In elementary school I learned “Never Eat Soggy Waffles” to remember the cardinal directions – North, East, South, West – I still think it in my head when I’m trying to orient myself. But no one gave me a guide for blotting directions – you might have heard of the western blot (I covered it last week). It’s a technique that’s used a lot to detect specific proteins. But did you know there’s also a Southern blot (which came first), a northern blot, and eastern blot, and some more copy-cats?! So today I present to you the bumbling biochemist’s “compass” for biochemical blots!
No matter what “direction” is in the name – the basic premise is the same – take a mix of biomolecules and separate (typically by size) then transfer to a membrane and probe to analyze. If the probe finds a match, a band you will see, but what kind of molecule will it be? For western blots, you’re looking for specific proteins, Southern blots look for specific DNA, and northern blots look for specific RNA. And they utilize different types of “probes” to do so – antibodies for western blots, and little pieces of sequence-complementing DNA for the others.
In last week’s Bri*fing, we looked at the western blot. The western blot is a way to test for the presence of specific proteins and it’s named after the Southern blot which tests for specific pieces of DNA – and the “Southern” name comes from the name of the scientist who invented the technique, Edwin Southern. This is why you capitalize the S but not the w.
The western blot isn’t the only technique to hop onto the direction name train – there’s a 3rd main type of blot called the northern blot and, as someone who studies small RNAs, I come across it in papers a lot. I haven’t actually done it personally because I do in vitro (in a test tube) work where I know exactly what RNAs I put into my reactions, so I can label all the RNA and know all the signal I see is from the one I’m interested in. But a lot of times scientists don’t know what and/or how much of a specific RNA is in some mixture – like the insides of a cell at different times. They could break the cell open (lyse it) and isolate all the RNA, but if they were to label all the RNA, they’d be labeling tons and tons of different RNAs, not just the one they want. Enter the northern blot.
The northern blot is a technique used to test for pieces of RNA that contain a specific sequence, utilizing the 1:1 base-pairing specificity of nucleotides (DNA & RNA letters). Since A binds U (T in DNA) and C binds G, you can design probes that bind the thing you’re looking for with specificity! (e.g. if you’re looking for the sequence 5’-GAUUACA-3’ you can design a probe that’s 3’-CTAATGT-5’ (if you use a DNA probe) or 3’-CUAAUGU-5’ (if you use an RNA probe). Thanks to the “RNA can pair with RNA or with DNA” thing, you can use either type of “oligonucleotide” probe, but DNA is more stable. Regardless of letter type, you usually you use longer sequences that are less likely to occur by chance (like the reason you should use longer passwords. The “oligo” in oligonucleotides refers to “few/several” but you don’t want too few!
The base-pair complementarity is super nice because it makes it WAY easier (and cheaper) to design probes for different sequences than it does to make probes for different proteins like those you use in western Blots – for proteins you usually use antibodies which are little proteins that recognize specific parts of other things – they can take a long time to produce and production usually involves injecting animals with the thing you want antibodies made to recognize, then letting the animal’s natural immune system select and amplify antibodies that recognize it as foreign. And then you have to purify those antibodies and make sure they’re specific for your protein (don’t cross-react with other proteins and give you false-positives), etc.
For antibodies, you often have to use a labeled secondary antibody to bind to the primary antibody (the one specific to your protein) in order to visualize it (the secondary antibodies are designed to recognize “generic” parts of the primary antibodies so you can use the same secondary antibody for multiple primary antibodies). But for RNA or DNA, you can just directly label the RNA/DNA – often through radiolabeling of the 5’ end (adding a radioactive phosphate group to it)
The basic premise is:
- RNA ISOLATION – isolate RNA (often from cells under different conditions you want to compare
- GEL ELECTROPHORESIS – run it through a gel (agarose works well for big pieces, polyacrylamide (PAGE) works well for smaller ones), using the RNA’s natural negative charge to motivate it through the gel’s mesh towards a positive charge at the bottom -> this gel-swimming’s harder for longer RNAs (they get tangled up in the mesh more) so they travel slower so the gel acts as a “molecular sieve” to separate the RNAs by their length
- BLOT – transfer the RNA out of the gel & onto a membrane
- the gel was great for the stage at which the goal is to separate – but if you leave it in there it can start to diffuse – even without the positive charge to motivate it downwards, it can still wander around randomly a little – so you want to get it stuck on something it can’t wander through
- BLOCK – you only want specific binding so you pre-hybridize the membrane with “generic” DNA like salmon sperm DNA – no joke!
- PROBE – hybridization – add labeled oligonucleotide probe that has a sequence that complements the sequence you’re looking for
- VISUALIZE –
- if you’ve used a radiolabeled probe you expose the membrane to a screen – as the probe lets off radiation, it’s captured on the screen above it and when you scan the screen you can see where the radiolabeled thing was – more here
- if you’ve used a fluorescent probe, you shine the right wavelength of light on it to excite the fluorophores and get them to give off light you can detect
The basic premise/workflow looks a lot like a western, right? Here are some key similarities differences
- SAMPLE ISOLATION – obviously if you’re doing a western, you don’t need to isolate RNA – you may need to “free” protein depending on where your sample’s coming from – if its from whole cells/tissues you’ll have to break those open (lysis) using detergents, mechanical disruption, etc.
- GEL ELECTROPHORESIS – instead of an agarose or urea PAGE, you typically run an SDS-PAGE. SDS stands for Sodium Dodecyl Sulfate and it’s a negatively-charged (anionic) detergent that denatures (unfolds) proteins, coats them to keep them “slippery” so they don’t unfold and negative so they travel towards the positive bottom of the gel.
- BLOT – be it protein or RNA, you still have the potential problem of things diffusing away – so you still need to transfer what you’ve separated to a membrane and get it to stay
- there are different ways to do this that typically involve using another round of electric motivating, here called “electroblotting” – this time in the “horizontal” direction (it’s harder to get those big, unfolded and trapped proteins out of the gel, so you need a little more electric oomph than you did for the RNA where, as we’ll see, you can sometimes just stick something heavy on it
- BLOCK – no sperm with the western, but you sometimes use something else you might not expect to find in a lab – milk! nonfat milk (typically from powdered milk) can be used as a blocker because it has a lot of the protein casein. Another common blocker used is the slightly more-expensive but more predictable bovine serum albumin (BSA).
- PROBE – for a western, you typically use 2 “probes”, both of which are antibodies – little proteins that have generic adapters and unique “ends” that recognize different things. You first probe with a primary antibody which has a unique part that recognizes the protein you’re looking for. And then you wash the unbound probe off and add a second antibody – a secondary antibody that recognizes and binds the generic part of the primary antibody. This is possible because the generic adapter parts are only generic for the species it came from so you can use things like “goat-anti-rat” antibodies that come from goats and recognize all rat antibodies. That’s what makes it possible – but what makes it desirable is that the secondary antibody can be conjugated (connected to) something detectable like a fluorophore or a “chromogenic” group that will give a colored product if you add some sort of “developer”
- VISUALIZE –
- if you’ve used a fluorescently labeled secondary antibody, it’s just like if you used a fluorescently-labeled DNA probe – shine light it likes at it and see where it shines back
- if you’ve used a chromogenic probe, add whatever developer you need
A couple of details for those who are interested and then I’ll get to some naming fun
As I mentioned briefly above, it can be a lot “easier” from the molecules’ point of view for little RNA to leave a gel for your membrane than it is for proteins. So, unlike with the western blot, where you use electricity (with the field now applied perpendicular to the original travel direction) to transfer the molecules to their new home, with a northern blot, if you’ve used an agarose gel, you can just stick something heavy on it – glass plates are often used as weight – if you put a stack of dry paper towels under the weight, capillary action will suck the liquid through the gel to get to those paper towels – and the RNA will get stuck to the membrane along the way. Or you can use a vacuum to help. Or, for polyacrylamide gels, which have tighter mesh that’s harder to escape from, electricity *is* used again (“electroblotting”)
But, ease of escape comes at a price – it’s also easier for the RNA to escape the membrane than it was for the protein to do do so. So, once it’s on the membrane, you want to make sure it stays there, so you can UV crosslink it (the UV light gives the RNA bases (especially the uracils) energy that excites them & gets them to react with amine (nitrogen/hydrogen) groups on the surface of the membrane – they form strong covalent linkages that keep them stuck there. Alternatively, you can heat the membrane – this evaporates the water that’s been hiding the hydrophobic (water-avoiding) parts of the RNA, so these parts can now interact with hydrophobic parts of the membrane.
The northern Blot used to be used a lot more, before the days of RT-qPCR, that technique we looked at yesterday which indirectly measure how many copies of a specific RNA is initially present by copying it into DNA form and making copies of those copies using POLYMERASE CHAIN REACTION (PCR) – and seeing how many copying cycles it takes to cross a threshold copy number (the more you start with the fewer copying cycles you’ll need). http://bit.ly/384xCo9
PCR works by using an enzyme (reaction speed-upper/mediator) called DNA polymerase (DNA Pol) to make lots of copies of pieces of double-stranded DNA. If you want to use it for RNA, you first have to Reverse Transcribe it – make a DNA version of it using a reverse transcriptase – and then you can make a complementary strand of that to get double-stranded DNA. Each round of PCR, you have a melt step where you raise the temp to “unzip” these strands – so each can be used as a template for making the other, complementary strand
You specify where you want to copy by providing short pieces of DNA called primers – 1 per strand – that serve as “Start points” for DNA Pol to travel along the strands, adding complementary deoxynucleotides (DNA letters) as it goes – and it goes until it falls off of the template it’s using which (after the first cycle) is where the other primer bound
With qPCR you usually use some sort of fluorescent dye or probe that gives a readout of how much DNA is present at the end of each cycle – if all goes according to plan, the amount of DNA doubles each cycle (exponential growth) until you run out of resources – so the more copies you start with, the more copies you’ll have at the end of each cycle. You can define a “noise threshold” and then see how many cycles it takes to get above it.
Normally the primers we use are ~20nt. And DNA Pol builds onto that. So it’s good for copying bigger pieces. But I’m interested in studying really small pieces of RNA, like microRNA (miRNA) which are short (~20nt long) pieces of RNA that complement sequences in mRNA “targets.” miRNA bind to a protein called Argonaute (Ago) & guide Ago to the target to prevent that mRNA from being used to make protein. Much more here: http://bit.ly/2R5soTe
These small RNAs are basically the size of a primer – so how can you make copies of it? There are special adapter tricks you can use, but, since the northern Blot doesn’t require copy-making, it’s an attractive alternative. Another advantage of the northern Blot over RT-qPCR is that you can detect differently-processed versions of RNA – “isomers” – so you can tell if RNAs have been modified by things like trimming or tailing.
There are benefits and drawbacks of the 2 techniques to consider:
RT-qPCR is more sensitive because you amplify the original sample – each cycle you double what was present in the previous cycle (since the newly made DNA pieces can act as templates). This amplification can be *good* because you can detect RNA that’s only present at really low copies. BUT this amplification can also be *bad” because when you’re amplifying the signal “artificially” you can get artifacts! When you do RT-PCR you usually assume that all RNAs get reverse-transcribed equally well and that all of the cDNAs made get amplified equally well and consistently – a doubling per cycle. but if this isn’t the case, it can seem like you started with more or fewer copies than you really did.
You don’t have that problem in northern Blotting, because you measure the original directly. But you’re far from out of the woods! With Northern Blots, you don’t have to worry about amplification problems, but you do have to worry more about the “opposite” – degradation!
RNA is really unstable – especially compared to DNA – in part because of its chemical makeup – RNA’s extra O makes it more prone to self-destruction – and in part because of enzymes called RNases (RNA chewers) to chew up RNAs that are foreign or no longer wanted. more here: http://bit.ly/2XHJKWa
So one of the benefits of RT-PCR is that you stabilize the RNA by turning it into DNA. But with northern Blotting you’re dealing with the actual RNA itself, so you have to make sure you don’t lose it!
Another “direction” you can take with blotting is the Southern Blot – the OG of the blotting world. It’s a lot like the northern blot (or I guess it’s more correct to say the northern blot is a lot like it!), except that you’re probing for DNA not RNA. It was used a lot when scientists were trying to locate genes, identify genetic rearrangements, etc.
Edwin Southern invented the technique in the 1970s because he wanted to be able to isolate the genes responsible for making 5S ribosomal RNA (rRNA) – most enzymes are proteins, but the ribosome (protein-making machinery) is a mix of proteins & RNAs, with the RNAs doing the brunt of the work.And Southern wanted to find the genes for these RNAs to give them their due credit! (and study how they work…)
Since DNA is usually present as big ole chromosomes you start by chopping it into smaller pieces using restriction endonuclease (DNA scissors that recognize & cut at specific sequences), then run it through a gel. At the time, the traditional way to go about it would be to extract all those cut up DNA pieces out of the gel (basically cut out the bands and melt away the gel around them). And then bind each of those purified bands to a filter and probe each of the filters with complementary DNA probes.
Southern thought – the probing part’s good, but why not just transfer all the bands to a filter at once? As long as you preserved their relative locations you wouldn’t lose any information – you’d lose less DNA since you didn’t have to go about purifying it first (where you’d inevitable lose some yield at each step) – and you’d save a lot of time.
Southern gives credit for the “blotting” part of the name to Frederick Sanger – the name might sound familiar because we looked at him a couple times – in addition to sequencing the first protein (insulin in 1955) he developed the “Sanger Sequencing” techniques for sequencing DNA using chain-terminating nucleotides – these dideoxynucleotides can get added but not added onto because they lack a 3’ OH and by adding some terminators of just a single letter you can puzzle out the sequence. http://bit.ly/2koF5el Sanger used electrophoresis to separate RNA fragments on cellulose acetate strips, then transferred the fragments out of the strips and onto DEAE paper by “blotting through”
It’s actually kinda funny because a bunch of different scientists wanted to join the directional naming fun – but there are only 4 cardinal directions – so you ended up with a ton of methods being called “eastern blotting” which turns out to not be so funny because it can instead be confusing… As well as as some other “hybrid” directions. Here’s a sampling…
Most of the techniques the eastern blot name refer to are variants of the western blot that look for post-translational modifications – “bells and whistles” or “cherries on top” added to the proteins after they’re made (translated) – things like phosphorylation (addition of negatively-charged phosphate groups), glycosylation (addition of sugar chains), or lipidation (addition of lipids (fatty things))
far-eastern blot: lipid analysis – this one’s a bit more different – instead of gel electrophoresis it uses TLC (thin layer chromatography) which basically uses capillary action to “suck” liquid through a special sheet that different components in the liquid interact with differently so they travel at different rates and thus separate as they go – kinda like when a piece of paper with a marker mark gets wet and the colors in the marker dye start to separate. After the TLC, the lipids are blotted onto a PDVF membrane
far-western blot: search for protein:protein interactions using a protein blot (the western part) and (non-antibody) protein probe (the far part), so basically do a western, but probe with a specific protein (which you can then detect with an antibody or via specific label it might have)
middle eastern blot: mRNA analysis – it’s basically a northern blot, except that the membrane is coated with “poly-U” so it specifically binds to polyadenylated (polyA-tailed) messenger RNA (mRNA) (this is the type of RNA that serves as a go-between between the DNA instructions for a protein and the actual protein). This is the same rationale behind using “oligodT” primers for the reverse transcription step of RT-qPCR, except in this blot, it’s used just to get the mRNA to selectively “stick”
reverse northern blot: use RNA probes to bind to DNA – this is the basic idea behind DNA microarrays, which have wells containing cDNAs and then you test to see if the complementary mRNA is present in a sample to see if a gene’s getting expressed – each tiny well on a microarray plate is like a mini reverse northern blot
northwestern blot: detect RNA-protein interactions by using RNA probes to (potentially) bind to proteins on a membrane (so basically start like a western, then probe like a northern).
southwestern blot: search for DNA-binding proteins using a protein blot (the western part) and DNA probes (the southern part), so basically do a northwestern, but probe with DNA not RNA
In looking into the history of the technique, I found that, in an early example of a “preprint” Edwin Southern gave a sketch of his Southern blot to Michael Matthews who brought the technique to the US – to my workplace, CSHL. http://bit.ly/343sKOe
there are other ways to measure various things, but these are some of the main ones.
This post is part of my weekly “broadcasts from the bench” for The International Union of Biochemistry and Molecular Biology. Be sure to follow the IUBMB if you’re interested in biochemistry! They’re a really great international organization for biochemistry.