Crude subcellular fractionation of cultured mammalian cell lines
© Horton et al; licensee BioMed Central Ltd. 2009
Received: 1 May 2009
Accepted: 10 December 2009
Published: 10 December 2009
The expression and study of recombinant proteins in mammalian culture systems can be complicated during the cell lysis procedure by contaminating proteins from cellular compartments distinct from those within which the protein of interest resides and also by solubility issues that may arise from the use of a single lysis buffer. Partial subcellular fractionation using buffers of increasing stringency, rather than whole cell lysis is one way in which to avoid or reduce this contamination and ensure complete recovery of the target protein. Currently published protocols involve time consuming centrifugation steps which may require expensive equipment and commercially available kits can be prohibitively expensive when handling large or multiple samples.
We have established a protocol to sequentially extract proteins from cultured mammalian cells in fractions enriched for cytosolic, membrane bound organellar, nuclear and insoluble proteins. All of the buffers used can be made inexpensively and easily and the protocol requires no costly equipment. While the method was optimized for a specific cell type, we demonstrate that the protocol can be applied to a variety of commonly used cell lines and anticipate that it can be applied to any cell line via simple optimization of the primary extraction step.
We describe a protocol for the crude subcellular fractionation of cultured mammalian cells that is both straightforward and cost effective and may facilitate the more accurate study of recombinant proteins and the generation of purer preparations of said proteins from cell extracts.
The expression of recombinant proteins in cultured cell lines has become an increasingly valuable tool for the study of protein trafficking and function. In many or most instances it is desirable to purify the recombinant protein to allow accurate study of such functions. While this can be complicated enough for a protein that is secreted into the culture medium of the cell system, it becomes even more daunting when the protein of interest resides within the cell, perhaps in a single specific compartment. In this case it is desirable to be able to isolate the target protein with minimal contamination by other proteins, potentially from compartments that the protein of interest may never encounter. In most standard cell lysis procedures a single lysis buffer, perhaps NP40 or RIPA is used to generate a total cell lysate essentially comprising a soup of proteins and creating abundant opportunity for non specific interactions.
Another consideration that arises when one is studying a mutant protein is the question of whether or not the protein is soluble in the extraction buffer used and thus whether or not all of the protein has been harvested, enabling an accurate assessment of the effect of the mutation on the intracellular trafficking and function of the protein. The use of more than one single lysis buffer can help to address this issue, particularly if the stringency of the additional lysis buffer is higher, thus increasing the chances of solubilizing any misfolded mutant proteins.
Additionally, many proteins do not reside permanently within a given intracellular compartment but rather shuttle between compartments to carry out their functions e.g. various transcription factors that shuttle between the cytosol and the nucleus. A simple subcellular fractionation protocol that allows easy and reproducible generation of fractions representative of specific subcellular organelles would facilitate the study of such dynamic changes in intracellular protein localization.
Here we present a simple protocol for the isolation of crude subcellular compartments from cultured mammalian cells, sequentially generating fractions enriched for cytosolic, membrane bound organellar, nuclear and insoluble proteins respectively.
Optimization of digitonin concentration for extraction of cytosolic proteins
The primary step of the protocol that we have developed is the permeabilization of the plasma membrane using digitonin to effectively release the contents of the cytosol. Digitonin binds to and forms pores in membranes by complexing with membrane cholesterol and other β-hydroxysterols . The extent of binding and permeabilization thus depends upon the accessibility of the membrane and its sterol composition . At low concentrations of digitonin the cholesterol rich plasma membrane can be effectively solubilized with little to no solubilization or permeabilization of intracellular membranes that are lower in cholesterol such as the endoplasmic reticulum (ER) and mitochondrial membranes. Similar concentration dependent effects on the degree of cell lysis by digitonin have also been shown in studies on parasites [3, 4].
Solubilization and extraction of nuclei and insoluble proteins
Following extraction in RIPA buffer the final extraction in enhanced RIPA buffer (E-RIPA) solubilized any remaining proteins as shown in figure 3C by Coomassie stain and subsequent silver stain. At least one distinct protein band is present in the E-RIPA extract relative to the preceding RIPA extract demonstrating extraction of previously insoluble protein. The Coomassie stain also shows the presence of Histones in the nuclear extract, as later confirmed by Western blotting. While no pellet was visible following the E-RIPA extraction, addition of 1× Laemmli sample buffer to the tube followed by boiling and analysis by SDS-PAGE and silver staining showed that no proteins remained after the E-RIPA extraction (Figure 3D). Also shown in this figure is the combinatorial approach from step 5 in the protocol where the RIPA and E-RIPA extracts are generated as one single concentrated extract, again showing that no proteins remain after these steps.
Demonstration of the advantages of crude fractionation over whole cell lysis and scale up of the approach
As proof of concept of the application of this method in the generation of purer preparations of recombinantly expressed proteins, we transiently transfected HEK293 cells with a construct encoding V5/6×-HIS tagged ERGIC-53. Cells were then either lysed directly in NP40 or RIPA buffer or were extracted sequentially according to our protocol. Subsequent analysis of extracts by SDS-PAGE followed by staining with Coomassie blue or by Western blotting and probing with an anti-V5 antibody demonstrated the advantages of crude subcellular fractionation over direct lysis. As can be seen in figure 4 purification of ERGIC-53 from the NP40 extract obtained in the stepwise procedure using Nickel Resin resulted in a purer preparation and higher yield of protein than from direct NP40 or RIPA lysis due to less interference by proteins from the cytosol becoming non-specifically bound to the resin/ERGIC-53.
To illustrate our concept that not all proteins are recovered in a single lysis step, particularly when overexpressed in a recombinant system, we transfected cells with constructs encoding WT and mutant (MUT) cartilage oligomeric matrix protein (COMP) using lipofectamine 2000 (Invitrogen) as per manufacturer's instructions . COMP is a secreted glycoprotein of the extracellular matrices of various skeletal tissues . As shown in figure 4C the majority of the protein resides in the NP40 extracted ER fraction undergoing synthesis and trafficking. However a significant amount of protein, particularly in the case of MUT COMP, was recovered in the E-RIPA extract indicative that it was misfolded and aggregated (Holden et al, manuscript in preparation). Thus our approach allows for the complete extraction of recombinantly expressed proteins which may be critical in determining functional consequences of mutations in any given protein.
Volumes of buffer required for protocol scale up according to culture vessel surface area.
Surface Area (cm2)
Buffer Volume (ml)
100 mm dish
60 mm dish
35 mm dish
6 well plate
12 well plate
24 well plate
Application of the approach to other cell lines
Continuation of the protocol using 100 μg/ml digitonin for the cytosolic extraction in each case demonstrated that, as was the case for HEK293 cells, distinct fractions comprising membrane bound organellar proteins, nuclear proteins and insoluble proteins could also be generated for HT1080 and HeLa cells (Figures 5B and 5C). Due to the weakness of the signal from Lamin A we chose to use an alternative marker of nuclear proteins, Histone H3 (ABCAM) which clearly localized predominantly to the nuclear extract. Once more, the silver stained gels in figure 5C reveal the presence of unique proteins in the E-RIPA extracts that are not present in the preceding RIPA extract.
We have developed a simple, inexpensive protocol for the isolation of crude subcellular fractions of cultured cell lines. The protocol uses simple buffers and equipment and requires only approximately 2 hours from start to finish. While here we have presented data from three commonly used human cell lines, HEK293, HeLa and HT1080, we anticipate that the technique can be applied to any cell line in culture with some optimization of the initial digitonin extraction step.
We envisage several useful applications of this technique; the generation of purer preparations of recombinant proteins from cultured cells, more thorough analysis of trafficking and secretion of wild type and mutant proteins and simplification of protein extracts for proteomic analysis.
The generation of purer preparations of recombinant proteins from cultured cells is highly desirable. One such advantage of having a purer preparation would potentially be in the context of the study of protein-protein interactions. Taking our expression and purification of ERGIC-53 as a example, if one was to attempt to study the interactions of this protein with other ER/Golgi proteins, our stepwise fractionation approach of essentially removing the bulk of cytosolic proteins prior to lysis of the ER, would potentially minimize non specific interactions between ERGIC-53 and cytosolic proteins. While we do not demonstrate this directly, figure 4B shows an approximately 50 kDa protein (Protein-X) that appears to co-purify with ERGIC-53 using our sequential lysis approach. While this band is also apparent when direct NP40 and RIPA lysis approaches are used, there are clearly other similar sized proteins also present which may complicate the identification of this band.
In the second case, studies of mutated recombinant proteins via transfection of cultured cell lines can often be misleading. In many cases, protein mutations lead to aberrant folding which can in turn lead to intracellular aggregation and insolubility in typical cell lysis buffers such as NP40. Using the approach described here, essentially all proteins from within a cell are effectively solubilized by use of buffers of increasing stringency, thus helping to ensure that every folded intermediate of a given protein is extracted which may help to give a more accurate picture of the effect of a mutation on the fate of a protein. We have ourselves, as shown briefly, recently used this approach in the study of mutations in cartilage oligomeric matrix protein that cause the dwarfing condition pseudoachondroplasia (Holden et al, manuscript in preparation).
While we are aware that the use of digitonin and detergent fractionation in the preparation of subcellular extracts is not a novel concept, as far as we are aware our protocol as presented here represents the first open access protocol that encompasses simple stepwise subcellular fractionation of cultured mammalian cells in totality, giving a complete extract of all cellular proteins. Given its simplicity, reproducibility and cost effectiveness we feel that this protocol will be extremely useful to scientists across a broad range of disciplines, particularly given its accessibility to all in this format.
List of abbreviations
Radio immunoprecipitation assay
Human embryonic kidney 293
Voltage Dependent Anion Channel
ER to golgi intermediate compartment
Cartilage oligomeric matrix protein
Laemmli sample buffer
This work was supported by grants from Shriners Hospitals for Children.
- Mooney RA: Use of digitonin-permeabilized adipocytes for cAMP studies. Methods in Enzymology. 1988, 159: 193-202. full_text.View ArticlePubMedGoogle Scholar
- Schulz I: Permeabilizing cells: some methods and applications for the study of intracellular processes. Methods in Enzymology. 1990, 192: 280-300. full_text.View ArticlePubMedGoogle Scholar
- Hide M, Ritleng A, Brizard J, Monte-Allegre A, Sereno D: Leishmania infantum: tuning digitonin fractionation for comparative proteomic of the mitochondrial protein content. Parasitology Research. 2008, 103: 989-992. 10.1007/s00436-008-1062-9.View ArticlePubMedGoogle Scholar
- Foucher AL, Papadopoulou B, Ouellette M: Prefractionation by digitonin extraction increases representation of the cytosolic and intracellular proteome of Leishmania infantum. Journal of Proteome Research. 2006, 5: 1741-50. 10.1021/pr060081j.View ArticlePubMedGoogle Scholar
- Huang Z, Li G, Pei W, Sosa LA, Niu L: Enhancing protein expression in single HEK 293 cells. Journal of Neuroscience Methods. 2005, 142: 159-166. 10.1016/j.jneumeth.2004.09.009.View ArticlePubMedGoogle Scholar
- Holden P, Keene DR, Lunstrum GP, Bächinger HP, Horton WA: Secretion of cartilage oligomeric matrix protein is affected by the signal peptide. The Journal of Biological Chemistry. 2005, 280: 17172-17179. 10.1074/jbc.M411716200.View ArticlePubMedGoogle Scholar
- Posey KL, Hecht JT: The role of cartilage oligomeric matrix protein (COMP) in skeletal disease. Current Drug Targets. 2008, 9: 869-877. 10.2174/138945008785909293.View ArticlePubMedGoogle Scholar
- Rasheed S, Nelson-Rees WA, Toth EM, Arnstein P, Gardner MB: Characterization of a newly derived human sarcoma cell line (HT-1080). Cancer. 1974, 33: 1027-1033. 10.1002/1097-0142(197404)33:4<1027::AID-CNCR2820330419>3.0.CO;2-Z.View ArticlePubMedGoogle Scholar
- Masters JR: HeLa cells 50 years on: the good, the bad and the ugly. Nature Reviews Cancer. 2002, 2: 315-319. 10.1038/nrc775.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.