vol. 1972 ordering mutant isoleucine- valine genes …procedure for fine-structure mapping thr used,...

JOURNAL OF BACEIWOLOGY, Feb. 1972, p. 730-740Copyright © 1972 American Society for Microbiology

Vol. 109, No. 2Printed in U.SA.

Ordering of Mutant Sites in the Isoleucine-Valine Genes of Escherichia coli by Use of

Merogenotes Derived from F14: a New Procedurefor Fine-Structure Mappingl

NANCY J. MARSH AND D. E. DUGGANDepartment of Microbiology, University of Florida, Gainesville, Florida 32601

Received for publication 4 November 1971

F-merogenotes derived from F" by transductional shortening have pre-viously been found to consist of the sex factor plus one or more of the ilv genes.It is shown here that they carry one or more ilv genes and a variable portion ofthe adjacent proximal ilv gene. This observation was used to develop a method,analogous to deletion mapping, for ordering mutant sites within the ilv genes.This method requires the use of a series of merogenotes each carrying an in-creasingly longer segment of the gene being mapped. A simplified method offine-structure mapping is also described which requires only one or two F'donor strains to map any one gene. This method is based on the large differ-ences observed in recombination frequencies for mutant sites at various dis-tances from the origin of the incomplete merogenote gene. The sequence of 25mutant sites within three of the ilv genes was determined by use of the simpli-fied procedure.

P1 lysates prepared on AB 1206, an F' strainof Escherichia coli K-12, which is haploid forthe region of the chromosome carried on the F"4merogenote (Fig. 1 and 2), were used by Pit-tard (11; Bacteriol. Proc., p. 138, 1963) totransduce the isoleucine-valine (ilv +) mero-genote genes into ilv- recipients. A small per-centage of the Ilv+ transductants could act asdonors in conjugation experiments. Most ofthese could transfer the ilv, metB, and argHgenes, which had all been carried on the orig-inal F" in the P1 donor strain. Pittard alsofound among the ilv+ transductants somestrains that transferred only the ilv genes inmating experiments. Among these latter, sometransferred either one gene, two genes, threegenes, four genes, or five genes (Fig. 3). Thisobservation was used by Ramakrishnan andAdelberg (14) to determine the order of genesin the ilv cluster. They found the order shownin Fig. 4. The sex factor of the F1 (and mero-genotes derived from F "4by transductionalshortening) was distal to ilvE but is probablyvery close (15).Our preliminary studies suggested that the

transductionally shortened merogenotes car-'Florida Agricultural Experiment Station Journal Series

4070.

ried one or more complete ilv genes plus a por-tion of the adjacent proximal ilv gene. Thiswas confirmed by more detailed study as out-lined here. The data obtained in this study ledto the development of two methods for fine-structure mapping of ilv mutations within agene. The order of 25 mutant sites within threeilv genes was determined by use of one ofthese methods.

MATERIS AND MEMEODSMedia. Haploid F" strains and all F- strains

were grown in L broth (10). Z broth was L broth with0.25 x 10'- M CaCl,. Merogenote-containing strainswere grown in a synthetic medium lacking isoleu-cine-valine. The synthetic medium was described byAdelberg and Burns (1). L agar and synthetic agarcontained 2% agar. Z agar contained 1% agar.

Bacterial strains. The bacterial strains used aredescribed in Table 1. The haploid F" strains, AB1206 and AB 1013, were used as the donor strains formaking transductionally shortened merogenotes. AB1013 differs from AB 1206 only in that it carries themutation iluvO113, conferring resistance to 10-2 Mvaline (Val-r), concomitant with derepression of thethree-gene ilv operon A (13).Each shortened merogenote derived from F" by

transduction was originally obtained in rec+ recipi-ents, either strain AB 2070 or AB 1450. The short-ened merogenotes were transferred by conjugation to

730

on April 3, 2020 by guest

http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

PROCEDURE FOR FINE-STRUCTURE MAPPING

thr used, each for different purposes. (i) Lysates used totransfer ilv mutations from the primary strains intoAB 2200 were prepared by infecting early-log-phase

sp/0/90 pro cultures grown in Z broth with Plvir (8) at an MOIarg £ /v80 10 of 0.01. The cultures were harvested after lysis by

r a> X \ exposing them to 0.25 ml of chloroform for up to 5min in ice, and then centrifuging them. The super-natant fluid was used for transduction at an MOI of0.1 to 1. (ii) P1 lysates of haploid F'4 cells were

_*70 20 made by use of the agar-layer method of Swanstromand Adams (16). Mid-log-phase cells grown in Zbroth were mixed with sufficient P1 to produce

strr try about 5 x 105 plaques per plate of Z agar. After 15min of preadsorption, the infected cells were added

\60 / to soft agar, poured onto fresh wet Z agar, and incu-arg GuJ6 3050 bated overnight (9 to 12 hr) at 37 C. The soft-agar

layers from 5 to 10 plates were collected in a 50-mlcentrifuge tube and homogenized with about 0.5 ml

t n y 50 40 ,/ of chloroform by use of a Virtis tissue homogenizerat very low speed. The tissue homogenizer was

rec A his equipped with special blades, obtained from TheFIG. 1. E. coli K-12 chromosome showing the Virtis Co., Inc., Gardner, N.Y., which did not whip

segment of the chromosome carried by the mero- air bubbles into the homogenized soft agar. Aftergenote in the haploid F14 strain AB 1206. incubation at room temperature for 20 min, the

homogenized soft agar was centrifuged at aboutilv pol A 5,000 x g for 20 min. The resulting supernatant

layer (about 15 ml from 10 plates) contained highmet E rho met o titers of P1 phage [10" to 5 x 1010 plaque-formingunits (PFU) per ml from P1 lysates of haploid F"WYAAA ' I I Ii~ .D strains, 2 x 10" to 5 x 1011 PFU/ml from most F-,

74 75 76 77 F+, or Hfr strains]. These lysates could be stored at

FIG. 2. Some of the genes known to be carried on refrigerator temperatures for long periods of timeF"4 without significant loss of titer. After 6 to 12F months, titers may drop about 50%.the recA recipient strain AB 2987 (Table 1). These Cross-streak test for complementation. Approx-F' recA strains were then used as donors in the mat- imately 0.1-ml quantities of the appropriate recip-ings to be described. ient strains [grown in L broth and resuspended inMost of the ilv mutations were obtained from the 56/2 buffer (1)] were distributed in a straight line

collection of E. A. Adelberg, and a few were obtained across a selective synthetic agar plate. The donorfrom K. Kiritani (University of Tokyo). Each ilv cultures were usually grown overnight in presteri-mutation was transduced into a common F- strain, lized Microtiter plates containing 0.1 ml of the syn-AB 2200 (Table 1). These resultant strains were thetic medium per well. The synthetic medium wasused as recipients in the matings to be described. designed to maintain diploid strains in the diploid

Bacteriophage. Two stocks of generalized trans- state by omitting the amino acids isoleucine and va-ducing phage were used. One was a single-plaque line. Sterile toothpicks were dipped into syntheticisolate of Plkc of Lennox (10) labeled P,8, which was medium cultures of the donor strains to be testedobtained from Roy Curtiss III (Oak Ridge National and streaked across the line of recipient culture de-Laboratory). The other was the virulent strain, scribed above. The mating plates were incubated forPlvir, of Ikeda and Tomizawa (8) obtained from 2 days and observed for the presence of coloniesHerbert Boyer. These phages will be referred to as along the streak. Complementation was indicated byP1 or Plvir. heavy confluent growth of large colonies. Recombi-Making transductionally shortened merogen- nation was indicated by a smaller number of isolated

otes. P1 lysates of haploid F" strains (AB 1206 or colonies. Cross-feeding between ilv mutants was rec-AB 1013) were used in these transductions. The pro- ognized as light creamy or smooth growth with nocedures of Lennox (10) were generally followed, and evidence of large colonies at the periphery of growth.a multiplicity of infection (MOI) of about 1 was The three types of growth were easily distinguish-used. The recipient F- strains carried an ilv muta- able.tion, and selection was for the merogenote gene ilv+. Mating procedure for ordering mutant sites.Ilv+ transductants were then tested for their ability Overnight cultures of the recipient strains wereto transfer merogenote markers by cross-streaking transferred into fresh L broth and grown until theappropriately marked F- strains. After characteri- recipient reached a turbidity of 250 Klett units (redzation, the shortened merogenotes were transferred filter; Klett Manufacturing Inc., New York, N.Y.).by conjugation into the recA ilvE strain (AB2987) Males were similarly grown in synthetic medium towith selection for Ilv+. 50 Klett units and mixed immediately in equal vol-

Production of P1 lysates. Two methods were umes with the female. The mating was terminated

VOL. 109, 1972 731


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

MARSH AND DUGGAN

CHN C

+ pyruvate Coo CH3-tCCH -C-COON--- ---- CN3-C-COON -NC3

II - CO23

2 OH NADPH NADP C

pyruvate 1 -aceto- t i, -I lactate isova

COCON

CmH-CN-CN-COOH ElOH NH2

L- t4ronine\ L- THREONINE

IDENSINGNZYME

REDUCTO -ISO MERASE

II

CHsN3CH3-CH C

CH2 CH-1N2 N3-'teucine HCNH2 HN

COOH

CH;CH-CH 4

t-ktoIlo- CH2 CH3-caproate C*O I

COOH

+otot\ //

N3 CH3,-OH - H CH -CH:O-H2-b C a O

:OOH COOH

di hydroxyorate

< - keto -isovaleral

_ t

DEHYDRASE

:H,3C-CH2OHC-OHCOOH

+2N

CH3C -CH2OHC.OCOOH

+ formate

pentelcacid

ketopetotlcccId

CHNCHi-CH

HHCNHI a

glut. (kg COOH

t velinet I

ITRANSAMINASE B

IDEAMINASE CH3 iN]1tCO* O CH3 CH3 CH3

CH -CH2-C-COOH CHN-CH2-C-COOH CHN-CH -C-ON CH -CH -CH CH CI-CH3 2 -Cl 3--2- 3 2 1 3 2 ~..C.2NO°X2 ON NADPH NADP HCOH CO lt. < g CNN2

COOH COOH COOH

A- aceto--- hydroxy-but yrate

(,- dghydroxy -

A -methylvelerete

FIG. 3. Isoleucine-valine biosynthetic pathway.

gim S ilv genes arg H

E D A 0 IC IB PTA DH TD R ICE

TD D.CE D RI DH D TA C

Ilv A ilv B ilv C ilv D ilv E

FIG. 4. Order of ilv genes and enzyme reactionsequence.

by vigorous swirling on a Vortex mixer (ScientificProducts, Evanston, Ill.) after 60 min at 37 C, anddilutions were plated on appropriate selective media.After 2 days of incubation at 37 C, colonies werecounted and recombination frequencies were calcu-lated on the basis of the numbers of participatingdonor cells.Enzyme assays. Dehydrase and threonine deami-

nase were assayed by the methods described byWechsler and Adelberg (18), except that cysteinewas used rather than NH4Cl in the threonine deami-nase assays. Reductoisomerase was assayed as de-scribed by Ramakrishnan and Adelberg (13), exceptthat phosphate buffer (pH 7.5, 0.02 mmole per ml)was used in place of tris(hydroxymethyl)aminometh-ane (Tris) buffer. (Tris buffer is inhibitory in thereductoisomerase assay.) Transaminase B was as-

sayed by a modification (in preparation) of the pro-cedure used by Ramakrishnan and Adelberg (13).The important modification in this method was inthe indirect method of keto acid determination ofFriedmann and Haugen (6). The hydrazone of themonocarboxy keto acid was extracted into toluene asit was being formed.

REJLTCharacterization of shortened merogen-

otes. When P1 lysates prepared on a haploid F14donor were used, about 4% of the Ilv+ trans-ductants were able to transfer, during conjuga-tion, several of the F1' genes (argH, metB, rha,metE, and ilv) at a high frequency. Thesesame transductants were not able to transferthe chromosomal genes of the transductionaldonor or of the transductional recipient strainsat high frequency. Among the Ilv+ transduc-tants were found some F' strains that wouldtransfer only ilv genes. Cross-streak comple-mentation tests showed that some transduc-tants would form large numbers of Ilv+ colo-nies only with the ilvE12 test recipient. (Themutant genes of all test recipients had pre-viously been characterized by enzyme assay.)

732 J. BACTERIOL.

ENZYME ASGNEVIATION gENETICE__N_Z_Y_1___________ Io LOCUS SYTMW

L-THREONINE DEAMINASE TD i/v A

CONDENSING ENZYME CE i/Ba

REDUCTOISOMERASE R I iivC

DEHYDRASE DOH i/D

TRANSAMINASE B TR ilvLJ

O - ketobutyrate - keto -

4- methylvelereteiolelelcie


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


TABLE 1. Escherichia coli K-12 strainsa

Strain Sex ilv met arg pro trp his thy str thi rec Remarks

AB 1450 F- D16 Bi Hi + + 1 + R - +AB 2070 F- E12 E46 + + 3 4 + R - +AB 2930 F- E12 + G12 + + 42 26 S + +AB 2987 F- E12 + G12 + + 42 + S + Al MA 1079x AB 2930 for thy+,

recAAB 2200 F- + E46 + + 3 4 + R - +MA 10796 Hfr + + + + + + + S - Al From B. LowAB 1206 F' 4 + + + 2 + 4 + R - + Haploid for the Fl4regionAB 1013 F'4 0113 + + 2 + 4 + R - + Derived from AB 1206, Val-r

phenotype

a Allele numbers are those assigned by the laboratory where the mutations were isolated.bMA 1079 is also serA and injects in the order +-thy-rec-his-trp-ilv.

TABLE 1A. F' strains constructed by conjugaltransfer into AB 2987a

Strain F-merogenote' Source

AB 2988 F310 (F ilvEDAC1) This laboratoryAB 2989 F'6 (F ilvEDA C2) AB 1528 (Pittard)AB 2990 F216 (F ilvEDAl) AB 1526AB 2991 F31 (F ilvEDA2) This laboratoryAB 2992 F312(F ilvEDA3) This laboratoryAB 2993 F313 (F ilvEDA4) This laboratoryAB 2994 F314 (F ilvEDA5) This laboratoryAB 2995 F2' (F ilvEDI) AB 1545 (Pittard)AB 2996 F311(F ilvEl) This laboratoryAB 2997 F316 (F ilvE2) This laboratory

a These strains differ only in the F-merogenotecarried. All are ilv+/ilvE12. Allele numbers are thoseassigned by the laboratory where the mutations wereisolated.

b The complete ilv genes transferred by these mer-ogenotes are given in parentheses.

Others would form large numbers of Ilv+ colo-nies only with the ilvE recipient and an ilvD16test recipient. Others transferred the ilvD andilvE genes and formed many colonies with anilvA201 test recipient but not with the ilvCstrain. Still others transferred the completeilvE, ilvD, ilvA, and ilvC genes. None has beenfound to transfer the proximal gene, argH, ofthe parental F14. It seems likely to us thatthese merogenotes may differ in length. Theycarry sufficient sex factor deoxyribonucleicacid (DNA) to replicate, to transfer, and toabsorb male-specific phage.To facilitate presentation and comprehen-

sion of concepts, we will refer to these mero-genotes by the complete ilv genes which theytransfer, F ilvE, F ilvED, F ilvEDA, and FilvEDAC, rather than by their assigned mero-genote numbers.

Identification of nonfunctional genes inthe ilv- recipients. The above methodsroughly characterized the transductionally

shortened merogenotes with respect to thenumber of complete ilv genes which they car-ried. By use of four selected merogenotes, oneof each type (F ilvE, F ilvED, F ilvEDA, and FilvEDAC), it is possible to characterize each ofthe unknown ilv mutants by cross-streak com-plementation tests. For example an ilvE mu-tant strain would be complemented and be-come Ilv+ upon receiving any of the above fourmerogenotes. An ilvD would be complementedby the presence of any of the merogenotes ex-cept F ilvE. An ilvC mutant would not becomeIlv+ after mating with donors carrying F ilvEor F ilvED, but would be complemented afterreceiving either of the other two merogenotes.In this manner, each of 25 ilv mutants wasclassified as being altered in the ilvE, ilvD,ilvA, or ilvC genes. No ilvB mutants wereused, as none suitable for these studies wasknown to us.Ordering of mutant sites. In cross-streak

complementation tests, the merogenote-recip-ient pairs formed large numbers of Ilv+ diploidcolonies if the merogenote complemented theilv lesion in the recipient, but no colonies ifthe merogenote did not carry the gene corre-sponding to the mutated gene in the recipient.Results were obtained with some transduc-tionally shortened merogenotes and some ilv -recipients which did not fit either of the abovepatterns. In these cross-streak tests, only a fewilv + colonies appeared along the cross-streakafter the donor had crossed the recipient. Itseemed likely that these results could be ex-plained if the merogenote carried less than thecomplete gene corresponding to the mutantgene in the recipient. Under these conditions,complementation could not occur, but recom-bination could occur between the endogenotegene and the merogenote gene fragment, if theexogenote carried the wild-type codon corre-sponding to the mutant codon in the endo-

VOL. 109, 1972 733


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

MARSH AND DUGGAN

ilv allele from AB 1514(Adelberg)

ilv amber antipolar muta-tion from this labora-tory

ilv allele from Kiritaniilv allele from AB 1285

(Adelberg)ilv allele from AB 1424

(Adelberg)ilIv allele from AB 1286











(Adelberg)ilv allele from Kiritaniilv allele from AB 1406




(Adelberg)ilv amber mutation from

this laboratoryilv amber mutation from

this laboratoryilv allele from Kiritaniilv amber mutant from

this laboratoryilv allele from AB 1419


(Adelberg)ilv deletion from Kiritani

aThese strains differ only in their ilv allele. Allelemembers are those assigned by the laboratory wherethe mutations were isolated.

genote. The recombination frequency would below enough to account for the numbers of colo-nies observed in the cross-streak test.

Such observations and explanations led us todevelop a method of ordering mutant siteswithin a gene. The initial concept was to con-struct, by transductional shortening, a series ofmerogenotes each carrying a different length ofthe gene to be studied. In mating these mero-genote strains with a series of ilv mutants ofthat gene, one would obtain ilv+ recombinants,depending on whether the merogenote carriedthe wild-type sequence of bases correspondingto the mutated bases in the endogenote. A dia-grammatic representation of the concept isshown in Fig. 5. Merogenote C could form re-combinants with all mutant sites within thegene except site 4. Merogenote B could formrecombinants with mutants 1 and 2, but not 3or 4. Merogenote A could form recombinantsonly with mutant site 1.

If several merogenotes, each carrying an in-creasingly longer segment of the gene, werecrossed with several recipients, each carryingan independent mutation of that gene, oneshould be able to order the mutant sites on thebasis of the presence or absence of recombi-nants arising from the various matings.Such matings were carried out with five

donor strains, each carrying a merogenote con-sisting of the sex factor, the UivE, ilvD, and ilvAgenes, plus variable segments of the ilvC gene.These merogenote strains were crossed withseven recipients, each carrying a differentmutation in the ilvC gene. The results of thesematings are shown in Table 2. The mutantsites are arranged in the order deduced fromthe data, and the merogenotes are listed inorder of size.Variable ilv gene segments on shortened

merogenotes. The merogenote F ilvEDAlformed recombinants with all seven of the ilvCmutations. The merogenote F ilvEDA2 formedrecombinants with five of the mutants. Whencrossed with the mutants ilvC49 and ilvC7, no

vwA~IW merogenote A

AA -- ---I- merogenote B

IAAO morogenote C

1 2 3 4_- ndogenote

mutant site within a gene

FIG. 5. Diagrammatic representation of the prin-ciple behind a method for ordering mutant siteswithin a gene by the use of a series of merogenotes ofdiffering lengths. The recipient is represented as ifthe several mutations existed on a single chromo-some.

TABLE 1B. F- strains constructed by cotransductioninto AB 2200a

Strain ilv allele Source

AB 3515

AB 3516

AB 2950AB 2970

AB 2971

AB 2972

AB 2973

AB 2974

AB 2975

AB 2976

AB 2977

AB 2927

AB 2978

AB 2979

AB 2980

AB 2981

AB 2949AB 2982

AB 2983

AB 2928

AB 2984

AB 2944

AB 2947

AB 2985AB 2945

AB 2986

AB 2929

AB 3590

ilvEI2

iluE316

ilvD109ilvD21

ilvD52

ilvD22

ilvD46

ilvD23

ilvD62

ilvD45

ilvD29

ilvD16

ilvD211

ilvD60

ilvD43

ilvD61

ilvD106ilvD39

ilvA24

ilvA201

ilvA44

ilvC285

ilvC288

ilvClOlilvC286

ilvC49

ilvC7

ilvDAC115

734 J. BACTERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


TABLE 2. Recombination frequencies of shortened merogenotes with recipients carrying a series of ilvCmutationsa

ilv allelesMerogenote

A201 C44 C285 C288 ClO1 C294 C49 C7

F EDAI......... 1.2 x 10 2 8.4 x 10 3 3.4 x 10 3 4.4 x 10 ' 4.8 x 10 4.5 x 10 5 3.6 x 10 6.4 x 10 71.1 x 10 2 1.3 x 10 2 2.0 x 10 3 4.5 x 10 4 1.9 X 10 1.9 x 10 5 4.0 x 106 5.0 x 10 7

Avg .......... 1.2 x 10 2 1.1 x 10 2 2.7 x 10 3 4.5 x 10 4 3.3 x 10 3.2 x 10 5 3.8 x 106 5.7 x 10 7

FEDA2 ........ 1.4 x 10 2 2.6 x 10 2 5.4 x 10 3 2.7 x 10 4 9.5 x 10 2.6 x 10 51.7 x 10 1.4 x 10 4.6 x 103 2.3 x 10 5.1 x 10 7.1 x 10 <1 x 10 -

Avg ......... 1.5 x 102 2.0 x 102 5.0 x 103 2.5 x 10 4 7.3 x 10 4.9 x 105

F EDA4. 1.2 x 102 1.7 x 102 8.4 x 10 7 3.5 x 10 7 2.0 x 1071.1 x 10 1.7 x 10 1.8 x 106 8.2 x 10" 8.7 x 10 < I x 10

Avg .......... 1.2 x 10 2 1.7 x 10 2 1.3 x 10' 2.1 1.0- 7 1.4 x 10 7

F EDA5. 1.3 x 10- 2 2.2 x 10 2 3.4 x 10-"1.3 x 10 2 8.1 x 10 3 6.7 x 10 7 <1 X 10 - _ _

Avg .......... 1.3 x 102 1.0 X 10 2 2.0 x 10 6

F EDA3 .... 1.4 x 10 2 2.5 x 101.2 x 10 8.7 x 10 4< Ix 10 _______ _Avg 1.3 x 10 21.7 x 10 I 3

a Recombination frequencies are given as the number of recombinants arising per donor cell in the mating mixture. Thedashes indicate < 10- s recombinants.

colonies were detected at the lowest dilutionfrom that mating mixture, and hence the re-sult is listed as a recombination frequency ofless than 108. Similarly, F ilvEDA4 formedilv + recombinants with ilvClOl but not ilv-C294, F ilvEDA5 recombined with ilvC285 butnot ilvC288, and F ilvEDA3 could recombinewith ilv C44 but not ilv C285. Complementation/recombination frequencies with ilvA201 arediscussed below.We have interpreted the data for the F

ilvEDA2 matings to indicate that the ilvCgene segment on this merogenote terminatesbetween the site of ilvC294 and that of ilvC49.Similarly, the merogenote F ilvEDA4 carriesan ilvC gene segment that terminates betweenthe site of the ilvClOl mutation and that ofthe ilvC294 mutation. The F ilvEDA4 termi-nates between ilvC285 and ilvC288, and FilvEDA3 terminates between ilvC285 andilvC44.Ordering mutant sites within the ilvC

gene. The confirmation of the proposedmethod of ordering mutant sites is evident.The mutant site ilvC7 is proximal (with re-spect to the origin of the merogenote) to theother six because it forms recombinants onlywith the longest merogenote. Similarly, ilvC49is proximal to ilvC249 since ilvC49 forms re-combinants with the longest merogenote butnot with F ilvEDA2. Similar reasoning wasused to establish the order of each of the re-maining ilvC mutants as shown in Table 2.

Transfer frequencies of these five merogen-otes into a strain carrying a mutation in the

ilvA gene, which is distal to the ilvD gene, areincluded to show Ilv+ colony frequencies whena merogenote gene complements the recipientmutation. We have no data yet to indicatewhat fraction of these Ilv+ colonies are recom-binants and what fraction are diploids. Thiswill be determined by comparing frequenciesof Ilv+ colonies obtained by mating with rec+and rec recipients and by testing the Ilv+ colo-nies among the rec+ recipients for ability toact as donors of the merogenote ilv gene. Thelatter will be complicated because the Ilv+colony may well contain F-, F', and Hfr types.The possibility that the F ilvEDA1 carries

some of the ilvB gene could not be tested di-rectly because of the dearth of mutants in thatgene. If it did carry any ilvB DNA, it wouldcarry all of the ilvC gene and would have com-plemented all mutants in the ilvC gene. Thiswould have resulted in Ilv+ colony frequency(diploids and recombinants) of 10-2 per partic-ipating donor, rather than the range observed.The ilvC44 mutant gave recombination fre-

quencies similar to complementation/recom-bination frequencies of the ilvA201 mutant. Theobvious question then is whether this mutationis in the ilvC gene or in the ilvA gene. Twopieces of evidence suggest that it is not in theA gene: (i) the shortest merogenote, F ilvEDA3,yielded 10-fold differences in ilv+ colony fre-quencies when crossed with recipients carryingilvA201 versus ilvC44 (Table 2); and (ii) strainscarrying ilvC44 do not grow in the presence ofisoleucine alone as do other known ilvA mu-tants. The recombination frequencies obtained

VOL. 109, 1972 735


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

MARSH AND DUGGAN

with the two merogenotes of differiwith ilvC44 have significance in themapping procedure discussed below.

Simplified method for orderingsites. When the results obtained wilvEDAl were examined closely, itserved that the recombination frequtained with the various ilvC mutants(to a surprisingly large degree) withtance from the origin of the merogesame was true of the recombinatiorcies of the ilvC mutants with the otgenotes.A simplified explanation for this

of recombination is shown in Fig. 6.immediately obvious, however, wshould be such large differences in rftion frequencies across a gene. A douover would be necessary to generate,bination, wild-type progeny from an i

gene segment and a mutant endogerOne crossover would be to the leftother to the right of the mutant siticase of an F merogenote, or any i

capable of circularizing, a singlecrossover to the right of the mutant sgenerate a wild-type gene, in this casan Hfr strain. The following explardiffering crossover frequencies wouldfor this special case.) If crossover freconstant per unit of DNA length, tilimiting conditions the frequency of (in recombinational events would bethe extent of available homologous DIparticipating strands on both sides otant site. In the limiting condition,Fig. 6, there is considerable homologto the left of mutant site 3, but vhomologous merogenote gene DNA toof site 3, that is, between mutant sthe proximal end of the exogencPairing and crossover become more

~~ ~~~

2 3

mutational sites within gene

FIG. 6. Diagrammatic representationplanation for the gradation of recombinquencies observed when a single m

carrying a segment of a gene is crossed wents which are mutant at different sites u

gene. The recipient is represented as if t

mutations existed on a single chromosometical dashed lines represent some of the pocrossovers.

ing length the mutant site is more distal to the origin ofsimplified the merogenote. The vertical dotted lines rep-

resent some of the possible first crossover pos-g mutant sibilities.rith the F It would seem possible, then, to order mu-t was ob- tant sites within a gene by use of only one orencies ob- two merogenotes. A merogenote carrying al-increased most all of the gene being mapped would betheir dis- used to determine the sequence of the majority,note. The of the mutant sites within the gene, but ini frequen- order to resolve mutants at the distal end of;her mero- the gene a shorter merogenote would be neces-

sary, as indicated above for the ilvC44 muta-gradation tion. The use of only one or two merogenotesIt is not would simplify the fine-structure mapping

rhy there procedure by reducing the number of matingsscombina- required. This simplified procedure was usedble cross- to order the mutant sites within the ilvD andby recom- ilvA genes.exogenote Ordering mutant sites in the ilvD gene.iote gene. Six selected mutants in the ilvD gene weret and the crossed with two independently isolated F ilvE;e. (In the merogenotes. The ilvD mutants were orderedexogenote according to their increasing recombinationreciprocal frequencies with the two merogenotes, as3ite would shown in Table 3. Increasing recombination3e making frequencies were taken as an indication of dis-iation for tance of the mutant site from the origin of thestill hold merogenote. No conflicts are evident in or-,quency is dering these mutants as determined with theien under two merogenotes used. Complementation fre-crossovers quencies are also shown for the reference mu-related to tation iIvE12 and the ilvE316 which was lo-NA in the cated in this gene by Wechsler and AdelbergIfthe mu- (18).s used in The methods that have been used to isolaterous DNA ilv mutants seemed to have favored mutantsrery little in the ilvD gene, since 16 of the 28 mutantsthe right were deficient in ilvD function. Recombination

ite 3 and frequencies of these 16 ilvD mutants with the)te gene. two F ilvE merogenotes are listed in Table 4.likely if All of these recombination frequencies, with

two exceptions, lead to the same gene order.merogenote The two conflicts are ilvD52 and ilvD46. These

are arbitrarily ordered as determined by theresults obtained with F ilvE2. Recombination

endogenote frequencies of these mutants with the F ilvElmerogenote suggest that each of the mutantsites should be exchanged with one of themutant sites listed below it in the column.

Of an ex- Since the mutant sites in this area of the genezation fre- seem to be very close, as indicated by veryerogenote similar recombination frequencies, it may beuithin that that these exceptions indicate that many repli-vhe several cates of the matings may be necessary to re-The ver- solve the order of such close mutant sites. We

ssible first are currently developing more sensitivemethods allowing a finer resolution. Until this

736 J. BACTERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


TABLE 3. Recombination frequencies of shortened merogenotes with recipients carrying selected ilvDmutantsa

ilv allelesMerogenote

E12 E316 D109 D62 D16 D60 D106 A201

F E2 ......... 1.6 x 10-2 2.3 x 10-2 1.1 x 10-4 4.1 x 10-2 6.9 x 10-' 1.6 x 10-6 1.7 x 10-' <1 x 10-11.9 X 1o- 2 2.0 x 10- 2 1.6 x 10- 4 6.0 x 10' 9.0 x 10- 7 7.4 x 10- 8.0 x 10-8 <1 x 10-8

Avg.1.8 x 10- 2 2.2 x 10- 2 1.4 x 10-' 5.1 x 10-5 3.9 x 10-6 1.2 x 10-6 5.0 x 10-8 <1 x 108

F El.2.0 x 10-2 3.5 x 10-2 4.3 x 10-4 3.1 x 10-2 3.1 x 10-8 1.9 x 10-6 8.0 x 10-8 <1 x 10-81.7 x 10-2 3.1 x 10-2 3.7 x 10-' 3.0 10-5 6.9 x 10' 3.9 x 10-1 8.0 x 108 <1 x 10-8

Avg ......... 1.9 x 10-2 3.3 x 10-2 4.0 x 10-4 3.0 x 10-5 5.0 x 10' 2.9 x 10-6 8.0 x 10-8 <1 x 108

a Recombination frequencies are given as the number of recombinants per donor cell in the mating mixture.

TABLE 4. Recombination frequencies of a series ofilvD mutants with two shortened merogenotesa

Merogenoteslv mutant

F ilvE2 F ilvEl

E12 ............ 1.8 x lo-2 1.9 x 10-2E316 ............ 2.2 x 10-2 3.3 x lo-2D109 ............ 1.4 x 10- 4 4.0 x 10- 4

D21 ............ 9.6 x 10-2 6.2 x 10-5D52 ............ 8.7 x 10-s 2.1 x 10-5D22 ............ 7.3 x 10-5 4.5 x 10-5D46 ............ 7.0 x 10-5 1.9 x lo'D23 ............ 6.7 x 10-5 4.4 x 1o- 5

D62 ............ 5.1 x 10-5 3.0 x 10-5D45 ............ 3.1 x 10-5 2.8 x 10-5D29 ............ 2.6 x 10-5 9.7 x 10-6D39 ............ 1.2 x 10-5 7.4 x 10-'D16 ............ 3.9 x 1o-6 5.0 x lo-6D211 ............ 3.3 x 10-6 2.5 x 10-6D60 ............ 1.2 x 10-6 2.9 x 10-6D43 ............ 7.0 x 10-f 3.3 x 10-8D61 ............ 7.0 x 10-f 5.0 x 10-8D106 ............ 5.0 x 10-8 8.0 x 10-8

a Recombination frequency is given as the numberof recombinants arising per donor cell in the matingmixture.

is done, we have shown many of these muta-tions as being in a limited region of the ilvDgene (Fig. 7).Ordering mutant sites within the ilvA

gene. Only two ilvA mutants were among the24 ilv mutants obtained from Dr. Adelberg.These two are ordered with respect to eachother and to the ilvD16 and ilvC7 referencemutations as shown in Table 5. Several moreilvA mutants have now been obtained for usein future mapping studies.On the basis of the recombination frequen-

cies obtained in the above crosses, we havemade a tentative fine-structure map of 25mutations in three of the genes in the ilvcluster (Fig. 7). The order of some of the muta-tions in the ilvD gene may be modified as

more detailed information is acquired. Theorder of the ilvE mutations is based on thedata of Wechsler and Adelberg (18).

DISTJSSIONAs has been previously shown, P1 lysates

grown on a strain haploid for the region of thechromosome carried by the F 1 merogenote areable to transduce merogenote genes into suit-ably marked recipients. Some of the transduc-tants carry shortened merogenotes consistingof only the sex factor plus one or more ilvgenes proximal to the sex factor. More specifi-cally, as we have demonstrated here, transduc-tionally shortened merogenotes carry one ormore complete ilv genes plus variable seg-ments of adjacent proximal genes. Merogen-otes carrying increasingly greater segments ofa particular gene were shown to be useful inordering mutant sites within a gene.

This new fine-structure mapping procedure,which makes use of several merogenotescarrying different segments of a gene, is analo-gous to deletion mapping (2, 3, 5) if one con-siders the shortened merogenote as being de-leted for the region lost in transductionalshortening. Thus, a merogenote carrying afragment of a particular gene will form wild-type recombinants with all mutant sites on theendogenote if the corresponding and homolo-gous nucleotides are carried by the "undelet-ed" portion of the exogenote. Obviously, nowild-type recombinants can be formed if themutant site on the endogenote lies in a regionthat is "deleted" in the exogenote.

After this "deletion" mapping of a series ofilv mutations in one gene, it was observed, asexpected, that wild-type recombination fre-quencies increased with distance of any muta-tion site from the origin of the merogenote.This observation led to a simplified method forordering mutant sites within a gene, a methodthat requires only one, or perhaps two, mero-genotes for each gene to be mapped. One mer-

737VOL. 109, 1972


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

MARSH AND DUGGAN

-I.

215222 45 16

12 316 1109 46 29 211 43 1061 2423 39 60 6162

ilv E ilv D

L4285 288 101 49 71294 '

ilv 0 ilv C

ilv A

-I.ilv B

ilv 0

tilv P

FIG. 7. Proposed order of 27 mutant sites in four genes in the ilv gene cluster.

TABLE 5. Recombination frequencies of a shortenedmerogenote, F ED1, with two ilv mutants and two

reference mutations

Recombinationfrequencya

D16 ......................... 1.2 x 1o-21.4 x 10-2

Avg ......................... 1.3 x 10-2

A24 ......................... 1.8 x 10-22.0 x 10-2

Avg ......................... 1.9 x 10-2

A201 ......................... 1.2 x 10-41.3 x 1o-4

Avg ............ .............. 1.3 x 10-4

C7 ................ ............. <1 x 10-8<1 x 10-8

Avg ......................... <1 x 10-8

a Number of recombinants arising per donor cellin the mating mixture.

ogenote should carry all but the very proximal(with respect to the origin of the merogenote)nucleotides of the gene being studied. Thatmerogenote gene segment should be neitherfunctional nor capable of complementing endo-genote mutants of that gene. This longer mero-genote would be used to order the majority ofthe mutant sites along the endogenate gene, bytheir recombination frequencies. The mutantswith the lower recombination frequencies arethose located near the proximal end of thegene; those with higher frequencies are distalfrom the origin of the exogenote. It was ob-served, however, that in matings with longermerogenotes the wild-type recombination fre-quencies of one mutant located most distal onthat gene were quite high. In fact, recombina-tion frequencies were indistinguishable fromdiploid/recombination frequencies resulting

from complementation or recombination, orboth, of a mutation in the adjacent distal gene.However, recombination frequencies obtainedin matings between the most distal ilvC mu-tant and merogenotes carrying the shortestsegment of that gene were different by anorder of magnitude from the diploid/recombina-tion frequencies found with the mutant in themore distal ilvA gene. Therefore, we concludethat a merogenote carrying a short segment ofthe gene being studied would be necessary toresolve mutant sites near the distal end of thegene.

It seems likely that recombination frequen-cies with strains carrying small deletionswould be lower than that of point mutationsand would therefore lead to erroneous mappingof deletions with respect to other mutationsites. We have no data to estimate this effect.Each of the mutants used in this study wascapable of reversion to wild type and so wasnot likely to be a deletion. No small deletionmutants of the ilvD, ilvA, or ilvC genes areknown to us. The ilvE12 [originally carried instrain 11A16 (10)] is said to be a deletion (H.E. Umbarger, personal communication), butwe have as yet no merogenotes carrying only asegment of the ilvE gene with which we couldevaluate the effect of this deletion on recombi-nation frequencies. It may be possible to iso-late this type of merogenote in the near future,as described below.The only other deletion mutation known to

us in the ilv gene cluster is one isolated andpartially characterized by Kiritani et al. (9) asbeing unable to form recombinants with anyilvD or ilvC mutants he tested. It seems ap-propriate to say something about this deletionhere, as its existence is not generally known.We have found by enzyme assay that strainscarrying this deletion (e.g., AB3590) maketransaminase B, but not dehydrase, threonine

738 J. BACTERIOL.

201 a9


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/


deaminase, or the reductoisomerase enzyme. Itdoes make the condensing enzyme (aceto-hy-droxy acid synthetase). It seems, then, to bedeleted in a portion of the ilvD gene, presum-ably all of the ilvA, and a considerable portionof the ilvC gene. An estimation of the extent ofthe ilvC segment deleted is based on the ab-sence of wild-type recombinant formation inmatings between this deletion and the FilvEDAJ merogenote which carries the longestsegment of the ilvC gene of those tested. Amerogenote carrying all of the ilvC gene (an FilvEDAC merogenote) will restore Ilv+ pheno-type to a strain carrying this deletion.

Kiritani (9) has carried out two-point trans-ductional mapping on a number of mutantsthat he had isolated in the ilvD and ilvCgenes. He found that ilvD109 was the mostdistal from ilvC and that ilvD106 was mostproximal to ilvC. Our data agree with hismapping, and this observation further supportsthe validity of the mapping procedure. Kiritanifound that ilvClOl was the most distal to ilvD.We found three mutants that map more dis-tally than his ilvClOl. Kiritani found that hisilvC102 mutation mapped most proximally tothe ilvD gene. Despite repeated attempts, wecould not transduce the ilvC102 into AB 2200.This may very well be because this mutation isa nonsense mutation of some type. The AB2200 strain was originally described as beingfree from amber and ochre suppressors since itwas derived from AB 2550 (4). It carries threeochre mutations (lac, trp, and his), and wehave been able to cotransduce into it sevendifferent amber mutations. However, we havebeen unable to transduce one particular ambermutation (ilvD283) into it (Duggan and Adel-berg, unpublished data). Further, GudmunderEggertsson has found that strain AB 2200 ispermissive for the T4 phage amber mutantN67 (personal communication).We have isolated several shortened merogen-

otes carrying four ilv genes. These merogenotescomplement a recipient carrying the ilvC7mutation. Presumably they carry variablesegments of the ilvB gene and could be ofvalue in ordering mutation sites within theilvB gene. As mentioned above, we are notaware of recessive mutants in this gene suit-able for mapping by this method. The ilvB196of Ramakrishnan and Adelberg (13) was de-scribed as growing slowly in the absence of iso-leucine-valine. At any rate, one could hardly"order" one mutant site without another sitefor comparison.The shortened merogenotes were isolated by

selection for Ilv+ transductants in an ilvE re-cipient because the ilvE gene is the closestknown site to the sex factor. By selectingtransductants in this manner, one can findshortened merogenotes of greater variety thanby selecting in recipients carrying ilv muta-tions more distant from the sex factor. Thisselection method, however, eliminates thefinding of shortened merogenotes carrying thesegments of the ilvE gene necessary to ordermutants in this gene. A recipient strain with amutant gene located between ilvE and the sitecorresponding to the 0-13 sex factor is neededto isolate merogenotes carrying segments ofthe ilvE gene. Recently Wu and Wu (19) re-ported that the glmS gene is located betweenilvE and phos, and is carried on F' . We arecurrently determining whether this mutation issuitable for selecting such merogenotes.One of the observations made in this study

was that recombination frequencies for muta-tions located closer to the proximal end of thepartial gene carried by the exogenote are lowerthan for mutations located more distally. Thisis reminiscent of the observations of Pittard etal. (12, 17). Their time of entry data were ini-tially interpreted to indicate an alteration inthe order of gene sequence on the F4 com-pared with its parent strain. Glansdorff (7)showed that the gene sequence was unaltered,but the most proximal gene (argH) appeared ata lower frequency in mating experiments thandid more distal genes. In our experiments, therecombination frequency seems to havereached a maximum level within one gene dis-tance (Table 2). We cannot explain the re-duced recombination frequency of the argHgene of the F'4 on the basis of there being onlya segment of the argH gene on the most prox-imal end of the F'4, because the primary hap-loid strain, AB 1206, is arg+. Further, in thediploid state, the F 14 complements the argHmutation in AB 1450. It is tempting to specu-late, however, that the reduced recombinationfrequency obtained for very early genes onlarge merogenotes and on Hfr chromosomesmay be the result of limited possibilities forcrossovers to occur in the proximal region ofthe first gene. This should be the case only ifthe recipient carried a mutation in the prox-imal region of that gene.The mechanism of formation of these short-

ened merogenotes is unknown at this time. Weare currently trying to determine whether themerogenotes are shortened at the time of pack-aging in the donor host or after injection intothe recipient.

VOL. 109, 1972 739


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

MARSH AND DUGGAN

LrlAURE CrrE

1. Adelberg, E. A., and S. N. Burns. 1960. Genetic varia-tion in the sex factor of Escherichia coli. J. Bacteriol.79:321-330.

2. Beckwith, J. 1964. A deletion analysis of the lac operatorregion in Escherichia coli. J. Mol. Biol. 8:427-430.

3. Benzer, S. 1955. Fine structure of a genetic region inbacteriophage. Proc. Nat. Acad. Sci. U.S.A. 41:344-354.

4. Eggertsson, G., and E. A. Adelberg. 1965. Map positionsand specificities of suppressor mutations in Esche-richia coli K-12. Genetics 52:319-340.

5. Franklin, N. C., W. F. Dove, and C. Yanofsky. 1965. Thelinear insertion of a prophage into the chromosome ofE. coli shown by deletion mapping. Biochem. Bio-phys. Res. Commun. 18:910-923.

6. Friedemann, T. E., and G. E. Haugen. 1943. Pyruvicacid. II. The determination of keto acids in blood andurine. J. Biol. Chem. 147:415-442.

7. Glansdorff, N. 1967. Pseudoinversion in the chromosomeof Escherichia coli K-12. Genetics 55:49-61.

8. Ikeda, H., and J. Tomizawa. 1965. Transducing frag-ments in generalized transduction by phage P,. I.

Molecular origin of the fragments. J. Mol. Biol. 14:85-109.

9. Kiritani, K., T. Matsuno, and Y. Ikeda. 1965. Geneticand biochemical studies on isoleucine and valine re-quiring mutants of Escherichia coli. Genetics 51:341-349.

10. Lennox, E. S. 1955. Transduction of linked characters ofthe host by bacteriophage P,. Virology 1:190-206.

11. Pittard, J., and E. A. Adelberg. 1963. Gene transfer by

F' strains of Escherichia coli K-12. fl. Interactionbetween F-merogenote and chromosome duringtransfer. J. Bacteriol. 85:1402-1408.

12. Pittard, J., J. S. Loutit, and E. A. Adelberg. 1963. Genetransfer by F' strains of Escherichia coli K-12. I.Delay in initiation of chromosome transfer. J. Bac-teriol. 85:1394-1401.

13. Ramakrishnan, T., and E. A. Adelberg. 1965. Regulatorymechanisms in the biosynthesis of isoleucine and va-line. II. Identification of two operator genes. J. Bac-teriol. 89:654-660.

14. Ramakrishnan, T., and E. A. Adelberg. 1965. Regulatorymechanisms in the biosynthesis of isoleucine and va-line. III. Map order of the structural genes and oper-ator genes. J. Bacteriol. 89:661-664.

15. Rosner, J. L., L. R. Kass, and M. B. Yarmolinsky. 1968.Parallel behavior of F and P1 in causing indirect in-duction of lysogenic bacteria. Cold Spring HarborSymp. Quant. Biol. 33:785-789.

16. Swanstrom, M., and M. H. Adams. 1951. Agar layermethod for production of high titer phage stocks.Proc. Soc. Exp. Biol. Med. 78:372-375.

17. Walker, E. M., and J. Pittard. 1970. Conjugation inEscherichia coli: interactions affecting recombinationfrequencies for markers situated at the leading end ofthe donor chromosome. J. Bacteriol. 103:547-551.

18. Wechsler, J. A., and E. A. Adelberg. 1969. Antipolarityin the ilv operon of Escherichia coli K-12. J. Bac-teriol. 98:1179-1194.

19. Wu, H. C., and T. C. Wu. 1971. Isolation and characteri-zation of a glucosamine-requiring mutant of Esche-richia coli K-12 defective in glucosamine-6-phosphatesynthetase. J. Bacteriol. 105:455-466.

740 J. BACTrERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

vol. 1972 ordering mutant isoleucine- valine genes …procedure for fine-structure mapping thr used,...

Documents