Summary: Genome Size Assessment

From: Claudio Vallan (claudio.vallan@dkf7.unibe.ch)
Date: Thu May 04 2000 - 07:32:05 EST


Many thanks to Nigel Blackhall, Gerhard Nebe von Caron, Doug Redelman and
Dennis J. Young for answering my questions about genome size assessment.
Here a short resume of their wisdom:



Summary of answers:

It is apparently not unusual to periodically see a shift in staining
intensity of PI stained samples. To control for this always include a
reference in the sample (e.g. chicken RBC nuclei) or better analyse the
samples together in the same tube, if possible. Use high concentrations of
PI for minimal effect of concentration disturbances. (20 - 50 ug/ml) To
dilute samples always use the staining buffer. Optimal flow rates
conditions: 500 bugs / sec.

It is very important that all samples have the exactly same treatement.  In
particular cross-linking agents such as formaldehyde reduce the labeling by
intercalating dyes such as PI. Be aware that  PI binds to chitin and other
things in the cell wall, which can alter signal intensity.





Original question:

Dear flowers,

I was asked to determine the genome size of Phytophtora porri, by measuring
the emission of its propidium iodide (PI) stained DNA and comparing it to
the emission of P.Infestans and E.Coli, for which the genome size is known.
My problem now is, that I do not trust the results for following reasons:

- The Phytophtora and the E. Coli were not treated exactly the same way.
Phytophtora was formaldehyde fixed prior  to 70% ethanol treatment  while
E. Coli was treated with 70% EtOH. only. Both were treated with DNAse-free
RNAse  and stained with 20ug/ml PI, though.
Might the difference in treatment lead to wrong results?

- When we measure cell cycle (e.g. in human and mouse cell lines), for
reasons that are beyond my understanding, now and then some samples which
were treated exactly the same way as the other samples will show a shift of
the G0/1 and G2/M peaks compared to the other samples. (The cells are
guaranteed to have the same amount of DNA. And I saw this with different
users and protocols.) The results are interpretable, but we have to change
the location of the markers (which sometimes is quite difficult, if you
have a shift of the G1 peak and you do not see exactly where the G2/M peak
is). This led me to the conclusion that PI staining might not be ideal to
assess the size of the genome, because of inconsistency of staining.

- When measuring the E.Coli-PI emission I had to dilute them (with BD FACS
Flow) because there were too many of them in the sample. The diluted
bacteria showed a shift to the left of almost one log in emission.  Now
either the FACScan will overestimate the fluorescence when charged with too
many cells or, more probably, the staining with PI is very sensitive to
dilutions. So the inconsistencies in the cell cycle explained above could
be due to a drop of sheet fluid falling into the sample.



Again many thanks for the answers

Claudio Vallan

===================================================
Claudio Vallan PhD                          Phone Lab:      031 / 632 88 76
FACS-LAB DKF                               Phone Office:  031 / 632 99 68
University of Bern                         E-Mail:  vallan@dkf7.unibe.ch
c/o Institute of Pathology
Murtenstrasse 31                          Insel hosptial area only:
3010 Bern                                     Beeper:           181 67 59
===================================================



This archive was generated by hypermail 2b29 : Sat Mar 10 2001 - 19:31:18 EST