Some Ab, especially after chemical modification, tend to aggregate with even a single freeze-thaw cycle. Sometimes this is obvious from observed cryo-precipitation, but sometimes it is subtle, causing incremental performance problems and poor reproducibility. These Ab must be stored at 4C.

Of course a pure protein in near-neutral buffer at 4C will grow psychrotroph microbes after just a few days. To prevent this add a preservative. I like sodium azide (0.01-0.1%), but note it is cytotoxic and inhibits HRP. Thymol also works at the same concentration range. It lacks toxicity to plants and animals, but it must be added from an alcohol stock solution due to low water solubility.

Also note Ab store best at high concentration, at least 0.5 mg/mL. At lower concentrations significant amounts of protein will stick to the storage container surface, resulting in decreased solution concentration. If you need to store it dilute, add an inert protein to the storage buffer - something like 0.1% BSA or gelatin - to occupy non-specific binding sites on the plastic or glass. Or at least use a siliconized container.

In reality, 4oC is just a default storage recommendation; rarely will any supplier have tested the effects of freeze-thaw and -20 storage in liquid form (i.e. with 50% glycerol). 4oC is just a reasonably safe option if you cannot/don't want to do any trials (assuming care is taken to avoid microbial growth as noted by John Carter). You also should recognise that each antibody is a unique entity, and that no single method will be perfect for every one of the millions of commercial antibodies. My own storage preferences are:

1. 4oC with azide (except HRP conjugates, use Proclin instead)

2. -20oC in 50% glycerol esp. for conjugates (allows you to avoid use of toxic azide too).

3. Snap freezing in liquid N2 in small aliquots and storage at -20 or -70*.

4. Freeze drying (but generally not worth the effort for research antibodies).

* It is important to appreciate that the rate of freezing can affect the outcome. A SLOW freezing rate is often damaging, as the solutes can become extremely concentrated as the solvent (water) freezes from the outside of the tube. If you do not have liquid nitrogen, SMALL aliquots in a -70 should freeze quite quickly, or you can try a dry ice/ethanol bath which, being liquid, allows rapid heat exchange with the sample.

Finally, whatever method you try, just assume the worst and test only a small portion of your antibody. Process the rest once the method has been validated.

As noted by John Carter above 4C with bacteriostat is preferable. If planned to be used in culture you can antibiotics - same as used in tissue cultures. Gentamycin and Kanamycin are stable antibiotics.

Alternatively, store at high osmolarity: sugar or salt. In the past researchers stored in 50% saturation of ammonium sulfate and diluted in water before use.