Let's talk about housekeeping genes - Specifically, in Bacteria (Aug/14/2007 )

Hello.

I am comparing growth of bacteria under different conditions and the relative expression of a gene in each condition. In other words, I am using RTPCR to ask how expression levels of this gene change with conditions.

I have developed primers for two different housekeeping genes for this project. One of these is the 16S rRNA gene. The other is the rpoD (sigma factor) gene.

• Q1: Are there better choices for bacterial housekeeping genes, and are there any drawbacks to the two I have chosen?

I have found in my preliminary experiments that the 16S transcripts (housekeeping gene #1) are very, very dominant in the cell. In other words, I get a signal for this gene in QPCR at about 8 cycles. My experimental gene requires about 20 cycles to detect.

• Q2: My instinct is that such high expression levels makes this gene less appealing as a housekeeper, because that (high) level of template in QPCR may introduce some problems. Are my instincts right? Can I get around this by diluting the reverse transcription reaction before the QPCR, or is the high level of the 16S transcript cauising a similar problem in the reverse transcription?

The second housekeeping gene, rpoD, gives a decent signal in QPCR around the same time as my experimental gene. this seems good to me....

• Q3: Should I toss the 16S gene and just use the rpoD gene? Or is it worth it to fiddle with the 16S gene so that I can calculate the expression of my experimental gene against two independent housekeepers?

thank you in advance -

Patty

---

I'm not so familiar with the selection of housekeeping genes in bacteria but i know from FISH that the ribosomal content (and therefore 16SrRNA) in bacterial cells can vary considerably dependent on the growth phase.

but if you want to study mRNA i think it is better to normalize to the mRNA of a housekeeping gene and not rRNA. these are different RNA fractions and you don't know if they behave the same during extraction, reverse transcription, etc... and again, the rRNA content does not necessarily reflect the amount of mRNA in the cell.

also the expression levels of target and HKG should be comparable. i would say, the difference between HKG and target should not exceed 10 Cycles. HKG should always be expressed higher.

finding and validating of an appropriate HKG can be more elaborative than the actual experiment.

i hope this was helpful.

---

Very helpful, yes.

Why should the housekeeper be higher than the target?

---

sorry, this was just out of habit because i'm doing my HKG and target pcr in the same tube and then you have to limit the primer of the more abundant species.

---

thanks again -

I now have data in hand and I'd like to make sense of it.

We used four growth conditions. The cells grew well in two conditions and poorly in two conditions. I got good RNA yields from the good growth conditions, and poor yields from the poor growth conditions. I then normalized all the tubes to the same concentration of RNA.

The 16S gene appeasr to be relatively the same level in all four tubes. In other words, it looks like the 16S rRNA is there, in decent amount, whether the cells are growing well or not. On the other hand, the rpoD gene (mRNA) falls off drastically (falls 20 fold) under poor growth conditions.

This sort of makes sense to me - and seems to argue that a mRNA housekeeper is better to use than a rRNA housekeeper. If mRNA levels drop due to growth conditions, you need to normalise to the drop. Yet in another thread someone recommended using rRNA "because it is constant."


Any more thoughts to add?

I have some more thoughts that I want to ask about, but I need to mull over the data a little bit longer.

---

maybe you could need this reference, i haven't looked into detail but it seems to handle your problem.

http://www.ncbi.nlm.nih.gov/sites/entrez?D...Pubmed_RVDocSum

---

Hello,

I'm starting to do experiments with bacteria and gene expression quantification too. I have reviewed the available literature and a few people use rrna gene whereas other use mRNA genes for normalisation. Furthermore I found that many other "putative HKG" have been used for normalisation: rpoS, rpoD, proC, relA etc. I believe that you may have to test different ones to find out which one is suitable for you assay setup. Have you read that one:

Savli H, Karadenizli A, Kolayli F, Gundes S, Ozbek U, Vahaboglu H.
Expression stability of six housekeeping genes: A proposal for resistance gene
quantification studies of Pseudomonas aeruginosa by real-time quantitative
RT-PCR. J Med Microbiol. 2003 May;52(Pt 5):403-8. PMID: 12721316 [PubMed - indexed for MEDLINE]

I would appreciate to share the findings about this issue.

regards,
See9

---

i think a good approach would be to take a couple of HKG (about 6-12), preferably acting on different cellular functions and analyze it with the geNorm applet, for instance, which can be downloaded for free.

patty, you wrote you are normalizing to RNA concentration if i have understood you correct? how did you determine the RNA concentration? spectrophotometrically? because i think this method only gives reliable results if very pure RNA is available. i think a good improvement would be to use an agilent bioanalyzer or bio-rad's experion system. if you have already applied such instruments just skip the last sentences

another thought which came up just in the moment: as about 95% of total RNA is rRNA aren't you in fact normalizing to rRNA? therefore it would be not surprising if the 16SrRNA content seems to be quite constant? or is there some error of reasoning in my considerations? I think, in the paper i have cited, they did it the same way. and last but not least: i think by normalizing to RNA concentration you can get only approximate results because you are not considering different efficiencies in reverse transcription. but ok. it should be adequate to exclude a HKG which expression is changed by 50fold in two different growth conditions.

---