Copyright © 1977 American Society for Microbiology Printed in U.S.A.
Genetic
Mapping of regA Mutants
of
Bacteriophage
T4D
JOHN S. WIBERG,* STEVEN L. MENDELSOHN, VIRGINIA WARNER,' CYNTHIA ALDRICH,2 AND THOMAS S. CARDILLO
Department ofRadiation Biologyand Biophysics, School ofMedicine and Dentistry, University ofRochester, Rochester, New York14642
Received for publication 22 November 1976
SP62, a mutant ofbacteriophage T4 shown by Wiberg et al. (1973) to be defectiveinregulation ofT4proteinsynthesis, wasshown by complementation
tests todefineanew gene,regA,and byintergenic mapping tolie betweengenes
43and62.ThemappinginvolvedcrossingSP62 withaquadruple ambermutant
defectiveingenes 42, 43, 62,and44,selecting allsixclasses of amber-containing recombinantscaused by single crossover events, andthen scoring the presence orabsence of SP62intheserecombinants.Inaddition, 15 new,spontaneous regA
mutants wereisolated, and13 oftheseweremapped against eachother;a total
ofeight different mutation siteswerethus defined. Most of the new mutants wereisolatedaspseudorevertants ofaleaky ambermutant ingene 62,according
to KaramandBowles (1974), whereas one was identified by virtue of the "white
ring"arounditsplaque,aphenotype possessedby all the regA mutants at high
temperature. SP62 was renamed regA1, and the new mutants were named
regA2,regA3, etc.
Aviablemutantofbacteriophage T4Dthat is
defectiveinregulationofphageprotein
synthe-sis, namely SP62, wasdescribed by Wiberget
al. (18); unpublished datawere cited to argue
that the mutant, SP62, defines a new gene,
regA, thatmapsbetweengenes 43 and62. The
present paper presents those data, which
in-volve both complementation tests and
inter-genic mapping of SP62. Also described is the
isolationandintragenic mappingofmoreregA
mutants.
MATERIALS AND METHODS
Bacteriophage and bacterial strains. Escherichia coli B and E. coli K-12 strain W3110, both nonper-missive for T4 amber (am) mutants, and E. coli CR63, permissive for T4 am mutants, have been described (8, 17). We isolatedaspontaneous mutant of E.coliBthat is resistantto 150pgof
streptomy-cinpermlandassigneditbacterialstock no. 150;it
exhibits drastically reduced ribosomal ambiguity suppression ofT4 amber mutants. Wild type T4D (T4+) and the unbackcrossed versions of the am
mu-tantswereobtained fromR.S. Edgar. SP62 has been described (18), and all studies on it in this paper
were done with the genetically purified version, SP62x3; this has been renamed regA1 (see below), andbothnames areusedinthis paper. The quadru-pleam mutant used for mapping (see Table 2) was constructed by standard crosses and identified by the spot-testprocedure describedbelow;itcontains
1Presentaddress: 20Offgrove St., E. Weymouth, MA 02189.
2Presentaddress:1857LindenSt.,E.Lansing,MI 48823.
the following mutations: amN55x5 (gene 42), amB22x5 (gene 43), amE1140 (gene 62),and amN82 (gene 44).
Growthmedia and chemicals. Most of these have been described (18). GCA medium is the
glycerol-CasaminoAcids medium of Fraser and Jerrel (5). Thermometers. All temperatures above 40°C were read on thermometers calibrated against a
thermometer certified by the National Bureau of Standardstobeaccuratewithin0.3°C.
Complementation tests. For complementation
tests (Table 1), E. coliB wasgrownat37°CinGCA medium to about 5 x 10f cells per ml and then chilled. For eachinfection,1ml ofcellswaswarmed
by shakingfor3 min at 44.8°C in a 10-ml, baffled
Erlenmeyer flask. Phage (0.1 ml) were thenadded
at atotal multiplicity of infection (MOI) of10(5 of each, whereamixtureoftwophagewasadded).At 50 min after infection, samples were diluted into broth saturated with CHCl3 (to lyse the cells) for determination ofphageproduction; platingwas on
E. coli CR63 at30°C, conditions permissive for all progeny. Bacterialsurvivorsof infection were deter-minedas colony-formers by spreading appropriate dilutionsonGCA agar,alongwithtwodropsof anti-T4 serum havinga K value of 260; theantiserum
preventskilling ofsurvivorsby phageproducedon
theplates. Controls showed that this level of antise-rum was adequate and harmless to the bacteria. Unadsorbed phage were measuredas
plaque-form-ers thatsurvive dilution through CHCl3-saturated broth; theywere sampledat 5 min afterinfection, before progenyphage appear.
Spot-test identification of mutant genotypes. Ourcurrentprocedureforspot-testidentificationof mutantgenotypes(Table 2)istheresult ofaseriesof
742
on November 10, 2019 by guest
http://jvi.asm.org/
regA MUTANTS OF 743
[image:2.501.53.246.106.341.2]modifications in this laboratory (8, 9, 17, 18) of the technique ofEdgar et al. (4). Each plaque, chosen at random from the progeny of a cross, was stabbed
TABLE 1. Complementation testsof SP62 against nearbymutantsa
Total phage pro- %of corre-Pha e duced/cellat
44.80Cb
Avg% spondingPhg of T4+ (T4++am)
Expt 1 Expt 2 cross
T4+ 74(100) 64(100) 100
SP62 15(20) 14(22) 21
SP62 +T4+ 65(88) 83 (130) 109
T4+ + 43- 86(116) 56(88) 102
T4+ + 62- 44 (60) 38 (59) 60 T4+ + 44- 83 (112) 45 (70) 91
T4+ + 45- 68(92) 38 (59) 76
SP62 + 43- 68 (92) 44 (69) 81 80
SP62 + 62- 37 (50) 25 (39) 45 75
SP62 + 44- 40(54) 49 (77) 66 73
SP62 + 45- 40(54) 29(45) 50 66
43- 0.16 0.17
62- 0.44 0.22
44- 0.18 0.09
45- 0.13 0.14
a Procedures aredescribed in thetext.Theresults
of twoindependentexperimentsareshown. Nomore than 1.5% bacterial colony-formers survived at 5
minafterinfection; this wasmeasuredonlyin the infections bysingle phage.Unadsorbedinputphage
were measured at 5 min after all infections and
representedlessthan0.7phagepercell;thesewere
not subtracted from the values shown. Thefigures
in parentheses represent the percentage ofphage
production relative to T4+ as 100%. The 43-, 62-,
44-, and 45- mutants were amE4332, amE1165,
amN82,andamE10,respectively.
bCell titerbasedoncolony-formerspriorto infec-tion.
oncewith a sterile, wooden toothpick; a toothpick canpickup over108phage. The toothpickwasthen rinsed in two drops of medium in one well ofa sterile, plastic Linbro plate (Disposo Tray, catalog no.IS-FB-96, clear, Linbro Scientific Co., New Ha-ven,Conn.).Wemarked offan arrayof56suchholes onthe Linbro plate such that thepatternfit comfort-ablyinastandardplastic petridish(15by100mm). About 1 ,ul of each plaque suspension was then transferredwithareplica-plating devicetothe sur-face of thetopagarlayer (3 ml) seededwithinthe preceding5min with108bacteria and5 x 107helper phage where appropriate. We now use a "floating-loop replica-plating device" (J. S. Wiberg, J. Appl. Bacteriol., in press), whichwas developed both to
givegreater accuracyandreproducibility of liquid transfer andtoavoidtearing the softagarlayer. It featuresanarrayof 56thin, wire loops mountedso that theyarefreetomove upand downuponcoming intocontactwiththeagar.To identify amber
geno-typesfor Table2,agarplateswereseededasfollows:
(i)E.coli CR63 (replica); (ii) E. coli B (no ambers grow); (iii) E. coli B plus helper amN55x5 (gene
42)-only candidatescontainingtheamN55x5 muta-tion will failtogiveapositive complementationtest; (iv, v, vi) same as iii, except substitute amB22x5 (gene 43), amE1140 (gene 62),oramN82 (gene44), respectively,ashelper. A positive complementation
testappears as ashotgunpatternofmanydiscrete
plaques, similarto spot B8of Fig. 2; these plaques result from complementation, which permits phage production, and (equallynecessary)from the
subse-quent replication of resulting wild-type recombi-nants.
Thepresence orabsence of SP62inthe six recom-binantclasseswastested by spot-plate procedure2
ofWibergetal. (18), whichmeasurestheabilityof the candidatetocomplement helper SP62onE.coli B atabout450C, where SP62 doesnotgrow. Since
temperature control is very delicate for this test (18), manyreplicaswereprinted, andoneach spot-plate several control spots ofT4+ and SP62 were included; the whole plate was rejected whenever these controlsgaveunsatisfactory results.
TABLE 2. Genetic mappingof SP62 relative to genes 42, 43, 62, and44a
No. inexpt1 No.inexpt2 Totalno.of TotalregA+
Amber genotype by gene: withregAgeno- with regA geno- Total regA genotype (%)
type type no.
screened
42 43 62 44 + - + - + - Found Predicted"
+ - - - 22 1 28 1 52 50 2 96 100
- + + + 1 14 0 33 38 1 37 3 0
+ + - - 41 5 20 9 75 61 14 81
- - + + 13 16 10 27 66 23 43 35
+ + + - 0 5 1 6 12 1 11 8 0
- - - + 16 1 8 1 26 24 2 92 100
aE.coli CR63 was grown at
370C
in GCA medium to about5x 108 cells per ml. One milliliter was infectedat37°C with0.1mlof a mixture of SP62 and the quadruple am mutant (genes 42, 43, 62, and 44) at an MOI of
5each.At60min, the culture was sampled into chloroform-saturated broth, diluted, and then plated on E. coli CR63at30°C,conditions permissive for all progeny. Plaques of these progeny were stabbed at random andspot-tested against helper phage to identify all amber genotypes, and then some of these were tested for the presence of the SP62 mutation (see the text).
b Predicted on the assumptions (i) that SP62 lies between genes 43 and 62, and (ii) that all recombinants scoredweretheresult of single crossover events.
VOL. 22, 1977
on November 10, 2019 by guest
http://jvi.asm.org/
[image:2.501.54.456.486.592.2]Single-plaque identification ofregA+
recombi-nants. Forsingle-plaqueidentification of regA+ re-combinants (Table 3), progeny of the various
pair-wisecrossesof regA mutantswereplacedin 3 mlof
topagar(8)with about3x 108cells ofE.coliBorE.
coli K-12-W3110, plushydroxyurea (11, 18),toa
con-centrationof62mM in the topagar;thiswaspoured
onto20 mlof bottomagar, and theplatewas incu-batedat42.8°C for about18 h.Both thehigh temper-ature(18) and thehydroxyurea (11, 18)contributed
to the result that the regA mutantsproduced
vir-tually no plaques, whereas T4+ produced small plaquesatanefficiency of plating (EOP)of around
40%;this EOPwasmeasured with eachsetplated, and a correction factor was applied. To maintain temperature towithin0.1°C,wemodifiedastandard Thelco model4 incubatorsothatnow (i)afanwas
mounted atthe bottom to circulate the air down-ward throughahole(11-cmdiameter)cut outofthe metal plate mountedovertheheating coils, upward through a 6-mm gap at each side formedby false walls of cardboard that extendedtowithin 15cmof the topof the chamber, and then down the central chamber where the petri disheswereplaced; (ii)the metal plate just above the heating coils was insu-lated with cardboard tominimize radiant heating; (iii) the thermostatwasreplaced with an Athena, solid state proportional temperature controller, model52-3(which featuresaslower heatingratethe
closertothe specified temperature),whose
thermis-tor sensoris mounted in the middle of the center
shelf; (iv) the convection ports at top and bottom
were plugged to minimize heat loss. For faster warm-up,petri disheswereneverstackedmorethan
two high. To minimize evaporation (which gave
very unevenplaqueappearance),close-fitting,
plas-tic petri dish covers were used rather than the common, ridgedtype that sit about0.5 mmabove the bottom dish.
RESULTS
SP62 defines a new gene. Since SP62 maps
between markers in genes 43 and 62 (see be-low), itseemed possible that SP62mapswithin
one of these genes. Totestthis,
complementa-tion tests were performed. Wiberg et al. (18)
showedthatphage production by SP62 is rela-tivelymoresensitivethan T4+tohigh
tempera-ture. Table 1 presents measurementsofphage
production at 44.8°C in mixed infections of E.
coliB bySP62 and ambermutantsingenes43
and 62,aswellasinthenearbygenes44and 45.
At this temperature, SP62 made only about
20%as many phageasdid T4+at50 minafter
infection; no conditions are known that
com-pletely suppress SP62 phage production
with-out drastically decreasing that of T4+. In the
controlcrosses of T4+ with each of theam
mu-tants,essentially full complementation(91 and
102%) was seen with the 44- and 43-am
mu-tants, respectively; T4+ complemented the
45-and 62-am mutants somewhat less efficiently
(76 and 60%,respectively).Asafurthercontrol,
it is shown that each of the am mutants
pro-duced fewer than 0.44phagepercell.
[image:3.501.61.452.406.597.2]The essentialdataofTable1arethoseinthe
TABLE 3. Recombination betweenregA mutantsa
Mutantatindicatedmap site
regA1, regA2,
Map site regA15 regA14 regA9 regA5 regA11 regA 12, regA4, regA8 regA13, regA6,
regA16 regA7
A B C D E F G H
A 0.88 (2) 1.07 (2) 0.91(2) 0.93(2) 0.13(2) 0.70(4) 0.59(2)
(0.56-1.2) (0.83-1.3) (0.90-0.91) (0.87-0.98) (0.10-0.16) (0.40-0.96) (0.56-0.61) B 0.50 (2) 0.26 (2) 0.21 (2) 0.074(2) 0.33(3) 0.24(2)
(0.24-0.76) (0.20-0.31) (0.19-0.23) (0.033-0.115) (0.17-0.47) (0.13-0.35)
C 0.20 (2) 0.23 (2) 0.11(2) 0.28(11) 0.47(2)
(0.19-0.21) (0.14-0.31) (0.09-0.12) (0.17-0.47) (0.34-0.60)
D 0.29(2) 0.17(2) 0.17(2) 0.24(2)
(0.27-0.30) (0.16-0.18) (0.15-0.18) (0.23-0.24)
E 0.049(4) 0.37(2) 0.32(2)
(0.033-0.088) (0.36-0.38) (0.12-0.52)
F 0.12(4) 0.31(2)
(0.08-0.20) (0.28-0.33)
G 0.13(2)
(0.12-0.14)
aE.coli W3110 at 5 x 108 cells per ml in GCA medium was infected at 37°C with amixtureof two phage at an MOI of 5 each.Progeny phagewere sampled 60minlater. The value (not in parentheses) at theintersectionof a vertical column and ahorizontalrow represents the average percent recombination (2 x percentage of regA+recombinants) among progeny of a cross ofthe mutants definingthat column and row. The value in parentheses at its right gives the number of crosses performed. The values in parentheses below that indicate the range of values ofpercent recombination. Where several mutants atthe same site were used in various crosses, the particular mutant used isnotindicated to simplify presentation ofdata.Background T4+ revertants in the single mutant stocks corresponded to apercentrecombination value of 0.05 for regA8, and less than 0.005 for all the other mutants; these values were subtracted.Otherdetails are given in the text.
on November 10, 2019 by guest
http://jvi.asm.org/
last column, where the degree of complementa-tion of each of the am mutants by SP62 is shownrelative tothat given by T4+. It is seen
that SP62 complements all to about the same
extent-66 to 80%-as wellas does T4+. Thus,
SP62 complemented am mutants in genes 43 and 62, itsimmediate neighbors, as well as it complemented thoseingenes44and 45, which
are more distant neighbors. Comparable
re-sults were observed with another regA mutant, regA 15 (datanot shown). This argues that SP62
was inneithergene 43 nor 62, and that,
there-fore, SP62 defineda newgene;wehave already named this generegA (18). Ambermutants in
genes 43 and 62 made essentially no phage
DNA; gene 43 is the structuralgene for DNA
polymerase,whereasthe exact function of gene
62is unknown (12).
Intergenic mapping of SP62. To locate the approximate genetic map position ofSP62, it was crossed at equal multiplicity with amber
mutants in various T4 genes, starting from
gene 49 and proceeding counterclockwise to
gene 60. The progeny were plated on E. coli CR63 at 30'C for total phage, and wild-type recombinants were scored on plates at 40'C con-taining E.coliBplus5-fluorodeoxyuridine and uracil (at 33 and 100
AM,
respectively, inthetop agar); these conditions were essentially
those of spot-plateprocedure 1ofWibergetal. (18). Under these conditions, amber mutants
didnot grow, SP62 gave tiny plaques atbest,
andwild-type recombinantsgavelarger plaques.
The results (not shown) revealed minimum
recombination frequencies in the vicinity of genes 43, 62, and 44.
To map SP62moreprecisely,atechnique was used thatisanalogoustothe secondprocedure used by Warner et al. (16) to map the dexA
gene;thisapproachavoids the scoring of
recom-binantsresultingfromdoublecrossoverevents,
thus avoiding the ambiguity often caused by
high negative interference (1-3). SP62 was
crossed atequal multiplicity withaquadruple
ambermutantdefectiveingenes42, 43,62,and 44;these four genesarelistedinclockwise order
on the T4 genetic map (13). First the progeny
were scored for thesixclasses of amber
recom-binants that can result from single crossover
events. These recombinants were then tested
for the presence orabsence of the SP62
muta-tion. The results of two such experiments are
shown in Table 2. They support a prediction that the SP62 mutation is between genes 43
and 62; 81% ofthe + +- - recombinants were
regA+, whereas 35% of the reciprocal recom-binants, - -++, were regA+. These two
val-ues should total 100%; that they total 116% is
doubtless due to statistical variation. If the values are normalized to a totalof 100%, then
70 and 30% of the + + - - and -- + + recom-binants, respectively, are regA+. This argues that the SP62 mutation is located about 70% of
the distance from am B22x5 in gene 43 to
amEl140 in gene 62 (Fig. 1). The two regA re-combinants found in the +--- and ---+ progeny, and the one regA+recombinant found
in the - + + + and + + + - progeny, were
prob-ably due to a low frequency of multiple cross-overs or insertions (2), and do not affect the
conclusion. Thetwo-factorcrossesof Fig. 1also
placeSP62about70% of the way fromam B22x5 to amE1140, thus providing independent con-firmation of the more rigorous conclusion from the experiments of Table 2. Figure 1 also shows therelative map positions of the four mutations used in thecrosses of Table 2.
Isolation of new regA mutants. We isolated
a number of new regA mutants by the tech-nique of Karam and Bowles (11). This tech-niquedepends on two observations: (i) amE1140
inT4gene 62isleakyon anam-restrictive host
because of high ribosomalambiguity, but isnot
quite ableto make a plaque; (ii) addition ofa
regA mutation suppresses amE1140, permit-ting formation of a tiny plaque, presumably because the low level of gene62protein is now increased. Six independent stocks (A through F)of amE1140 wereprepared, one from each of
six separate plaques on E. coli CR63. These unmutagenized stocks were plated at various
gene42 amN55x5
.. 9.7
gene43 gene regA gene62 gene44 amB22x5 amEll4Ox3 amN82x3
14.3 4.8
SP62
9.0 3.8
FIG. 1. Relative map positions of SP62 and the amber mutants used in the multifactor crosses of Table2.Percent recombination (=2xpercent T4+)
is indicated between markers and represents the
av-erageoftwoindependent experiments;variationfrom
the average was less than 13%ofthe values given. The arrow indicates the position ofSP62 derived from the data of Table 2, i.e. 70%ofthe distancefrom amB22x5toamE114Ox3. Toone mlofE. coli CR63 (5 x108 cells per ml) inGCA at37°C was added 0.2
mlofphageinGCA(MOI of eachphage,5); progeny phage weresampled60 minlater. Totalphage pro-duction (150to200phage percell) was determined
onplatesat30°Cseeded with E. coli CR63. For the
am x am crosses, T4+recombinants werescoredas thosephage givingplaquesonE. coliBat30°C.For the am x SP62 crosses, T4+ recombinants were
scored as thoseplating at 43°C on E. coli B plus
hydroxyurea (see text).
I..
VOL. 22, 1977
on November 10, 2019 by guest
http://jvi.asm.org/
[image:4.501.255.447.407.473.2]746 WIBERG ET AL.
dilutionson E. coli B at 30and40'C, andtiny
plaqueswere selected. These plaques were
di-luted and replated, and single plaques were
stabbedtomake stocksonCR63at30'C. Those thatfailedtogrow onbacterial stockno. 150,an
am-restrictive, streptomycin-resistant (strr)
strain, were retained for further testing; the
strrstraineliminates the ribosomal ambiguity
on which suppression by the regA mutation
depends (11). Of these, mostfailed to
comple-mentSP62 (regA 1) in liquid culture on E. coli
B at44.50Candwerejudgedto containan regA
mutation. The gene 62 amber mutation was
then bredout ofall by crossing with T4+. The resulting regA single mutants were given the
followingdesignations, where the letter in
pa-renthesesisthe amE1140 stockinwhicha
mu-tant arose and the temperature is that of the
plate on which itwas found: regA2 (A, 300C);
regA 6,regA7 (A,400C);regA 3(B, 300C);regA 8,
regA9 (B,
400C);
regA10 (C,400C);
regA11 (D,40C);regA4 (E, 30C); regA14(E, 40C); regA 5
(F, 300C) regA 12, regA13, regA15 (F, 400C).
Comparisonof thisdata with themappingdata
of Table 3 shows thatregA 2, regA 6, andregA7
are at site G, and all arose from the same
plaque isolate ofamE1140; also, regA12 and
regA13 mapped atsite F and arose from a
dif-ferentplaqueisolate. The rest of the mutations
aroseindependentlyof these and of each other.
Although mutants that arose in a common
plaque isolatewereprobably identical,wegave
themseparatenamesbecausethey maynot be
identical; one may be an ambermutation,
an-othermay be an ochre at the same codon; or,
they could be mutant at different, but very close, sites indistinguishable in our
experi-ments.
One other mutant, regA 16, was detected
solelyonthe basis of its white halo (see below
andFig. 2). It was apparentlyarare
contami-nant in the same T4+ stock from which SP62
(regA 1) arose. This and the fact thatregA1 and
regA16 map atthesame site (Table3)suggests
thattheyareidentical.
Intragenic mapping of
regA
mutants.ManyF 1
- t
8 3O
}1
*
v
v* *
2w
J-15 0 s <
T4+
reqAI
FIG. 2. Whitehalo phenotype of regAmutants.ForallbutF, phage were dilutedto109lmlindilution broth (0.9%nutrientbroth[Difco]and 0.5%NaCl)andprinted, with the floating-loop printer, onto plates seeded with0.15mlofE. coli Batabout2 x109cells per mlinGCA medium by using modified (9) GCA bottom and top agar. Theplates were then placed at the temperatures indicated for the periods indicated. All but Dwere
photographed under dark-field illumination, on a Bactronic colony counter, model C110, New Brunswick
ScientificCo.;D wasphotographed with direct lighting from behind,on alight boxdesigned for viewing
X-ray films. (A) T4+ and all the regA mutants, incubated at 42.8°C for 21 h; regA3 contains a 'junk"
temperature-sensitive (ts) mutation in an unknown gene, whereas this ts mutation has been removed in
regA3xl. The names of the mutants have been abbreviated fromregAl,etc.to1,etc.,andT4+ istermed+.In B through E, the phageare inthe sameorderasshown here. (B) Same as A, except incubatedat 43.9°C for21h.
(C)SameasA, except incubatedat38°C for20 h. (D)Same petri dishas inA, but lightis directlybehind dish. (E) SameasA, except incubatedat44.6°C for21 h.(F)Single plaques ofT4+andregAl incubatedat
43.5'Cfor21 h.
J.
-IM.
-V
vp3, fv
....i
b-lll.
0
Iw--:
n.1,
0
4-:" ASK
-.,ftW.;&,%Ae.
46
0
*
O-W.,
0
0
.4lb
0 10
0
a.
9
40-41,
0
0
0
0
0-
0
w 40
on November 10, 2019 by guest
http://jvi.asm.org/
[image:5.501.64.453.295.542.2]regA MUTANTS 747
pairwisecrosses oftheregAmutantswere
per-formed in an effort to construct an intragenic
map. The results are shown in Table 3. The
most important conclusion from the table is that themutantsfellintoeight distinct sites,A
throughH. SitesFand Gweredefined by four and five mutants, respectively. Initially these
mutants at sites Fand Gwereassignedtotheir
respective sites because each failed to
recom-bine withoneother(tester) mutant inthesite.
However it seemed possible that ifthe tester
mutation were a small deletion, other (point)
mutants atthatsite might define twoor more
close but different sites within the deletion
re-gion. Thus, within eachsite, all possible
pair-wise crosses were performed; no recombinants
were detected, which argues that themutants atsitesFand Grepresenteitherall point
muta-tions or (less likely) all deletions.regA mutant
R9 of Karam and Bowles (11) is included in
Table3andwasseen to map at siteG.Mutants
regA3 and regA10 were notincludedinTable3
becausetheyweretooleakyto map.
We tried to assign an unambiguous, linear
mapordertothesitesbutwereunabletodoso; the data of Table 3 were simply not precise
enough, and we did not find conditions that eliminate thevariation. The primaryfactor in
these large variations in recombination fre-quencies for replicate crosses is undoubtedly the fact thatnocompletely selective,restrictive
conditions for regA mutants are known. For
example, theplatingconditions usedtorepress
growth of regA mutants decreased the effi-ciency ofplatingof T4+toabout 40%; this was
measuredineachexperimentand corrected for
(see above). Other factors that maycontribute
to the variation are the following. (i) Perhaps
the "wild" recombinants differ in their EOP, depending on the cross, because of
unrecog-nized "junk" mutations; we did, in fact, find
thatregA3, as originally isolated, containeda
temperature-sensitive mutation in someother
gene (Fig. 2B). (ii) These same plating
condi-tions aredifficulttoreproduce precisely, partly
because thehydroxyurea sensitivity and
tem-perature sensitivity among the regA mutants
differ slightly; sometimes duplicate plates in
thesameincubatordiffer somewhatin
appear-ance. (iii)It ispossible thatinthissmall gene
there aresignificant recombinationalhot spots
orsite-specific effects that disturbalinear rela-tionship betweenphysical distanceand
recom-bination frequency.
We did not succeed in orienting the most
separated regA mutants with respect to genes 43 and 62, primarily because the regA gene is
apparentlysosmall. We find that the distance
between the closest known gene 43 amber
(amE4306) and the three known gene 62
am-bers (which map within 0.3% recombination frequency [RF] of each other) is about 11%
RF, and the regA mutants (about 1% apart)
map roughly in the middle of this gap. Thus,
any differences among the regA mutants in
their RFs with the gene 62 or43 ambersare not
convincing,especially considering the problems
of reproducibility in scoring regA+
recombi-nants among many regA mutants (just
dis-cussed).
White halo phenotype of regA mutants. We
observed that regA1 produced a white halo around the centralzoneof clearinginbacterial lawns, both on spot plates and as single plaques, whereas T4+ didnot, under the same
plating conditions (Fig. 2). Allof the regA
mu-tantsexhibited the white haloto somedegree,
and high temperature appeared to be
neces-sary:theywere seen at 43.9and 42.8°C (Fig. 2A
and B), but not at 38 (Fig. 20) or 30°C (not
shown), even ifthe plates were incubated for several days. The white haloappearedto
repre-sentlargerbacterial microcolonies thaninthe
phage-free areas of the bacterial lawn, as
judged byexaminationwithamicroscope. Also supporting this view is the observation that,
withlight directly behind the plaque(Fig. 2D), ascontrasted with dark-field illumination (Fig. 2A, B, C, E,F),the white halo appearsdarker than the adjoining bacterial lawn.
Itis notclear why the white halooccurs,but
ithasprovided useful confirmationof the usual
spot-test identifications ofthe regA genotype.
Occasionally, the white halois not seenaround
regA zones athigh temperature; this sporadic
ficklenessmaybecaused byvariations in ageof inoculating bacteria, dryness ofagar, duration ofincubation of the seededplate,etc.Forall the halos shown (Fig.2),theapplied phagewere in abroth medium; substitution of GCA medium had no effect on development of the halo. We
were tempted to view the white halo as
diag-nosticforregA mutants and weresupportedin
thisview by the discovery of regA16 solely on
thebasis ofitswhitehalo(seeabove).However,
we recently found that ifthe temperature is
raised to 44.6°C, where the regA mutants are
essentially dead, T4+ now produces a white
halo (Fig. 2E). The significanceof this halo is
discussed below.
DISCUSSION
The fact that the phenotype of SP62 is very
differentfromthat ofmutantsintheadjoining
genes, 43and 62, already suggestedthatSP62
defines a new gene, but thecomplementation
experiments of Table 1 constitute direct
evi-VOL. 22, 1977
on November 10, 2019 by guest
http://jvi.asm.org/
748 WIBERG ET AL.
dence for thatconclusion. Our location ofSP62 between genes 43 and62 (Table 2) documents
our preliminary mapping report (18). Karam
andBowles located theirregA mutant R9 also between genes 43 and62, by using adifferent approach fromours (11).
Untilnow itseemed possible that SP62was a
special kind of missensemutantand that there-fore theregA phenotype might be the result ofa
partial loss, or subtle modification, of regA function. However, our observation that the
mutants define eight different sites, spread
over a distance ofover 1% recombination
fre-quency, suggests that no subtlety is involved
and that the phenotype is due to total loss of
regA function.
The whitehalo seen on petridishes
contain-ing regA mutants is clearly due to faster
growth ofnearbybacterial cells. One
explana-tion may be the release of some factor from
regA-infectedcellsthat promotes thegrowthof
uninfected cells. We are awareofonlyone
pre-vious report ofa white halo phenotype forT4
mutants. Halletal. (7)showed thatT4 mutants
defective in dihydrofolate reductase and, to a
lesserextent, thymidylate synthetase, produce a bright white halo, but only under special
platingconditions. The key factorintheir
plat-ingconditions is apparently auracil-requiring bacterial host. Hallsuggested thatasignificant
amountof dUMP accumulates incells infected
by thesemutantsand, uponlysis,isreleasedas
deoxyuridine or uracil (6). The genefor dihy-drofolate reductase was originally named wh
(for whitehalo) (7)buthas sincebeenrenamed
frd (10, 14);tdisthegene forthymidylate
syn-thetase (14). Inthiscontext, it isworthnoting
that regA1underproducesthymidylate
synthe-tase at
370C
(18);perhaps highertemperaturescause greater underproduction. For this to be
anexplanation of the whitehalo,wewould also
have to propose that, athightemperature, E.
coliBdevelopsatleastapartialdependenceon
exogenous products (e.g., dUMP) thatmay
ac-cumulate when thymidylate synthetase
activ-ity is low. We have no way of knowing at this
point whetherthe white halo seen withT4+,at yethigher temperatures, iscausedbythe same
mechanism aswith regA mutants.
In preliminary experiments, we recently
found that extracts of E. coli B cells infected
with most ofthe regA mutants lack a protein
havingamolecular weight oflessthan 12,000,
when displayed on sodium dodecyl
sulfate-polyacrylamide gels (Landry, Cardillo, and
Wiberg, unpublisheddata). Thus, regAmaybe
the structural gene for this protein. If so, we
may be able to detect nonsense fragments (if
someofthesemutants areindeednonsense
mu-tants) and order them bysize. This should aid
in constructing a rigorous map of the regA gene. Further,if nonsense-suppressor strains of
E. coliconvert someof thenonsensefragments
toafull sizeprotein, this willprovethatthese
mutants are nonsense mutants.
ACKNOWLEDGMENTS
Our thanks toDwightHall forvaluable discussion about thewhite halo phenotype.
This paper is based on workperformed undercontract with the U.S. Energy Research and Development Adminis-tration at the University of Rochester Biomedical and Envi-ronmental research project and has been assigned report no.UR-3490-1034. This work wasalso supported by Public Health Service grant GM-21999 from the National Institute ofGeneral Medical Sciences.
LITERATURE CITED
1. Berger, H., and A. J.Warren. 1969. Effects of deletion mutations onhigh negativeinterference in T4D bac-teriophage. Genetics 63:1-5.
2. Broker, J. R., and I. R. Lehman. 1971. Branched DNA molecules: intermediates in T4 recombination. J. Mol. Biol. 60:131-149.
3. Chase, M.,and A. H.Doermann. 1958. High negative interference over short segments of the genetic struc-ture ofbacteriophage T4. Genetics 43:332-353. 4. Edgar, R. S., G. H. Denhardt, and R. H. Epstein. 1964.
A comparative genetic study of conditional lethal mutations ofbacteriophage T4D. Genetics 49:635-648.
5. Fraser,D., and E. A. Jerrel. 1953. Amino acid composi-tionof T3bacteriophage.J.Biol. Chem. 205:291-295. 6. Hall, D.H.1967. Mutantsofbacteriophage T4 unable to induce dihydrofolate reductase activity. Proc. Natl. Acad. Sci. U.S.A. 58:584-591.
7. Hall, D. H., I. Tessman, and 0. Karlstrom. 1967. Link-age of genescontrolling a series of steps in pyrimidine biosynthesis.Virology31:442-448.
8. Hercules, K., J. L. Munro, S. Mendelsohn, and J. S.
Wiberg.1971.Mutantsinanonessential gene of bac-teriophage T4 which aredefective in the degradation of Escherichia coli deoxyribonucleic acid. J. Virol. 7:95-105.
9. Hercules, K., and J. S. Wiberg. 1971. Specific suppres-sion of mutationsingenes46and47bydas, a new class of mutations in bacteriophage T4D. J. Virol. 8:603-612.
10. Johnson, J. R., and D. H. Hall. 1973. Isolation and characterization of mutants of bacteriophage T4 re-sistant to folateanalogs. Virology 53:413-426. 11. Karam, J. D., and M. G. Bowles. 1974. Mutation to
overproduction of bacteriophage T4 gene products. J. Virol.13:428-438.
12. Morris, C. F., N. K. Sinha, and B. M. Alberts. 1975. Reconstruction ofbacteriophageT4DNAreplication apparatus from purified components: rolling circle replication following de novo chain initiation on a
single-strandedcircular DNAtemplate. Proc. Natl. Acad. Sci. U.S.A. 72:4800-4804.
13. Mosig, G. 1976. Linkage map of bacteriophage T4, p. 664-676. In G. D. Fasman (ed.), Handbook of bio-chemistry and molecular biology, 3rd ed., Nucleic acids, vol.II.CRCPress,Cleveland,Ohio. 14. Simon, E. H., and I. Tessman. 1963.
Thymidine-requir-J. VIROL.
on November 10, 2019 by guest
http://jvi.asm.org/
regA 749
ing mutants ofphage T4. Proc. Natl. Acad. Sci. U.S.A. 50:526-532.
15. Tessman,I., and D. B. Greenberg. 1972. Ribonucleotide reductase genes ofphage T4: map location of the
thioredoxingenenrdC. Virology49:337-338. 16. Warner,H.R., D. P.Snustad, J.F.Koerner, and J.D.
Childs.1972.Identification and genetic characteriza-tionofmutantsofbacteriophage T4defectiveinthe
abilitytoinduceexonuclease A. J.Virol. 9:399-407. 17. Wiberg, J. S. 1966. Mutants of bacteriophage T4 unable
tocausebreakdown of host DNA. Proc. Natl. Acad.
Sci. U.S.A. 55:614-621.
18. Wiberg,J.S., S. Mendelsohn, V. Warner, K.Hercules,
C. Aldrich, and J. Munro. 1973. SP62,aviable
mu-tantofbacteriophageT4Ddefective inregulationof phageenzymesynthesis. J. Virol. 12:775-792.
VOL. 22, 1977
on November 10, 2019 by guest
http://jvi.asm.org/