The number of diagnosed cases of autism has increased sharply in recent years. The cause of that remains something of a mystery, and suspicions of environmental factors, most notably vaccines or the preservatives used in vaccines, have not been borne out by further scientific investigation (though they have succeeded in some areas in creating an additional public health problem: parental resistance to childhood vaccinations).

Recent reports suggest that most or all of the increase might be due to better diagnoses rather than a true increase in the number of cases.

Whatever the cause, there now are close to 1 million children diagnosed with autism or autism spectrum disorders in the U.S., but no medications are approved for the disorder. In contrast, seven medications are approved for the treatment of attention deficit hyperactivity disorder, some for children as young as 3.

That is partly due to the fact that autism is a developmental disorder, and partly due to its complexity; a disorder with the hallmarks of language and social dysfunction presents daunting challenges to attack pharmacologically.

"It's not possible to create a pill that can substitute for two or three or five years of complex experience and brain development," Ruth Carper, post-doctoral fellow at the University of California at San Diego, told BioWorld Today. She added, "What may have the most promise is very early pharmaceutical treatments in combination with behavioral therapy."

While it would be difficult to develop drugs that address the dysfunctions of autism, it might be possible to pharmacologically address the underlying cellular defects. That, too, has its challenges. Carper pointed out that "what is so particularly challenging in the treatment of autism is that it is a developmental disorder. That is, something goes wrong with the brain while it is developing so that it doesn't construct itself properly in the first place," perhaps as early as prenatally.

Nevertheless, any progress on understanding how nerve connections are formed might shed light on the cellular basis of autism. Two recent papers, in the Feb. 25, 2005, issue of Science and the Dec. 20, 2004, issue of Cell, report progress toward such understanding.

Neurons can communicate via excitatory and inhibitory signals to prevent neuronal conversation from being a series of non-sequiturs. Presynaptic (sending) and postsynaptic (receiving) structures need to be anatomically aligned with each other and in agreement on which neurotransmitter they will use for signaling purposes. The overall ratio of excitation and inhibition also needs to be fine tuned; otherwise, the receiving neuron will be either unable to pass the signals it receives downstream if it is too strongly inhibited, or unable to shut up if it has too many excitatory inputs.

In the Science paper, researchers from Columbia University showed the involvement of a protein family known as neuroligins in the formation of both excitatory and inhibitory synapses.

The neuroligin family consists of three known members to date, all of which live together in the membrane of postsynaptic receiving structures of neurons during development. It already was known that neuroligins direct the assembly of sending structures on neurons. When the scientists overexpressed neuroligin 1 in cell cultures of the hippocampus, they found that such overexpression also led to an almost 70 percent increase in the formation of receiving structures, and that those structures had receptors mainly for excitatory neurotransmitters. Experiments with truncated neuroligins showed that for those receptors to cluster at the synapses, it was necessary for neuroligins to interact with complementary proteins, neurexins, on the sender neurons.

Peter Scheiffele and his colleagues in the Science research next inhibited the neuroligins via RNA knockdown, either alone or in combination. Simultaneous inhibition of all three neuroligin types resulted in a greatly reduced number of excitatory synapses, though the reduction in excitatory current was not nearly as strong, suggesting that the neurons may have made up for the synapse reduction by somehow increasing transmission efficiency. Neuroligin 2 knockdown affected inhibitory synapse development more strongly than knockdown of the other neuroligins, though the authors noted that all three neuroligins contribute to the development of both excitatory and inhibitory connections.

The findings of Scheiffele's group come on the heels of a Cell paper on synapse development. In that publication, senior author Anne-Marie Craig and her colleagues focused on proteins located on sender neurons, the neurexins, and their contribution to synapse formation on the receiving neurons.

Like Scheiffele's group, the authors of the Cell paper, who are at Washington University Medical School in St. Louis, used cell cultures - in their case, hippocampal cells and fibroblasts - to study the cellular underpinnings of synapse formation. The authors found that neurexin and its interaction with neuroligins contributed to the development of both excitatory and inhibitory synapses. The finding that neurexins are involved in the formation of inhibitory connections came as a surprise to the authors - indeed, the original plan had been to use inhibitory connections as controls. Instead, Craig and her colleagues found that neurexin interaction with neuroligin 2 led to the development of inhibitory synapses.

Previous research has shown that mutations in the genes for neuroligins 3 and 4 are associated with autism, as well as mental retardation; the papers suggest a possible cellular basis for those findings, as well as possible pharmacological approaches to the disease. Until then, the name of the game remains behavioral therapy.