Electrons rotating about a nucleus are governed almost entirely by electrical attraction which is billions upon billions of time stronger than gravity...gravity can be generally ignored in regard to particle interactions.
The basic source of KE is acceleration even though acceleration is not displayed in
KE = 1/2mv2. Another way to say the same thing is that a net force has acted on a mass m to accelerate it; work is done and that is reflected in the motion of the mass.
If the magnitude of the velocity remains constant, v = c, there is not change in KE and the work done by any force is zero. In uniform circular motion the speed of the particle (magnitude) remains constant and perpendicular centripetal 'force' does not work on the particle. A force at right angles to the direction of motion changes direction not magnitude.
Gravity is in general related topotential energy, that is, energy of position. By convention, the gravitational potential energy of a particle is taken to be zero at infinite distance from the center of attraction. So a particle drawn from outer space to Earth's surface would have a negative potential energy, negative work is done by gravity...due to the attractive force of gravity...positive work is required to remove the particle from the influence of earth.
Any introductory college level physics text should give a more complete description...or try Wikipedia for things like "kinetic energy" or "gravitational potential energy".
Gravity causes clumping of gas cloud material to form stars,planets and solar systems and galaxies as entropy increases; dark matter spider webs may aid that formation. I suspect it is not the cause of motion between adjacent galaxies since that gravitational attraction is small. Maybe random motion from formation momentum?? Over vast intersteller distances, expansion of the universe causes galaxies to in general move apart from one another, but that effect is minimal on closely spaced galaxies.