The sample processing for SDS page usually involves heating in SDS and DTT or BME buffer. So any aggregates should be dispersed prior to loading. However, you could try adding fresh DTT to 10mM prior to concentration.

But LB is indeed a lousy medium to use for this and I suspect that this is causing your problems. Have you tried transferring LB overnights  to minimal media plus some carbon source (sucrose or glycerol) for expansion?

Thanks for your reply Geoff.

Some aggregates apparently resist even repeated boiling in protein loading buffer, so I need to somehow prevent their formation. Including BME or DTT in the LB solution was one of the ideas I had in mind, how about some detergent like Tween 20?

What about the temperature? I read contradictory recommendations in this aspect.

I combine this experiment with additional ones performed in LB so, bad as LB might be for this purpose, at the moment I´d rather not change the growth medium.

 First assume that you are running protein MW standards, and they run fine on the same gel that your samples do not "and" the MW standards have the same SDS/PAGE sample buffer added to them as do your protein samples.

1) If you are running the gels just for analysis purposes, then make the samples 8 M urea "before" adding the SDS/PAGE sample buffer and if you heat them do not go above 95oC.

2) For proteins that have the tendency to aggregate during concentration, try using the Millipore tangential flow filtration (TFF)-based filters, they reduce the generation of a steep concentration gradient during concentration and so reduce aggregate formation, the alternative is to use a tangential flow filtration (TFF)-based concentration cell, but then you need a N2 tank with regulator, though the cell will eliminate any concentration gradient from forming.

3) Have you treated the samples with a nuclease?, and analyzed the protein sample for DNA/RNA? (A260/A280 ratio). DNA can contribute to aggregation of proteins at all steps of purification and concentration.

Aupa Mike que tal tio?? Como te va seguro que bien. Mira he visto que la trehalosa evita fenomenos d eagregacion proteica en procesos de concentracion congelacion y deshidratación. 

Te dejo aquí el link por si te sirve de algo.

http://www.espatentes.com/pdf/2194911_t3.pdf

Si necesitas mi receta de Sample Buffer me dices. a mi nunca me ha pasado nada parecido

Un saludo amigo

http://www.espatentes.com/pdf/2194911_t3.pdf

Try BJP clinical concentrators (http://www.vivaproducts.com/bjp-clinical-concentrators/) which uses a gentle static absorption method to concentrate protein samples and prevents over-concentration and aggregation.

@ Gary: Thanks for your reply. Markers and non-concentrated samples show normal migration, so it is indeed an issue of protein aggregation while concentration. Thanks for the very valuable tips. I will consider advice number 1 and 2 for future experiments. Regarding the DNA, considering I am working with the secretome (I pellet down cells and 0.22 um-filtrate the supernatant before starting the concentration) I believe DNA should not be an issue in this case.

@ Giltzero: Eskerrik asko, le echaré un vistazo. Escríbeme por medios menos formales y nos ponemos al día ;)

@ Avinash: thanks for the reference, I will consider those columns in case I decide to replace mine.

Our lab's proteins are highly prone to aggregation, which is a focus of our research. We find

(1) just eluting the protein during purification at a higher concentration is so much easier, thereby avoid concentrators. This creates it's own challenges for storage, which we are continually experimenting with improvements with that regard but do often use the protein fresh from purification.

(2) A large tag like MBP can dramatically increase solubility and slows down aggregation.

(3) We find that RNase/DNase (or benzonase) treatment does often reduce aggregation and RNA is known to promote aggregation even for proteins that do not bind RNA. The mechanism behind that is an active area of research right now.

(4) We find that relatively low concentrations of urea (1 or 2M) is generally not sufficient to unfold your protein and also dramatically reduces aggregation (if you can find in the literature that someone has done urea melts on your protein, you can do this with much more confidence that your protein is fine). We often dialyze to other buffers before experiments even though find our protein doesn't behave much differently in the urea containing buffer. We use this for lysis, elution, and storage. High salt (1M KCl) is also rumored to help, which we use but haven't demonstrated that it is in fact responsible for reduced aggregation.

(5) Dissolving your protein in formic acid is a common practice for eliminating aggregates before loading on a gel (as is RIPA buffer but is less effective). We've had this work in many but not all cases, but you have to precipitate protein to wash away formic acid before the gel. Sonication also helps eliminate aggregates.

You may find some useful information on reduction of protein aggregation in some of our issued patents, including Israeli patents  No. 151065 and No. 193365 and U.S. patents No. 6,783,965 B1, No. 7,927,852 B2, No. 8,921,064 B2. The most recent U.S. patent in this family is attached. Also, please see EP 2 196 538 B1.  You can contact me at [email protected] if you need more advice.

What volume are you working with during this step? Someone mentioned Millipore TFF systems and I second this option. Easy to use, quick, and not much loss.