Does viscosity result in Ni-NTA low affinity?

[louhazosc]

6xHis-Tag protein purification under denaturing conditions.
General Protocol is:
Sonicate in tris/NaCl - spin - Sonicate pellet in 2M Urea - Spin - Incubate pellet for hours in 6M GuHCl/20mM Tris/500mM NaCl/1mM b-mercaptoethanol pH 8 - spin - Filter supernatant - run filtered supernatant on Ni-NTA.

What I have been seeing for a while and just dealing with it is a very very low affinity for Ni_NTA, somewhere around 50-75mM imidazole elution.

To answer the obvious questions : yes, I have extensively evaluated the solubility and 6M GuHCl is required and the pi of the protein is around 5 and my Ni-NTA is equilibrated in the pH 8 GuHCl buffer mentioned above. According to sequencing following cloning there is a 6xHIS tag

The only thing I can think of lately is that the sample is of course still has a level of viscosity. I mean, it's not very high but still present.

Could it be that or is there just something else I am missing? I mean this is a very standard protocol.
The only thing I am not doing that many are now is on-column refolding. I wish I could but I am actually working on several proteins of the same family and some just love to aggregate on the column - I have tried.

Any ideas, opinions, comments, questions?
Thank you very much