Hi, you could also use sonification to homogenize the aggregates. I had the same problem and applied homogenization of aggregates in a water bath. That technique also helped me to setup dilution series and allowed me to use simple OD measurements. Of course, it depends on the stability of your aggregates so you should play a bit with the sonification intensity but the viability of the cells was still there. If you want I can send you a short description how to do it.

Hi Anna, we have just published a method of using metabolic dye (TTC) to measure growth and had sucess with that in Campylobacter jejuni growht in media which particulates. The paper is on my profile if you think it might be useful. Tween 80 does affect the viability of C. jejuni but I think you could test this in your system simply by plating out some bacteria from the cultures and comparing them to the untreated flasks (or use a baclight kit to detect viability changes). Hope that helps you a bit?

Hi Anna. That's a tough one. If you could pipette representative samples you could do a protein or DNA assay as a proxy for growth. If you can't, you'll need an assay that can be performed on the entire culture, that doesn't affect growth and which isn't affected by the clumping. The clumping problem rules out pretty much every photometric assay. The only solution I can think of is to measure the depletion of some essential nutrient like free phosphate or the nitrogen source. Not easy, because you will need to be very accurate in the early growth stages. A good analytical chemist would be a boon.

Hi, you could also use sonification to homogenize the aggregates. I had the same problem and applied homogenization of aggregates in a water bath. That technique also helped me to setup dilution series and allowed me to use simple OD measurements. Of course, it depends on the stability of your aggregates so you should play a bit with the sonification intensity but the viability of the cells was still there. If you want I can send you a short description how to do it.

Hi, sometimes the use of a different growth medium is the solution of the problem.

Hi Anna; you could try sonication at different power intensity, or use a tissue homogenizer. Another method is to add glass beads to the culture and vortex before sampling. The use of dry weight, protein, DNA, RNA, depletion of a nutrient like glucose etc. can be used to replace ODs. Since you can't pipette a representative sample to do any of these assays, I suggest running multiple tubes of the growing culture and pull out duplicates over the time course of the experiment and use the whole sample to do the appropriate assay.

Hi Anna,

there is a recent paper about the aggregation of Methylobacterium strains ( https://www.researchgate.net/publication/235376061 ) in which the authors state that "Aggregation occurred ... under high C/N growth conditions". As Hans-Jürgen Busse mentioned, you can try to use a different growth medium. I used to work a lot with CHOI-medium (see https://www.researchgate.net/publication/246994521 ) or the derivates Medium 3&4.

You can also try to use another carbon source than methanol.

Article Aggregation of selected plant growth promoting Methylobacter...

Article High-cell-density production of poly-β-hydroxybutyrate (PHB)...

For mycobacterium, we add either 0.05% tyloxapol or 0.05% tween-80 to all growth media. It prevents the aggregation.

Thank you all for the helpful answers! I was thinking of trying sonication although wasn't sure how does it affect cell viability. I will try also adding Tween 80 to the liquid media. I noticed that aggregation depends on the media used, however, formation of clumps was observed both in rich (NB, R2A) as well as in various methanol-supplemented media and it apparently depends also on the specific strain that is cultured.

As I want to grow different Methylobacterium strains, all in the same culture conditions for further metabolite analysis, I am looking for some universal and non-invasive technique of estimating bacterial growth.

Thanks again for all the hints!

You could try antifoam A (Sigma), it is a non-toxic silicone emulsion used to prevent foaming in E. coli cultures grown in baffled shaker flasks. I am thinking it may help prevent aggregation "if used from the start of a culture". Just make sure you include it in your reference media cuvettte since it does affect the OD a little.

Using a concentration of Triton X 100 that does not affect the cell growth might help, i guess, in preventing aggregation.

Hi Anna,

Hope these protocols help. Cheers,Elma

Try to vortex the liquid cultures, albeit smoothly.

I've used something like Lee Klinkenberg's, Gary Laco's, and Sandeep Gaudanas techniques: To determine viable count of Bacillus spores, which clump well, I've done serial dilutions with Tween 20. You could try the other detergents and any Antifoam as well, depending on whether or not the detergent lyses the bacteria and/or you need to keep the samples viable. FTR, Antifoam 204 works well with Bacillus spp. As stated above, for O.D., do include the detergent in the blanks.

Hi Anna! If you do not want to change your growth conditions I suggest you add a defined quantity of sterile glass beds an use a homogenizer like a vortex, in full power, for at least 30 seconds prior to the OD measurements. With sonication can cause the rupture of the cell wall and cell lysis.

Thank you for all the answers! I think I have to find out what type and concentration of available detergents will work the best with my system when added to the media prior to bacterial inoculation. This will allow to culture all the strains in the same growing conditions, independently on their tendency to clumping.

Hi all! I tried vortexing bacteria with the small glass beads, and it seems working and breaks the clumps in the solution nicely. However, does anyone has any experience to tell if this technique is safe and does not destroy the cells as well? I want to keep cell pellet intact as much as possible, but I am not able to compare viable cell number before and after vortexing. This is why I would rather add detergent to all the media, as in this case all the cultures will be affected in the same way, in contrast to vortexing only the clumped ones.

Anna, you could try a detergent then compare counts with and without vortexing with glass beads. I'm dead certain you're lowering your cell count, because we use "bead-beaters" to get DNA out of bacterial spores to determine genomic equivalents (GE). If you don't know, GE is 1 - 2 orders of magnitude greater than cfu for bacteria, and about 3 orders of magnitude greater than pfu for some viruses.

I have used a very similar method to that described by Madalena Lemos, for breaking up sputum and suspending the bacteria. I used a bead beater with large silicon-zirconia beads. Saline was the diluent. To check if the homogenization conditions are damaging, see if the CFU/mL decreases with increasing homogenization time. So run a time course, measuring CFU/mL (or CFU/g) from the point when the suspension first appears homogenous. Possibly a combination of this method with a surfactant like Triton X-100 will be the answer. That said, it surprised me how effective was saline alone.

Thank you Roger, for your response. From my experience so far I can tell that shaking the tubes energetically before OD measurements to some point helps in dispersing the clumps. However, increase in the culture optical density not always was correlated with the exact cell number in the culture, as measured by direct plate count technique. My thinking is that breaking the clumps, even if it does not strongly affect cell viability, may release some cell wall compounds, used by bacteria to build the "bridges" between cells, to the medium which in consequence will make the spectrophotometer reading false.

I experienced similar in relation to P. aeruginosa biofilms. It was incredible to see that increasing serial dilution, out to 10^-12 or more was still getting a count. There should have been much less than 1 cell at this dilution. It was clear that the biofilm was not being disrupted. That is also when using glass beads with bead beating was found to be effective.

I suggest that if you can actually disrupt the clumps the standard curve will work just fine, as the concentration of the cell wall compounds you describe will change in the same way as the numbers of the bacteria. That is part of the reason that when you make a standard curve it should be used for only one type of bacteria at a particular stage of growth.

Measuring numbers by RT-PCR of a housekeeper gene is another method that I have used for these types of samples. Though I would not recommend this if you can avoid it. And at some point you will still need to make a standard curve against known bacterial numbers

Anna, please let us know which method works best for you. If you could without compromising original work, please include relative results. This problem goes well beyond Methylobacterium spp.

Roger, I would be interested in the gap between GE and cfu when you bead-beat the Ps. aeruginosa. Have you looked at that? When I worked in the clinical lab, we had some very rubbery growth of that bug on plates. Such growth, rarely in the form of separated colonies, was hard to suspend. ( Because the growth was always slow, and these isolates were usually multiresistant, we suspected they were porin mutants.)

Hi, you may try the oldest approach, to weigh the cells by filtration them on membrane (0.22 um poresize) and dry-out. However, you must have enough cell number (culture volume) for that since one cell may only weight 10^ -12 or -13 g and you should have a good balance to determine around 10^ -2 g level of biomass and to be precise at least at 10-4 level. Wash the cells before filtration by DI water since the medium may contain salt to interfere the weighing.

Hi, if you have a single source of Carbon in the medium, you can try to measure its concentration. It will correlate with cell density till the bacteria stop to grow exponentially. All other ways will change your conditions of growth.

Hai Anna, My suggestion is before you are going to be measure optical density you just sonicate the culture and then take a OD, it will separate the clumping of bacterial culture. and if your bacteria produce any exopolysaccharide?

Isothermal microcalorimetry (IMC) provides a means to accurately measure growth of bacteria which clump in solution. Clumping does not matter. IMC works by measuring continuously in real time the amount of metabolic heat produced by the bacteria. The amount of heat (uJ/s = uW) is proportional to the number of bacteria present. The aggregate heat (uJ) produced vs. time is proportional to bacterial growth rate. For a complete explanation with lots of literature citations, see:

https://en.wikipedia.org/wiki/Isothermal_microcalorimetry#Microbiology

and see the complete article for background on IMC principles.

https://en.wikipedia.org/wiki/Isothermal_microcalorimetry

I would recommend to determine protein concentration in clustering bacterial suspensions to monitor their proliferation instead of OD measurement. There is no way to transform a clustering bacterial suspension in a really homogenic one because the cells in clustering cultures have high tendency to re-cluster again.

Thank you all, for the valuable answers!

Igor Brown and Alex Ignatov, I would really appreciate if you provide some more details on the protocols that can be used to calculate protein and carbon concentration (the sole carbon source in the media I am using is 0.5% MeOH) in growing bacterial culture. Also, I am wondering how to correlate the measured intensity of the metabolism with the accurate cell number in the culture?

Thank you, once again.

Hi Anna , I don't know if this problem is still up to date, but I think your first Idea with tween is not so bad. You can use a similar amount of carbopol this usually prevents filamentous ungus from forming mycells, so you can have a bacteria like shaking- or fermenting culture. Have a look to this patent: US 8343741 B2

You can use an anionic detergent like Carbopol 934P or similar, the Tweens ar non-ioonic, this is why you do not get a similar effect

Correlating metabolic heat of bacteria in culture (measured by IMC) with the number present is straightforward. In IMC the bacteria and culture media are typically in a 2 ml sealed ampoule. At any time t, note the cumulative amount of heat (J) up to time t. Remove the ampoule from the IMC instrument. Stop further growth by any convenient means. Assay the liquid for e.g. total bacterial protein. Doing this on specimens removed from IMC at a series of times gives calibration curve for cumulative heat vs. cumulative number of bacteria present vs. time.

Dear Anna,

Hopefully you have already solved your challenge. If not, check the attached paper for aggregation characterization on antibodies. I know that detection of bacterias with the biosensor is straight forward and maybe the assays can be combined to both measure the total density and to obtain aggregation characterization.

Feel free to contact me directly if you have any questions.

Best regards

Teodor

Article Improving the Developability of Biopharmaceuticals

I will soon try tween and I'll tell you. I know some work that has been done in Mycobacteria.

As I said before, isothermal microcalorimetry (IMC) is an ideal method for your purposes, because it does not matter whether your bacteria are planktonic, attached to solid surfaces, floating on the surface of a nutrient broth, etc. Here is a link to pertinent section of the Wikipedia article on IMC, and below that is what I wrote before. https://en.wikipedia.org/wiki/Isothermal_microcalorimetry#Microbiology 

Correlating metabolic heat of bacteria in culture (measured by IMC) with the number present is straightforward. In IMC the bacteria and culture media are typically in a 2 ml sealed ampoule. At any time t, note the cumulative amount of heat (J) up to time t. Remove the ampoule from the IMC instrument. Stop further growth by any convenient means. Assay the liquid for e.g. total bacterial protein. Doing this on specimens removed from IMC at a series of times gives calibration curve for cumulative heat vs. cumulative number of bacteria present vs. time.

Tween will help, but may not completely eliminate aggragation. Running a sample several times through a thin tube (e.g., blunt needle) will break up

aggregates with shear, and is gentler than sonication or bead vortexing.

It is not necessary to use an extremely narrow tube, a 21 gauge needle is

small enough. You can assess how many times are necessary by initially

measuring OD after, say, 4 times, 6 times, etc. and noting when the OD

no longer changes. Blunt needles can be obtained inexpensively from

suppliers for ink jet printing cartridge accesories.