I'm currently working on a research project which aims to identify cyanbacterial species present on moss leaves through morphological and molecular methods (polyphasic approach to cyanobacterial identification). The genera in question are Nostoc, Stigonema, Cylindrospermum, and Calothrix.

My question is, is anyone aware of a technique which can be used to separate bacteria based on size, charge, or other morphological characteristics?

We have been attempting to culture the various cyanobacteria on agar, which has its drawbacks. I'd like to find way to identify the actual cyanobacterial species/strains growing epiphytically on the moss, as opposed to working with the isolated species/strains on the agar. 

The difficulty I'm having in the experimental design is that if I were to use a denaturing gradient gel electrophoresis (DGGE) or other molecular method for microbial community analysis, I would not be able to connect the genetic identity to a morphological one (as far as I know). My thought has been that if I can find a way to physically separate the various cyanobacterial morphologies in large enough amounts, then I can run PCR on each of the morphological groups. Is there a method of using gel electrophoresis or other filter?

Any information on the topic will be greatly appreciated!

I should have included that I plan to detach the cyanobacteria from the moss leaves into sterile water by vortexing in a plastic tube. 

Morphological separation of bacteria is difficult. Automates systems like cell sorters do not work well on that scale and generally require some fluorescence markers. Theoretically, a variety of single-cell microfluidic systems could be used, but many are in the concept stages and require a lot of development. Setting up a working system can easily turn into a PhD thesis. 

The traditional approach would be to use plating to eventually isolate pure cultures and characterize those pure strain thoroughly.

You could look into using Raman spectroscopy. The question is whether it will be sensitive enough to differentiate between them. Also it does not (yet) include morphological information, only biochemical data.

6 hours ago, Prometheus said:

You could look into using Raman spectroscopy. The question is whether it will be sensitive enough to differentiate between them. Also it does not (yet) include morphological information, only biochemical data.

AFAIK similar to MS techniques they require a significant amount of pure cells.

You can get a signal from a monolayer but more cells would improve it. Could also use Surface Enhanced Raman Spec. People claim to be able to detect single cells. I don't know anything about looking at cells though, just suggesting alternatives.

First of all, can you get pure samples of each genus?

If you can get hold of pure samples then you can do some old fashioned protein SDS PAGE made against secreted antigens from each genus.  These are probably few enough so that you end up with a reasonable protein profile.  If you make radiolabelled antibodies against the secreted antigens, you have a fine control of the secreted proteome from each genus.

Then you can use some fancy reverse genetics to make a DNA probe for each genus and then finally use your chosen method to identify genus/species specific DNA.  I am assuming that there is enough genetic difference to classify the cyanobacteria into distinct criteria.

Quote

Distinct Differences in Repertoires of Low-Molecular-Mass Secreted Antigens of Mycobacterium aviumComplex and Mycobacterium tuberculosis

1. Ingrid Olsen1,2,^*, 
2. Liv J. Reitan1, and 
3. Harald G. Wiker2,3

http://jcm.asm.org/content/38/12/4453.full

Quote

Small secreted proteins enable biofilm development in the cyanobacterium Synechococcus elongatus

• Rami Parnasa
• , Elad Nagar
• , Eleonora Sendersky
• , Ziv Reich
• , Ryan Simkovsky
• , Susan Golden
•  & Rakefet Schwarz

• Scientific Reports 6, Article number: 32209 (2016)
• doi:10.1038/srep32209
• Download Citation
• • Biofilms
  • Prokaryote

Received:

15 April 2016

Accepted:

03 August 2016

Published online:

25 August 2016

Abstract

Small proteins characterized by a double-glycine (GG) secretion motif, typical of secreted bacterial antibiotics, are encoded by the genomes of diverse cyanobacteria, but their functions have not been investigated to date. Using a biofilm-forming mutant of Synechococcus elongatus PCC 7942 and a mutational approach, we demonstrate the involvement of four small secreted proteins and their GG-secretion motifs in biofilm development. These proteins are denoted EbfG1-4 (enable biofilm formation with a GG-motif). Furthermore, the conserved cysteine of the peptidase domain of the Synpcc7942_1133 gene product (dubbed PteB for peptidase transporter essential for biofilm) is crucial for biofilm development and is required for efficient secretion of the GG-motif containing proteins. 

https://www.nature.com/articles/srep32209

Edited January 3, 20187 yr by jimmydasaint