Question and Answer Section

Would you recommend a control plasmid for the beginner to use when starting the technique?

There are two control constructs that I suggest using for learning the technique. The marker gene in both plasmids is a bright version of GFP, so the efficiency of transgenesis can be assayed by simply visualizing the embryos under fluorescence. For ubiquitous expression, I suggest using CSGFP3, which contains the CMV promoter driving GFP. This promoter is expressed in all cells starting at the mid-blastula transition (St. 8). For tissue specific expression, I suggest using pCarGFP2, which contains the cardiac actin promoter driving GFP. This promoter is expressed in all the muscles of the embryo. If you would like to use either of these plasmids, contact me and I will send it to you.

Is there a way to freeze sperm nuclei and still get some normal development?

In the early days I tried using frozen sperm nuclei, and the answer is that you can get normal embryos from frozen nuclei, but the efficiency was significantly lower, so we use fresh nuclei for our experiments now. The nuclei prep takes about 1.5 hours and the nuclei can be stored in the refrigerator for two to three days with no loss of transplantation efficiency or normal development.

Have you ever used other methods for preparing sperm, e.g. Percoll gradient centrifugation? Is it possible that the sperm is damaged during preparation or injection?

My guess is that sperm nuclei are more likely to be damaged after placing in extracts than before, therefore I doubt that they are being damaged during the sperm nuclei prep. I have not tried Percoll or sucrose gradients for making nuclei. A potential benefit of using these type of preparations is that they are less likely to be contaminated with blood nuclei, which when transplanted are unlikely to give you normal development (i.e. they are diploid and not prepared for the rapid divisions of the early embryo). To decrease contamination from erythrocyte nuclei, I try to remove as much of the clotted blood from the testis as possible before macerating. It is not worth removing the blood vessels from the testis, though. Instead I often poke holes in one or two of the larger vessels and then push the some of the blood out through the holes.

Could you please tell me the reasons for using DTT in the sperm preparation buffer? What is the role of spermine and spermidine?

The sperm nuclei prep that we use is pretty much the same one described by Andrew Murray in Methods in Cell Biology, (B. K. Kay, and H. B. Peng), ed., Vol. 36, pp. 581-605 (1991). It in turn is based on a very old method from John Gurdon. My understanding is that the DTT helps match the reducing conditions of the cell and the spermine and spermidine are added to stabilize the chromatin. I haven't tried to leave either out.

As you mentioned in your protocol, the needles pulled by the standard puller have a steeper slope near the tip and are more difficult to use. I have been using needles pulled directly by standard puller. Have you compared needles of different shapes and could this be the reason for my low success rate?

Lately we have been using a Sutter puller, which only has one pull. If the settings are right, you can get a pretty nice needle point. The thing to consider is that the steeper the slope, the harder it is to clip the point at the right diameter, and secondly the bigger the hole that you make in the egg, (because as the the needle goes in, the hole gets bigger). For these reasons, it is preferable to pull needles with shallow slopes at the tips.

What's "Versilube F50" / what is it good for?

Versilube F-50 is an oil with the special property that it is heavier than water but lighter than eggs. Therefore by layering the oil on the eggs and spinning them at low speed (i.e. 1000rpm), one can essentially remove all of the liquid between the eggs. The oil sweezes all of the water out, and you end up with a layer of water in the top of the tube, the Versilube oil next, and the packed eggs at the bottom. After removing the water and most of the Versilube, the eggs are crushed at 10K. Therefore the value in Versilube is that we can make cytoplasmic extracts with absolutely no dilution (i.e. the extracts are at the same concentration as the cytoplasm in the eggs).

The main problem is that Versilube can only be purchased in relatively large quantities (I think it is 1-2 liters), and it is quite expensive (several hundred dollars). So the question is whether Versilube is really essential? I don't know the answer, since I have always used Versilube. I know that many labs studying cell cycle and nuclear assemby in vitro routinely make extracts without Versilube. So most likely it is not necessary.

Do you have to exclude all dead/lysed eggs from the extract prep or can you tolerate a certain percentage of sick looking ones?

In general it is a good idea to try to minimize bad eggs from the extract prep as much as possible. The reason for this is that lysing eggs seem to compromise the quality of the remaining ones. It is the proverbial bad apple spoils the bunch.... In my experience it is not worth trying to remove the dead and lysing eggs because it seems that as the bad ones are removed, more spontaneously appear. So the best approach is to collect eggs from frogs that lay exclusively good eggs and throw all the rest away. This seems wasteful, but worth it. On the bright side, once you have a good prep frozen, you will probably not need to make another prep for several months.

You have used high speed extracts. Would for instance, low speed supernatants also work? (Because, low speed extracts also decondense sperm chromatin and assemble nuclei).

Yes, low speed extracts definately work. In fact I used low speed extracts in the early days of this technique. The main problem with low speed extracts (in my experience) is that they are full of actin. Most people add cytochalasin to depolymerize the actin, but if you do that to extracts used for nuclear transplantations, the resulting embryos have difficulty dividing. So I leave out the cytochalasin. Therefore the low speed extracts have polymerized actin and form a gell, like toothpaste. This makes it difficult to use because the sperm nuclei do not mix well in the gelled extracts. Also low speed extracts promote DNA replication and cell cycle progression. All of these things are not a problem with high speed extracts.... The actin is spun out, also ribosomes and membranes, therefore cell cycle does not progress (i.e. the nuclei are suspended in interphase with no DNA replication, etc.) and the nuclei are easier to mix.

Does it make any difference what restriction enzyme is used in the REMI reaction, and should the enzymes used for linearization of the plasmid and for the REMI reaction necessarily be the same? (There is some exonuclease activity in egg extract, so I would guess that the enzymes do not necessarily have to be the same.)

This is a question that I don't know the answer to. We know that adding restriction enzyme is not absolutely necessary for transgenesis, so at least some integration is certainly not true ligation of sticky ends but I don't know whether any ligation of sticky ends is not occuring. I suspect that perhaps the efficiency might decrease, but to what extent I don't know. Up to now I have always used the same enzyme to linearize and to use on the REMI reaction.

Update!! (April 14, 1997) I have now done the experiment (by mistake), and the answer is that the enzyme used for linearization does NOT need to be the same as the one used on the sperm nuclei. We added a Not I linearized plasmid to sperm nuclei that we added Sal I, and we got a high frequency of transgenic embryos.

Have you tried to use an oil-filled Drummond injection system?

Yes, you can use a Drummond. If you use a Drummond, you must still back-fill the needle as described in the procedure until the needle is full. Then insert it into the Drummond. Since the needle is full of liquid (the sperm suspension), you don't need to add oil. The Drummonds don't necessarily need oil. They just need to be fluid filled. After my limited experience using a Drummond for nuclear transplantations, I much prefer using an infusion pump instead.

I have been having difficulties getting embryos to neurula and beyond Recently, I cut back to a 1:20 dilution of restriction enzyme which has given me some success to produce more advanced stage embryos. However, I am seeing what you describe as haploid embryos (ie shortened, thick trunks,curled tails and pigeon chests ) that are regionally expressing a cardiac driven GFP in a non-mosaic manner. It can't be haploid and transgenic. Right? Any ideas on how this is happening? Should the enzyme be even more dilute?

The description of your embryos suggest that they are severely aneuploid. Therefore my guess is that your problem is definately due to trashed nuclei, but I am not sure that you are trashing them with too much enzyme. There are many ways of trashing nuclei (i.e needle tips that are too small, keeping decondensed nuclei on ice prior to transplantation, keeping nuclei too long before transplanting them, front-filling nuclei into the needle). The non-mosaic expression of the cardiac GFP construct is probably due to either non integrated DNA (depending on how much DNA you are adding to the REMI reaction), or integrated but trashed chromosomes which do not partition correctly at mitosis.

We prep our DNA with Qiagen. Would you still recommend gene-cleaning it?

I also do Qiagen preps. I then linearize the constructs, run them on a TAE gel, isolate the bands and gene clean them. I am not absolutely sure that all of this is necessary, but I have been somewhat influenced by the mouse transgenesis lore that says that DNA must be very clean for viability after injections.

We are currently trying to produce transgenic frogs according to your protocol. One point that concerns me is the sperm centriol which is required for production of the sperm aster and of the cytaster. Did you ever check whether the centriol adheres to the isolated sperm nuclei? Are there critical steps in the nuclear preparation that might remove the centrioles?

At the beginning I was very concerned about the centriole separating from the nucleus during the preparation. In fact at the time I was getting a lot of haploid syndrome, and I figured that one possible explanation was that the centrioles were falling off and I was essentially injecting isolated centrioles. At it turned out, the reasons for the haploid syndrome was mainly that I was using needles that were too small. It is rather amazing but swollen sperm nuclei can be as big as 60 microns in the long axis. So when I started using needles tips that where at least 60 microns inner diameter, then I no longer had problem with haploid syndrome. To answer your question more precisely, my understanding is that the centriole adheres relatively strongly to the sperm nucleus and it is not generally a concern when isolating sperm nuclei by the gentle approach that we use. I know that the success or failure of the transgenic technique relies mainly on how intact the nuclei are when they are transplanted into the eggs. If the nuclei are not intact then two things happen: 1. If the nuclei are totally trashed then probably the centrioles are stripped and many of the resulting embryos show haploid syndrome. 2. If the nuclei are somewhat trashed, then the result is suprisingly worst!! What I believe happens is that the resulting embryos are severely aneuploid, which is worst that being haploid. These embryos tend to arrest at gastrulation. From the comments that I have gotten over the months of people trying the technique, this is the most common outcome of their attempts (i.e embryos that arrest at gastrulation).

Could superficial cleavage result from defective centrioles rather than being a consequence of the haploid syndrome?

If by superficial cleavage you are talking about what I call "pseudocleavage", then no, I don't thing that they result from defective centrioles. Embryos with "speudocleavage" have cleavage furrows that start to form but give up before completing. The embryos end up with dark lines but no cleavage. This is what you get if you just prick activate an egg, so it has nothing to do with damaged centrioles. The other time I have seen superficial cleavage furrows was in the early days when I added cytocholasin to the egg extract prep. In this case I think it was due to affects on actin. Until St. 28 or so there is no easy way of distinguishing embryos with haploid syndrome from normal embryos.

How long can you use the sperm/DNA/egg extract-mixture after the 10 min incubation? Should one make a fresh one for each filling of the needle or store it undiluted in ice?

The longer the nuclei are kept after decondensing before transplanting the worse they get, so I try to transplant within 30 minutes of decondensation. I usually load one needle per reaction. It usually lasts the 30 minutes or so that I transplant, so I rarely need to fill another needle. Decondensed nuclei should NEVER be placed on ice.

From my experience with in vitro nuclear transport assays I would expect that the sperm nuclei cannot really round up and assemble a membrane in the extract described. What is a criterium for the quality of the extracts? Have you ever measured or documented otherwise how much the chromatin should have decondensed? We saw lengthening and broadening in initial experiments.

Lengthening and broadening is all that you should see in high speed extracts. As you mentioned high speed extracts cannot assemble a rounded up nucleus since the membrane fraction has been removed. In fact, any rounded up nuclei that you might see are likely to be contaminating erythrocyte nuclei. Nevertheless, high speed extracts should be able to decondense the chromosomes. After 10 minutes, one should be able to see a lengthening and broadening of the nuclei and if digested with micrococcal nuclease, one should see a nice nucleosomal ladder of the chromatin. For the purpose of transgenics, this is sufficient. I presume that soon after transplantion, the nuclei continue to assemble a proper nucleus, fuse with the maternal nucleus, etc.