 |
 |
 |
 |
Thanks to everyone who responded. The answers are somewhat mixed, so we
decided to screen a couple of the hybridomas to see what kind of
fluorescence profiles would be generated. The first gave a broad
fluorescence profile, but the second had no discernable fluorescence over
control. The user confirmed that there was indeed quite a bit of secretion
(by ELISA, I believe), so perhaps this line was an exception to the positive
results that some have experienced.
Unfortunately, I'm not going to find out. Based on the negative outcome
(and probably since I would have charged to do the sorting), the user
decided to manually re-clone the 5-7 hybridomas in question.
My lab is now planning to look into some of the secretion-capture flow
methods that have been posted - and my thanks to those who posted answers to
that recent query as well (even though I was not the person who asked).
Sure is great to have this resource!
Regards to all,
Neal Benson
--------------------------------------------------------------------------------
Q:
One of my lab's users has several hybridoma cell lines that they would like
to re-clone to improve antibody production. It was suggested that they use
cell sorting to select the "high-expressers". They propose to label the
cells with a FITC-labeled anti-mouse IgG (subclass specific), then sort
those in the upper tail of the resulting fluorescence histogram.
In my limited knowledge of this area, I seem to have the impression that
expression of antibody on the cell surface does not necessarily correspond
to the quantity of antibody secreted from a hybridoma cell. I'd be very
interested to obtain opinions as to whether sorting of these cells based on
high surface expression is a reasonable thing to do for improvement of
antibody production.
--------------------------------------------------------------------------------
A: My lab, and a few other laboratories in this institution, have had
great success in sorting the high expressor hybridomas. Invariably, we found
these cells to be high secreters as well.
--------------------------------------------------------------------------------
A: You're right...in our experience, there is little correlation
between cell surface Ig and the level of secretion of Ab by hybridoma
cells, although there may be some hybrodomas that can be selected this
way. The best bet to improve Ab production by a hybridoma is to
reclone with limiting dilution (or Autoclone/ACDU if you've got one)
and to screen clone supernatants by ELISA. Of course, you have to
control somewhat for the rate of cell growth of each clone, but good
high-production clones usually make themselves known fairly quickly by
this method.
--------------------------------------------------------------------------------
A: From time to time we do this sorting of the highest expressors of
hybridomas and indeed it does seem to work. We normally set the "cut-off" as
being in the top 2% on FITC then sort straight into 96 well trays.
--------------------------------------------------------------------------------
A:
Actually (and somewhat surprisingly), sorting for high surface
expression does often yield high secretion as well. We have, over the
years, found that this procedure can work wonders.
You can take two approaches. (1) bulk sort the 0.5% highest
expressing cells, keeping the population as a "mixture" (i.e.,
epigenetically-variant clones). (2) single-cell sort the very highest
expressors and then test the grown-up wells for secretion. #2 is the
preferential method for getting stable high-producers, however, some
hybridomas don't clone very well. Try both.
--------------------------------------------------------------------------------
A:
Based on Marder et al.'s report of success (Cytometry 11:498, 1990), I
sorted one lackluster hybridoma for high surface Ig expression. Two
separate batches of ascites of the original line gave half-saturation
at 1/200 and 1/240; culture supernatant at >1/2. I stained with
FITC-anti-IgG and sorted for the brightest few percent. The staining
in this case did not show any Ig negative cells, so I was simply cutting
off the tail of the bell-shaped curve. A pristane-primed mouse injected
with 1,000 of the brightest cells eventually produced ascites with 60-fold
more antibody than before (half-sat 1/12,000). We were delighted!
A big advantage of this method is that you can start antibody production
immediately. If you have pristane-primed mice available, you can
sort in a single day (NON-STERILE if you wish!) and inject the sorted
cells directly into mice. This doesn't guarantee clonal stability
(since you haven't cloned) but gets you lots of antibody in the least
time (saving the weeks needed for limiting dilution enrichment, screening,
limiting dilution cloning, re-screening, and clone expansion).
I'd be interested in the basis for Rob Chervenak's statement:
"in our experience, there is little correlation between cell surface Ig and
the level of secretion of Ab by hybridoma cells". Was I just lucky?
--------------------------------------------------------------------------------
A:
Herzenberg and Parks worked on this topic in the late 70's or early 80's
and they found NO CORRELATION beteween antibody expression and antibody
production.
Your best bet would be to encapsulate your cells and allow them to divide
for several days and then work out a staining protocol to quantitate the
amount of produced antibody...
A lot of work, picking clones seems the fastest way out if you plan to do
this once or twice.
--------------------------------------------------------------------------------
A:
We have used this technique to obtain 'higher' Ig secretors for
several years. In response to Rob Chervenak's statement: "in our
experience, there is little correlation between cell surface Ig and
the level of secretion of Ab by hybridoma cells" I would say that
there is NOT an absolute correlation between the two, but in many or
even most instances, there is! Our experiences are outlined in
Cytometry 11:498, 1990.
Like many things in the literature, there are a few tricks. First
and foremost, since hybridoma cells are not plasma cells, the Ig on
their surface is probably due to passive adsorption. To detect
these small amounts, you have to use a high concentration of
anti-Ig-FITC (~100 mcg/ml). Normal concentrations (~10 mcg/ml) do
not detect the slight differences very well. Even at that, some
hybridomas still do not behave well. Our greatest successes have
been in resurrecting cultures that have quit secreting antibody,
altogether. We stain and clone the few bright cells present, grow
them up and again have a secreting culture. In other instances, we
have been able to separate cultures containing mixed secretors (IgM
& IgG's) into clones secreting a single isotype (using
isotype-specific antibodies).
Even though, there is not an absolute relationship, we routinely
use the procedure. Our reasoning is that the procedure adds very
little work and the potential gains are well worth the effort.
--------------------------------------------------------------------------------
A:
I have (long ago) used surface expression to obtain high producers with some
success. The current fashion is to use secretion with trapping of secreted
product on the cell surface or in microdroplets. You might contact:
One Cell Systems, Inc
100 inman St
Cambridge Mass 02139
617-868-2399
617-876-6468 FAX
I am not sure of the commercial status of their droplet maker, I haven't talked
with them for some time. Their technology is similar to that published by
James
Weaver, ie agarose microdroplets.
Andreas Radbruch has published some work on surface trapping of secreted
products, but I don't have any of those references.
--------------------------------------------------------------------------------
|
|
|
|