LED dimming with PWM + nMOS

HINT: You can probably ignore the whole circuitry at the top of the 
attached image. I highly doubt it is of relevance to my problem.

Hello,
I've got a problem with this project. It is supposed to be a driver for 
a high-powered LED (bottom right IC) with a dimming function using a PWM 
generator (bottom left IC, equivalent to a regular NE555). I also added 
something to monitor my input voltage level (top IC) since it'll be 
running off a battery pack.

The whole things works quite brilliantly. No excessive heat build-up on 
any of the parts except the load resistor which I substituted my LED 
with.

------ THIS IS THE PROBLEM ------

The dimming is not working because the gate pin of my nMOS is being 
pulled to approx. 11V (input voltage is around 13.2V). The 555 is 
producing a rect signal of 4.8V at 1kHz (22% duty) at its OUT pin but 
immediately after D3 I'm meassuring roughly 11V. The AC signal is still 
there, superimposed on a 10V DC signal, with a 0.8V amplitude, same 
frequency, same duty cycle - but of course the nMOS is permanently open 
under those conditions.

I have tried pinpointing the source, even cut off the connection to the 
LT3478's PWM pin because I suspected it might be supplying that voltage 
level, but it didn't.

None of the diodes D1, D3 or D4 are anywhere near their reverse 
breakdown voltages. I meassured the MOSFET's on/off resistances to 
confirm whether or not it may be broken but all is in order.

I probed my PCB for hours, looking for any printing errors or 
short-circuits and found none. Everything looks exactly as I designed 
it.

------ END OF WALL OF TEXT ------

I'm pretty lost here. I'm doing electronics as a hobby and I never 
studied this stuff so I might be missing something elementary. Or maybe 
this project hates me.

If you see something that could be the cause or have an idea what I 
could try to locate it, please share it. Anything, and I mean ANYTHING, 
is better than staring at this board for another week every evening.