A paper in the March Nature Neuroscience suggests that it does. Researchers from the National Institute of Mental Health in Bethesda, Maryland, report that immediate versus gradual application of brain-derived neurotrophic factor (BDNF) has “dramatically” different results for cultured neurons, the authors wrote in an e-mail to ARF. First author Yuanyuan Ji and principal investigator Bai Lu have since moved to GlaxoSmithKline in Shanghai, China.

BDNF acts by binding to the TrkB receptor, activating downstream pathways including the MAP kinase, PI3K and PLC-γ signaling cascades. It has diverse functions in development, learning and memory, and pain (reviewed in Binder and Scharfman, 2004). “It is mind-boggling how one molecule could do all that,” said Helen Scharfman of the Nathan Kline Institute in Orangeburg, New York, who was not involved in the study. “Bai Lu’s paper helps address that complexity.”

In neurodegenerative disease, BDNF mRNA levels are reduced in the hippocampus in human Alzheimer’s samples (Phillips et al., 1991), and the protein is decreased in the substantia nigra of people who had Parkinson’s (Howells et al., 2000). Its expression is regulated by huntingtin (see ARF related news story on Zuccato et al., 2001). These and other data (reviewed in Zuccato and Cattaneo, 2009) suggest that BDNF therapy might be beneficial; the current work indicates that the delivery timing could be crucial.

Lu was inspired to study the role of timing in BDNF treatment by a brush with lymphoma in 2003, he told ARF in an interview. The standard chemotherapy was a “bullet” treatment. This single, high dose led to an initially toxic concentration in the blood, dropping off to a lower plateau level. Lu participated in a clinical trial with a different approach: he wore a pump, which delivered a steady infusion of chemo, all day for five days. The hope was that slow perfusion would prevent toxicity and yield a higher concentration of drug in the blood. But Lu, now in remission, wondered if the different drug delivery might lead to different biological activity, too.

Various labs working with BDNF often get conflicting results, Scharfman said: “There is a huge variability in what people have seen…we have never really understood that.” Lu, Ji, and colleagues investigated the effects of BDNF, in cultured cells and brain slices, by varying the treatment kinetics. In the acute, “bullet” condition, they added BDNF to a final concentration of 1 nM in cultured neurons. In the gradual mode, they increased the BDNF concentration 10-fold every half-hour, starting at 0.0001 nM and working up to 1 nM.

Acute BDNF application to cultured rat hippocampal cells turned on TrkB quickly, with TrkB phosphorylation peaking after just 15 minutes, but it dropped back to baseline levels within two hours. Gradual BDNF treatment caused long-lasting TrkB activation, persisting for up to eight hours. Phosphorylation of Erk and CREB, downstream of TrkB in the MAP kinase pathway, was similarly affected. CREB regulates transcription of the genes Arc and Homer1. Levels of these proteins barely budged in the acute condition, but rose and stayed high with the gradual treatment. The researchers also found that acute or gradual treatment modes caused brief or sustained activation, respectively, of both the GSK3 and PLC-γ1 pathways.

BDNF is involved in dendritic growth and morphology, so Ji and colleagues examined the shape of the cultured cells over a three-day course of treatment. Bullet BDNF treatment caused neurites to elongate and spine heads to enlarge; gradual treatment caused neurites to branch and spine necks to elongate. These morphological changes may influence synaptic plasticity and memory formation (reviewed in Yang and Zhou, 2009).

The researchers used hippocampal slices from mice to examine how synaptic signaling responds to acute or gradual BDNF treatment. Researchers in the field agree that BDNF influences long-term potentiation, but there has been disagreement over its effect on basal synaptic transmission, said Margaret Fahnestock of McMaster University in Hamilton, Ontario, who was not involved in the study. Lu and colleagues compared a high BDNF perfusion rate of 240 ml/hour to a slow rate of 25 ml/hour. They performed their experiments with both developing hippocampus from two-week-old mice and adult hippocampus from eight-week-old animals. They found that slow BDNF perfusion leads to long-term potentiation in the young slices, whereas fast BDNF enhances basal synaptic transmission in adult samples. “It resolves a longstanding controversy,” Fahnestock said.

It is presently unclear how BDNF concentrations vary in vivo, as it is difficult to directly assess its levels in the body (see ARF comment). The authors theorize that the acute delivery corresponds to regulated BDNF secretion, for example, in response to neuronal activity. In contrast, the gradual infusion may mimic constitutive BDNF secretion, or BDNF diffusing from a distant source.

Lu suggested that the difference between fast and slow treatment might be a general principle, with implications for both cell biology and medicine. In the typical cell culture study, he noted, “you basically dump the growth factor into the dish.” But that treatment may not reflect what happens in a living organism. And in the therapeutic arena, altering treatment kinetics may change outcomes. A gradual treatment may not necessarily be better, he noted, but could be different. Slow-release formulations of drugs for various conditions exist, and they are a research priority in some neurodegenerative conditions such as Parkinson’s. And researchers attempting BDNF therapy for amyotrophic lateral sclerosis have used implanted pumps (Ochs et al., 2000). Lu’s work suggests these strategies may be valuable for more reasons than previously appreciated. “You may turn on different genes, you may have different biological function,” Lu said. “It is a fundamental difference.”—Amber Dance.

Reference:
Ji Y, Lu Y, Yang F, Shen W, Tang TT, Feng L, Duan S, Lu B. Acute and gradual increases in BDNF concentration elicit distinct signaling and functions in neurons. Nat Neurosci. 2010 Mar;13(3):302-9. Epub 2010 Feb 21. Abstract