The bane of my existence for some time now.

I dont talk much about my research here, but its safe to say I consider myself an inventor who works in virology. People say "Damn. Its so hard to do ___. I wish there was an easier way..." and Im like "... I can do that."

Gimme a plasmid, a primer (just need one), a paperclip, and a package of 'Fruit Stripe' gum, and I can make your retroviral dreams come true.

I make stuff, and then pass it off to other people to like, do research with it.

So I made myself a couple new toys. I then have to characterize my toys so other researchers know how to use them (basically write the product stats and instruction manual). The 'instruction manual' has been done for months, but there was one thing holding the publication up.


Nef is not a 'core' retroviral component (gag, pol, env, LTRs)-- its an 'accessory' HIV-1 (and some of its relatives) has. Has some neat features from the perspective of a virus and from the perspective of evolution as a whole. But like, I fidgeted with it a little bit, and I really thought it should have been there... Yeah...

I couldnt find it ANYWHERE.


Well, first of all, its hard to *find* Nef in the first place. Nef is this teeny tiny protein. Its not an enzyme, so you cant test for its presence with some kind of simple assay (substrate + enzyme? = yes or no). It doesnt glow. It pulls cellular proteins off the surface of infected cells, but its kinda hard to really 'see' that in an assay, and then there are other HIV-1 proteins that do the same thing. Even if you saw an 'effect' of down-regulation in infected cells, you couldnt be sure it was BECAUSE of Nef. It wouldnt be as cut-and-dry as say, a reverse transcriptase activity assay (either RT is there and reverse trascription occurred, or it isnt and nothing happened).

What I needed were antibodies to Nef.

Antibodies with enzymes glued to them, that were specific for Nef proteins, could be used to detect the Nef protein in a Western Blot. Or, my go-to-protocol-- make everything radioactive in a cell with radioactive sulfur (35S. every methionine, the start of every protein, will have a radioactive sulfur, with means all of your proteins will be 'hot'. cysteines are included too, but methionine is the big one). Anyway, when all your proteins are 'hot', you can use antibodies to Nef to purify the Nef protein, run it out in a gel, see a radioactive band the size of the protein you are interested in, YAY! Its called an immunoprecipitation. I mean seriously, this is like *my* thing-- when in doubt, make it radioactive! YAAAAY!

Minor problem: I have done approximately 5 million Western Blots and 10 million IPs.

No Nef.




If your antibodies dont see the protein, you dont see the protein.

I ordered antibodies from EVERYWHERE. EVERYWHERE. They didnt WORK. So finally I did what I should have done right off the bat-- contact my go-to-reagent company.

Me: Hey, hey man. Can you hook me up with some anti-Nef?
Tech Dude: Yeah! We got that.
Me: What kind you got?
Tech Dude: We got some hyperimmune sera. Rabbits.
Me: *shudder with ecstasy* Does it work on IPs?
Tech Dude: I dont know about that. But it works in Westerns 1:8000 primary. And in ELISAs.
Me: 1:8000 PRIMARY?? *shudders again* Hey, I dont suppose you could hook me up with some of your immunogen for a positive control. I just need a little, man. Common.
Tech Dude: Sorry, we dont sell that. Its just in house.
Me: *shamelessly begging for 45 minutes*

See, the antibody was just one side of My Problem. For all I knew, the antibodies I was using were fine, but my timing with the virus was off. I had a 'positive control' for Nef-- just a regular ol lab strain of HIV-1. But Nef is only made very early in infection, and then HIV-1 moves on to other proteins. If my timing was off, if I was too late, I would never see Nef, even if my antibodies were fine.

I needed some pure, free base Nef for a positive control...

So I had to make that. Amplify the nef gene, plop it into an expression vector-- a DNA sequence that you can put into cells that just screams "GO!! GO!! GO!! GO!! GO!! GO!! MAKE NEF!!!!"

But how do you know you made Nef from that unless you have an antibody that you know works??? Well, then I had to put a little tag on the Nef that could be seen by an antibody I know works. So if I could see Nef with the tag antibody, but I couldnt see it with the Nef antibody, then I knew the protein was fine, but the antibody sucked. And if I didnt see either one, the protein sucked.

So I got THAT to finally work.

So I finally had ALLLL the controls I needed to do the one experiment I needed.

And Nef wasnt there in my new toy.

I mean, thats okay, its just something I need to tell everyone in the instruction manual. But all that time. All that effort. All that frustration... and Nef was never there.

I told one of the undergrads that this had to be an analogy for something.

And then it hit me:


Nef is like my Jesus. People pouring their hearts and souls out to him, their time and money into him, desperate for him, and he was never there.


Or its like Kenny.



But I definitely killed it. Now I can move forward.

More like this

Wow, biochemistry even hurts when I'm reading about someone else doing it.

My lab just got in some new antibody we had made... in chickens! They sent us 12mgs... sooooo excited.

LOL! Nooo! I love it! My IPs could be framed in The Louvre!

Its just when things dont work... *blink*... :-/

This is the best post I have read in ever.

Also, my life. At the moment I'm so deep in controls for controls for controls I can't even remember why I'm doing anything.

That was awesome. I love reading someone passionately geeking out over their work.

By the way, if you want people outside your field to commiserate with regarding building your own tools just to find out you were wrong in the first place? Programmers. We live and breathe that shit.

By Bryan Elliott (not verified) on 06 Jul 2011 #permalink

Also, my life. At the moment I'm so deep in controls for controls for controls I can't even remember why I'm doing anything.

But the second you dont include all 47 billion controls you need, you do an experiment, get wonky results, and you have to do it all over again with the 47 billion controls anyway.

There are no shortcuts :(

There are no shortcuts :(

...I am so not gonna be done in three more years. Today I got sequences back from a gel extraction. Simple, right? Right sized band, working primers, just have to verify I'm hitting what I mean to hit...aaaand it seems to be cloning vector. Except the right size. With primers for a protein coding gene not in cloning vectors. Back to the drawing board.

Truth. There are no shortcuts.

I work with another very notorious bug you can get if you arent careful - Chlamydia. No genetic system. No knock outs. No point mutations. More aggravation for me as I struggle to design experiments any other bacteriologist wouldn't need to think about.

You know, I bet every time something like this happens, you'd start thinking "Why didn't I just do X first?" where X is what solved the problem.

Of course, the sensible part of your brain (most of it) would know you'd gone about things in a sensible way, but frustrated ERV would still be screaming.

By hellocthulu (not verified) on 06 Jul 2011 #permalink

I feel we must have had the same chat with the same tech dude from a major antiBoDy supplier that I won't mention, of course. Begging got me nowhere. Hunting down phosphorylated endothelial nitric oxide synthase in pulmonary myofibroblasts has been the bane of my existence for the last year and a half.

By eNOS, but not … (not verified) on 06 Jul 2011 #permalink

Well, I certainly under the process of iteration and tons of time and effort only to find out the pathway one is heading down has been completely a goose chase. However, I'm not a biologist, so that was mostly lost on me.

Alas, I'd be a shitty biologist. My mind just doesn't seem to work that way.

If you keep writing like this I'll likely ditch math and physics. Seems there's more going on at biology's house.

By bibliovore (not verified) on 06 Jul 2011 #permalink

I share Western blot woes... multiple mysterious bands that won't go away because it's some weird antibody from a tumour cell line that appears not to be monoclonal, but it's the only antibody for that protein in existence being made by one man in one lab... And I will cry if I have to grow, harvest and qPCR another cell line that I'm *promised* is EBV- that turns out to be EBV+. Especially as one of the ones I would use normally is XMRV/XMLV+ *twitch*

But the second you dont include all 47 billion controls you need, you do an experiment, get wonky results, and you have to do it all over again with the 47 billion controls anyway.

Not necessarily, you could always join the WPI!

The radio said "No Abbie. You are the nef"
And then Abbie was a virus.

@Assays: What about mass spectrometric methods?

NJ: Do not try and find the Nef. That's impossible. Instead... only try to realize the truth.

ERV: What truth?

NJ: There is no Nef.

*Mystic gesture* These are not the proteins that you are looking for.

By Stephen Wells (not verified) on 07 Jul 2011 #permalink

I so feel for you. The bane of my grad school existence was the elemental analysis of a single polymer. My boss expected half a percent max deviation to allow publication, and I was off by 1.5%. The C/H ratio was right, but not the total percentage. TGA - no residue (aka no inorganic crap trapped in it), NMR -every single peak assigned, all integrals as good as you can get, redisolved, reprecipitated a half dozen times, still off. Until I got desperate and ordered an all-elements analysis (we got C/H in house for free, those cost real money), and found a trace of Cl in the material. Who'd thought that 3 months in a 1 Torr vacuum oven at 100C doesn't get rid of 40C boiling methylene chloride if it really loves you material.

I love the part about using antibodies. It is another example of the specificity of antibodies. AIDS Denialists claim that HIV tests are useless because they test for antibodies and not the virus itself. Could you do a post speaking to the specificity of antibodies and how they are more than adequate for HIV ELISA and WB tests? I have used examples of vaccinations and even ABO Rh blood group but I canât get thru to them. How about it? Would you write a post about the specificity of antibodies?

I love the part about using antibodies. It is another example of the specificity of antibodies.

Unless you purchase them from Santa Crap, hah! But that can happen with any antibody once you bring it out and try to use it in an assay. Then the antibody is like "BSA? OMG, I effin' LOOOOVE BSA! Who's that other protein? Sod him for a laugh."

I really despise how companies now routinely refuse to even tell you the sequence of the epitope they've used. Really? If I wanted to inject rabbits for 6 months, I wouldn't bother buying your shit in the first place.

@11 At this point I think 7-11 may even sell antibodies more effective than what I'm using.

@22 HA! The same goes for goat serum. "OMG, something to bind to!!! OMNOMNOM, I don't care about anything else!"

By eNOS, but not … (not verified) on 07 Jul 2011 #permalink

When I was in grad school, I remember one of my committee members saying "You'll look back on this as the best time of your life." And I remember thinking "What the fuck are you smoking? I'm depressed, I'm exhausted, I'm doing failed experiment after failed experiment and I'm no closer to publishing than I was a year ago! What the FUCK are you smoking?"

Now I have my long sought-after TT position and I'm dealing with grant-writing bullshit and teaching whiny students and fucked-up departmental politics, and I'm finally beginning to understand what he meant.

I'd trade places with you in a heartbeat. Good luck, Abbie!

By Amenhotepstein (not verified) on 07 Jul 2011 #permalink

Ha! Reminds me of my personal quest to find the S2 ORF of EIAV (which sits where vpx/vpu/vpr would be in the hoo-man virus). No antibodies to pony viruses... in the end we had to 6XHis tag the damn thing because it's tee-weeny tiny. Thought about an in-frame tag clone?

Yeah, I anticipate your first Ab was a Santa Cruz... Research Russian Roulette! My kingdom for a specific monoclonal!

By c0nc0rdance (not verified) on 11 Jul 2011 #permalink

I don't understand most of what you wrote there, but I still find it awesome. Kinda like when I read about researchers coming up with a method of stabilizing a table they were using for an experiment in the middle of the US. They were stabilizing it against the vibrations from the waves hitting the coasts! We humans are awesome!

By Robert Thille (not verified) on 26 Jul 2011 #permalink