It is possible, and it depends on the ligand (ie, the agonist) and the receptor. The ligand may effect other receptors (as agonist, antagonist, anti-agonist (eg, a blocker)), etc which may induce side effects. Or it just metabolizes in humans dangerously. Or it agonizes in a way that impedes a receptor's other roles. (Many receptors have complex roles.) People have to understand that a receptor is not a "lock" and a ligand is not a "key". This is a mental analogy that is far from accurate. A ligand's molecular shape, charge location, affinity strength, binding site, etc all are factors in how it affects the receptor: they determine whether and how it agonizes, antagonizes, blocks, deactivates, etc the receptor. Some ligands can do combinations of several effects, the affect depending on physiological conditions. So beware of folks who talk of agonists as being the magic bean that unlocks the theorized powers of a receptor. They most often are no such thing.
This is why I point out that what matters is 2-73's clinics usefulness (does it work in humans in the way that is needed and under the right circumstances). S1 agonists are not all the same. That is a mantra for us AVXL investors to remember, and one which the company should make clearer to the market.