Gilbert, 1966). The predator itself produces different morphs according to the quantity of prey it perceives in the environment (Gilbert, 1980). It is believed that both the predator-induced defense in the first case and the prey-induced polyphenism in the second, are transmitted to the offspring via some maternal cytoplasmic factors but the factor(s) and the steps of transmission of the information on the presence of predator from the animal’s CNS to its eggs are still unknown. However, we can safely assume that whatever the route might be (visual, olfactory, or via the interoception), the sensing and/or perception of the predator/prey takes place in the CNS, implying that there is a causal chain extending from the CNS to the synthesis and deposition in the egg of the maternal cytoplasmic factor.

The offspring of the cotton aphid, Aphis gossypii, consists of four different phenotypes: alatae (winged), normal light green apterae (unwinged), normal dark green apterae (unwinged), and dwarf yellow apterae (unwinged) each with distinct life histories. When on predator-free plants, apterous and alatae aphids produce primarily offspring of apterous and alatae types respectively, but under increased predation risk (for example, when exposed to search track of the ladybird beetles, Hippodamia convergens) the aphids produce a greater proportion of alatae for several generations (Mondor et al., 2005).

When attacked by its predators or when visually detects their presence, the pea aphid, Acyrthosiphon pisum (Harris, 1776) emits a volatile compound, the alarm pheromone, (E)-beta-farnesene (EBF), which is the only volatile compound emitted by these aphids in the presence of their predators (Kunert et al., 2005). The synthesis of the alarm pheromone is induced by pheromonotropic neurohormones (Masler et al., 1994). The alarm pheromone is received by olfactory neurons and perceived in the brain of conspecific aphids inducing an immediate behavioral change, walking or dropping off the plant. In response to the alarm pheromone or to the presence of predators (ladybirds, hoverfly larvae, lacewing larvae, etc.), these aphids also increase the proportion of winged individuals in the offspring (Dixon and Agarwala, 1999; Kunert and Weiser, 2003). This shift in morphology does not protect the prey from the predator, but allows it to avoid the predator by flying to other plants (Weisser et al., 1999).

Recently, investigators have demonstrated that exposure of mother aphids to the alarm pheromone alone also increases the proportion of winged morphs in the offspring (Kunert et al., 2005; Podjasek et al., 2005). A. pisum also produces a greater proportion of alatae individuals when exposed to its natural parasitoid enemy, Aphidius ervi, but it does not if is parasitized (Sloggett and Weisser, 2002).

There is another impressive example of transgenerational developmental plasticity, that is considered to be a genuine case of speciation by many authors. It comes from experiments conducted by G. Kh. Shaposhnikov with the parthenogenetic aphid, Dysaphis anthrisci majkopica reared on the unsuitable host plant, Chaerophylum maculatum, leading to almost 100% death rate. After 8 asexual generations, however, a proportion of aphids changed their morphology (their body size and rostrum), became able to survive and adopt as their host the ex-hostile plant (Shaposhnikov, 1965).

In response to deterioration of food quality, crowding, and other adverse factors, female individuals of some parthenogenetic aphids, such as Sitobion avenae Fabricius (Watt et al., 1981), Acyrthosiphon pisum (Shaposhnikov, 1965), and rosy apple aphid, Dysaphis devecta (Forrest, 1970), switch to production of alatae (winged) offspring, but in absence of these factors gives birth to apterae (wingless) offspring. The change has adaptive character.

Another aphid, Aphis fabae, experiences very high mortality when reared on the garden nasturtium, Tropaeolum majus. However, after 3 generations of rearing on this inhospitable plant, the aphid is able to use that plant as its specific host.

 

Phase Transition in Locusts

 

Under natural conditions, the locust species Schistocerca gregaria (Forskål) and Locusta migratoria, occur in one of two behaviorally and morphologically alternative and reversible states, known as solitary sedentary and gregarious migratory phases.

Most of the time in the wild, these locusts are in the solitarious phase with a population density of ~3/100m². When solitarious larvae of these locusts experience crowding, they dramatically, within 4 hours, change their behaviour from the tendency to stay away from, or avoid, other locusts (solitarious phase) to a strong bias of associating with other conspecifics (gregarious phase), reaching a population density of up to 100,000/100m².

These locusts have an extremely strong ability that, in response to a social cue (living alone or in crowd with other conspecifics), an olfactory, visual, tactile, or auditory stimulus, to switch between the above two phases. This phenomenon of phase transition is characterized by a number of behavioral, physiological and morphological changes, including several changes in body color, head size, wing size, the number of sensilla, etc., as well as morphometric and behavioral characters. All these new characters are maternally transmitted to the offspring.

Transition from the solitary to the gregarious phase is also associated with a change from the cryptic green to dark coloration. Administration of the brain neurohormone, [His⁷] corazonin changes to dark the body color of the albino nymphs. Gregarious locust nymphs, also switch to the solitarious phase, with all the accompanying behavioral and morphological changes in reverse, when are kept isolated.

From field observations it has been concluded that the color of locust hatchlings is determined by the degree of crowding the locust mothers have experienced rather than by volatile chemicals in pods (Bouaichi and Simpson, 2003). A factor from accessory glands of the adult females determines the maternal inheritance of gregarious behavior; by washing eggs of gregarious locust females with a saline solution, a solitarious offspring is produced, but gregarious behavior may be restored by administration of extracts of gregarious females (Hägele et al., 2000).

A crucial proximate inducer of transformation of the solitary form into the gregarious form is JH (juvenile hormone). The hormone has a significant effect on the external coloration of locusts (Tawfik et al., 1997; figure 12.2). In this context, it is important to remember that the JH secretion by corpora allata in insects is under strict cerebral control of stimulating neuropeptides (allatotropins) and inhibiting neuropeptides (allatostatins) secreted by secretory neurons in the insect brain (Stay et al., 1996) as well as via nerves innervating corpora allata (nervi corpori allati I and nervi corpori allati II) (Kou and Chen, 2000). 

Figure 12.2 . Photographs showing the effect of [His⁷] corazonin on the body color in L. migratoria (A and C) and S. gregaria (E and H). Locusts were injected with the hormone in 4 µl of oil or with 4 µl of oil alone at day 3 of the third instar and photographed 2 days after the subsequent ecdysis. Hormone was injected into L. migratoria and S. gregaria, respectively. (A) Oil-injected crowded albino nymph. (C) Hormone-injected crowded albino nymph. (E) Oil-injected isolated nymph. (H) Hormone-injected isolated nymph (From Tawfik et al., 1999).

Besides the above controls via the brain-corpora allata axis (JH secretion) and innervation of corpora allata, the CNS exerts a humoral control on the process of phase transition by secreting in hemolymph the neurohormone DCIN (dark color-inducing neurohormone), a 11-amino acid residues long peptide, also known as [His⁷]-corazonin (Grach et al., 2003), whose primary structure, to a large extent degree, is similar to the vertebrate melanophore-stimulating hormone. Administration of [His⁷]-corazonin in solitary nymphs of both sexes induces morphometric changes, characteristic of the gregarious form, especially F/C (length of the hind femur/head width) and E/F (length of fore wing/length of the hind femur) ratios (Maeno and Tanaka, 2004). Administration of [His⁷]-corazonin in nymphal solitary locusts also induces a gregarious type of reduction in the number of antennal sensillae at the onset of adulthood (Maeno and Tanaka, 2004). Injection of the neuropeptide [His⁷]-corazonin alone, or combined with JH, in migratory locust instars (Tanaka, 1996; Tanaka, 2000a; Tanaka, 2000b; Yerushalmi and Pener, 2002) produces various patterns of coloration, depending on the time of administration and the dose. Differences between the two phases are also observed in the number and amount of neuropeptides secreted by corpora cardiaca in the hemolymph, suggesting that these neuropeptides may also play a role in the transition to the gregarious phase (Clynen et al., 2002).

Summarizing the above evidence, it may be said that besides the brain neuropeptide [His⁷]-corazonin and corpora cardiaca neuropeptides, which communicate to the target tissues messages for phase transition, JH also is a conveyor of CNS gregarizing messages.

Aggregation behaviour in the desert locust Schistocerca gregaria, is elicited by aggregation pheromones. Of special interest is to know how the external gregarizing stimuli are related to secretion of neurohormones ([His⁷]-corazonin in the brain and various neuropeptides in corpora cardiaca) and to neural signals (allatostatins, and allatotropins), which regulate JH synthesis/secretion.

At this juncture, for our purpose, it is helpful to switch to the reverse approach, i.e., to follow the chain of events in their natural course, as they unfold under the influence of the external stimuli. External (visual, olfactory, auditory, or tactile) stimuli are received by the sensory neurons (exteroceptors), converted into specific electrical spike trains, and as such transmitted to the respective neural circuits in the central nervous system, via interneurons. By processing the input of electrical signals, neural circuits generate their chemical output, mostly in the form of neurotransmitters and neuromodulators, which stimulate secretory neurons to release neurohormones [[His⁷]-corazonin, allatotropins and allatostatins].

A recent study shows that vast differences exist in the amounts of 11 out of 13 analyzed potential neurotransmitters and neuromodulators in the brain and the thoracic nerve cord of nymphs and adults of the solitary and gregarious phases. Crowding of solitary larvae leads to some drastic changes in the levels of neurotransmitters that are characteristic of the long-term gregarious forms (Rogers et al., 2004).

In the case of the desert locust, Schistocerca gregaria (Forskål), the olfactory stimuli (aggregation pheromones) are converted into electrical signals, which are transmitted to interneurons of the frontal antennal lobe for further processing and then to the mushroom body and the lateral protocerebrum. These interneurons respond with different specificity to different pheromone stimuli (Anton and Hansson, 1996). The CNS response to aggregation pheromones also depends on the presence of JH as it is demonstrated by the fact that in allatectomized or old individuals, which do not secrete JH because their corpora allata are atrophied, the neural circuits do not respond to the pheromone (Ignell et al., 2001). Tactile stimuli (touch on the outer side of the upper portion of a hind leg, for instance) from mechanosensory trichoid sensillae on the hind limb, via metathoracic nerve 5, are transmitted to the CNS (Rogers et al., 2003). Tactile stimuli induced in head hair receptors by wind are transmitted to specific interneurons and then to thoracic motor centers inducing flight behavior in gregarious locusts (Ayali et al., 2004).

Many studies have shown that phase transition in locusts is also related to differences in the neuronal function. Analysis of responses of the descending contralateral movement detector interneuron have shown that in solitary locusts habituation to approaching objects is five times stronger than in gregarious locusts (Matheson et al., 2004). Changes related to behavioral phase transition are observed especially in the activity of flight-related neurons. So, e.g., TCG (tritocerebral commissure giant) interneurons have weaker wind-induced spiking activity in solitary locusts than in gregarious locusts and the spontaneous activity of the TCD (tritocerebral commissure dwarf) interneuron in darkness is significantly weaker in solitary than gregarious locusts (Fuchs et al., 2003).

The darker color of gregarious locusts results from deposition of melanin in the cuticle, which is cerebrally regulated by secretion of  PBAN (pheromone biosynthesis-activating neuropeptide), also known as MRCH (melanization and reddish coloration hormone). The neurohormone is also known to regulate secretion of sex pheromones in moths (Altstein et al., 1996).

While the olfactory (gregarizing pheromone) and visual (crowding) stimuli can induce gregarious behaviour when combined, tactile stimuli per se, uncombined with other stimuli, are capable of inducing that behavior even when applied to individual locusts (Rogers et al., 2003). Repeated mechanical stimulation in the hairy femoral region leads to transition of the solitary locusts to the gregarious behavior within a few hours. Electrical stimulation of the metathoracic nerve 5 is very effective in inducing gregarious behavior. However, its branches, the nerves 5A and 5B are less effective in that respect. Electric stimulation of the 5B nerve branches, 5B1 and 5B2, can induce gregarious behavior when applied on both nerves simultaneously but not separately (Rogers et al., 2003; figure 12.3).

 

Figure 12.3. Ventral view of the thorax and proximal hind leg dissected to show the metathoracic ganglion, nerve 5, its principal branches and the extensor tibiae muscle. Values in bold are median P_sol values obtained after electrical stimulation at each location. Nerve 5A runs between two muscles in the coxa, hence its apparent sudden termination (From Rogers et al., 2003).

The neuronal circuits that integrate mechanosensory gregarizing stimuli should combine exteroceptive signals from the anterior surface of a hind femur with a specific proprioceptive signal that naturally results from the inward displacement of the leg on contact with another locust. These neuronal circuits are the first central elements of the pathway that initiates the rapid and widespread neuronal plasticity that underlies the behavioural phase change. (Rogers et al., 2003)

Electric stimulation of the nerve 5B causes gregarization but stimulation of neither 5B1 nor 5B2 alone leads to gregarization. The nerve 5B1 innervates tactile hairs on the anterior face of the femur and is provider of the exteroceptive input, whereas the nerve 5B2 innervates hairs on the posterior face and is the main source of the proprioceptive input. Only combination of the exteroceptive and interoceptive input serves as gregarizing signal (Rogers, 2003; figure 12.4).