elst March,1952

Dear Lederberg,

Your typescript@and your subsequent letter of March 10 have
been received. I have had a talk wit.. Hayes in London,and a few crosses have
been going on in my absence,so that it will take me a long letter to deal
with everything.

1) your typescript. I agree with it and have added a few notes to
it on a separate sheet.

2) its publication. I doubt that we can publish exactly the same
paper on two separate journals,but if one of the two journals were a bacterio-
logical and the other a genetical one,it may be sufficient —- and sdvantageous
too - to write the sa.ie things in the different styles suitable for the two
classes of readers. Hayes told me he would send a paper about his new findin-
g8,stating in it that he had heard of our pievious work,and without pressing
for publication. He is likely to publish on J.Gen. Mier. ,and we might send
the Buropean edition of our joint paper to the same journal,asking for its
inclusion in the same issue as Hayes's paper. I ama member of the Soc?Gen,.
Micr. I shall write to nHayes about it as soon as 1 hear from you. Another,
alternative,way of publishing in furope would he on our a een Reh tale
bacteriological journal ,which would certainly take a paper in English. As
to the American edition, I should be very pleased if it were on Genetics,
and I agree entirely about api@aships:authorships. Also,the idea of keeping
further developments for future paperfs seems quite good to me, the effect
on segregation of F+ (on which 1 also had some data,see later) and the rela-
tions between nfr and F+ being some of the obvious ones. But 2 hint at some
of such developments in the first paper/s mignt seem advisable.

3) Hayes. His work sees good, but he has undoubtedly messed the
whole problem by drawing too simple conclusions and using improper termino-
logy. His trobble is that he is not a geneticist. He thinks of "sametes" as
of non-filterable phaZe particles. His streptomycin experiment was probably
meant to show that "cell-free" suspensions ( rather,suspensions free of li-
ving cells) but containing the pnage particles able to effect the genetic
transfer,would be fertile. When he heard of self-incompatibility he truought
it should mean the loss of the self-reproducing pnageslike gamete,and that
tois - and therefore fertility - could be reitfroduced by infection. I have
tried to explain to him the difference between a "gamete",representing the
nucleus of a cell, whatever this gamete is, and the capacity of forming
gametes, and have suggested to him a fuller use of the markers to get a clear
cut difference between infection for ¥+,and xeesmbinakion: gamete forpati ne
This advice hs probably been unfortunate,since it turneds out later’that Fr
affects segregation,and he may by now be confronted with a similar puzzle to
the one that now confronts us,wxkk However, 1 shall see him in veptember and
have another discussion whth hin.

4) F+ effect on segregation. It is quite possible ,4as far as my
data go, that the make difference between BM and TLB, lines is wholly due
to the + effect,as you say. My approach to the problém was a different one,
as I was interested in mapping separately,the BM and 'LB, lines and compare.
their maps. Three alielic markers snowed #4 Same order in BM x BM or TLB, x
TLB, crosses,but this is not the main thing. The puzzle is represented by
the results of two crosses of type TLB-S* sugars negative x TLB,+ sugars
positive, the formerly "forbidcen"cross,now made possible by F+ transduction.
Data on such a cross wkkk ‘n the coupling and repulsion phase in respect to:
P+, i.e. Fe x F- and F- x F¥ were made,and although giving the same (linear!)
order,i.e. (TLBl1 %) Ara Lac Gal Xyl wal §$

a b c d e f c.O. re: ions. —_——>
2

the c.o. values are widely altered in the two crosses,apparently with
according to some sort of gradient along the chromosome. .Cross 1 is TLB,-s su-
gars neg.F- x TLB)+ sugars + F+ , Cross II is MxF. Triples,not given

in detail, are roughly proportional to the expecéasions according to singles,

C.o.region Cross I Cross II Cross Tit

a 1 1 2

d 73 15 50

c 31 5 32

d 109 120 27

e 31 97 24

f 16 80 5
triples 38 10 25

1+ thus seems that the position of F+ in the cross affects pairing, or c.0o.
with some sort of gradient along the chromosome, we perhaps wg reducégg or
increastgg either pairing or c.0. in one arm, but the coupling and repulsion
effect of F+ @oadet indicatessome asymmetry which is not easily explained.
Perhaps F+ determines the direction of the cross,in the sense that it deter-
mines formation of gametes,and these may not be representative of the whole
nuc@@us but of a fraction of it,as if the chromosome had a tendency to get
broken before the free end and thus loose some of the terminal markers, al-
though the point of breakage may vary fxanxgamekextexgamekex in various “male
gametes. Also,one might assume that F+ gets a locus on the chromesome,and
the strand carrying F+ haz is selected for. There is still ample room for
imagination. It would be important to know what happens in a F+ x F+ cross
in t.e TLB, line. If, as you assume,only F+ x F- c.osses are permissible,
and some of the cells of an F+ line are phenotyp¥cally F- ,then such a cross
Mauka should give data intermediate to those of cro: ses T and IT. Cross III
is a substitute for such an F+ x F+ cross,in that there the TLB,-S” sugars
fe@ative F+ strain was crossed to K-12. It does behave as somewhat interme-
aidte between crosses I and II,and does not show major deviations from the
linear order, pointing out that probably there is no mgjor cytogenetic change
in either BM and TLB, lines. A fourhh cross, i.e, TLB]-S" F+ x TLB)+F+ will
say much about it. It wil- soon be made, amxwakiszas or rather repeated, as it
was accidentally lost. I shall also be looking for the F+/P- C and R effect
in the BM line, and checking some other points, i.e. whether independent

F+ strains wheth have got their F+ by transduction, behave differently—-Oece,
Gwo independent ones behaved exacity alike) ; and also whether tne origin of
F+ has any importance in it.

5) I shall be collecting some data on Hfr,and 1 hope you will do the
same. It seems to me that it does affect the F+ story to a lawge extent, Hfr
being perhaps a mutant of the F+ particle. The symbol F™ might iy such a
case be useful. After all,this would be merely a reversal of terminology,
pecause,if I remember correctly our early correspondence,the first F- strain
you and I came across were called Ofr. Can you confirm that Hfr does not
transduce r+ in infection experiments,while it can do so in sexual propaga-
tion, and that some F+ (transduced) TLB|- give a high frequency of recombi-
nation with Hfr,waile others do not ?
6) If you sre looking for F- mutations’in other lines, it may inte-
rest Nou that Hayes tested some 120 colop,ies from an& F+ strain and found
none F-., Why not try nitrogen mustard resistance ? Out of three strains
thus selected in 1949, one was Hfr,the other normal,the third F-.

7) I think it would help me in the work on isolation of F+ out of
cells to have Maas's strain. Would it be possible,and shall I ask Dr.Maas
directly.

8) Could you give e at yoir earliest convenience an answer on these
two points . (i) Have you anything against me saygng a few words about this
Fy story at a local national congress of ‘ierobiology. I am not keen myself
about it,but have been asked to do so. Anyhow 1 can easily give it up at tnis
stage. A short summary should be published in the proceedings,in Italian.
Would you prefez coauthorship or your contribution being cuoted at its full
value in the text. I should not take this as the Zuropean edition of our
joint paper,since the paper must be in Italian and is hound to be buried,in
a highly condensed form,in the proceedings »which should have only local
readers, The congress is to take place on April 15 and { should give the tit +e
and 10 lines of summary before April 5th or so,but the summary to be published
( some two or there sheets ) can go in later,so that you covld well see it
before it wase sent for publication. I hope you will be entirely frank about
it. (ii) I should give a takk on resistance to antibiotics at the Baris
congress, July. I should ¢3ke to quote your method and results with the
replica-plating, which s#ax ia be the most convincing evidence against Hinshel-
woodian objections. Could you let me have some more details about it:antibio-
tics used,bacteria,etc. I am expected to send tne manuscript by the 15th of
April.

Yours sincerely

Levee
4

Notes on the paper : SELF-incompatibilit» in E.ooli and genetic infection
for fertility ?

N.3B. I am not meaning that these thingy should be added,or ddded as such
to the manuscript. “hey are just data,or considerations,relevant to the
various items.

page 2, top. An independent occurrence of a BM-F~ was found at Cambridge
during selection for nitrogen mustard resistance,which implied a long expo-
sure to the drug. The strain did not cross to TLB -,but it was found later
that it would cross to other,non TL3, stocks and to filial TLB)-.

Later,line 8. The 3 marker has apparently disappeared from most stocks which
carried it at the beginning, presumably by back mutation and selective advan—
tage of B+, so that all BM stocks will grow witn th. aduttion of methionine
only,and mmxRx the addition of B has no effect on back mutation on minimal of
3M=- strains. However,the !] marker is an exceedingly stable one,and its hack
mutation can be secured only by the addition ofk limiting amounts of methio-
nine (slighlty more than 107-) to the minimal medium,and,in some instances,
with the additional help of UV. Three independent single-step reversions were
obtained, called BM+ . Tne crosses BM-ST x BM ( symbol 3B kept to indicate
the origin of the streins) are always fertile,except that they need previous
incubation imxkx of the mixture in broth before plating on minimal St,in
order to secure a decent amount of prototrophs. Incubation in a "rolling"
apparatus (Ryan, MGB) determines an increased yield and shortens the necessary
time, so that 3-5 nours are amply sufficient for a high yield ,when starting
from non-rolled ("depp"{grown cultures). Starting from rolled cultures there
is almost no fertility (later,aereation effect).

Two independent reversions of TLB -, i.e, TLB|+,obtained in three steps, were
Era tested far. in crosses to (parental) TLB -~ST,anc never was a Binagle recom
binant prototroph recevered; but a full yield was obtained when crossing to
filial TLB)-S".(five different’ filial strains tested,differing by recombined
markers). Also,parental TLB,-SY crossed freely with filial TLB)4+ (prototrophs
from BM- x TiB]- crossjeaxrxxing eight different ones tested,cariying all the
e ght possible combinations of three markers).

It may be mentioned thet the cross TLB,-st x Pr,when fertile, does not re
quire previous incubation of the mixture in broth, before plating on minimal st
although incubation would inerease the yield by @ factor of 10x or more.

Heterothallism : see discussion .

Trnesmission of F+,page 3,

In bouilion, incubation at 37°, initial amounts of either "+ and F- types about
107/ml. After 4h, 13 out of 14 had become T+ ; after 8, 19 out of 16 ;2fter
24h, 15/15 ; after 48° 16/18. No egohange of markers bbserved in these strains
to which F+ had been transduced. The transmission can occur also in the pre-
sence of streptomycin when tne F+ donor strain is s§; the yield is still high,
but there was no wk@ high bactericidal action of streptomycin in these condisti
tions,

Adding raw DNA-ase tou bouillon,then 42 ineubation of mixture: 14/36 infected.
Controls without DNA-ase, 25/36.

Experiments with raw DNA-ase with and without citrate 1%. With citrate:
transduced F+ 4 out of 7. Without citrate 1/8 (difference not Significant).
5

in a similar experiment the treatments were started before Iixing F+ and F-
cells,and incubation together shortened to 34, Results with citrate 1/10
infected and without citrate 6/10. These experiments should better not be
quoted until they will be repeated with purified DNA-ase,now available.

Trials to get F+ off cells :

filtering of broth or minimal cuitures : ineffective.

grinding F+ cells with alumina ,with and without prior UV irradiation ,then

filtering (membranfiltres): ineffective.

heating F+ cells at 60° half en hour,and also various other times and tempera-

tures,then 3; Pr.

1) adaing directly to F- celis: doubtful results (cultures not
sterilised)

2) killing with chloroforn, ineffective.

3) adding streptomycin (also after treatment at 508,a/2 hour)

ineffective,

growing with citrate,or grsenate at various conens.,then killing with chlo-

roform : ineffective.

penicillin lysis,then filtering : ineffective.

Discussion,

1) inheritance of self-compatibility appears as extranuclear or as infective
according to the angle under which it is looked at. The ménage a trois exp.
might have been interpreted as due ‘to 1) F+ hormone, 2) transduction ; but
the fact that inheritance of F+ is extranuclear,zs snown by filial strains,
that filtrates are ineffective,that the infection is possible with a very
high yield it seems that there is no need of assuming an F¥ hormone in aduitio
to F+ infection.

2) nature of infective agent,? DNA-ase experiment should,i think,be carried
out in both laboratories,as it seems an essential point.

3) Are @11 falialistocks from an F- x Ft cross F+ ? If s0,Since transduction
in minimal is limited , + transduction must play an essential role in the
eross, Pe.haps if DNA-ase sensitivity is confirmed, F- prototrophs might be
ohtained crossing in presence of DNA~ase.

4) However,the possibility of having cells with F- phenotype and F+ genotype
(aereation, pantothenicless) shows that fertility and presence of F+ are not
necessarily the same thing.

5) The problem of heterothallism is not settled » in your/t#Eters you assume
it possible that maky some of the celis of an F+ stock are phenotypically Fe,
and that only F+ x P- crosses are pernitteds then we are exactly in the same
position .s Paramecium, variety 1. The comparison of crosses 1,II,III above
seems to support to some extent the idea that only F+ x F- crosses are per-
missible,

6) Correlation of F+ with other effects :

a) Hayes's effect I (St-resistance of crossability in an F+ strain).It
may be wotth seeing whether this ef ect is extended to all F- stocks,
Also,if an exp where St-treated F+ x St-treated F+ is non-fertile ( T
asked Hayes kgxdexikk} if he could do it) then it must be assumed that
the contributions o° the two parents are different,in the sense that
there is anisogamy, a male-like and a female-like gamete(The latter may
be the normal cell), It would also lead to conclude that only F+x F-
eross:s are permissible, i.e. K-12 is heterothallic,
b)

c)

qd)

f)

: 6
Hayes's effect II (UV enhancement of #x fertilityh infn F+ strain).
Again,it is possible that F+,Hayes's effect I (is it what you call Gs?)
and Hayes's efiect BI may be the same thing but a more extensive
testing of F+ and F- stocks would be needed, ;
there may be a slight difference in cultural behaviour after UV of F+
and F- strains,the latter tending to form longer snakes ( I am not
quite sure of this).
derological data. Spicer (unpublished) found a serological difference
between W677 and 58-161,
smooth-rough or similar changes are not morphologically apparent, but
Maccacaro (unpub.) finds that some F- strains (W 677 and related)
are strongly agglutinated by NaCl 5% while some F+ are not. TI have
suggested that he trices acriflavine as well.
would it not be better to give a hint that F+ affects segregations in
some orderly,not fully explained way.