Science Watch

Babies' developing brains have a lot of responsibility. Besides soaking up the outside world and keeping their bodies away from danger, they also contend with environmental factors that can influence them for the rest of their lives. Bacterial infections, nutritional deficiencies and maternal interaction all change the way a baby's brain genes express themselves, even decades later—a concept known as perinatal programming.

In the late 1980s, physician and epidemiologist David Barker, MD, PhD, discovered that babies with low birth weight were more likely than normal-weight babies to develop coronary heart disease later in life. At the time, the idea that prenatal or infantile experiences could influence adult diseases was unheard of. But as evidence mounted, physicians and researchers gave in to the idea that early-life environments could program later physiological changes.

As the effects of perinatal programming have expanded beyond these ailments into behavioral effects on maternal care and memory, they underscore the idea of developmental plasticity, the concept that more than one phenotype can shake out of a single genotype depending on the environment it faces.

In fact, as many as half of adult diseases may have their origins in development, many researchers believe. By better understanding these developmental roots, researchers, physicians and therapists will be better positioned to perhaps prevent or delay the development of disorders, says Christina Williams, PhD, a behavioral neuroscientist at Duke University.

Sick minds

One such disorder is memory impairment brought on by childhood bacterial and viral infections. A few years ago, scientists noticed this correlation but were at a loss to explain it. Now, perinatal programming is emerging as a likely explanation, says Staci Bilbo, PhD, a developmental psychobiologist also at Duke, who pioneered some of the first infection-memory loss studies.

In a 2006 study in Behavioural Brain Research (Vol. 169, No. 1), she and her colleagues injected 4-day-old and 30-day-old rats with E. coli. She injected a control strain with a saline solution. Then she waited for the two groups to grow up. When they were 3-month-old healthy adult rats, Bilbo injected them with the harmless molecule lipopolysaccharide (LPS), which is found on the cell membranes of E. coli. Both the E. coli-exposed and control rats' immune systems sprang into action, mistaking LPS for actual E. coli and triggering inflammation. A day later, after the inflammation subsided, the rats were placed in a cage with an electrified floor and immediately given a shock.

The next day, they were placed back in the same cage. A little more than half of the control rats froze in anticipation of the coming shock. The results were about the same for the rats exposed to E. coli when they were 30 days old. But with the neonatal E. coli-exposed rats, only about 15 percent froze. They seemed to be forgetting their previous-day experiences, Bilbo says.

This suggests to Bilbo that there's a critical window in rat brain development. Because the neonatal period is one of immense brain growth and development, an infection during that time can have long-lasting effects on memory. In particular, the early infection may alter the development of baby rats' hippocampi, the brain region that controls learning—especially learning about new environments, such as the electrified cage. Bilbo explains that the hippocampus is highly populated by proteins, known as cytokines, which help cells coordinate the body's immune system response. If the hippocampus is infected early enough, she suspects, the infection can permanently alter its cytokine production. Then if the rat is infected again, its overactive cytokine production could interfere with its hippocampus's memory formation. By looking at which cytokines are highly active during these immune responses, researchers have fingered one in particular known as interleukin-1 beta (IL1b) as the most likely offender. Recently, Bilbo tested her hypothesis further by seeing whether an enzyme, caspase 1, which disrupts IL1b production, could prevent infection-related memory loss. She repeated her previous experiment but this time injected a strain of neonatal E. coli-infected rats with caspase 1. These rats froze when placed in the electrified cage just as frequently as control mice.

Human brains may be similarly susceptible to early infections—and E. coli is the second most common cause of early infection in U.S. newborns. "Modern medicine has done a great job increasing survival in these populations," says Bilbo, "but the incidence of neurodevelopmental disorders in these individuals, including memory disorders, is extremely high."

Unfortunately, a solution isn't as easy as knocking out IL1b early in life; that would cripple an immune system. But Bilbo hopes that by further exploring how increased cytokine production blocks the formation of new memories, researchers will be able to develop targeted treatments and preventions.

Paul Patterson, PhD, a biologist at the California Institute of Technology in Pasadena who specializes in immune system research, says that Bilbo's line of work is "very worthwhile." His own work dovetails with Bilbo's: He is examining how pregnant women's infections can influence their children's development. Women who get the flu during pregnancy are more likely to have children who develop schizophrenia and autism, Patterson says, and that could tie in with the same kind of cytokine response suggested in Bilbo's work (Journal of Neuroscience, Vol. 27, No. 40).

Like Bilbo, he's so far only seen results in mice, but his recently begun monkey trials will provide a clearer picture of whether the results apply to humans, too. The sooner the better, he says, because both his and Bilbo's work could give scientists greater insight into how our childhoods influence the rest of our lives.

"Since childhood infections are so frequent, it's extremely important to investigate their long-term effects on the brain," Patterson says.

Memory for a lifetime

Infections make the brain more vulnerable to memory loss, but nutrition can program it to resist that damage, Williams says. For more than a decade, Williams and her collaborators have focused on the developmental effects of the nutrient choline on memory and neural plasticity.

"Most of the choline we use in our bodies and brains comes from our diet," Williams says.

It's found most abundantly in egg yolks, soy and liver. Choline is necessary for many biological processes, she says, including producing the neurotransmitter acetylcholine, which is important for memory formation. Because of variations in choline in mothers' diets and genetic variations in choline metabolism, fetuses may be exposed to different levels of choline in utero, Williams says, which can lead to great variability in choline levels among humans.

In an experiment, Williams found that the adult offspring of mother rats that were fed extra choline—about 4.5 times the normal level—during a critical period of brain development showed increased memory capacity and precision in a maze navigation task. But what astounded Williams was that this memory improvement lasted into old age. As the other rats grew older and experienced typical age-related memory loss, the extra-choline cohort's memories stayed as sharp as ever.

More recent studies show that increased production of new neurons in a hippocampal region called the dentate gyrus accompanies this improved memory (Brain Research, 2008, Vol. 1237). In normal-choline mice, this area stops making new neurons as they get older. In the extra-choline mice, though, new neurons continue to be born in this region throughout life, although it declines in old age. Surprisingly, Williams says, it doesn't take much choline to produce these effects.

"Amazingly, rats that were supplemented with choline for only seven days during fetal life continue to show higher rates of neuron proliferation even in very old age," she says.

All you need is love

Less tangible influences also produce significant changes in behavior, says behavioral psychologist Frances Champagne, PhD, at Columbia University. Her work looks at the developmental impact of a mother's love: specifically, how maternal licking and grooming in rats affects that behavior in later generations.

Rats that are frequently licked and groomed as pups have lower stress, are more social, have greater cognitive skills and are more responsive to reward motivations than those that don't get much motherly affection, Champagne says. The rats seem to pass on the love, or lack thereof, to their own offspring.

But so far it's been unclear why that is. Even in genetically similar rat families, there are huge differences in licking and grooming behavior among them, she says. What's more, if you take a young female rat from a less-affectionate family and raise it with a highly affectionate mother, the young rat will grow up to engage in more licking and grooming than her sisters that were left with their original family. So the answer is likely to lie beyond simple genetics and points to alterations in gene expression, Champagne says.

Champagne thinks she finally has the answer in oxytocin, one of the hormones that encourage maternal behavior. All rat mothers produce oxytocin, but in negligent rat mothers, the hormone isn't able to communicate with their brain cells' DNA. Sometimes an organic compound called a methyl group—which normally helps regulate gene expression—gets in the way by attaching itself to the DNA first.

In a study published in Endocrinology (Vol. 147, No. 6), Champagne analyzed the methyl levels of various rat families and compared their licking and grooming behaviors. Sure enough, she found much higher methylation in the rats that showed little maternal care and lower methylation in the affectionate rat families.

It's unclear exactly how licking and grooming prevents methylation, but Champagne thinks it goes something like this: Maternal grooming triggers the production of proteins called growth factors. These growth factors bind to the same DNA regions as methyl groups, preventing the methyl groups from shutting those genes down.

"Essentially, it's a competition between growth factors and methylation in the genes," she says.

Because methyl groups and growth factors weave themselves into the rats' DNA, those effects pass down to their progeny.

"As generations go on, it becomes a self-perpetuating pattern," Champagne says.

That's an intriguing finding, Williams says, because it illustrates how early environmental influences carry over to offspring's behavior outside of genetics or learning alone. If there are similar effects in humans, it could help explain the very diverse range of human phenotypes given our relatively similar genotypes.

"It's the coming together of nature and nurture," Williams says. "Or rather, how nurture can become nature by programming permanent changes in the biology of the brain."