Jump to content


  • Content count

  • Joined

  • Last visited

About Jens@mabtech.com

  • Rank
    Advanced Member

Profile Information

  • Gender
    Not Telling

Recent Profile Visitors

878 profile views
  1. Blank areas in IFNg ELISpot with human PBMCs

    Oh, I see! It looks like the cells have been "pushed" to the side of the well. That might happen if you add the cell suspension first and the stimulus after. In what order are you adding reagents? The best way is to first add medium including (or not including) the stimuli and the cell suspension last. Use for example 50% of the final medium volume for the stimuli (or just medium for negative control) and 50% of the final volume for the cell suspension.
  2. Blank areas in IFNg ELISpot with human PBMCs

    Hej Lena! It looks like you are experiencing the exact same issue as has been discussed in this thread: Briefly, probably some reagent have not reached the bottom of the well. We have seen these blank wells a couple of times, and we have yet to experience a blank well that cannot be "rescued" by repeating the detection steps again. Thus, the problem is not a lack of coated capture antibodies, but a one failed detection step. Please repeat the detection steps in the blank wells as follows: 1. Wash the blank wells 5x with 200ul PBS. 2. Add detection antibody at 1ug/ml in PBS containing 0.5% FCS and incubate for 2h. 3. Wash the blank wells 5x with 200ul PBS. 4. Add SA-ALP in PBS containing 0.5% FCS and incubate for 1h. 5. Wash the blank wells 5x with 200ul PBS. 6. Add substrate and incubate for 10min. I would be surprised if you didn't get results looking like this: So what to do to avoid blank wells in future experiments? Well, if you are too careful with emptying the pipette, there is a risk for a bubble formation in the bottom of the well, that blocks the following reagent to reach the bottom. Solution: •Look carefully at the plate for odd looking wells after addition of liquid. •Remove bubbles by tapping the plate against the bench. •Do not touch inner wall of the well with the pipette tip. Hope you manage to resolve the issue! Please come back to us if you don't (or, yeah, please do if you resolve it as well, would be nice for us to know :))
  3. Stimulation with target cells (co-cultures)

    Hej Esther! Co-cultures of e.g. cancer cells and effector cells is a quite common setup, although we hardly ever perform these experiments in-house. The important thing is that you find a suitable target:effector ratio and to include control wells with e.g. only the target cells to make sure that they don’t produce the cytokines of interest themselves. For protocol inspiration, you can have a look in this paper by Zuber et al 2005: https://www.ncbi.nlm.nih.gov/pubmed/16005014. Although we do not have extensive experience of using adherent cells in ELISpot/FluoroSpot we have not encountered any problems in the few experiments we indeed have made. The cell types we have analyzed include different epithelial cell lines where we have either looked at the production of cytokines (e.g. IL-6 and TNF-alfa) by these cells or have used these cells as APCs. When looking at things secreted by epithelial cells you actually get very nice spots looking very much like spots produced by lymphocytes and there seems to be no interference from possibly remaining cells sticking to the membrane. To my knowledge, there is no special protocol for using adherent cells in ELISpot. We have just used our regular protocols provided in the kits. However, the risk I think one may anticipate when using adherent cells is that they will stick more firmly to the membrane and by not being washed off properly will cause background problems. As said, we have not changed the washing protocol for in our in-house experiments with adherent cells, but logically you could think that some more harsh washing procedures would be required. We normally wash with only PBS in an automatic ELISA washer but one could consider using PBS with 1 mM EDTA and leaving it for some time in the well (most adherent cells would detach under these conditions). I know that some groups also include Tween in their washing buffer but we have never seen any beneficial effect of this but rather that you may sometimes see increased background. Hope you at least got some ideas! Don’t hesitate to come back if you have further questions. Kind regards,Jens
  4. Directly labeled antigen detection

    Hi again Josh, There is actually a paper, Luque et al 2018, in which the authors utilize a one-step detection system for FluoroSpot: https://www.ncbi.nlm.nih.gov/pubmed/30075182 However, they don't detect fluorophore-labeled antigen, but fluorophore-labeled HLA mulitmeres (thus looking for anti-HLA IgG in a transplantation setting). The HLA multimeres might have a higher binding capacity and bind more fluorophore molecules than a single antigen would, potentially increasing the signal, but at least it's some data showing the feasibility of a one-step detection system. Kind regards, Jens
  5. Directly labeled antigen detection

    Hi Josh, Welcome to the forum! Sounds like an exciting project! I can't say for sure that it would be impossible, but I think the signal loss might be substantial if you were to utilize direcly fluorophore-conjugated antigens. We have done similar experiments in-house, some of which have been published in Jahnmatz et al 2016: https://www.ncbi.nlm.nih.gov/pubmed/26930550. In this paper, we applied a two-step detection system using peptide-tagged antigens and subsequent anti-tag detection with fluorophore-conjugated mAbs (see attached image). The peptide tags - whose amino acid sequences are published in the paper - were recombinantly expressed with the antigen and we developed the fluorophore-labeled anti-tag mAbs, now QC:ed and readily available. We did not even consider testing directly conjugated antigens for three reasons: 1. There is a risk that the fluorophores hide the relevant epitopes of the antigens, essentially rendering them useless. Perhaps that is the issue you have been seeing in the past? A peptide tag can be positioned specifically, e.g. N-terminally and thus won't hinder antibody binding. 2. It can be hard to get consistent fluorophore conjugations on different batches of antigens. With our setup, we can work with large batches of QC:ed fluorophore-labeled antibodies instead. 3. With a peptide-tagged antigen, there is no need to purify the antigen; you can just use the antigen-supernatant. To my knowledge, there are only a few other papers utilizing similar setups, among them Hadjilaou et al 2015: https://www.ncbi.nlm.nih.gov/pubmed/26320246. The authors of this paper used fluorophore-labeled antigen-specific mAbs, and thus also they used a two-step detection setup. Two reasons for taking the Jahnmatz 2016-approach instead is that (i) tagged antigens circumvent that very need for antigen-specific detection mAbs, and (ii) that anti-tag detection mAbs won't compete for binding the same epitopes that the plasma cell secreted antibodies might recognize. The anti-tag mAbs are not yet officially launched, but send me an email (jens@mabtech.com) if you want to continue the discussion and/or would like to try those anti-tag mAbs. Kind regards, Jens
  6. No spots with human perforin ELISpot kit

    Hi E, Great to hear that you have good viability of the cells, and that you are willing try the BCIP/NBT-plus substrate as well as pre-coated plates. Please share your results if you’re up for it! Kind regards, Jens
  7. No spots with human perforin ELISpot kit

    Hi again E, Thank you for all the details. So, Christian and I have discussed your issue and come up with the following comments: • Perforin is a rather tricky cytokine to detect with good results. If you look in the ELISpot-well photo in this page, you see a visual example of how this system looks after 72h stimulation with a strong CEF peptide responder: https://www.mabtech.com/products/human-perforin-elispot-basic-kit-alp-3465-2a-0. You do get spots after 24h with a good positive control, but for antigen specific responses it is recommended to use 48h or 72h incubation. • You are using our ELISpotBASIC kit and thus coat the ELISpot plate with capture antibody yourself; what amount of capture antibody (Pf-80/164) are you adding to each well? Since perforin is a quite weak system, you really have to use the recommended concentration 30 ug/well. This is double our normal concentration and it is really necessary. If you dilute the capture antibody and use e.g. 10 ug/ml, you will get blanks. • Christian doesn’t think one can expect good results with Cellulose plates for this system. A plate with PVDF membrane, which is pre-treated with EtOH, is essential. The sensitivity of MAHAS4510-plates is lower, and there is a strong chance they won’t work with perforin. • When you pre-wet the PVDF plate with EtOH in your Test 3, does it go well? Have you seen our tutorial on YouTube?: • PMA and ionomycin are very strong stimuli, but they can seize to function for example after light exposure. There is also quite large variance in how well they work, at least in Christian’s experience. As a result, we would recommend to use anti-CD3 and anti-CD28 mAbs as positive control. The anti-CD3 mAb from Mabtech is very reliable, and together with anti-CD28 the stimulus is even more potent • For weak cytokines like perforin, we really recommend using the Mabtech ELISpot substrate. We only use BCIP/NBT-plus for ALP. The "plus" is important as it is much stronger than normal BCIP/NBT. You can see the difference in the bar graph here: https://www.mabtech.com/knowledge-center/tutorials-and-guidelines/enzymes-substrates/elispot-substrates • As you might have gathered from other threads in this Forum, most problems in ELISpot can be traced back to issues in capture antibody coating. As a result, Mabtech have pushed harder and harder for pre-coated ELISpot plates. They are more expensive, but they are produced under optimal conditions at Mabtech HQ in Stockholm, and in addition, pre-coated plates allow you to be spontaneous in that you can just take the plate out of the wrapper and start doing your assay. When coating yourself, you have to do it the day before, often when you are the most tired in the end of the day. So, please consider trying our pre-coated ELISpot perforin-kit. • Viability of cells is essential for ELISpot, and for perforin it is even more crucial than for other cytokines. For IFNg, you can still get some spots although you have only e.g. 60% viable cells at 100 000 cells/well, but with perforin the plate could be blank at that level of viability. According to Smith et al (2007) (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1865640/), 89% viability is a sort of "threshold" for getting reliable antigen specific responses in ELISpot. Below that, and you see a significant drop in spot counts of IFNg. Had the this study been done for perforin, we are pretty certain the viability demands could have been even higher. So, can you tell us something about your source of your cells? Are you assessing viability using trypan blue exclusion? Fresh isolated or thawed PBMC? Kind regards, Christian and Jens
  8. No spots with human perforin ELISpot kit

    Hi E, Would you please provide us with some more information of your protocol to allow us to trouble shoot this problem? I understand that you followed our protocol, but there are details in it that we leave for you as a researcher to decide. For example, what substrate did you use? How long did you develop spots for? How many cells did you add to each well? What was the viability of the cells? What medium did you use? What did you wash the plates with? Kind regards, Jens
  9. How can I wash adherent cells completely?

    Hi T, I think one should avoid using trypsin or accutase as they are both proteolytic enzymes and can lead to the digestion of both the antibodies and the analyte. Instead, we recommend a wash protocol including EDTA. In short, adherent cells are removed by first washing 3x with PBS, and then a 10 minute incubation at 37°C with 100ul PBS supplemented with 1mM EDTA in each well. After that, you just wash once again with PBS, and the membranes should be clean from cells. Please find the steps below and attached as a pdf. Mabtech_ELISpot_adherent_cells_protocol.pdf 1. Wash the cells 3x with 200ul PBS. If the cells should be recovered, this step should be done sterile in a sterile hood. If the cells will be discarded. The washes can be done in an automatic washer. NB! Wash vigorously to make sure that all medium (containing Ca2+ and Mg2+) is washed away before adding the 1 mM EDTA in step 2. 2. Add 100 ul PBS supplemented with 1mM EDTA. If the cells should be recovered, this step should be performed sterile. Incubate at 37°C for 10 minutes. NB! Since EDTA is more effective at 37°C it should ideally be pre-warmed before being added to the plate. For difficult cells, the time of incubation could be increased to 15 minutes. For further effect, you could also tap the side of the plate with your hand or put the plate on a lab shaker. 3. If the cells are to be recovered, hit the plate against the palm of your hand a few times and re-suspend vigorously with a pipette. Remove the cells to another plate and continue with your ELISpot/FluoroSpot assay. If the cells are not to be recovered, wash them away using 1x 200 ul PBS and then continue with your ELISpot/FluoroSpot assay as usual. NB! Do not touch the membrane with the pipette tips! Kind regards, Jens
  10. Maximum sensitivity in ELISpot?

    Hi M, The maximum sensitivity of ELISpot is hard to say, because in theory it’s unlimited. Let’s say you have an ELISpot well big as a bucket, and you add 100 billion cells in there, and add 1 cell you know are responding to your antigen, then yes, you would theoretically actually be able to detect it. It is a bit dependent on the cytokine secretion over time, as you point out, and that in turn depends on the particular cell and antigen. Unfortunately, we can’t say for sure exactly how much cytokine must be produced to see a spot. But normally, for low-frequent antigen-specific T cell IFNg-responses, many papers show a handful of antigen-specific spots in a population of several hundred thousand PBMCs. Of interest, we have a customer that FACS-sorted exactly 1 cell into an IFN-gamma/IL-8 FluoroSpot plate well: https://www.ncbi.nlm.nih.gov/pubmed/28754679 The authors of the linked paper were by that procedure able to first confirm that recent thymic emigrants secrete only IL-8 and not IFN-gamma, and then – by removing the still alive cell with a pipette and transferring it to another plate for cloning – show that the very same cell transitioned into an IFN-gamma producing Th1 cells. That’s not 1 cell in a population of 600 000 cells, which is what you are asking for, but at least it shows that it is possible to detect one single responding cell using FluoroSpot (and FluoroSpot and ELISpot have similar sensitivity.) ELISpot and FluoroSpot are the most sensitive assays out there, so if you are looking for a needle in a haystack, then these assays are the way to go. Kind regards, Jens
  11. False positve in IFN gamma ELISpot

    Hi Flavia, Sorry for the delay in response, you caught us in the middle of our midsummer's holiday. So, before in the experiment with primed+boosted mice you saw false positive spots in the "Ctrl vh +pept" wells, which are now gone. But instead now you have spots in the "Ad-GFP -pept" wells, a type of control you didn't include in your first run, correct? In essence, it looks like the T cells are secreting IFNg without the need for peptide re-stimulation. Could that be a true finding? Yes, it is not surprising. With the prime and the boost you are stimulating the T cells in vivo, and many of the cells are probably still activated and secreting IFNg even after splenectomy, cryopreservation and thawing. I don't know how long after prime or boost you are taking the spleens, but if we are talking days after immunization then what you see is most probably correct. We see this from time to time, especially in the context of Adenovirus which usually gives the T cells a real jolt in terms of IFNg-production. As you can see in your picture, the peptide re-stimulation thus however increase the number of spots, at least in exp 1. It's great to see also that you have solved the issue with false positive spots in the "Ctrl vh + pept" wells. //Jens
  12. False positve in IFN gamma ELISpot

    Great, Flavia! Good luck with your second experiment and please keep us posted! Kind regards, Jens
  13. False positve in IFN gamma ELISpot

    Hi Flavia and welcome to the Mabtech Forum! I think you are addressing the issue you're having in the best possible way. My guess is that it is the added IL-2 that is causing you false-positive background, so in my view it's good that you are removing it. You shouldn't need extra IL-2 for a short incubation like this anyway. Did you include it in your original protocol for a particular reason? The 1 million splenocytes/well is on the brink of being too many, so you're doing the right thing of decreasing that number to 250k and 500k cells/well. If you didn't have the no-stimulation-condition before, it's good that you have chosen to include that negative control now. Incubating for a shorter time, e.g. your suggested 16h-18h is also wise. The kinetics of IFN-gamma secretion is quite fast, and although you might catch a few extra responding cells at 24h, you also increase the risk of background. Whether to use serum-free AIM-V or RPMI1640 supplemented with FBS – I don't think it will make much of a difference. We always use RPMI+FBS in-house, but then again we have tight control of the FBS we are using. If you are unsure of the quality of your FBS, please go ahead and use AIM-V instead. In our hands, that is the serum-free medium that works the best. Finally, the words "transparent" (or "clear) and "white" in the plate names refer to the plastic of the plate, not the PVDF membrane. The membrane is always white. You can see the difference in plastic of two different MSIP-plates, transparent and white (your strip-plate is similar to the transparent one), in this video at approx. 50 seconds:
  14. Red blood cells in ELISpot

    Hi Lily, I hope another Forum-user than me can shed light on your quest for hard evidence papers showing the effect of RBC on ELISpot. As I wrote above, there are theoretical ways of investigating this, but I have yet to see them published in practice. But regardless, for sure, an RBC lysis of whole blood before an ELISpot assay could work out ok, and we have performed it occasionally in-house. But after talking to some of our veteran ELISpot experts here at Mabtech, our official answer is that we would advice against RBC lysis and thus still recommend the Ficoll separation method. Because: with Ficoll you also get rid of most granulocytes. Granulocytes may, just like Julia writes above for contaminating RBC, compete for space on the PVDF membrane, perhaps reducing the contact between e.g. APCs and T cells and thus lowering the sensitivity of the assay. In addition, contaminating (and activated) granulocytes may directly interfere with T cells, reducing their CD3z-chain expression, especially in blood stored for a long time before PBMC separation. I made a small "review" on the literature surrounding granulocyte contamination and how to avoid it, and the take-home message from that literature overview is that it if you're not preparing PBMCs from the blood immediately after collection, it is recommended to dilute the blood 1:1 in PBS or medium before storage. By an unclear mechanism, the dilution procedure reduces the negative effect that activated granuolcytes may have on T cell responses (at least true for IFN-gamma). I am attaching the literature overview here in its entirety, but please note that the content of this ppt-file is my personal take on the issue, and not an official Mabtech document. Granulocyte contamination in ELISpot.pptx
  15. Number of T cells per well

    Hi Sylvain, and welcome to the Mabtech forum! It sounds like a wise choice to go for ELISpot to detect tumor-antigen specific T cells. Here are my answers to your questions: 1. Given that the antigen-specific T cells are very low-frequent, I think you're doing the right thing sorting the T cells first. Normally in ELISpot, you plate between 50k and 500k cells, and in your case I would plate 300k - 400k T cells per well. That way the cells aren't too crowded but you still have plenty of purified T cells to evaluate in each well. 2. The phenomenon of T-to-T cell presentation do indeed exist in the context of peptide pulsing (although I'm not entirely sure how it works? Is the exogenous peptide added in such an excess that it replaces the endogenous peptide in the MHC groove at the cells surface? Or does it work by another mechanism?): Both other customers and ourselves have successfully pulsed total CD3 as well as pure CD8 T cell cultures with peptides resulting in good IFN-gamma secretion. I cannot tell if a pulsed T2 cell line would present the peptides in a more satisfactory way. You would have to evaluate that in your own setting. But for sure: It is possible to perform the test using only T cell cultures. 3. When talking IFN-gamma secretion, our experience is that most activity is over by 24 hours. Thus, to be sure you're getting the most sensitivity our of your assay, incubate the ELISpot plate for 24 hours. (If someone else that Sylvain reads this, please note that other analytes than IFN-gamma due to kinetics might benefit from longer or shorter culture times.) 4. The pre-coated ELISpotPLUS or ELISpotPRO kits won't give a higher sensitivity vs the classic ELISpot, but the pre-coated plates generally give lower background and less variability. Since high background was one of your concerns, I would go for either of those kits. Good luck with your experiment, and please come back with any further questions you might have! Kind regards, Jens