Cleaning worms with antibiotics besides streptomycin

Erich Schwarz schwarz at
Tue Nov 10 13:55:42 EST 1998

    Worm strains aren't always effectively decontaminated by
streptomycin (Str), and can be inefficiently decontaminated by
bleach -- either because the strains are poorly fertile, or because
one has many dozens to decontaminate, or both.  This motivated me to
ask bionet.celegans if other antibiotics besides Str work.

    First, my own brief take on all the responses:

    Ampicillin works sometimes, and doesn't hurt the worms.  You can
feed them on XL-1 carrying pBluescript, and XL-1:pBlue will grow (a
bit slowly) on NGM with Amp.  If you are in a hurry to get rich XL-1
lawns, try the "enriched peptone plate medium", plus ampicillin. En.
pept. medium is described by Lewis and Fleming in Ch. 1 of the _C.
elegans_ methods book (Academic Press).

    Ampicillin doesn't always work.  Another option is to grow the
worms on MYOB (described by Eric Lambie in WBG 13(2): 12a).  It is
very restrictive and seems to only grow _E. coli_.  It is not
totally reliable -- some worm strains die on it -- but worth trying
if Amp fails.

    A final option is tetracycline, which is very powerful at
cleaning up slimes but also somewhat toxic to worms.  Again, you can
grow XL-1 upon NGM + Tet or en. pept. + Tet, and then grow the worms
on that.  One might want to try using a LOW concentration of Tet (1
ug/ml instead of 10 ug/ml) to select against slime while trying not
to kill the worms too.

    Next, a lengthier and detailed summary of the responses:

    1. If you want to be maximally rigorous, use bleach.  John
Thaden <jjthaden at LIFE.UAMS.EDU> emphasized this point, noting that
protocols exist that allow bleach to be used on a small number of
gravid hermaphrodites.  For example, one is available at:

under the header, "Cleaning Worm Stocks: Michael Koelle".

    2. Ampicillin (Amp): Raffi Aroian <raroian at> reports
that this works well, given _E. coli_ carrying pBluescript.  I've
also found it apparently effective in stopping at least one slime
contaminant.  However, Thaden found Amp to be ineffectual against at
least one contaminant in his lab.

    3. Tetracycline (Tet): Peter Barrett
<BARRETT at A1.TCH.HARVARD.EDU> reports that this is the *one*
antibiotic effective against the most persistent contaminants.  One
caveat reported by Barrett and Thaden is that Tet visibly reduces
brood size and can prevent worm growth completely.  Another possible
caveat is that Tet is antagonized by magnesium.  Mg^2+ is a major
component of NGM, but not of MYOB or NGM Lite (more below).

    4. Gentamycin (Gen): Thaden finds that 60 and 200 ug/mL
concentrations inhibited growth of an Amp-resistant contaminant,
while lower concentrations failed.  Gen did not visibly harm worms
at 200 ug/mL.

    5. Pen Strep aminoperin (PSa): Thaden finds that PSa is harmless
to worms, and inhibited growth of at least one contaminant.

    6. Chloramphenicol (Chl): Several XL-1-Blue derivatives from
Stratagene are resistant to Chl at <40 ug/mL, so Chl might be worth
trying, especially since its mechanism of action (inhibition of the
bacterial 50S ribosomal subunit) is like that of Tet (30S
inhibition). However, like Tet, Chl may well be toxic to worms (it
hasn't been tested by anybody as far as I know).

    7. Changing the pH: Barrett reports that this can eliminate
unwanted microbes picked up at pH 8.0 in NGM.  An easy way to change
pH from 8 to 6 is to switch the worms to Eric Lambie's MYOB medium,
described in WBG 13(2): 12a.  MYOB is harmless to most, though not
all, mutant _C. elegans_ strains;  Hodgkin has observed one mutant
strain totally unable to grow on MYOB.  Where MYOB is unusable one
can instead try Lambie's "NGM Lite", described in WBG 13(5): 11.

    8. Antibiotics versus E. coli: For exotic antibiotics for which
a plasmid conferring resistance to E. coli isn't available, one
might plaster OP50 or HB101 onto the antibiotic plates in the form
of a thick paste.  Thaden: "A typical lawn on a normal 10 cm NGM
plate is about 18 A600 optical density units, so that is about how
much I would spread.  We use a stock slurry of OP50 in S buffer that
is usually 80-100 A600 units per mL, as determined by reading a
serial dilution at about 0.3 A600 units."  This would sidestep the
need for resistant _E. coli_; HB101 might be preferable, being less
sticky and easier for worms to eat.  Alternatively, one might plate
OP50 on several antibiotic plates, select rare resistant colonies,
and grow them as an OP50-derivative strain.

    9. XL-1 and XL-2 are consumable by worms, and are commercially
available as competent cells.  Also, XL-1 and XL-2 have intrinsic
Tet and Chl resistance.

    10. Not all contaminants are created equal.  Thaden found a
contaminant that was resistant to Amp.  In general, no one
antibiotic is guaranteed to remove all contaminants or any given

--Erich Schwarz
  schwarz at

More information about the Celegans mailing list