There's a similar page for SDS page here http://www.ruf.rice.edu/~bioslabs/studies/sds-page/sdsgoofs.html

Plenty of crazy examples!

In order to explain the most mysterious lightning from the previous picture there is an another example. This time the lightings appeared on the rolls of the X-ray film developing machine thus you observe this nice bush growing up. It happens when the air is dry, you wear synthetic clothes, especially in the winter at the temp. outside -20 degrees. This is problem with film only, probably you noticed this one is dirty due to not proper storage.

This Ponceau-S stained nitrocellulose shows the transfer result of two gels, in duplicate (out of 8 done by this transfer), where the samples were in the same order. Should look the same… The lower gel is an extraordinary example of not proper rolling the sandwich during the semi-dry-transfer. The rest-over of the buffer in between the sandwich layers led to the protein random electrophoresis. They grey smear in the upper left is very acceptable for me, since this is only water in the scanner. Please do not be afraid to do Ponceau-S staining since:

- it will not destroy your membranes,

- will take you 10 minutes with scanning,

- will exclude transfer as a problem in case you have no bands,

- is an prefect loading control in case of my mammalian tissues shown on this blot

Promised one;)

You can enter these in an art gallery.

Exhibition will be only if your pictures Dear Marcia and others will be posted. All of you are strongly invited to this artistic event. I really want to have here a huge international art collection. Anyway at the time of gel/transfer running I will go through my mess to pick-up some next curiosities you can expect.

I don't have any, but I will encourage others to participate. Take care, marcia

Since the antibodies are not always perfect, here you will find the opinions about the tips before you order:

https://www.researchgate.net/post/How_do_you_choose_a_good_antibody_for_western_blot

There's a similar page for SDS page here http://www.ruf.rice.edu/~bioslabs/studies/sds-page/sdsgoofs.html

Plenty of crazy examples!

Thank you Natalie for this link, already depressed me since the author of this page did what I wanted. Solid description with the native English, actually colleagues recognize mine language as Janglish. The perfection of this page is impressive, this level can be reached here only if I will have the support of my RG-colleagues, will add some photos please… Anyway this link concerns SDS-PAGE only, thus I feel it make sense to go on with WBs.

This picture shows WB with the antibody previously was recognizing our protein as an single band of the mass just over blue one in the marker. We took the next aliquot of this AB, repeated the previous “single-band” procedure with unexpected increase of the bands of the mass not corresponding to desired one should be just above. This is typical example of not specific AB binding, due to their disruption or rather the phase of the moon.

What I’m afraid is that such an artifacts I’m finding in prestige journals nicely shaved like on the panel below. Just above, just below doesn’t matter so much let’s pick-up the one fits our theory better. This is becoming a modern way of preparing the figures. I’m kind of old-fashioned guy, likes the art definitely, but this new trends are depressing me again, especially when I’m trying to repeat the results. If you see in the results section no 1st&2nd antibody concentration, nicely shaved photo, it could mean 1:100 for the first and 1:100 for the second. This unique combination will show you whatever you want in most of cases…

Very nice photo and 39 answers in order to find the APS was old. Please change your buffers/chemicals first in case you have problems...

https://www.researchgate.net/post/After_transferring_on_to_a_membrane_step_in_western_blot_my_marker_and_protein_bands_are_smeared_Why_does_this_happen2

For the artistic purposes I have adjusted the color composition of this X-ray film. This is typical outlook of WB with anti-hPNPase antibody used (done by myself). The protein was sliced out of the gel, than I have obtained 200 ml o the rabbit anti-hPNPase serum. Because of this huge volume I had a lot of material to set up WB, took me one year. After I have found accidentally, the tricks applied are useless, since the rabbit produced the antibodies against acrylamide… The dark areas around the gel are acrylamide dust on the membrane recognized by this serum. These AB was published twice (PMID: 12798676, PMID: 16505900) with conclusive results. Since the gel piece with the protein is still being used for immunization you can observe similar acrylamide spots on your membrane. The tricks I have applied are:

- Just pre-wet Whatman paper in water/PBS/TBS (or liquid you prefer) and wash the blot after transfer like with the sponge. Only problem is, that you need to be delicate not break the blot, and powerful to make sure you have removed all the dust, at once...

- Titration of the anti-acrylamide antibodies with acrylamide. Regular acrylamide you supplement with PBS/TBS add Temed and APS, break in smallest possible particles and incubate your AB O/N with agitation.

The same anti-hPNPase antibody as above, dark spots nicely repeated with measurable statistical significance. Color/saturation adjustment applied.

I want to focus here on additional effect of the grey pathways between the lines. No protein is there, thus this increased signal means, the loaded samples have blocked already the membrane. On other hand it shows, the blocking procedure was done badly. I do the blocking 1h 10% milk, for some antibodies 10 min. is enough, for some you need to do it O/N.

Here is a very mysterious gel made by one of my studenta. The lower parts of two blots were cut off and immunostained for tubulin-beta as loading control. Ponceau staining looked OK with no visible changes in protein separation and transfer. Both parts of the blots were incubated with antibodies against tubulin together and the antibody mix was used before a few times and always looked OK (after this blot we have thrown it away). The upper parts of the blot were immunostained for another protein and looked OK (but were unusable due to the lack of the proper loading control). Upper blot is showing five lysates and lower blot six lysates made from the same cell line (B16F10 melanoma) at the same time. Cells were treated with the factors which do not influence significantly viability of the cells. However, the look of the tubulin bands in lines no 3 and 5 in the lower blot are still a mistery for me.

Dear Patrycja, you are the first brave scientist, showing we are the community of the enthusiasts, unfortunately 95% of our results is placed into the trash bin. Thank you. Line 3 looks like protein degradation. If you still have the tube 5 open it to check if it smells with BME/DTT added by your student.

The same buffer was added to all the samples (for sure with beta-mercaptoethanol) with an anti-protease coctail from Roche. But yes, I agree, it really looks like our student was able to degrade the sample in spite of this coctail. I suppose it could be an effect of some saliva/snot in the sample due to the sneeze, so the coctail was not able to block everything? And the line no 5 looks like something landed inside the well which decreased the migration of the sample (some lint, a piece of paper?), especially that some leakeage is visible at the ends of this band. Nobody should underappreciate the ingeniousness of the students.

And thanks for the nice photo of the blocking procedure done badly. I will use it as example for some students who want to shorten the blocking time. :-)

These FPLC size-exclusion fractions were resuspended badly after TCA precipitation. The smear going up and the spot-like bands are typical outlook of not properly dissolved proteins are dissolving during the electrophoresis. The arrow indicates the protein of interest, should be observed in 3 fractions at least. The right panel shows the same preps after all the tricks applied to rescue them :( Such an aggregates will rather not dissolve, better is to do the next preparations. For TCA precipitation we are usually successful, when we wash 2X with 70% ethanol, 100 mM TRIS pH (7.5 – 8.5), bromophenol blue (BB, light blue in the solution:) Strong Tris is buffering the acid, you can monitor it with the color of the BB (blue, when properly washed, yellow means the additional wash is required). Also essential is to precipitate no more than 50-100 ug of the proteins in one eppendorf, since huge pellets are hard to wash and resuspend. This is the theory in this case was not working at all…

Nice photo. ;-) I was doing a lot of TCA precipitations and for preparation of WB samples we were doing it a very easy way. To the precipitated sample the lysis buffer was added together with sample buffer. The samples were of course very yellow. Then we prepared the supersaturated Tris (Tris was dissolved in water up to the moment some of the cristals were unable to dissolve) and then it was added small drop by the small drop into the sample up to the moment it turned blue. Then the sample was mixed, usually 5-6 times every 2 minutes on ice and then boiled. They looked beautifully., so I never seen such interesting pictures.

But I got very interesting smears, quite similar ones, when I got very big, almost completely insoluble inclusion bodies from my overexpressed protein. Pity, I was a student at that time and I have not thought about making the photos from such gels (it was expensive then, so the photos were made only from the good gels).

Adding to the collection of ugly blots

https://www.researchgate.net/post/Western_blot_not_looking_good

At the link above you will find well described by Jagdish conditions of this experiment. In the advices you can find the suggestions, what can be wrong, but to be honest today, there is no simple trick can be applied. Can anyone help, please?

@ Jagadish: Like I said in you other post look at the Jan gel here marked "These FPLC size-exclusion fractions were suspended badly after TCA .........."

There is clear protein precipitation.

The problem described by Mats appeared at another post:

https://www.researchgate.net/post/Western_blot_debugging_Cause_of_circular_grid_pattern_on_PVDF_membrane_after_transfer

Up to now it seems to be the only WB problem without clear solution. Please find it, since for this discovery huge prestige is expected. Or alternatively you can switch to semi-dry transfer if it is possible in case of your protein...

Modification done by me two days later:

On the post above there is full agreement the pads should be washed properly. This simple rule concerns all the WB procedures. For example my blot fall down in the dark room into developer on the floor gave no conclusive results...

First attempt at unfixing and blotting samples from PFA fixed brain tissue. Clearly did not unfix the tissue all the way because the protein wouldn't reduce. Such a mess. Problem was resolved on the 3rd attempt with a longer unfixing protocol.

How about when the bands look like this?

Dear Clare,

Your bands shape is due to co-migration with abundant protein or contaminants. Since these bands are sharp, the "W" shape of them is convincing me, this is true W-B, not the result of Photoshop...

Best - Jan

I have had this problem on and off when doing western blot and was wondering if I can get any insight into why this happens? My lanes look like they are swimming out of their lanes. I run transfers at 130 mA for 3 hours in a 4 C cold room (with an ice pack in the transfer box and a stir bar).

Dear Akiko,

Seems your sandwich was not enough tight pressed. Thus you observe the diffusion during the transfer trough the buffer before reaching the membrane. If you have this problem on regular bases, semi-dry apparatus cost is 1000 USD, will let you to spare your time. Anyway for avoiding of frustration during W-B I would pay more;) Apart of your transfer problem I suspect your antibody is giving non specific signal, since it appears from the top till bottom of the lanes. You are welcome to show more pictures, would be happy to help you.

Best - Jan

This gel was negatively verified before W-B procedure by stain-free procedure. This is the serious of ammonium sulphate total protein pre-purifications. 50% ammonium sulphate in samples gives the artistic effect of nicely diffused smears. Not rehydrated proteins you can load at the concentration of 500 micrograms/per lane. Than you will obtain an unique effect of acrylamide destruction at the entrance of the gel. Overloading with the proteins gives all the alternatives shapes to the the well known, already not modern “protein bands”. Many thanks to my beloved students, their artistic way of doing science gives me this art-grade paintings every day. Stain-free procedure gives you the opportunity to decide to do science or art with the gel.

Here's a pretty terrible Western blot (and Coomassie stain of the membrane afterwards). My conclusion was over-transfer (using a Bio Rad wet tank system), unless anybody else has input? It seems like the protein is not degraded, but the lower half of the membrane (lower MW bands) seem to have disappeared. 

Dear Urszula,

It seems your protein preps are terrible... Or too concentrated, or not sonicated or both... Please write your procedure of it's preparation to find the mistake.

Best - Jan

Hi Urszula:  Like jan says: 

1.  I wish we could see the actual lanes.  Since if the top of the lanes are very dark then your protein did not enter the gel.

2.  The streak would mean that there is lot of precipitate, or DNA, and the streak is just small amounts of stuff getting in the gel.  (whatever little is soluble).  The stuff is forced into a thin line because the natural tendency of the focusing effect of the technique + the small amount of the gel that the sample can access since the well is so blocked with ppt.

3.  Another reason could be your sample really did not enter the gel properly - not enough or wrong sample buffer, bad stacking gel, and other incorrect running conditions.

Although, this pattern is quite typical of DNA or precipitation.......

Dear Urszula and Prem,

The protein marker is fine, thus the gel and it's running doesn't seem to be guilty;)

Yes, it is in the sample lanes.

On ponceau in the lanes 3&5 a very dense preps vere runed. Was there any sugar or glycerol aded at high concentration? What dou mean by ppt?

best - jan

Can't decide which one's the worst :D I feel like I am going backwards with every western blot. :( I have no clue why does this keep happening. Miraculously, once, there weren't any random non-transferred regions and I thought I finally mastered the technique of not letting air bubbles creep in between my NC membrane and the gel. But, the next time I done it the same way, voila, my bad westerns were back! And I was back at the troubleshooting, and now I am out of options and ideas. The only thing that might be happening is that the gel is kind of getting heated up, sticking to the NC membrane, as I leave it for an overnight wet transfer, but as I study proteins which are around 110-200KDa, overnight wet transfer were the conditions that I optimized for myself. 

Hi Shruthi-

These artifacts are definitely not air bubbles. You are correct, the gel is getting too hot and sticking the the membrane, interfering with the transfer. I had the exact same problem a few years ago with the exact same 'splotches'. You will need to adjust your transfer time, voltage or current. What are your current conditions? 

For reference, here is my question:https://www.researchgate.net/post/Can_anyone_explain_these_small_colorless_splotches_on_my_nitrocellulose_membranes?_tpcectx=profile_highlights

 @Grant T Corbett: I am now sure that these are not air bubbles and as you said, the gel melted and interfered with protein transfer or the gel melted and got attached to the NC membrane, and hence when I dissemble the transfer setup, the transfer proteins might have been pulled off. I run an overnight transfer in Bolt and Mahoney buffer at 25V, 300mA, constant voltage at 4C. 

Glad you got it Shruthi!

Sometimes I used the 12% acrylamide SDS gel but it couldn't separate my proteins below 28 kda. All the proteins below 28 kda (including the ladder, the lower red one is 28 kda) were in the same line with the dye. However when I prepared the gel several days before and stored it in the complete running buffer 4°C, the gel worked fine. It could be the pH problem? Both the gel and the buffer contained 0.1% SDS. Thanks for your comments !!

It seems that the problem is with protein separation basically with the sample. Make sure that the concentration of ions is minimized in the sample. For this the protein precipitation will help.

The protein was isolated from kidney was runed  in 12% gel and got smeared band after 30KDa , B-Actin is perfect thin band but the desired protein is 14kDa badly smeared. 

Dear Raj, your blots looks nice, but which is actin, and where is marker?

Dear Imran,

Such a conditions does not exist in my eyes since for the range 2-150kDa you need to adjust SDS level from 0%-0.1%.

best- jan

Dear Raj,  I agree with Jan.  The marker positions would be useful on the whole gel.  b-actin (actin?) is around 42 KDa.  If the fat band is b-actin, and the "smear" 14kDa is just the one below it, then you have a situation where you are at the dye front, and your gel % is not appropriate for accurate separation of these.

Dear Imran, Do you mean 20 kDa and not 2 kDa.  2 kDa is not possible on conventional SDS-PAGE.

 Dear Jan piwowarski,  that marker is the first one which is 14 kDa , first one on row.i have added the b-actin band. i want to know why the  band is so thick and smeared ? 

Dear  Prem subramaniam,  B-actin is not that band, I have added another one picture of B-actin, that is last picture. The first two picture has  two band is LC3-I And LC3-II which is 14kDa

Dear Imran,  abeta = amyloid fragment??  OK, so what you are seeing is close to the dye front using 16-18% gel.  If you do not let the dye run out, and if your Ab is good you can of course detect the proteins in this area.   But generally Ab like clone 82E1 is quite specific.

Firstly, why not use a 4-20% gradient gel with overnight transfer.  You can get all ranges transferred.  Since your proteins are so far apart, this may be a solution.

1.  I have myself compared the iblot and traditional WB (Invitrogen XCell II Blot Module), and in my expts, the iBlot gives much lower signals that traditional WB for the same protein (50 kDa) + Ab combination.  Several time lower - need much longer exposure times. I have gone back to traditional WB.  7 mins vs 90 mins is not a big deal for good transfers. 

2. So, in my opinion, the best is to do simple overnight transfer on the same gel  (like 4-20%) and detect both simultaneously.  Since you have a 110 kDa maybe you should just use traditional WB.  The 4 kDa will transfer also.

3.  Instead of overnight run, why cannot you run traditional WB at 100 V for 1.5 hrs - 2hrs for old full immersion type, or 30-40V (start 170 mA) for

Dear Raj,

So the fat band is 14KDa ?? Is the last pic b-actin, and the exposures are the same?  You seem to have low b-actin signal compared to first two pics, or you exposed for lower time?

EDIT: Are you doing autophagy work? LC3-I signal is strong, and the "smearing" is actually saturation of the film.  If you look at the first lane the LC3-1 is not much smeared.  So some lanes have a lot more LC3 than others and the film is oversaturated.  Are you sure the fainter band below is LC3-II. ?

Also, my exposure question stands.

Dear,  Prem Subramaniam  yes the fat band is 14kDa, The exposure of B-Actin is more then the lc3. Yes i am sure the fainter band is lc3-II. Actually the marker is also smeared .

Hello everyone, I have ponceau stained image of transfer. In last lane, one could see the rocket like band from down end. Can anyone suggest reason for it?

hi ravindra, i would suggest you change your stacking buffer solution..ph6,8. don't know why but when this buffer gets old, we sometime observed this. we solve this by preparing stacking buffer and store 50ml aliquot at -20 , and we keep the workingsolution at 4°C.. give it a try and see

 Dear Ravinda,

What kind of samples are you loading?

best - Jan

Hello Jan, These are glioma cell line samples.

Dear all,

I transferred pure BSA protein from a 2D SDS-PAGE gel  onto nitrocellulose membrane. Protein transfer was succesful, but my blot looked terrible after antibody incubation. I performed this experiment before with success. So I know the protocol works well. I don't have any idea why it failed this time. Has anyone an idea what was going wrong? (The blot represents the part of the 2D gel where the BSA protein spot is present, not the entire 2D gel was blotted.)

Best wishes,

Valérie

any suggestions?

For this blot, I used heart probes. The first three are the controll groupes but a few weeks later, I also didn't get a usable band for this group. Other people in our lab using the same buffers didn't get this kind of result.

Thanks in advance.

So these are my saline-treated cerebellum samples. Lanes 1-9 are loaded with equal amount of protein. This membrane was stained for b-actin only, and as you see some bands are missing (my protein of interest, which is 96kDa is absolutely fine). What happened to some of my samples? Could it be that lower MW protein just went missing during homogenisation/heat treatment?

I am quite new to running western blots. The last few times that I have done western, although I am getting the desired results, there are lot of background signal that overpowers the signal for the protein. I am clueless as to how these appear on the blot and how to remove them.

I resolve the protein on a 12% polyacrylamide gel, transfer it overnight at 4 deg C on a PVDF membrane. Do a ponceau stain, wash it with PBST. The ponceau looks clean and stains properly. This is followed by blocking with 5% skimmed milk for 5h at 4 deg C and 30 mins at RT. I use a primary anti-biotin antibody incubation, overnight at 4 deg C and HRP conjugated secondary antibody incubation at RT for 2h.

Dear Prerona,

You have classic dust spots of unknown origin. Try to wash the blot with delicate sponch or Whatman paper using PBST/TBST, proir the blocking.

best - jan

Hello,

Just wanted to update that the issue has been sorted. There are lot of reports suggesting that milk has endogenous biotin, hence there are chances of cross reactivity with the primary anti-biotin antibody, resulting in spots and patches all over the blot. I have switched to using 3% BSA in PBST for blocking and it works. Here is the result. Hope this helps someone facing similar issues. :)

This is my 2 western blot nitrocellulose membrane (one membrane cut in 2, stained with AB1 and p-AB1).

Can anyone figure it out, what went wrong in this W-B?? I was supposed to get band at 65 Kda. But it is so weird.

Whole lysate from MCF10A.

Running 10% Gel, stacking gel 5%. Run the gel at cost 25mAmp 1h 30min. Transfer at 100 Volt cost, 1h 30min 4'C.

Heart shaped sample, was it the buffer? or too much protein in the sample?

I transferred protein from a 1D SDS-PAGE gel  onto nitrocellulose membrane. low molecular weight Protein transfer ( from 66kda to 14 kda) )was succesful, but high molecular weight protein didn't transfer at all...what can be the reason ?