Delivery of follistatin (FST) represents a promising strategy for both muscular dystrophies and diabetes, as FST is a robust antagonist of myostatin and activin, which are critical regulators of skeletal muscle and adipose tissues. FST is a multi-domain protein, and deciphering the function of different domains will facilitate novel designs for FST-based therapy. Our study aims to investigate the role of the N-terminal domain (ND) of FST in regulating muscle and fat mass in vivo. Different FST constructs were created and packaged into the adeno-associated viral vector (AAV). Overexpression of wild-type FST in normal mice greatly increased muscle mass while decreasing fat accumulation, whereas overexpression of an N terminus mutant or N terminus-deleted FST had no effect on muscle mass but moderately decreased fat mass. In contrast, FST-I-I containing the complete N terminus and double domain I without domain II and III had no effect on fat but increased skeletal muscle mass. The effects of different constructs on differentiated C2C12 myotubes were consistent with the in vivo finding. We hypothesized that ND was critical for myostatin blockade, mediating the increase in muscle mass, and was less pivotal for activin binding, which accounts for the decrease in the fat tissue. An in vitro TGF-beta1-responsive reporter assay revealed that FST-I-I and N terminus-mutated or -deleted FST showed differential responses to blockade of activin and myostatin. Our study provided direct in vivo evidence for a role of the ND of FST, shedding light on future potential molecular designs for FST-based gene therapy.