The first big space-specific issue is actually air quality. You can't just open a window to air out the molten-ABS smell from the ISS!
FFF printers put out fumes and nanoparticles. In a space station, the same air gets recycled over and over, and the air purification systems have a specific set of contaminants that they are optimized for, as well as a design capacity for air turnover and chemical removal rates that won't be adjusted just because somebody's printing a space-ratchet today. Protecting cabin air quality is a huge design factor for any experiment that goes into space.
The Made in Space printing experiments on the ISS to date were performed in one of the vacuum experiment chambers, so any unfiltered fumes (or fire flare-ups) could be vented directly to space if required. In the long run, this isn't going to work -- other experiments may need the vacuum chamber, or "production" printers may be too large to fit. So the printer needs to have its own internal air purification system.
Another MAJOR design constraint is launch survival. Rocket payloads must be designed for extreme g-forces without 1) damage, or 2) significant internal shifting of mass which would affect the payload center of gravity.
Total payload weight is also quite important here: lifting mass to low Earth orbit is EXPENSIVE.
Surprisingly, the microgravity environment itself isn't that big of a deal. Molten plastic is highly viscous and pretty much stays where you put it long enough to solidify, as long as it's sticking to something. But two impacts do come to mind.
- First, an unsecured filament spool will try to unwind itself. Gravity won't provide the contact friction we usually rely on to keep spools from bird's-nesting. (Think about it: a tightly-wound spool is literally a giant coil-spring.)
- Second, heat flows are different in microgravity -- you can't rely on passive convection to cool the print or the motors. Accommodations must be made for sufficient forced airflow and heat-sinking on anything that requires cooling. And that includes the enclosure itself, since, as mentioned above, the print chamber must be sealed up tight for air quality control.
Finally, reliability is critical. Amazon doesn't deliver to the ISS (yet). Even a single stripped screw may take the printer out of commission for months until a replacement part can be fit into an upcoming supply launch. Having the printer catch on fire because something shorted would be catastrophic.
So, really, it's all about making a printer robust enough to make it up there, operate safely, and never break. Printing upside-down is trivial in comparison.