In classical mechanics bonds cannot break or appear. Unless you apply morse potential or qmmm, this is probably an issue of your visualization software.

I'm using PyMol .!!! .And every thing is right before minimization or simulation .

Gromos Force field .

GROMOS 53a7 and ligand was parameterized using ATB server .

parameters :

title = Minimization

integrator = steep

emtol = 1000.0

emstep = 0.01

nsteps = 50000

energygrps = Protein ligand residue name

nstlist = 1

ns_type = grid

rlist = 1.0

coulombtype = PME

rcoulomb = 1.0

rvdw = 1.0

pbc = xyz

Pymol guesses bonds based on interatomic distance. It may be a result of minimization that the phosphorus and oxygen get so close that pymol thinks they are bound even though there is no bond connection in the force field.

Now the question remains, why they get so close. What about bad charges? Wrong repulsion? Etc.

This is the molecule :

http://compbio.biosci.uq.edu.au/atb/molecule.py?molid=18784#files