Screening for Amyloid Aggregation by Semi-Denaturing Detergent-Agarose Gel Electrophoresis



SDD-AGE is a useful technique for the detection and characterization of amyloid-like polymers in cells. Here we demonstrate an adaptation that makes this technique amenable to large-scale applications.

Cite this Article

Copy Citation | Download Citations | Reprints and Permissions

Halfmann, R., Lindquist, S. Screening for Amyloid Aggregation by Semi-Denaturing Detergent-Agarose Gel Electrophoresis. J. Vis. Exp. (17), e838, doi:10.3791/838 (2008).


Amyloid aggregation is associated with numerous protein misfolding pathologies and underlies the infectious properties of prions, which are conformationally self-templating proteins that are thought to have beneficial roles in lower organisms. Amyloids have been notoriously difficult to study due to their insolubility and structural heterogeneity. However, resolution of amyloid polymers based on size and detergent insolubility has been made possible by Semi-Denaturing Detergent-Agarose Gel Electrophoresis (SDD-AGE). This technique is finding widespread use for the detection and characterization of amyloid conformational variants. Here, we demonstrate an adaptation of this technique that facilitates its use in large-scale applications, such as screens for novel prions and other amyloidogenic proteins. The new SDD-AGE method uses capillary transfer for greater reliability and ease of use, and allows any sized gel to be accomodated. Thus, a large number of samples, prepared from cells or purified proteins, can be processed simultaneously for the presence of SDS-insoluble conformers of tagged proteins.


Part 1: Preparing the gel

  1. Assemble the gel casting tray. Standard equipment for horizontal DNA electrophoresis can be used. For large numbers of samples, we prepare a 20 cm x 24 cm slab with up to four 50-well combs. Make sure the slab does not have scratches, as these can distort the blot image.
  2. Create a 1.5% agarose solution (medium or high gel-strength, low EEO) in 1X TAE. The volume should be enough to completely submerge the comb teeth -- you will want to load as much sample as possible to maximize detection. Microwave the mixture until the agarose is completely dissolved.
  3. Rapidly add SDS to 0.1% from a 10% stock. Swirl to mix. If some agarose solidifies during this step, redissolve using a hot plate and be careful to avoid boiling.
  4. Pour the solution into the casting tray. Use a comb to rake out any bubbles, as they could later interfere with transfer.
  5. After gel has set, remove combs and place the gel into the gel tank. Completely submerge the gel in 1X TAE containing 0.1% SDS.

Part 2: Preparing samples

  1. For high-throughput analysis of yeast lysates, we start with 2-ml cultures grown overnight with rapid agitation in 96-well blocks. In this case, each culture is overexpressing a protein of interest. When analyzing low abundance proteins, larger culture volumes must be used
  2. Harvest cells by centrifugation at 2000 RCF for 5 min at room temperature.
  3. Resuspend cells in water and centrifuge again.
  4. Resuspend in 1 ml spheroplasting solution. Incubate for approximately 30 min at 30°C (you may confirm spheroplasting efficiency by microscopy).
  5. Centrifuge at 800 RCF for 5 min at room temperature. Completely remove supernatant.
  6. Resuspend pelleted spheroplasts in 100 ml lysis buffer.
  7. Cover the block with tape and vortex on high speed for 2 min.
  8. Pellet cellular debris at 4000 RCF for 2 min.
  9. Carefully remove supernatant to a new container, e.g., a 96-well PCR plate.
  10. If desired, determine the protein concentration of the lysates.
  11. Add 4X sample buffer to the samples to generate lysates containing 1X sample buffer. Incubate for 5 min at room temperature.
  12. Load gel. If desired, save half of the sample volume and boil it for SDS-PAGE analysis. In order to monitor the extent of transfer later on, include a pre-stained SDS-PAGE marker. Additionally, a molecular weight marker consisting of very large proteins (e.g., chicken pectoralis extract) can be used to estimate sizes of the resolved complexes.
  13. Run at low voltage (≤ 3 V/cm gel length) until the dye front reaches ~1 cm from the end of the gel. This will take several hours. It is important that the gel remain cool; otherwise diffusion of low molecular weight protein (e.g., monomers) can limit their detection.

Part 3: Transfer

  1. Cut a piece of nitrocellulose to the same dimensions as the gel.
  2. Cut 20 pieces of GB004 and 8 pieces of GB002 blotting paper, to the same dimensions as the gel. Cut an additional piece of GB004 to be used as a wick; make it about 20 centimeters wider than the gel.
  3. Immerse the nitrocellulose, wick, and 4 pieces of GB002 in 1X TBS.
  4. In a plastic container, assemble a stack of papers as follows: 20 pieces of dry GB004, then 4 pieces of dry GB002, then one piece of pre-wet GB002. Lay the nitrocellulose on top of this stack.
  5. Rinse the gel on the casting tray briefly in water to remove excess running buffer. Then, carefully begin to slide the gel off the tray onto the stack. While sliding the gel off the tray, douse the membrane with TBS as necessary. The extra buffer helps prevent bubbles from becoming trapped under the gel. A transfer pipette works well for this purpose.
  6. After the gel has been moved to the stack, check thoroughly for bubbles. If any are present, lift the edge of the gel and reapply buffer until the bubbles can be worked out.
  7. Put the remaining three pre-wetted GB002 pieces on top of the gel. Ensure thorough contact between all layers by rolling a pipette firmly across the top of the stack.
  8. Flank the transfer stack with two elevated trays containing TBS. Drape the pre-wet wick across the stack such that either end of the wick is submerged in TBS.
  9. Cover the assembled transfer stack with an additional plastic tray bearing extra weight (e.g., a 500 ml bottle of water).
  10. Allow the transfer to proceed for a minimum of three hours, or overnight.
  11. After transfer, the membrane can be processed by standard Western blotting.

Spheroplasting Solution
1.2 M D-sorbitol
0.5 mM MgCl2
20 mM Tris, pH 7.5
50 mM BME (add fresh)
0.5 mg/ml Zymolyase 100T (add fresh)

Lysis Buffer
100 mM Tris 7.5
50 mM NaCl
10 mM BME (add fresh)
protease inhibitors (add fresh)

4X Sample Buffer
20% glycerol
8% SDS
bromophenol blue to preference


SDD-AGE was first reported by Kryndushkin et al.1, to study SDS-resistant complexes of the [PSI+] prion in yeast, and has since found widespread use studying both prion and non-prion aggregates 2-9. However, transfer of the proteins to a membrane following electrophoresis in an agarose gel is problematic, and can result in a distorted blot image 5. Additionally, the submerged electroblotting technique most commonly used introduces practical limitations for the size of the gel and thus the number of samples that can be processed. We have addressed these problems by employing downward capillary transfer 10, a simple procedure which uses a stack of dry blotting papers to transfer proteins from the gel to a nitrocellulose membrane. Capillary transfer prevents distortion and allows large gels to be processed easily. There are a few things to consider before using SDD-AGE. For crude samples (e.g., lysates), immunodetection of specific proteins is necessary. SDD-AGE does not fully denature the protein complexes of interest, so the protein(s) to be detected must bear an epitope tag outside of the amyloidogenic region. Lysates can generally be prepared as they would be for a normal SDS-PAGE, with two important differences. First, increased care must be taken to prevent degradation by proteolysis. The partially denaturing conditions used here are not sufficient to inactivate proteases, and can also make target proteins more susceptible to proteolysis. Use a complete protease inhibitor cocktail at at least two-fold the recommended concentration. Second, heating the samples should be avoided. If an all-monomer negative control is desired, for instance to confirm that high-molecular-weight species are not due to covalent modifications, a 10-minute incubation at 95°C can be used, which will restore most amyloids to monomeric protein.


We thank Simon Alberti for assistance with developing this protocol. This work was supported by a grant from the National Institutes of Health (GM25874), a Howard Hughes Medical Institute Investigatorship (to S.L.), and a National Science Foundation predoctoral training grant (to R.H.).


Name Company Catalog Number Comments
zymolyase 100T Seikagaku Corporation 120493-1
Halt Protease Inhibitor Cocktail Thermo Fisher Scientific, Inc. 78429
Hybond-C extra nitrocellulose Amersham RPN303E
GB004 blotting paper Whatman, GE Healthcare 10427926
GB002 (3MM Chr) blotting paper Whatman, GE Healthcare 3030-917
Agarose (UltraPure) Invitrogen 15510-027



  1. Kryndushkin, D. S., Alexandrov, I. M., Ter-Avanesyan, M. D., Kushnirov, V. V. Yeast [PSI+] prion aggregates are formed by small Sup35 polymers fragmented by Hsp104. J. Biol. Chem. 278, 49636-49643 (2003).
  2. Alexandrov, I. M., Vishnevskaya, A. B., Ter-Avanesyan, M. D., Kushnirov, V. V. Appearance and Propagation of Polyglutamine-based Amyloids in Yeast: TYROSINE RESIDUES ENABLE POLYMER FRAGMENTATION. J. Biol. Chem. 283, 15185-15192 (2008).
  3. Allen, K. D., et al. Hsp70 chaperones as modulators of prion life cycle: novel effects of Ssa and Ssb on the Saccharomyces cerevisiae prion [PSI+]. Genetics. 169, 1227-1242 (2005).
  4. Aron, R., Higurashi, T., Sahi, C., Craig, E. A. J-protein co-chaperone Sis1 required for generation of [RNQ+] seeds necessary for prion propagation. The EMBO journal. 26, 3794-3803 (2007).
  5. Bagriantsev, S. N., Kushnirov, V. V., Liebman, S. W. Analysis of amyloid aggregates using agarose gel electrophoresis. Methods in Enzymology. 412, 33-48 (2006).
  6. Borchsenius, A. S., Muller, S., Newnam, G. P., Inge-Vechtomov, S. G., Chernoff, Y. O. Prion variant maintained only at high levels of the Hsp104 disaggregase. Current Genetics. 49, 21-29 (2006).
  7. Salnikova, A. B., Kryndushkin, D. S., Smirnov, V. N., Kushnirov, V. V., Ter-Avanesyan, M. D. Nonsense suppression in yeast cells overproducing Sup35 (eRF3) is caused by its non-heritable amyloids. J. Biol. Chem. 280, 8808-8812 (2005).
  8. Tank, E. M., Harris, D. A., Desai, A. A., True, H. L. Prion protein repeat expansion results in increased aggregation and reveals phenotypic variability. Mol. Cell. Biol. 27, 5445-5455 (2007).
  9. Douglas, P. M., et al. Chaperone-dependent amyloid assembly protects cells from prion toxicity. Proc. Natl. Acad. Sci. USA. 105, 7206-7211 (2008).
  10. Nagy, B., Costello, R., Csako, G. Downward blotting of proteins in a model based on apolipoprotein(a) phenotyping. Analytical Biochemistry. 231, 40-45 (1995).



  1. Thanks. This is the best form to communicate results, I have this idea long time ago.. I would like to publish my results like this soon. Please let me know How I need to do it. COngratulations!!!   Dr. Gabriela A. Balogh Genetic Laboratory CERZOS-CONICET BAHIA BLANCA-ARGENTINA EMAIL: tel: 54-²91-48611²4 Int:188

    Posted by: Anonymous
    October 6, 2008 - 7:53 AM
  2. Hi Gabriela, I agree this publication format is very useful.  It really lowers the activation barrier for people considering using a new technique, and it reaches a large audience.  There is an online submission process that you should start with when you are ready to submit.  If you are concerned about how to make the video, just email the editors.  They are very helpful.  For my submission, I emailed the editor with a detailed protocol, then worked with them to make a script.  A videographer was provided by JoVE to record the video, but I'm not sure if they can do this for all shoot locations.  Overall they made the submission process very easy.  Good luck! Randal

    Posted by: Anonymous
    October 6, 2008 - 12:56 PM
  3. very cool work done

    Posted by: Anonymous
    October 21, 2008 - 10:20 PM
  4. This definitely is a new format of publication. It's so cool. Not only a combination of video and paper. But also a very good method to show the ideas of authors. Also I learned a lot tips from this great work that I can not found from the pdf files.   Haitao Zhang Department of Cell Biology and Molecular Genetics Microbiology Bldg University of Maryland College Park, MD ²074² 301 405 0547

    Posted by: Anonymous
    October 21, 2008 - 10:33 PM
  5. This new way to share methods and protocols is very nice, though I think that  it will not be simple for people around the world to produce high quality videos. I've a question for Randal, regarding  the statement that "a 10-minute incubation at 95°C  will restore most amyloids to monomeric protein". Is it a personal observation or is it based on published data? Isn't the resistance to the denaturation in boiling Laemmli buffer (which contains much more SDS than your partially denaturing buffer) usually reported as a typical feature of many amyloid aggregates?   Stefania

    Posted by: Anonymous
    November 27, 2008 - 10:07 AM
  6. Hi Stefania,   While the concentration of SDS in the gel and running buffer is only 0.1%, it is ²% in the sample buffer. Previous work with Sup35 prion amyloids has shown that they are denatured under these conditions (²% SDS, 10’ at 95C). For an example, see: Mechanism of Cross-Species Prion TransmissionAn Infectious Conformation Compatible with Two Highly Divergent Yeast Prion Proteins.  Cell ²005, Volume 1²1 , Issue 1 , Pages 49 - 6² M . Tanaka , P . Chien , K . Yonekura , J . Weissman   Amyloids do vary in their stability when boiled in SDS, as aβ and polyglutamine fibers have been reported to resist such a treatment (see work by R. Wetzel). But in my experience, the vast majority of intracellular amyloids (including all aggregates visualized in the results section of this protocol) are readily solubilized at 95C in ²% SDS.

    Posted by: Anonymous
    December 6, 2008 - 4:38 PM
  7. thanks alot for sharing methods. I have a question, can I make the lysis buffer and Spheroplast solution without BME? DŒs BME play an important role in this procedure?

    Posted by: mahshid R.
    February 21, 2009 - 8:01 AM
  8. thanks alot for sharing methods.I have a question, can I make the lysis buffer and spheroplast without BME?

    Posted by: mahshid R.
    February 21, 2009 - 8:10 AM
  9. thanks alot for sharing methods. I have a question, can I make lysis buffer and spheroplast solution without BSE?

    Posted by: Anonymous
    February 23, 2009 - 2:30 AM
  10. thanks alot for sharing methods. I have a question, can I use lysis buffer and spheroplasting solution without BME? thank you

    Posted by: Anonymous
    February 24, 2009 - 4:17 AM
  11. Hi Randal, do you have a feel if this would work for amyloids in brain homogenate? Do you have a feel if the amyloid associated proteins in a plaque (e.g., stuff bound to abeta) would also be resistant to solubulization in SDS? I was wondering if this would perhaps be a way to enrich plaque proteins for LC-MS analysis.

    Posted by: Roger M.
    June 29, 2009 - 9:54 PM
  12. Hi Rojelio. I've never tried it, but I imagine SDD-AGE will work just fine for brain homogenate. I know that people are successfully using it with cell culture lysates. It's definitely worth giving it a try on a small scale. I doubt that non-amyloid plaque associated proteins will remain structured in the SDS treatment, however. Also, in thinking about it, I think you'd have to find a way to solublilize the SDD-AGE-resolved aggregates prior to LC-MS. If you find a way to couple this with identification of amyloid proteins I would be very eager to know! Good luck!

    Posted by: Anonymous
    July 5, 2009 - 11:21 PM
  13. Thank you for posting this article! I have been using agarose gels to resolve aggregates from multiple Huntington's disease models including cells and mouse brain tissue. I am strugging to find a good high molecular weight marker. I have recently tried using the CPE, but I haven't really had any luck resolving the titin band. Do you have any suggestions about a marker to use? Thank you!

    Posted by: Anonymous
    January 28, 2010 - 3:05 PM
  14. Hi Emily. Generally we don't use the high molecular weight marker, as it only provides a rough estimate anyway (the amyloids likely migrate differently than large soluble proteins). We often include a normal SDS-PAGE marker (highest band at ²50kD) to verify transfer and show at least that proteins are running higher than ²50kD. I've also tried Invitrogen's NativeMark protein standards and found that some of the proteins do not denature on SDDAGE and can provide an alternative to CPE. However, I think the best marker may be von Willebrand factor, a serum protein that forms covalent multimers up to 1.8mD, although I haven't tried it yet. Here's a reference you can start with:
    Bone Marrow Transplantation (²007) 40, ²51&#x²013;²59; doi:10.1038/sj.bmt.17057²4; published online 4 June ²007
    Prophylactic fresh frozen plasma may prevent development of hepatic VOD after stem cell transplantation via ADAMTS13-mediated restoration of von Willebrand factor plasma levels
    M Matsumoto, K Kawa, M Uemura, S Kato, H Ishizashi, A Isonishi, H Yagi, Y-D Park, Y Takeshima, Y Kosaka, H Hara, S Kai, A Kanamaru, S Fukuhara, M Hino, M Sako, A Hiraoka, H Ogawa, J Hara and Y Fujimura.
    If you obtain a source of von Willebrand Factor (either purified or crude, plus antibody), you'd be all set.


    Posted by: Anonymous
    January 30, 2010 - 3:51 PM
  15. Hello Randal, I have also had issues with CPE, so am looking into using VWF, as you suggest. However, VWF runs as many different size multimers, so how do you suggest we determine the molecular weight of each band if we plan to use it as a ladder?

    Posted by: Nayan L.
    May 1, 2014 - 9:42 AM
  16. Dear Dr Emily
    Im a pHd student from Portugal. The reason why I send you this e-mail is because im planning to perform Semi-Denaturing Detergent-Agarose Gel Electrophoresis in brain samples. In the article from Randal Halfmann they do not refer the protocol to prepare protein extracts from mouse brain.
    Can you give the protocol that you use?

    Thank you

    Posted by: Anonymous
    February 22, 2010 - 5:35 AM
  17. A protocol on tissues or mammalian cells would be most useful! Or a reference would help. Would it be enough to break the cells in the lysis buffer, spin down and use the supernatant?

    Posted by: Anonymous
    March 31, 2011 - 10:45 AM
  18. To answer my own question, I tried it on various preps and it worked fine!

    Posted by: Anonymous
    July 28, 2011 - 11:00 AM
  19. Hello! We are attempting to perform this protocol on yeast to test to see if a natively expressed protein aggregates at mid-log phase under different genetic backgrounds. For this, we are collecting 50 mL cells at OD 0.5, flash freezing in liquid nitrogen, then lysing as described in the (excellent) protocol above. However, we are not getting very good gel results. Do you have any advice? We are using GFP tagged protein and running our gels for about 8 hours at 3 V/cm in the cold room and performing the capillary electrophoresis as described. Any help could be greatly appreciated. Thanks!

    Posted by: Kevin M.
    October 14, 2016 - 2:52 PM

Post a Question / Comment / Request

You must be signed in to post a comment. Please or create an account.

Usage Statistics