Autism is a diverse disease that causes a range of symptoms, from repetitive behaviors to extreme sensitivity to noise.

This week, two separate research teams are shedding light on two different proteins that seem to play a role in causing some autism symptoms. Both believe their findings could point to new treatment strategies.

Scientists at the Gladstone Institutes zeroed in on tau, a protein that’s most closely associated with Alzheimer’s disease. The researchers were investigating a connection between Alzheimer’s and epilepsy when they made a surprising discovery that reducing levels of tau in mouse models of autism prevented three of the main symptoms of the disease: repetition, difficulty socializing and impaired communication. It also prevented seizures, which affect about one-third of autism patients, they reported in the journal Neuron.

The Gladstone team had previously discovered that reducing tau prevents seizures in mouse models of Dravet syndrome, a form of childhood epilepsy that sometimes also causes autism symptoms. So, first they deleted copies of the gene that encodes tau in mouse models of Dravet and observed that even a 50% drop in protein levels prevented core autism symptoms.

Then the researchers repeated the experiment in a different mouse model of autism. They measured changes in autistic behaviors by observing self-grooming behavior, the ability to build nests and complete maze exercises, and social responses such as the tendency of pups to vocalize when they’re separated from the group. Reducing tau levels prevented deficits in these behaviors, they reported.

The team went on to discover that reducing tau levels boosts the activity of PTEN, an enzyme that can tamp down an autism-related signaling pathway. They now hope to develop small-molecule drugs that will either reduce tau or boost PTEN, they said in a statement.

A second autism discovery was reported by neuroscientists at the Massachusetts Institute of Technology and Brown University. They discovered that mice lacking a protein called Shank3 are overly sensitive to touch on their whiskers—a behavior that corresponds to the hypersensitivity to light, noises and other stimuli seen in people with autism. They reported the discovery in the journal Nature Neuroscience.

The MIT team had previously shown that mice lacking the gene that produces Shank3 display many of the other core symptoms of autism, such as repetitive behavior and social avoidance. They wanted to measure the effect of lowering Shank3 levels on sensory behaviors, so they decided to focus on the whiskers, which are key to the ability of mice to maintain balance and navigation abilities.

They trained both Shank3-mutant mice and normal mice to indicate when they felt their whiskers were being touched. The mutant mice were “very sensitive to weak sensory input, which barely can be detected by wild-type mice,” said Guoping Feng, Ph.D., professor of neuroscience at MIT, in a statement. “That is a direct indication that they have sensory over-reactivity.”

The researchers went on to use advanced imaging techniques to measure neural activity in different cell types in the mice that were hypersensitive to sensory stimuli. They focused on excitatory neurons and the inhibitory neurons that modulate their activity.

In a series of mouse studies, they discovered that low levels of Shank3 in inhibitory neurons prevented them from regulating excitatory neurons. That, in turn, caused sensory sensitivity.

Feng’s team is now planning further studies to determine the timing of inhibitory defects and the emergence of sensory symptoms in mouse models of autism. He believes the findings could guide the development of drug treatments that target defects in inhibitory neurons.

This article originally appeared in FierceBiotech and was written by Arlene Weintraub