WordPress.com



Genome Engineering in E. coli Using pSMV3J. Jeffrey Morris, modified 12/6/14The most recent version of this protocol can be found at the morrislab.protocols.Here’s the citation for pSMV3:Saltikov, C. W., and D. K. Newman. 2003. Genetic identification of a respiratory arsenate reductase. PNAS 100:10983-10988.Overview: I’ve successfully used this system to delete genes in E. coli MG1655 and REL606, Shewanella oneidensis MR1 and Synechococcus elongatus PCC7942. Briefly, the process for E. coli follows these steps:The desired mutant allele is cloned into pSMV3 using Gibson Assembly (GA).The plasmid is delivered to the strain you want to mutate using conjugation.Mutants that have lost the plasmid are selected using sucrose.STEP 1: Gibson Assembly (GA)In this step, you will generate the mutant allele using PCR; you will also PCR amplify the pSMV3 vector. GA will then be used to stick it all together into one mutagenesis vector.Design PrimersThere are three basic things you can do with this method (Fig. 1). First, you can replace a section of DNA with a mutant allele of your choosing. Second, you can delete an arbitrary amount of the chromosome. Third, you can insert completely novel DNA anywhere you like. Figure 1 gives a run-down for how you’ll design your primers based on which thing you’re trying to do.The easiest way to design primers for GA is to use the NEBuilder tool on NEB’s website () but you can use any tool you’re comfortable with. As Figure 1 shows, each primer has a region (arrows) that anneals to its template as well as about a 20bp “tail” that anneals to the next fragment in the final construct. One simple way to design GA primers is to just design normal primers for the regions of interest, and then stick the reverse complement of the opposite-direction primer from the next fragment in the construct on the 5’ end of each primer. However, NEBuilder is generally much more efficient than this in its primer design and will save you a few dollars by keeping your primers shorter.I always clone into an area between the BamHI and NotI sites of the pSMV3 multiple cloning site (Fig 1). Note that the pSMV3 primers I use don’t have overlaps – that means they can be re-used for any construct I want to put into pSMV3. The overlaps are added via the insert primer sets.Figure 1. Red and pink regions in all figures overlap the opposite sides of the BamHI/NotI site in pSMV3 and are the reverse complements of the primers used to amplify pSMV3. Primers and overlap tails in the other schematics are colored the same as the regions they overlap. Primers in the right-hand set of illustrations are for screening the post-GA transformants: they will only give bands if the GA was successful. Note that these primers can also be used to screen for the desired mutants after sucrose selection. In Method 1, any kan-sensitive cells that contain the barcode primer must be mutants. In Methods 2 and 3, mutants will give very different amplicon sizes than wild-type cells.Run PCRsNext, amplify the desired fragments from their templates by PCR. I use Phusion DNA polymerase, but any proofreading polymerase should be acceptable. I just use the cycling parameters recommended by the manufacturer’s online protocol for my Phusion reactions. Importantly, melting temperatures (Tm’s) for primers are dramatically higher when using Phusion, so make sure you account for this. If you use NEBuilder to design your primers, you can tell it which polymerase you’re using and it will calculate Tm’s for you. Otherwise, I suggest using NEB’s Tm calculator tool (). Also, extension times are much quicker with Phusion: 30 s per 1 kb amplicon size is sufficient. I have amplified 10kb products with Phusion without any tweaking of the protocol at all.It is important to start the GA reaction with a relatively large amount of DNA. For this reason, I like to increase my PCR yield with 2 simple steps. First, I use more template DNA than I usually would in a PCR. 3 L of a plasmid miniprep or 5 L of a typical small-scale genome prep (e.g. a Wizard prep) works well. Second, I will do 2 (or more) 50 L PCR’s, then pool them together in the next step. So, an example recipe would be as follows:5X Phusion HF Buffer20 L20 mM dNTPs1 L10 M fwd primer5 L10 M rev primer5 LTemplate DNA3 LPhusion polymerase1 LMilli-Q H2O65 LSplit into 2 50 L reactionsRun 2 L of each PCR reaction out on a gel to confirm it’s the right size and it’s clean (i.e. there are no undesired bands). If there are prominent undesired bands (don’t worry too much about faint ones) you have two choices: mess around with the PCR parameters (raise Tm or use more MgCl2) or gel-purify the desired band. I usually opt for the latter. This protocol doesn’t cover that method however.DpnI digest and purifyIn order to reduce the opportunities for weird and undesirable GA constructs, we will now destroy all DNA except what we’ve made using PCR. The restriction enzyme DpnI attacks methylated DNA and non-specifically chews it up. Only cell-made DNA is methylated; PCR products are safe. Therefore, we’re killing the templates and leaving our products unmarked.50 L PCR product5.67 L NEB CutSmart buffer1 L of NEB DpnI enzymeIncubate at 37? C for 3 h.If you did more than one PCR reaction per product, pool them here and scale the amount of buffer accordingly. However, 1 L of enzyme is more than sufficient to do the job in 3 h for any reasonable amount of template DNA.After the 3 h is up, run the whole thing through a single PCR cleanup spin-column. There are lots of varieties of these kits but all of them work more or less the same. Regardless of how much product you put through the column, elute in ~ 20 L final volume of nuclease-free water or TE buffer. This will concentrate the product and improve success in GA and transformation.Quantify DNA concentration in all fragments.Many of us get used to NanoDrop-style spectrophotometric DNA quantification. It’s great for telling whether or not you’ve got DNA, or even whether you have “a lot” or “a little” DNA. It stinks for telling exactly how much you’ve got. The success of ligation reactions often depends on getting molar ratios of fragments right. So we use a much more accurate fluorescent quantification protocol for DNA. My protocol for using PicoGreen to quantify DNA is available at morrislab.protocols. Use it to measure the precise concentration of each of your fragments.Gibson AssemblyNow it’s time to stick all the products together. First, we must calculate how much of each product to use. This is surprisingly complicated. What we got from Step 4 was the weight/volume concentration of DNA in the sample. However, the molar amount of fragment varies dramatically between a 500 bp fragment and a 10 kb vector. But the molar concentration is what determines the probability of the fragments encountering each other, and so we have to keep that in mind.Generally, if all the fragments I’m assembling are about the same size, I add them all at the same molar amount. However, if one fragment is obviously a vector, i.e., it’s a lot bigger than all the other fragments, I will add the other fragments at a 3-fold molar excess relative to the vector fragment.The tricky part is getting all of these moving parts to work together and to result in us having 10 L of DNA-containing liquid (the required amount for the GA reaction). I put together a formula that works pretty well. Actually, it’s more of an algorithm, but let’s call it a formula anyway. For each of N fragments, calculate the effective concentration Ci:Ci=DiLiMiwhere Di, Li, and Mi are, respectively, the weight/volume DNA concentration of the fragment (determined in step 4), the length of the fragment in base pairs, and the desired molar ratio of the fragment in the assembly (probably either 1 or 3). Now we can calculate the volume Vi of each fragment to include in the GA:Vi=CminCiVTi=1NCminCiwhere VT is the total volume desired (usually 10 L) and Cmin is the effective concentration of the most dilute fragment. In the end, Vi=VT should be true or else you didn’t do something right.So presuming I haven’t scared you off from GA with math, combine all your Vi’s together into a PCR tube, add 10 L of NEB Gibson Assembly Master Mix, and incubate at 50? C for 1 h. Transform immediately or freeze at -20? C.TransformationpSMV3 has the R6Kγ origin of replication, which means it can only stably replicate in a host with the pir gene. So we must transform our GA reactions into such a host. Invivogen sells chemically competent E. coli GT115 cells that are pir+. You can also use E. coli UQ950 (available from me as JJM157). With JJM157 I use the following heat-shock transformation protocol that works well most of the time:Grow an overnight culture of JJM157 in LB at 37? C.Transfer 100:1 into fresh LB the next morning and grow with 120 rpm shaking for 3 h.Centrifuge 5 mL of culture at 6500g for 5 min.Decant supernatant. Remove as much as possible using a pipettor if necessary.Resuspend pellet in 100 l ice-cold 0.1 M CaCl2. Transfer to a sterile 1.5 mL eppendorf tube.Add 5 L of GA reaction and incubate 30 min on ice.Heat-shock in a 42? C water bath for 1 min. Immediately return to ice.Add 400 L room-temperature SOC medium.Shake horizontally at 250 rpm for 1 h (this is the recovery step).Many other transformation protocols are equally acceptable. After the recovery incubation, plate cells on LB + 80 g/mL kanamycin agar plates that have been pre-spread with 7 L of 20% IPTG and 40 L of 20 mg/mL X-GAL. I will usually make 3 plates: one with 25 L of the recovery culture, one with 100 L, and one with the remainder of the culture. Incubate plates overnight at 37? C.ScreenHopefully you’ve got some colonies now. Weird things happen with molecular biology so we need to confirm that transformants are what we think they are. We will do this two ways.Sometimes despite all our efforts we will get false positives from cells transformed with re-circularized or not-quite-digested native pSMV3. Fortunately, pSMV3 has the fragment of lacZ that allows for blue/white screening. Colonies will be blue if they don’t incorporate anything in the multiple cloning site; otherwise they will be white. Thus, probably your desired colonies will be white and blue ones are right out.Grow a few (~4) white colonies in LB+kan overnight. The next day, run a PCR on the cultures using M13 primers (commonly available sequencing primers, found outside the MCS in pSMV3). These primers will give a product regardless, but if nothing was inserted into your vector, the product will only be ~ 200 bp long.If you’ve got positives, AWESOME! Freeze stocks in glycerol. Do a miniprep and send the thing off for sequencing – GeneWiz will sequence the whole plasmid for you if you want them to for $0.10 per bp. Once you’re confident your plasmid is put together properly, move forward to step 2, scientist!STEP 2: Bacterial RomanceIn this step you’ll first put the pSMV3 construct into a conjugation-proficient E. coli (WM3064, aka JJM167) that requires 2,6-diaminopimelic acid (DAP, SIGMA 33240) to survive, then use this strain to put the plasmid into your organism of choice. After the mating, you’ll take away the DAP to kill off the donor, then select for homologous recombinants of your recipient by plating on kanamycin. Since the plasmid requires the pir allele to replicate, the recipient can only grow if the plasmid has recombined into the chromosome.Preparation of WM3064 Donor Strain:Grow a 10mL overnight culture of E. coli WM3064 in LB + 100 L of 60 mM DAP.Use the heat-shock protocol described above to transform WM3064 with the desired pSMV3 construct plasmid. In step h (the recovery step), add 5 L of DAP along with the SOC medium.Plate entire mixture on LB+kan+IPTG+X-gal+100 L DAP.Incubate overnight at 37? C.Pick a white colony (shouldn’t be any blue ones) into LB+kan+100 L DAP.Freeze glycerol stock of WM3064+pSMV3.Mating:Make overnight cultures of the WM3064 donor and the recipient you want to mutagenize. Make sure to include kan and DAP in the donor culture!In a 1.5 mL eppendorf tube, add 750 L each of donor and recipient. Centrifuge at 6500 rpm for 2 min and decant supernatant.Resuspend pellet in 100 L sterile saline.Plate entire mixture on antibiotic-free LB supplemented with 100 L of 60 mM DAP.Incubate at 37? C for 3 h.Add 1.5 mL of saline to the plate and scrape up as much of the culture as you can using a cell spreader. Pipet this medium into an Eppendorf tube.Note: it is often okay to just streak directly from the mating plate onto a kan plate.Plate serial dilutions of this mess onto LB + kanamycin but WITHOUT DAP.Re-streak four or more colonies onto a fresh kan plate. This can be done easily with toothpicks.IMPORTANT: the colonies picked off of this kan plate are the most logical choices for biological replicates in any downstream fitness experiments you want to do with these mutants. If secondary mutations are generated in this protocol, they will be randomly assigned to these mutants and each will be different; therefore if they all end up with the same fitness, then we can disregard the effects of secondary mutations on our results.STEP 3. Sucrose selection.In this final step you’ll select for secondary recombination that removes the plasmid, leaving behind only one of the two alleles; either the wild-type or the deletion mutant you constructed. You’ll then screen by phenotype or PCR to find an appropriate clone.Streak a colony from each replicate in step 2 #14 onto LB with NO NaCl supplemented with 5% sucrose. Incubate overnight at 30? C.IF you have a phenotype that you can detect on a plate, go ahead and streak or patch colonies from these sucrose plates onto that screening medium. A couple of dozen should give you your desired mutant.Otherwise, restreak 8 colonies from each replicate onto fresh sucrose plates. Using toothpicks, you should be able to do 8 colonies on a single plate. Incubate overnight at 30? C.Repeat step #3 on fresh sucrose plates.Pick a colony from each re-streaking into both LB and LB+kan. This can be done conveniently in 96-well plates. Grow overnight at 37? C.The next day, do PCR on the cultures using an appropriate primer set (Fig. 1). If you get a band on the PCR AND the cultures are kan-sensitive, you’ve got a winner! Grow overnight in 10 mL LB and freeze a glycerol stock.Congratulations, you’ve got mutants! Remember, the appropriate level of replication in these experiments is at the initial antibiotic selection in your target strain. We are still working on figuring out how many discreet mutants you need to be able to disregard the influence of secondary mutations, but our preliminary data suggests that 4 should probably do it. Stay tuned for better estimates! ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download