UNCLASSIFIED

AD4 1 4 8 2 8

DEFENSE DOCUMENTATION CENTERFOR

SCIENTIFIC AND TECHNICAL INFORMATION

CAMERON STATION. ALEXANDRIA. VIRGINIA

UNCLASSIFIED

NOTICE: When government or other drawings, speci-fications or other data are used for any purposeother than in connection with a definitely relatedgovernment procurement operation, the U. S.Government thereby incurs no responsibility, nor anyobligation whatsoever; and the fact that the Govern-ment may have formlated, furnished, or in any waysupplied the said drawings, specifications, or otherdata is not to be regarded by implication or other-

wise as in any manner licensing the holder or anyother person or corporation, or conveying any rightsor permission to manufacture, use or sell anypatented invention that may in any way be relatedthereto.

C..) Pin& ucal Rehnol kport

zPMoD Ovm:r

I Daomber 1962 - 1 Suptembcr 1)63

C5 IZNC=PAL IISYZBTG0MR

Robert~ Marls~i Wood, Pha. D.

0 0 GoorL-P Vashinatn Univi-rcoty School1 of h'~i.ic-, '-.'hln ton 5, D. C.

==CT@' or REPOE

0 0 Stilea an~ Lb-j GmAics of ftrulunca La Catrler~ 3cwra

D3*496193-N0-2376

USC AV*ZUKZLZTZ OWSAW

q ~ ~ asli ld bv.vati~atcrae ?IV Ubtaln Copiea cof 2Lds eort From& O

UaslasiftD DDC

TI~SIA E

Preparing !wtltution: Ocorge Wau1ingtor, UnLvrity Saibool of IWdicizieVwha1vgtoz,5, D.C.

Ti.tle of liuport. Studiera on i.Jii Ccn -tics of ViTuteco in EnUkrlc hctwta

Principal 1nct,ator: 1kwburt Charl,! L'(od, Ml.D.

trattionA =dc Date. lO,pae, 5 tab3er,1 Scpteazbor 1 63

Contract A1krber:Dh9142?

Suzxertfed I17: U. S. kn~y M~~dctA PQacl and DevelaOr~4J Co~aaudDvpa-twuc of a-" &--tcwhine~ton 25" D.C.

Vo t,- ~ Ccrpic of *Jhio -,r-L rt . rilt4 ith tbu 1vf~inoc

AL-muadri, VirL:%a, anl =W be. cbta±x.';, fro:,, ihu'.- iiucy 1 y qua:.ific IInv -.Atfra -mriziurj ' r Lmv

treated t uina 14tqa 'u1I -.x - LbridizuAvr. by raeu.1cuha -v' &.12 I~r t'raina.

Gaeritic rtud.. ir, . 1 :-4, .c AMXV3cU;.a 1..,4 ,' tlX-C.. UW4;c V%) CpyiAvu.

,i. coli X E. ccvic ,c*. r:nsv t:,".uf-r of Jw' EvuV-A4itu1 %1mmu was

(dcUtA~d Only OCaGIOnally'. Vir-CrUCC t.-BA Cn U 2&* 14 k lds carrvla ov r

bobten thc& w' ll -yni., whicLU t~ wA -3&ent. I for virleose. *bAri4o

carryin* th,c~~in ~_jl-rhLq Eacht44ha. rutQu so a V :Gviu,,ut cxuyIwtLe 41spl~d

into rus-Aiate 4drulce; 1aloi scam azmta --tich ~t1i UL c=&iflotu~ wem 1U11

vixra2."zt, vidlc ow. hlu14l Lth ad Lworporat. * it proived to 14c airulent.

PMUM1IUM7r tuste on a Muoua-TIzuleut &Srw- ujl bkbri4a1 by an

swirulant Galmw~ a~~o SacaUti t2'..n, -, Cta Wax~its QleQp WlzUoef

mw be lot ca substitution of the +-i _s . Ubm1a

$train 643W r , hybridized by mating with an R ooli t stalsa, vas fR to

exhibit enlaumed recipient ability on remating with V. oli donors% hof lee e

chromosoma] region matched the enetie segmnt inugrated by the hybrid.

I, * torstlos In the ykvalonce of ESaherichlar &lnea hybwl4 for the

G0lzhe PIC.

1. har-a* the poet you this Inwestigatio has foeuooed on the genetic basis

of vlrult-c, i u teric bacteria. Owe Imprtact finding has been the lose of

Culnea pie, vIralclw, In the 2a straft of SIMUS flxnr after specific sub-.

atIitton of 3sc2hrtobla .011 genetic materialp as a result of mating vith

S. cODI 9-12 wr strains.

2. Oezetlc r'ombinaton betzicn Wtr straius of Bocherickla coil 1-32 and

nectre of tZhe !M crcm was UIaily temonstrate4 by larls mi harrowi

(1107). btouuivcs thiougi Incoiiwetep Le-et1c banoMF 6w found to exst

betveen these t;.* erew Fowmoa t al. (1903C, 199) dexonetwsted a fatal en..

teric iaccttsu In eitber Otaw~d or cafw tetractlonide pzeted C1nzea

VIL5 chal.1eased orally w4th 3. flepor 2a. .S Watl proiue4 leslons it

both this X2ce mul amUn lbx4 Aiab rescduld lbeft w~on In bas es of

sInbplosis. Ma wstilllty of both a V~reitc trw~fer systemad an ISfoom

tion =W,1 *sn eabovl texemiutio of alter&tMs In the v~iaje of a.

ftgwl 2 forprereaed S~ws "qft~lvis co~d~aim ith3. eoli 3-32.

3. Uiyoe R. oa*t Z-22 ftv strains vere employed In the pwntle cosses:

Ifl95i lAA&c twmsfcer Its cluaoms3 wrs In the ovdar origin# Imaetafe

Sull pduelon(~~)xyab se utiliatan (. ) rhamlse Uuilation (')

utdol vtMon (io) 3313 (owSlIA U0_ * , ... zba'); Ut1&t

3 kw

.20

(origin, aa1 Im 6 AM .. J.a) kt1W~ bei~ven these S. ooll donors

ad the 2a . eFkloari e3tain tilleated that a grosi humloig exsts between

the chramonee of theau tim speci. Roeverp resombination freqencies were

lo,6v thau thiose obtaitricd tu Lar~ale E. colt X S. doll crosses (Table 1).

Ptather, i- yredcJ~ncmt RJcl hbbr~d class wo9i?1-d only the selected gams.

tic mwcer froz tho Z. aoll parent; ext -natwu trofcr at the Ezscherldhis

ceomu vwe detvtotcd ouly occaitonally (table 2).

4. A deftutc pos~ ruilatSov betmen deae! an repneI o

alm~ c~~*W I~n the aral (B1~.D.1tration o~f 91uU totarved or carbon

ttralriL t@4A' CU- c VaIa A p (ftrM4a Ut a. 19 ~59). nSM It was,

not possibli. to Qra)p Loy LD 5) dose w~ a estimt of viaulenee. Sovvew,

it wau obsermL that a deft of 5 X 10 to 3. Z 1.0 ora~mzms aftalsterad

oraWly uzM'.-r our (uQerL..!ta coxwiltIous& cossistovutly kille a large jr-

tIom or mktmla. Aca-ojout of yindu-.nj of vw-si =tlA bsbrld strolm,

th' r Ibre, wa de.u~n prltdly on sn "all or t%'1,(; typ otivowu.

N. Y tcstlm WfL-t.Jii '.3ridsarrying overluppLac seuatg of tho

Es~rih r.now, It va bvP..d that the :An~Im num""" 'r cbxtzoeinoa 70-

Clain ruipnlblt gor lik virult-nov of' fl3& ft~r V. pretrwested Ouluoa pM=U(t " SWrntU~o1 iJ. r virulee dA In GWAriz,-1 In, IOU 3. Jbrma

hav~ig r.tcivr loo- + Le- jdl. vta* m4 ~ ? te Z. COil 3Wut mric

as '4imlrjut 40 tbW- '&-U Parami&l sia-aft. Uhowe Lldts vftdb SWPSOA __

I*W ver m4torgaly avimumet -d MA MI~ hburift wvt etbar

eoIitU,2v vlleu&t v1 UY9l1t In i@Ut oaal Almon. himid4:4a m

fuc* bytula vW virlezat, the rmasl lag bybAU4 teeted# as wvl m all+ + * . loo? mm- ct tflterv*a U,1 villacam-0

I.e. * Y amwLisotsar U31e4 a 3m~ vat ammfSomt xprwt~o Ot animal.

In &4.ttic. brybrU %Aildb WM rebe Ow 3 mtoa tam 2. coUlaa44cpl d thi

.3-

6. Examination of hybrids possessing the characteristics of segregating

partial diploids strongly indicated that the observed alterations in virulence

were due to the transferred Zscherichia material. Several rha - 11.4

p+hybrids, in spite of repeated purification, segregated clones (about 1 In

10 cells) which exhibited the parental Shigella phenotype. When tested in the

guinea pig (Table 4), these partial diploids were found to be of intermediate

virulence, while their haploid segraents, which had completely lost the in-

jected Zscherichis genes, were Indistinguishable from the parental Shizella strain

in their virulence. On onsoccesion a haploid segregant was isolated in which

the previously oxogenotic fragment had become integrated. This segregant

proved to be avirulent.

11. Alterations In the virulence of Solmonell - Salmonella hybrids for the

mouse.

1. The Indleation of a genetic locus, or loci, centrelling the virulence

of fliselJM for the pretreated guinea pig has led us to Investigate the possible

existence of a similar factor, or factors, In Salmonella. Hybridization of

Salmonella by 5.. 5"aU was initially reported by Sron, Carey and Spilmon (1958).

Subsequently, it has been found possible to produce Salmonela possessing doao

ability, either by hybridization for a terminal 2. colt Ifr morker (which of ten

results In integration of the sox factor, 1), or by infection with F, followed

by selection for Rfr sastents. The use of such dono in crosses with Salmnell

,a natural pathogen for the mense, presents am excellent system br

the gonetic study of virulence.

-4-

2. In preliminary experiments, S. typhimurium strain C-5, genetically

marked and virulent for the mouse, was mated with an avirulent S. ebony donor

strain, and several hybrid classes from this mating were tested for mouse viru-

lence. The initial findings indicate that virulence may be lost in the hybrids

when substitution occurs in the xvl+ - rha+ region. The correspondence of this

preliminary data with that obtained from the experiments with the Shiaella hy-

brids is most encouraging. However, further studies will be necessary to con-

firm these findings, and to establish whether other chromosomal regions might

also be involved in the virulence of S. typhimurium for the mouse.

I1. Behavior of E. col - S. typhosa hybrids on remeting with Z. call Ifr strains.

1. From the outlet of this investigation our hope has been to transfer

the property of virulence from a virulent genetic donor to an avirulent genetic

recipient. As yet, this has not been accomplished, owing to the difficulty of

obtaining virulent donor str~ins. Salmonella, either virulent or avirulent,

generally are poor recipients in gentic crosses with S. coli K-12 donors. Trans-

fer of Hfr donor ability to virulent Salmonella strains in these cases is very

hard to achieve. However, our studies on the behavior of ji. tynhos hybrids

produced by meting with j.-.col have pointed out a method by which the genetic

recipient ability of Salmonella my be increased.

2. It has been previously observed that S. t hybrids from

genetic crosses vith 3. coli Ifr strains exhibit increased fertility when re-

mated with the Z. S&U parent. This has been shown to represent selection of

pro-existing, high frequency recipients from an otherwise sterile population by

the first round of mating. However, upon examination of this phenoueon with

j. tIbna s the genetic recipient, we discovered that, in this spolees, soleo-

-5-

tion of high frequency recipients was not involved. Our studies showed, in

fact, that the increased recipient ability of the S. typhosa hybrids was due to

the presence of integrated I. coll genetic material.

3. S. typhosa strain 643WSR was mated with the E. coli Hfrr W1895

(origin, lac+, are+, . . l+ ) with selection for the lead marker lac'.

Lactose positive S. typhoss hybrids, when remated wibh W1895 for the more dis-

tal marker aLe + , showed frequency increases of 200 times the normal frequency

for transfer of this gene to previously unmated 2. typhosa. It was noted that

many lag+ S. typhosa hybrids were heterogenotes which continually segregated

lactose negative (lac ) clones. When a number of these lac' segregants, which

had lost the E. oli genetic material, were remated with the E. coli parent,

none showed any increase in recipient ability. This provided the initial evi-

dence that the fertility increases of the hybrids occurred as a consequence

of the transferrel E. coll -enetic material.

4. E. .li Hfr strains W1895 and Hayes (transferring origin, Oae, lac

. . . rho+) were crossed with S. typhosa 643 WSR, with selection for the single

markers lac+ and asra. Four classes of 643WSR hybrids were obtained which were

labelled as follows: those receiving lac+ alone or arm + alone from W1895 were

designated WSR lac+(95) and wSR ara+(95), respectively; those receiving these

individual markers from Hayes were labelled wsR jj+(H) and wSR ara+(H). These

four hybrid classes were then reasted with each of the two E. colf Hfr strains

in a series of eight reciprocal genetic crosses. These crosses, shown in Table 5,

demonstrated that each hybrid class was capable of exhibiting significant in-

creases in recombination frequency wherever the leading chromosomal region of

-6-

the 3. coil Hfr used for resating matched the 5. uol genetic segment previously

integrated by the hybrid. When the alternate Rfr (which did not inject its

chromosome into the region of artificial genetic homology) was employed, no

increase in recombination frequency was observed. An exception was noted in

the Hayes X USR lsc+(H) matings (Table 5, cross 6), where a 13-fold increase

occurred when the E. col segment integrated by the hybrid does not appear to

match the lead chromosomal region of Hayes. Probably, these hybrids have inte-

grated E. colt material nearer to the Hayes lead region than the absence of

the or+ marker would indicate.

5. Thus, it is possible to increase significantly the recipient ability

of 1. typhosa in crosses with 9. coit Hfr strains by transfer and integration

of strategically located 1. coli chromosomal segments. The studies of Zinder

(1960) and Falkow, Rowd and Baron (1962) have shown that the homology which

exists between Ischerichis and Salmonella is incomplete. Presumably, then,

prior establishment in S. typhose of an Eacherichia genetic segment homologous

with the leading chromosomal sOction of the E. Sol donor facilitates early

genetic pairing and integration of the transferred material. Removal of the

initial barrier to integration of lead Hfr markers would increase the chances

of integration of more distal markers, thus accounting for the observed in-

crease in recombination frequencies.

SumMr

1. Three E. col K-12 Hfr donors were employed in genetic crosses with

the 2a strain of Shizellsa flexneri. The results of these mtings confirmed

the existence of a gross homology between the chromosomes of these two organisms.

-7-

However, recombination frequencies were lower than those obtained in comparable

E. coli X E. coli crosses, and extensive transfer of the Escherichia genome was

detected only occasionally.

2. Shigella hybrids carrying overlapping segments of the Escherichia

genome were compared with the parental Shimella strain with regard to their

virulence for starved or carbon tetrachloride treated guinea pigs. Hybrids

which received lac+, lac - ara , rha and mal from the E. coli parent were

as virulent as the parent strain. Hybrids carrying rha+ - ind+ were uniformly

avirulent, while xyl and xvI - mial hybrids were either completely virulent

or avirulent in about eiual proportions. With one exception, fuc+ , nic +

fuc - ni. c+, and fuc+ - nic+ - lac+ hybrids, as well as those receiving the

pili antigen from E. col , were of intermediate virulence.

3. Shigella rha+ - in d+ - xyl + - mal+ hybrids, possessing the character-

istics of segregatiug heterogenotes, were found to be of intermediate virulence,

while their haploid segregants, which had lost the E. coli genetic material,

were fully virulent. One haploid segregaut, which had integrated the pre-

viously exogenotic E. coli fragment, proved to be avirulent.

4. In preliminary experiments, a virulent Salmonella typhigurium strain,

C-5, was mated with an avirulent Salmonella ebony donor strain. Testing of

several hybrid classes produced by this mating indicated that mouse virulence

may be lost in the hybrids when substitution occurs in the l+ - the+ region.

5. Lactose positive Salmonells typhosa hybrids produced by mating with

E. coli Hfr W1895 showed significant increases in frequency when remated with

-8-

this Hfr. However, lactose negative segregants from Iac+ S. typhQsa hetero-

genotes displayed no fertility increase, indicating that the integrated Escher-

ichia genetic material as responsible for the enhanced recipient ability.

6. E. coli Hfr s;trains W1895 and Hayes were mated with S. typhosa 643WSR

with single marker -46ectlon for lac+ and ara . Four hybrid classes were ob-

tained, each possessing a single marker derived from one of the E. coli parents$

In a series of eight reciprocal genetic crosses, each hybrid class was remated

with each of the two E. coli Hfr strains. With one exception, recipient ability

was increased in the hybrids only when the E. co!i genetic segment previously

integrated matched the proximal region of the remating Hfr chromosome. We

have interpreted this as establishment in S. tSyhosa of an aritficial chromo-

somal homology, which removes the initial barrier to inte ;ration of E. coli

lead markers, thereby increasing the integration chances of more distal Hfr

markers.

Discu!;ion

The genetic .tudies presented in this reiort confirm and extend the obse,-

vations of Luria and Burrous (1957) that E. coli K-12 and Shilella exhibit genetic

homology. Our finding that Siveila hybri& ,how alterations in virulence as a

consequence of integration of segments of the E.,cherichia chromosome is mo.st sig-

nificant. Of equal importance is the fact that some hybrids retain virulence

even after incorporation of up to 15 per cent of the Escherichi s genoe.

The intermediate virulence of these byhri,ls which had received one of two

widely divergent chromosomal regions is not immediately apparent. However, it

-9-

is encouraging that analysis of over one-half of the chromosome has revealed

only one region which appears to be essential for the virulence of S. flexneri

2a for the guinea pig. The segregation of virulence among the =1 and

ZXv + - mal+ hybrids, and the complete loss of virhmsce in all hybrids carry-

ing the closely linked ind+ and rla + - ind+ markers indicates that we may be

dealing with a single gene, or gene cluster, located between rha+ and xyl+ .

The results obtained with partial diploid strains are strongly suggestive

that the observed virulence alterations of Shiijella hybrids are the direct re-

sult of the presence of Escherichia geactic material, rather than some non-

specific phenomenon. In the partial diploid, the hybrid retains its full

complement of ShiLella genes, but is still significantly less virulent than

the parental dysentery culture. The strain may return to comlete virulence

by eliminating the Escherichia chromosomal fragment or become avirulent by

incorporation of the fragment. The reduced virulence of the diploid cells

indicates that one or more determinants of the endogenote important for viru-

lence are recessive to, or may be replaced by, determinants on the Escherichia

exogenote. In addition, the data obtained with diploids suggest that the viru-

lence loss associated uith the indorporation of the rha+ - .xvi region is very

likely not due to deletion or unequal crossing over of Shigella material.

As indicated previously, evidence is accumulating to show that the

chromosomes of E. coli, Salmonella, and Sh!ella are grossly homologous. This

gross homology could indicate that identical chromosomal regions responsible

for virulence (or lack of virulence) may be identified in all three groups of

organisms. In fact, our preliminary data with _. ty:nhimurium indicates that

this is the case. Further studies will, of course, be required to verify these

-10-

data, and to establish the precise chromosomal location of the genetic deter-

minant (or determinants) responsible for virulence in these organisms. In

addition, it is anticipated that the information derived from our studies on

Salmonella recipient ability will enable us to obtain virulent Salmonella and

Shigella donor strains. The employment of these strains in future studies

would provide significant information with regard to virulence, genetic homol-

ogy, and evolution of species within the Enterobacteriaceae.

Literature Cited

Baron, L. S,, W. F. Carey, and W. H. Spilman. 1958. Hybridization ofSalmonella species by mating with Escherichia colt. Proc. Intern.Coagr. Microbiol., 7th Stockholm, p. 50.

Falkow, Stanley, R. Rownd, and L. S. Baron. 1962. Genetic homology betweenEschtatchia coli K-12 and Salmonella. J. Bacteriol. 84: 1303-1312.

Formal, S. B., G. J. Dammin, E. H. LaBrec, and H. Schneider. 1958. Experi-mental Shigells infections. Characteristics of a fatal infection pro-duced in guinea pigs. J. Bacteriol. 75: 604-610.

Formal, S. B., GO. J. Dammin, H. Schneider, and E. H. LaBrec. 1959. Experi-mental Shiaeila infections II. Characteristics of a fatal entericinfection in guinea pigs following the subcutaneous inoculation ofcarbon tetrachloride. 3. Bacteriol. 78: 800-804.

Luria, S. K., and J. W. Bourous. 1957. Hybridization between Bscherichiacol. and Shltella. J. Bacteriol. 74: 461-476.

Zinder, N. D. 1960. Hybrids of lscherichia and Salmonella. Science 131:813-815.

BiblioAraphy

Baron, L. S., I. Ryman, S. Falkow, and V. Krishnapillai. 1963. Geneticrecombination between species of Salmonella. Bacteriol. Proc.

Johnson, ,. M., Stanley Falkow, and L. S. Baron. 1963. Recipient abilityof Salmonella typhosa in genetic crosses with Echerichia colt K-12.Bactariol. Proc.

Table 1

Transfer of Genetic Characters from 2. ccii K-12 to S. flexneri. 2a strain 2457T

Cross Selection Frequency ofRecombination

E. coli Hfr W 1895 lac + I X 10.3+

x Ora 5 X 10 "4

S. flexneri 2a F- 2457T rha + 3.5 X 10- 5

xvl+ 2 X 10"5

reelI + 2 X 10 5

fu.C+ 5 X I0o6

nit*" 10-7

. Soil Hf AB 313 xV1+ 3 X 10. 3

X Mal+ 1 X 10, 3

S. flexeri 2a F- 2457T nic+ 5 X 10-6

J colt Hfr 1362 M..l + 2 X 1o- 3

x xvl " 1 X 10- 3

S. flexnerl 2a ?- 2457T lac + 7 x 10- 5

The frequency of recombination is expressed as the number of hybrids isolated

per donor cell.

040

u AC

~1iI S am

00 -0

oil h bi- h i

Table 3

Suwmary of Hybrid Virulence

Hybrid Class No. hybrid o95% Confidencestrains teted T I ..ort.lity Limits

+ 3 19/24 79.16 62.90-95.42

.UK+ - A+ 2 61/77 79.22 70.16-88.28

Ale - p i1 3 16/55 29.09 17.10-49.08

r b-+ 2 18/24 75.00 57.95-92.05

L_. + 3 20/25 80.00 64.32-95.68

xl+ 3 1/36 2.77 -2.59-8.13

xk + - mae. 4 39/60 65.00 52.93-77.07

x .+ - M-I+ 2 2/39 5.12 -1.79-12.03

al+ 5 62/92 67.39 57.81-76.97

+- x + - inl 3 4/56 7.14 0.42-13.86

rho, .-I+-n_-ni . + 4 9/103 8.74 3.54-13.95

LUA+ 1 17/25 68.00 49.78-86.82

li+ 3 11/33 33.33 17.26-49.40

nic+ - 2 6/20 30.00 10.0t-49.99

m + - - 1 8/21 38.09 18.30-57.88

Controls

,1. fljxl 2457T 1 120/157 76.43 6979-83.07

.1. SAi V1895 1 1/48 2.08 -1.96-6.12

Table 4

Virulence of PFrtial Diploid HybridsOf S. flexnori 2a and Their Haploid Segrelants

For the Guinea Pi,,

Genotype of Hybrid Deaths % Mortality 957 ConfidenceTotal Limits

Diploidrhe,-indoxvml-ma1 11/29 34.48 19.55-49.61

Haploid Segregantrha'-ind'-xyl"-mal" 24/29 87.76 63.54-9?.82

Haploid Segregant.h_+-_i+ _-v+- a + 2/26 7.69 -2.60-17.98

0 LMOS . E4 B

46.4 0.4

~C I 0 0 a Go 0 CA LA

I

48a +0 + 1 4. +4 4 41 e

400o0 '4.66 6 .

BA CJ- . 4 4 .

* 0 In - a *, *

C.: Cd - A *Y~. % C

0 40

C a

C 00

' .0 AI 6. 0 0-'3 a a 0 w4 Vi

0%~ 0% 0% 0

e.c '

Iw x N K hi xin4S

BA~ ~~~ 0iU U S U A B@4~~~s CP6 ~ O 0 a~S-4 ~4IV

- Seatle$ of Y1~qsIQ~o t

2. GewtLc

-of An3911C Lu :'wt'rlc batc-ria

3. ftmgunm