JOURNAL OF BACrERIOLOGY, Jan. 1976, p. 340-345Copyright 1976
American Society for Microbiology
Vol. 125, No. 1Printed in U.S.A.
Genetic and Physiological Classification of
Periplasmic-LeakyMutants of Salmonella typhimurium
R. A. WEIGAND* AND L. I. ROTHFIELD
Department of Microbiology, University of Connecticut Health
Center, Farmington, Connecticut 06032
Received for publication 29 July 1975
Mutants of Salmonella typhimurium that leaked periplasmic
proteins wereisolated. Four classes of mutants were identified by
their increased sensitivity todyes, detergents, or antibiotics.
Conjugation studies indicated that representa-tives of two classes
mapped in the proA-galE region of the Salmonellachromosome and two
in the cysI-argE region. According to their
bacteriophagesensitivity pattern, all four of the mutant classes
appear to retain the smoothlipopolysaccharide characteristic. One
class of mutants has an abnormal cellenvelope in which the outer
membrane balloons away from the murein layer.
The periplasmic region of the cell envelope ofgram-negative
bacteria contains a number ofproteins that do not escape into the
mediumunder normal conditions, although they can bequantitatively
released by treatment of wholecells by the cold osmotic shock
procedure of Neuand Heppel (10).To study the role of cell envelope
components
in preventing release of periplasmic proteins innormal cells, we
previously developed tech-niques for the isolation of mutants that
sponta-neously leaked the periplasmic enzyme ribonu-clease (RNase)
I into the medium during nor-mal growth (8).
In the present report we describe furthergenetic and physiologic
studies of a large num-ber of periplasmic-leaky mutants of
Salmonellatyphimurium.
MATERIALS AND METHODS
Organisms and media. The S. typhimuriumstrains used in this
investigation are listed in Table 1.Cells were grown in proteose
peptone-beef extractmedium (PPBE) (17) or minimal medium (M9)
(1).Solid media (PPBE agar and M9 agar) also contained1.5% agar.
PPBE-ribonucleic acid (RNA) plates con-tained 20 ml of PPBE agar
with a 4-ml overlay of 1'7yeast RNA (pH 7) (Sigma) in PPBE agar.
Minimalmedia were supplemented with 0.4% glucose (Baker)and when
appropriate: amino acids (Sigma), 204g/ml, except for serine, which
was 200 Ag/ml; nucleo-sides (Sigma), 5 Ag/ml; lipoic acid
(Calbiochem), 0.2jg/ml; and streptomycin (Sigma), 200 gg/ml.
Isolation of mutants. Cultures of S. typhimuriumSA722 were grown
to stationary phase in M9 mediumcontaining serine, adenine, and
thiamine. After thecells were pelleted and suspended in M9, they
weretreated with 4% ethyl methane sulfonate (EastmanOrganic) for 20
min at 37 C. After the cells werewashed three times with M9,
survivors (28w survival)
were immediately plated on PPBE and were cloned bystreaking
before further study. The resulting cloneswere screened for leakage
of RNase by transferringthem onto PPBE plates and onto PPBE-RNA
plates.After 15 h at 37 C, the PPBE-RNA plates wereflooded with 8
ml of 0.5 N HCl for 3 min. Mutantsthat leaked RNase were identified
by the large halosaround the colonies.
Conjugation technique. The donor strain wasgrown to
midexponential phase in PPBE and thenwas mixed gently with a
10-fold excess of an exponen-tial-phase culture of the F- recipient
strain in PPBE.After 30 min to 3 h at 37 C, the mixture was
diluted10-fold with saline, followed by disruption of themating
pairs with a mechanical mating interrupter(9). The mixture was
immediately spread on selectiveplates, and recombinants were scored
after 72 h.Mapping of RNase-leaky loci. Matings were per-
formed as described above with the RNase-leakymutants as donors
and a series of auxotrophic F-strains (Table 1) as recipients.
Matings were inter-rupted 30 to 60 min after the first appearance
ofrecombinants for a selected marker, and 100 to 400recombinants
were isolated for each of the selectedmarkers indicated in Table 1.
Recombinants werecloned, and each clone was tested for RNase
leakageon PPBE-RNA plates. For each selected marker, theratio of
RNase-leaky recombinant clones to totalrecombinant clones was
determined, and the ratiowas plotted against the genetic map
position of theselected marker (Fig. 1). If the selected
auxotrophicmarker is transferred long before the Iky mutation,very
few of the recombinants would be expected toreceive the Iky
mutation. If the selected marker istransferred after the Iky
mutation, about half of therecombinants should leak RNase. When the
selectedauxotrophic marker is very near the Iky mutation, ahigh
percentage of the recombinants are likely to leakRNase. Each Iky
mutation was assigned a maplocation corresponding to the point at
which the ratiofirst reached its maximal value. For map positions
ofthe selected markers, see reference 18.
Analytic methods. Sensitivity to inhibitors was
340
on May 17, 2018 by guest
http://jb.asm.org/
Dow
nloaded from
http://jb.asm.org/
VOL. 125, 1976
TABLE 1. Genotypes of Salmonella typhimuriumstrains
Strain Selected markersa Other markers
SA722b Hfr K10, serA, purSA571b metA, trpB, hisF, xyl, strA,
malA
ilvA, pyrESL761c galE, proA purC, fla, iM, strA,
fim, ileSB100c tyrA, proA purI, purC, iM, fla,
strA, fim, ile, rhaSA1124c cysC,D deletion metGSB1301c cysI,
lysSU277b tyr, argE serASA342b lip str, araaroEb aroESA225b
purEleuA' leuA
a Auxotrophic markers that were used to selectrecombinants in
the conjugation studies described inFig. 1 and in the text. Gene
symbols used are asdefined by the donors of the strains and follow
thenomenclature of Sanderson (18). All strains otherthan SA722 are
F- strains.
bStrain kindly provided by K. Sanderson of theUniversity of
Calgary.
cStrain kindly provided by P. Hartman of TheJohns Hopkins
University.
assayed by colony-forming ability on inhibitor plates.The
inhibitor plates contained 20 ml of PPBE agarplus one of the
following inhibitors: methylene blue(Baker), 100 ,ug/ml; acridine
orange (Sigma), 100ug/ml; rifampin (Sigma), 8,g/ml;
ethylenediamine-tetraacetate (EDTA), 1 mM; deoxycholate
(Calbio-chem), 10 mg/ml; sodium dodecyl sulfate (Sigma), 10mg/ml.
Clones were transferred from solid media ontothe inhibitor plates,
and growth was scored after 16 hat 37 C. Sensitivity to
bacteriophages was measuredby spot-testing bacteriophages on a
freshly pouredlawn of cells in enriched soft agar. Cyclic
phosphodi-esterase was assayed according to the procedure ofNeu and
Heppel (10) with the bis-(p-nitrophenyl)-phosphate (Sigma) assay
(0.2-ml volume). Acid hex-ose phosphatase activity was measured in
an assaymixture (0.2 ml) containing 50 mM sodium acetate(pH 5.8), 5
mM EDTA, 0.5 mg of bovine serumalbumin per ml and 5 mM
[14C]glucose-6-phosphate(G-6-P) (1.8 x 104 dpm/mM) (New England
Nu-clear). The reaction was terminated by heating at100 C for 3
min. The liberated ['4C]glucose wasseparated from [14C]G-6-P on a
BioRad AGW 1-X8column in a Pasteur pipette by elution of
the['4C]glucose with 3 ml of water and counted in amixture of
toluene, Liquifluor, and Biosolv (toluene:Liquifluor [New England
Nuclear ]:Biosolv [Beck-man], 267:8:25). The RNase assay mixture
(0.4 ml)contained 30 mM potassium phosphate (pH 7.0),1.5 mM EDTA,
and [14C]polyadenylic acid (0.68mmol of phosphate/ml; 4 x 104
dpm/mmol of phos-phate) (Miles Laboratories). The reaction was
termi-nated by adding 0.1 ml of 0.2'S. yeast RNA, 1 ml of
LEAKAGE OF PERIPLASMIC PROTEINS 341
cold 3% HClI4, and chilling 15 min. The mixture wascentrifuged
at 10,000 x g for 10 min, and 1 ml of thesupernatant was counted in
the toluene, Liquifluor,and Biosolv mixture. Glucokinase was
assayed in amixture (0.2 ml) containing 100 mM
tris(hydroxy-methyl)aminomethane-hydrochloride (pH 7.5), 10mM
adenosine 5'-triphosphate (Sigma), 5 mM MgCl2,2.5 mM NaF, and 0.5
mM D-[14C]glucose (40 mCi/mmol) (New England Nuclear). The reaction
wasstopped by heating at 100 C for 30 min; the productswere applied
to a BioRad AGR1-X8 column in a Pas-teur pipette; and the column
was washed successivelywith 4.5 ml of water, 6 ml of 5 mM HCl, and
3 ml of 50mM HCl. The unreacted glucose was eluted in thewater
wash, an undetermined radioactive contami-nant was eluted in the 5
mM HCl wash, and the G-6-Pwas eluted in the 50 mM HCl wash. The
[14C]G-6-Pwas counted in the toluene, Liquifluor, and
Biosolvsolution. G-6-P dehydrogenase was assayed in a mix-ture (1
ml) containing 87 mM glycylglycine (pH 8.0),15 mM MgSO4, 10 mM
G-6-P (Sigma), and 0.4 mMnicotinamide adenine dinucleotide
phosphate(Sigma). The reaction was started with the additionof
nicotinamide adenine dinucleotide phosphate andinitial rates were
measured at 340 nm in a Gilfordrecording spectrophotometer.
Ribose-binding proteinsand histidine-binding proteins were assayed
(2) by R.Aksamit and D. Koshland of the University of Califor-nia
at Berkeley, on cultures grown for 24 h at 30 C inVBC medium (20)
containing 0.4% ribose. Sodiumdodecyl sulfate-polyacrylamide gel
electrophoresis bythe method of Neville (11) was carried out on
isolatedcell envelopes. The pellet of the 360,000 x g
centrifu-gation, according to the procedure of Osborn et al.(16),
was suspended and used as the source of cellenvelope material.
Electron microscopy. Electron microscopy wascarried out as
described previously (21).
RESULTSMutant isolation and mapping. One hun-
dred mutants that leaked RNase were inde-pendently isolated as
described above. Twentyof the mutants that showed the most
markedleakage were selected for further study.
Since we have been unsuccessful in develop-ing a positive
selection technique, the muta-tions responsible for RNase leakage
wereroughly mapped by conjugation with the iso-lated mutants acting
as chromosome donorsand well-marked F- strains as recipients;RNase
leakage was scored as an unselectedmarker. The loci responsible for
RNase leakagefell in two widely separated regions of the
S.typhimurium genetic map. Mutants thatmapped in the proA-galE
region were defined aslkyCD and those mapping in the
cysI-argEregion as IkyAB (Fig. 1). As described below,differences
in phenotypes among the mutantssuggested that each region might be
furthersubdivided. Precise definition of the map loca-tions was
complicated by the relatively poor
on May 17, 2018 by guest
http://jb.asm.org/
Dow
nloaded from
http://jb.asm.org/
