Hi everyone, not sure if this is the right sub-forum to post this question but does anyone have any experience in extracting DNA from mice fed high fat diets? I have been using a chloroform/phenol/alcohol extraction protocol to get some DNA but my yields are always unreasonably low leading to poor sequencing data. I suspect the high lipid content of the stool some how interferes with DNA extraction. I’ve seen the Powersoil kit mentioned quite a few times in the literature, does anyone have any experience using this and what kind of yields were you getting? Are there other good protocols I could use to circumvent this problem?
Hi @vpandyar ,
I’m not sure the high fat content of the stool is causing the low yields (though maybe?), but more likely has to do with your extraction protocol itself. How are your yields with regular diet fed stool samples?
In my previous lab I had seen a variety of high fat diets fed to mice and we never had a problem getting good DNA yields from their stool. Back then we were using either Qiagen mini stool, then Mobio Power soil (which is now owned by Qiagen again under a different name) kits. I’ve also seen other kits used with high fat diets as well, so there are certainly other options. Column-based extraction methods tend to be a lot more efficient than just chloroform/phenol extractions, at least in my experience. But if purchasing those kits is not an option, you can always try a few things like 1) increase your physical disruption step (increase bead-beating time) 2) add a detergent during cell wall disruption, 3) add/increase a heating step (make sure you don’t boil though).
The advantage of the kits are that they are standardized, proven to work, and you can always work with the company to get the protocol to work, usually with free additional kits if they fail.
Thanks for your response. The normal diet animals have excellent yields using the same protocol (which includes a column purification step). I’ve tried different cohorts of animals fed high fat diets and all pretty much come back with the same result which makes me think there is something in the high fat stool which prevents my particular protocol from working. I may just have to move on to a kit because as you say they are standardized. Of note I am trying to get fungal data which is the data set that always comes back poor.
Hi @vpandyar ,
How are you evaluating your fungal DNA yield and at what step? Post sequencing? Pre PCR, post PCR? Total DNA? amplified ITS?
Fungal DNA is a tiny fraction of overall stool DNA so no surprise that it would have very low yield but there should still be enough to amplify and sequence.
I use a primer called fungiquant typically after purification of total DNA and prior to sequencing. I usually get Cq values in the range of 25-35 which seems to indicate acceptable amounts of fungal DNA is present but I never seem to get a ton of diversity after processing the data. For example some samples look like only 1 or 2 species are present when I know this cannot truly be the case based on published literature.
I’m not familiar with that primer but I’d look into its limits of detection and specificity to your target community.
Also, are you testing these against a relevant positive control? Say a real mouse gut sample that you know has reasonable amount of fungal DNA from other tested methods? Additional DNA purification (following initial extraction) is not uncommon for these “difficult extraction” samples as well, in fact most kits have additional instructions for difficult extractions.
On the analysis side of things, how are you processing your sequences and how are you determining species per samples? Though I’m guessing your issue is at the wet lab side of things, as you know, there are a lot of potential steps during the bioinformatics steps to also accidentally discard real biological reads. Though looking a closer look at that would be better for a new topic on the fourm.
Thank you again for your response. The Fungiquant primer I use is a “universal” fungal primer that can be used to quantitate fungi. See FungiQuant: A broad-coverage fungal quantitative real-time PCR assay by Liu et al. Admittedly I have not tried this against a positive control but using this PCR primer I am getting Cq counts in the 20-30ish range which should be more than enough DNA for detection by NGS. I currently use a DNA purification kit after my phenol/chloroform extraction to really clean up my DNA. My main problem stems from the fact that after DNA extraction and submission to my sequencing core I get low read counts. I’m not sure what is typical but I’m getting in the range of 200 - 50000. After processing this of course lowers my usable reads significantly. I am beginning to suspect that a PCR inhibitor may be coming through in my extraction process which is affecting the quality of sequencing runs although why it doesn’t affect my bench top PCRs I’m not quite sure. Any thoughts?
Hi @vpandyar ,
Alas, these are very common issues of low biomass samples and there are lots of literature on the topic. I also recommend speaking with both the company you purchase your kits from as well as your sequencing center. They should have much more specific ideas to your issues.
In the meantime here are some thoughts from the top of my head:
- If you have more than 1,000 sequences per sample in mouse gut samples for fungi, I would say you are fine. There is very low diversity of fungi in their gut and you should be able to capture the overall diversity with that.
- What are your amplicon concentrations prior to sending them off to the sequencing center? If at that point you have lots of pure target DNA then the sequencing shouldn’t be an issue. You may have an issue if there are other non-target DNA in there though, or some other contaminants.
- How do your libraries look on something like a Bioanalyzer? Are you getting bands in the ranges you expect?
- Testing against a sample that you are sure that has lots of fungal DNA would certainly be a good idea.
- Is it possible that your Fungiquant primer is picking up fungi that are not ultimately amplified by your target gene target (ex ITS2?)