While concentrating the protein using centricons (using appropriate cutoff filter) generally lower mol. wt proteins/impurities goes off. Since your target protein is over 100kDa and impurity is around 30kDa, the range difference is lot. So this method might work !

One of those contaminants could be E. coli wandorous histidine rich protein, which sometime requires 140 mM imidazole to remove it. Since ur protein of interest is way larger than the contaminants, concentrate ur fractions and perform Gel permeation using superdex-200 column. It will surely work for u. All the best.

Since the relative abundance of these low MW contaminating proteins correlate with the total mount of your target protein, these low MW proteins may be the products of partial proteolytic degradation of your target protein. If you have polyclonal Ab to your protein you can test it in western blot. Try to use protease inhibitor cocktail at all steps extraction / lysis and purification.

I agree with all of the above comments and suggestions. 

I wouldn't worry that the smaller impurity in your elution fractions seems to be at a higher concentration when compared to your crude fraction. It is possible this smaller protein is interacting with your protein of interest (or is a degradation product with the His-tag still attached) since the intensity of the band increases with the intensity of your protein of interest. Based on your gel I don't think you are increasing the amount of contaminant protein relative to your protein of interest. 

The above suggestions to remove the smaller protein using a concentrator with an appropriate MW cutoff are good, but I also suggest using a size exclusion column if you have one available to you. By monitoring the absorbance of your protein peak of interest, you can also check to see if your protein has misfolded or aggregated. 

I assume you are washing your affinity column after loading your crude sample and before eluting your protein of interest. How much salt is in your buffer? You could try increasing the NaCl concentration to reduce nonspecific binding to your protein or column. I've used as much as 1M NaCl in a wash buffer to remove nucleic acids from binding to my protein. Adding a low concentration of imidazole to your lysis buffer should also prevent non-specific proteins from binding to your column, although if your protein elutes at only 30mM Imidazole that wont work for your protein. 

I hope one of these suggestions helps. Best of luck!

How about changing your expression system to mammalian cell since your target protein is over 100kDa?

Hi Guru Krishna Viswanathan, I tried using the spin column with 30kDa cutoff filter membrane, but unfortunately it didn't remove the impurities at all. I'm doubted why? 

Hi Bhanu, thanks for the suggestion. I think I will tried out for this.

Hi Kaitlyn, I'm worried about this as I would need to proceed with crystallization. I'm worried that the impurities might interrupt the crystallization of my protein. I'm using 300mM of NaCl. How actually do the salt help in non-specific binding? Will it cause any effect to my protein?

Hi Fengjuan. I have not tried that before. Maybe can try that out. 

Hello,  Having contaminants on Nickel columns after elution is quite frequent, it can be due to a too large amount of nickel resin (but this you can read in the trouvbleshooting part of the manual from your provider). Sometimes increasing a bit the salt concentration helps to remove non specific binding, adding glycerol up to 10% can also helps -of course this can be used only if your protein stand these conditions. Increasing imidazole concentration in the washing steps helps a bit  (50mM why not 140mM the risk at this concentration is that your protein begins to elute depending on the tag accessibility etc...) and extending the washing step as well .Then some proteins stick to the nickel resin because they may agregate on it and are hardly removed

Then if you couldn't get rid of the contaminant on concentrators with a cut off of 50kDa it is probably because this contaminant sticks to the concentrator membrane or is not a globular protein and cannot pass properly through the pores.  

Last thing is if you can't get rid of this contaminants after one step of purification add a second one GEl filtration seem to be indicated as was proposed previously  Superdex 200 or superose 6 may be convenient. Then another technique could be ammonium sulfate precipitation It's an old method but that works. You add ammonium sulfate powder step by step and the proteins precipitate at different concentrations of it. You have to analyze the pellet and supernatants for the first time but once you found the conditions where the contaminants precipitate and not your protein (or the reverse) you get rid of it. If your protein precipitate before the contaminant  just spin the pellet discard supernatant and resupend your pellet in a buffer with no ammonium sulfate. And dialyse to totally remove it. the protein should be good.

If nothing of these work changing tag is another option. Hope this helps.

Hi Alex,

Do you have any information about your protein regarding DNA binding ?

One step purification is not enough for most of the proteins, so may be you could try heparin and other ion exchange chromatography before going for SEC

In this way you could characterize completely and also you can use SEC as a method for checking the oligomeric state of your protein 

I hope this information could helpful for you 

Perhaps you need to look into what the bands actually are, since from the looks of your gel, it looks to be consistent with your main protein of interest. One possibility is that the smaller bands that you get are actually breakdown products of your protein of interest, and hence containing your his-tag.

Hi Edward, I had compare the crude portion, and eluent portion of my protein in both SDS-PAGE and Native-PAGE. You seem to be correct. There is no impurities present in native at about the same size but do present in SDS. I had attached the comparison below. There is 3 band present in Native-PAGE and my protein is highlighted in the red-box. The Swiss-model software predicted that my protein is monomer, but how come it appear 3 band there (the green box)? My protein size is ~113kDa. What can I actually do to confirm that they are my protein breakdown?

Simplest way would be protein Id by mass spec. If you see the same peptides appearing, and some segments of the protein are missing in the smaller bands, then that is a clear sign that there is breakdown products. Alternatively, multiple bands appearing could also be from interacting partners of your protein. If you really want to get rid of it, perhaps purify your protein in denaturing conditions, or add in some detergents/salts to break the interaction. 

Depending on your expression system, multiple bands might also be post translational modification, splice variants, or protease degradation. Can't say much since I don't know your protein.

If you want to be very quick you could also check with his tag antibody but the perfect solution could be cut the bands trypsin digest and run mass spec analysis. If you dont have the facility i would suggest you check with WB 

If you look carefully at your protein of interest, you have  multiple bands there. You didn't write, what expression system you used. So, the small proteins (25kDa) could  be  degradations of your protein, or some chaperone.  I agree to do MassSpec of these bands. You can get more pure protein using ion exchange chromatography. The chaperones bind to anion exchange resin.

Hi Edward and Jana,

I'm using pET32-b in BL21(DE3) expression system. My protein is actually a Geobacillus sp. DNA polymerase. I using Swiss-model to predict the structure of my protein and it seem to be monomer. 

Usually the protein containing some of histidines can bind nickel columns. The problem in which non specific binding occured may be due to the low expression level of your protein, so you can reduce impurities in eluate to increase the expression level of your protein. The higher expression level of protein, the higher purity you will get. (use other bacteria strains such as RP or RIPL, adjust culture condition(temparature, time, medium...)

Second, generally several kinds of purification method constitute the purification steps to guarantee to purify protein having  high level purity above 95%

First step : capturing(Histag, GST-tag etc) -usally there are lots of impurities

Second step : purification (ion exchange or Hydrophobic interaction chromatograpy)

        You can remove misfolded proteins, degraded proteins at this step

Last step: polishing(size exclusion chromatography) - high level purity

If yo follow these steps you will get pure protein you are interested in.

My recommendation considering that is a DNA polymerase, you could be use a heparin resin preferently or a exchange ionic resin. Because is common obtain protein contaminants. Using a NaCl gradient I think you can hide this. I hope you have good experiment

You can go for two way purification methods also.. Otherwise, different concentration of imidazole in wash buffer (20mM, 50mM) will also works :)