Dear Friends of Chlamydomonas:
I have gotten a number of interesting responses to my query about experiences using regulatable promoters and antisense for manipulating gene expression in Chlamydomonas.
There have been a number of attempts at antisense by various people, some attempts having been more extensive than others, and I think there is reason for some optimism. I suspect that the success of this approach may depend strongly on the particular gene whose expression is being repressed.
On the potentially negative side, Steve Gantt writes:
“I tried to use antisense RNA to block plastid chaperonin (cpn60- alpha) gene expression. We created several antisense constructs consisting the nitrate reductase promoter fused to different regions of the chlamydomonas cpn60-alpha cDNA sequence. These constructs were co-transformed along with the arg-7 gene into arginine auxotrophs. We selected for arginine prototropy on arg minus, ammonium-containing plates and isolated several hundred colonies. These colonies were then replica plated onto arg minus plates that either contained ammonium or nitrate as a nitrogen source. I expected that cpn60 antisense co-transformants would display altered growth characteristics (slow or no growth) on the nitrate plates. Unfortunately, I could not convince myself that any of the putative transformants grew slower on the nitrate plates.
This was a one-shot experiment and shouldn’t be taken to mean that it can’t work. We never even examined the transformants for the presence of the cpn60-alpha construct or antisense transcript. We also don’t know what the phenotype of a cpn60- alpha mutant would look like. I have assumed that a cpn60-alpha null mutation would be lethal and cells containing low amount of the protein would grow slowly. Similar work has been done in yeast with the mitochondrial chaperonin.”
Stefan Fabry has made rather extensive attempts to manipulate ypt gene expression in Chlamydomonas and Volvox. His experiences are somewhat encouraging for antisense expression, I think, and his lack of success with expressing epitope-tagged ypt proteins may mean that such expression would be deleterious. His description follows:
“For some time we have been trying to express antisense as well as HA-epitope-tagged Ypt genes (encoding small G proteins) in Chlamy (and Volvox).
(a) HA-epitope-tagged proteins: In the beginning, we used ypt cDNA for HA-fusion, and the ars promotor from John Davis (as you intend to do). We introduced the construct by cotransformation together with pArg7.8 into an Arg- strain. We verified successful cotransformation by PCR, induced the promoter by sulfur deprivation, and looked on Western blots for protein synthesis with the anti HA antibody in protein crude extracts from arg+ cotransformants, but could not detect any signal. We then continued by adding an intron to the 3′ region of the gene construct; however, it also failed. We then assumed that intron-free coding regions may in general not be expressed in Chlamy (is that true? are there any other experiences?), and reconstructed the whole system with a genomic DNA. This was also not successful. Then we speculated that the ars promoter was not reliable enough for our system, and used the rbc promotor (as described in Kozminski, Diener, and Rosenbaum, 1993, Cell Motil. and the Cytoskeleton, 25, where they expressed a HA-tagged alpha tubuline; actually we got exactely their plasmid for our purposes). Again, however, we did not receive any signal up to now, even after checking now nearly 100 co-transformants.
(b) antisense: In principle, similar experiences were made with antisense ypt experiments. We only used the constitutive rbc promoter (even risking that this will result in lethality of positive cotransformants). However, we observed one time (and only with very fresh transformants!) that a respective Ypt protein was in fact totally absent from the protein extract of two transformants (on the same blot, lots of positive controls were included. However, after repeating the experiment (after growing the strains for some further time), the protein reappears. At moment, we speculate that we had observed a transient antisense expression, even though the gene construct seems to be stably integrated within the genome.
Finally: Currently, we are having very similar negative experiences when trying to express (tagged and non-tagged) histone genes in Volvox.
So to summarize, there appear some general facts, as far as we see: While in both algae, the cotransformation rate is quite good, the nonselectable marker of the cotransformation experiment, though inte- grated in the genome, is not expressed! The only exception I know up to now is the successful HA-tagged-alpha-tubuline experiment published by Rosenbaum and coworkers. My question therefore is: does anybody have similar or other experiences? If my speculation is true, namely that if you do not s e l e c t for the cotransformed gene, it will “disappear” in a silent region of the genome with high probability, then antisense expression of genes (which will also be nonselectable in most cases) may cause considerable trouble!”
The most encouraging news comes from Bernhard Scheidlmeier in repressing arylsulfatase gene expression in Volvox. His description follows:
“To evaluate if antisense technologies are working in Volvox, I decided to `knock out’ the expression of the Volovox arylsulfatase gene. The antisense vector consists of the Volvox ars promotor and a 2.3 kb genomic ars-gene fragment (n-terminal region) cloned behind in inverse orientation. So far, I have analyzed the ars enzyme activity of 7 transformants compared to that of the non-transformed recipient strain. The results look promissing: One transformant showed no detectable ars enzyme activity at all, the other ones exhibit inhibtion of ars from 60 to 95 %. I have to repeat these mearurements some more times. So please, take these data only as preliminary results.”
My own experience with a variety of plastocyanin mutants in Chlamydomonas is that transformants carrying mutations that do not eliminate gene/protein function are easy to recover. However, using genes with more extensive deletion mutations, we have been unable to recover transformants that accumulate RNA (or protein). This suggests a number of possibilities. One possibility is that expression of the mutant gene product is deleterious to the cells and that 1) transformants carrying the DNA are not recovered at all or 2) there is strong selection for transformants that do not express or have silenced the DNA. Further analysis should distinguish these possibilities.
I hope that we can continue this dialog and get around these troubling technical difficulties. At the very least, a reliable, regulatable promoter would be useful.