Scientists have little trouble accepting the idea that ontogeny is a process rich in interplay between genetic and environmental factors, or that environmental insults ranging from malnutrition to toxins during the embryonic and early postnatal periods can harm development. But what about the argument that environmental insults suffered early in life could predispose the person to adult brain deficits, or even neurodegenerative disease, without any obvious impact on normal development? Most scientists give little credence to talk of such far-reaching consequences. And yet, several articles published in recent weeks on perinatal infection and memory, on lead exposure and APP expression, make enough of a case to stimulate serious thought.

Perinatal Programming and Memory Loss
The notion that perinatal insults can predispose organisms to later brain dysfunction is not at all new. For example, a large body of epidemiologic evidence links infection, fetal malnutrition or hypoxia, and other obstetric complications to schizophrenia. But schizophrenia, which typically appears clinically in late adolescence or early adulthood, can plausibly be classified as a developmental disorder. Linking perinatal environmental perturbations to disorders of later adulthood, or even neurodegenerative disorders, would seem to be a larger leap.

Researchers in cardiovascular and hypothalamic-pituitary-adrenal (HPA) axis disorders have been making this connection for many years. A seminal study by David Barker and colleagues at England’s University of Southampton (Barker et al., 1989) demonstrated an inverse relationship between birth weight and the incidence of cardiovascular disease in a cohort of British men. The Barker hypothesis, as it has come to be known (for review, see Barker, 2004), proposes that the perinatal environment “programs” the fetus or young animal for later susceptibility to disease. A large body of subsequent clinical and experimental data have supported this hypothesis, particularly for intrauterine nutritional effects on disorders of the cardiovascular system and HPA axis.

Several groups have shown evidence in animals that perinatal pathogen exposure, and subsequent temporary inflammation, permanently reorganize physiological systems such as the peripheral immune system and the HPA axis. The notion that pathogens (e.g., Chlamydia pneumoniae bacteria or herpes simplex virus) play a role in AD has been a minority opinion for some time; however, this research focuses mainly on pathogen infection during adulthood (see ARF Live Discussion). By contrast, perinatal programming suggests that pathogen exposure during development effects a change in physiology or gene expression that persists long after the body has dealt with the original infection.

In a paper published in the February issue of Behavioral Neuroscience, Staci Bilbo and colleagues at the University of Colorado, Boulder, tested the hypothesis that perinatal exposure to an infectious agent can affect memory processes by changing how the nervous system responds to a later immune challenge. Bilbo and colleagues exposed rats to E. coli on postnatal day 4, then tested their memory in adulthood with a fear conditioning test that required the animals to learn to anticipate an electric shock. Rats exposed to E. coli as pups learned just as readily as control animals that being placed in certain test chamber meant a shock might be coming. However, in a second experiment, some of the rats were exposed to a "fake" infection with bacterial lipopolysaccharide (LPS), eliciting an immune response before they were exposed to the shock. These rats subsequently failed to recognize the chamber in which they had been shocked. Control rats who were infected with LPS before the testing, but who had not been exposed to E. coli as neonates, were not impaired in learning this task.

What could have changed in the neonates exposed to E. coli? The impaired process was that of memory consolidation, making hippocampal alteration a prime suspect. The authors found that neonatal pathogen exposure resulted in decreased numbers of hippocampal astrocytes in adulthood, but increased hippocampal astrocyte reactivity and decreased brain IL-1β levels following adult LPS exposure. "Taken together, these data suggest that neonatal infection facilitated the ability of a subsequent immune challenge in adulthood to interfere with memory consolidation," Bilbo and colleagues write.

Among the ways by which an infection might acquire such a long reach, Bilbo and colleagues mention possible changes in neuronal or synaptic architecture, citing in particular the work of Ben Barres and colleagues at Stanford University. And it so happens that Barres has a new study that offers up a candidate mechanism for astrocytic involvement in synaptic development just after birth.

Writing in the February 11 issue of Cell, Barres's group, along with collaborators at several other institutions, reports that thrombospondin (TSP)-1 and -2, extracellular matrix proteins secreted by astrocytes, are critical, and sufficient, for the proper assembly of synapses in an in-vitro preparation of rat retinal ganglion cells. Interestingly, the thrombospondins control only the assembly of ultra structurally normal and presynaptically active synapses. The last step, generating postsynaptic activity, depends on other, as yet unidentified, astrocyte-secreted factors.

APP Transcription Spike Long After Lead Exposure?
Falling even closer to the immediate interests of AD researchers was a paper appearing in the January 26 issue of the Journal of Neuroscience. Nasser Zawia of the University of Rhode Island in Kingston and colleagues have produced a body of work on the effects of metals on gene expression, especially during pre- and postnatal development (for review, see Zawia, 2003). In their present study, this group, with collaborators at Indiana University in Indianapolis, tested the hypothesis that postnatal exposure to lead (Pb) could have implications for AD and Down syndrome, a neurodevelopmental disease that involves AD-like amyloid deposition in early adulthood. First author Riyaz Basha and colleagues exposed rat pups to 200 ppm Pb-acetate in drinking water for the first 20 days of their lives. Relative to untreated animals, this produced a transient rise in APP expression, and, more surprisingly, a second, later wave of increased APP expression at 20 months of age. Expression of control genes did not change. Gross phenotypical characteristics such as body weight or survival rate at 20 months were not affected by the postnatal Pb exposure.

Looking for possible mechanisms for this delayed APP effect, Basha and colleagues used a microarray screen of transcription factors and found that Pb induced high levels of the APP transcription regulator specificity protein 1 (Sp1). There is evidence that Sp1 promotes production of APP, perhaps via BACE (Pollwein et al., 1992; Querfurth et al., 1999; Christensen et al., 2004; Ge et al., 2004). Basha and colleagues found that Sp1 DNA binding in the cortex of Pb-exposed animals correlated over time with levels of APP transcription. Indeed, Sp1 also showed a "startling rise in activity" 20 months after the Pb exposure. By contrast, none of these effects were seen following adult Pb exposure between the ages of 18 and 20 months.

Next, the researchers tested whether Pb might be inducing APP upregulation via Sp1 in PC12 cells. They found that Pb stimulated both activity of a transfected human APP promoter and Sp1-DNA binding, while reducing Sp1 activity with short, interfering RNAs reduced Pb effects on the APP promoter.

"[O]ne interpretation of these data [is] that Pb has permanently interfered with the programmatic settings of the regulation of the APP gene," the authors write. Such a developmental event could contribute to amyloid-related neurodegeneration later in life, they suggest. However, Basha and colleagues note that any effects of Pb on gene transcription would need to escape recognition by DNA repair mechanisms and suggest that such stealth effects could include epigenetic phenomena such as DNA methylation patterns, histone deacetylation, and chromatin restructuring.—Hakon Heimer


  1. My speculations about AD include the thought that AD has an extremely long presymptomatic period, which may extend into childhood. This is based on data such as the Braaks finding that 20 percent of people in their 20s have neurofibrillary tangles (NFT) and senile plaques (SP), though not meeting the criteria for AD. It is also based on our knowledge that NFT and SP represent only the most obvious of a sequence of more subtle changes in tau conformation and other post-translational modifications that certainly precede the formation of frank NFT; similarly for APP and Aβ, and SP.

    Whether the metals and infectious agents mentioned in this news review are related to this sort of speculation is not at all certain. There are a number of subtle parameters that could be looked at that might be informative.

  2. Infants and young children are subjected to a battery of infectious agents, including human herpes virus 6, herpes simplex virus type 1 (HSV1), adenoviruses, respiratory syncytial virus, rhinoviruses and probably enteroviruses, and these, as well as overt neonatal infections, might have long-term effects on the host. Further, these pathogens (like many others, e.g., poliovirus, West Nile virus, and the bacteria that cause TB, or stomach ulcers), can elicit a whole spectrum of responses depending on the host, infecting vastly more people than they seemingly affect, and some can remain lifelong in the host. Thus, even slight damage might be cumulative and long-term, even in people who appear asymptomatic. In other words, surreptitious persistent infection/disease needs to be considered as well as overt diseases of infancy. The same point relates to the hypothesis put forward by Finch and Crimmins, 2004, linking longevity to the prevention of childhood illnesses due to improved public health, etc., in the last half-century or so: Only certain major diseases have been prevented.

    An interesting paper by Wang et al., 2005, may be relevant to the above and to our data implicating HSV1 in brain together with ApoE-ε4 as a risk factor in AD (Itzhaki et al., 1997; Dobson et al., 2003). Wang et al. find that peripheral infection with West Nile virus triggers a transient change in the permeability of the blood-brain barrier (BBB). Perhaps other pathogens cause such changes in BBB; certainly there is evidence that systemic infections lead to cytokine entry into brain, which in turn causes inflammation and cognitive decline in the elderly (Holmes et al., 2003), and we have suggested (Itzhaki et al., 2004) that the inflammation could reactivate latent HSV1, thereby augmenting the damage, just as subclinical prion disease in mice augments the brain damage caused by LPS injection (Combrinck et al., 2002).


    . Inflammatory exposure and historical changes in human life-spans. Science. 2004 Sep 17;305(5691):1736-9. PubMed.

    . Toll-like receptor 3 mediates West Nile virus entry into the brain causing lethal encephalitis. Nat Med. 2004 Dec;10(12):1366-73. PubMed.

    . Herpes simplex virus type 1 in brain and risk of Alzheimer's disease. Lancet. 1997 Jan 25;349(9047):241-4. PubMed.

    . Do infectious agents play a role in dementia?. Trends Microbiol. 2003 Jul;11(7):312-7. PubMed.

    . Systemic infection, interleukin 1beta, and cognitive decline in Alzheimer's disease. J Neurol Neurosurg Psychiatry. 2003 Jun;74(6):788-9. PubMed.

    . Infiltration of the brain by pathogens causes Alzheimer's disease. Neurobiol Aging. 2004 May-Jun;25(5):619-27. PubMed.

    . Peripheral infection evokes exaggerated sickness behaviour in pre-clinical murine prion disease. Neuroscience. 2002;112(1):7-11. PubMed.

    View all comments by Ruth Itzhaki

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Weight in infancy and death from ischaemic heart disease. Lancet. 1989 Sep 9;2(8663):577-80. PubMed.
  2. . The developmental origins of chronic adult disease. Acta Paediatr Suppl. 2004 Dec;93(446):26-33. PubMed.
  3. . Transcriptional involvement in neurotoxicity. Toxicol Appl Pharmacol. 2003 Jul 15;190(2):177-88. PubMed.
  4. . The expression of the amyloid precursor protein (APP) is regulated by two GC-elements in the promoter. Nucleic Acids Res. 1992 Jan 11;20(1):63-8. PubMed.
  5. . Enhancer function and novel DNA binding protein activity in the near upstream betaAPP gene promoter. Gene. 1999 May 17;232(1):125-41. PubMed.
  6. . Transcriptional regulation of BACE1, the beta-amyloid precursor protein beta-secretase, by Sp1. Mol Cell Biol. 2004 Jan;24(2):865-74. PubMed.
  7. . Functional characterization of the 5' flanking region of the BACE gene: identification of a 91 bp fragment involved in basal level of BACE promoter expression. FASEB J. 2004 Jun;18(9):1037-9. PubMed.

Other Citations

  1. ARF Live Discussion

Further Reading

Primary Papers

  1. . Neonatal infection induces memory impairments following an immune challenge in adulthood. Behav Neurosci. 2005 Feb;119(1):293-301. PubMed.
  2. . Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell. 2005 Feb 11;120(3):421-33. PubMed.
  3. . The fetal basis of amyloidogenesis: exposure to lead and latent overexpression of amyloid precursor protein and beta-amyloid in the aging brain. J Neurosci. 2005 Jan 26;25(4):823-9. PubMed.