PUBLIC HEALTH LABORATORY SERVICE
(Directed by the Medical Research Council for the Ministry of Health)

Telephone: COLINDALE 7041 (8 lines)

COLINDALE AVENUE,
Telegrams: DEFENDER, NORPHONE, LONDON.

Lonpon, N.W.9.

 

10th February, 1953.

Dear Dr. Lederberg,

I must apologize for not having written before to thank you for
your letters of the 5th and 23rd January and the attached Summary of the
paper by lirs. Lederberg and yourself on "Genetic Studies of lysogenicity in
Escherichia coli", The impressive collection of reprints has also arrived
safely. I am rather slow in writing letters and, in addition, I waited a
few days to have some of the old cultures ready for despatch before writing
to you.

Naturally, I studied the contents of your two letters with great
interest and very much enjoyed some of your remarks. I must edmit that I do
not feel competent to follow you into the vast realm of genetics, but somehow
I feel entitled to dislike and distrust Luria's "parasitism at a genetic
level", To my simple mind it eppears that what heppens at this level are
reactions that will be clarified one day by enzymologists rather than by
virologists of the type I knew in the older generation. I know I shall not
live long enough to witness this development. On the other hand, your ow
work end that of your school is advancing so rapidly that I seem to have a
good chance of reading one day that the parasite theory of bacteriophege is
untenable. I do believe that the only basis on which bacteriophages may be
“specific self-reproducing units which form part of the genetic make-up of
the bacterium", is the one assuming that they ere "endogenous products of
the bacterial cell." I do not believe that "the question of endogenous vs.
exogenous devolves mostly on e question of how long ago" (see page 2 of your
letter), or that there is any tfurther evolution of phage into a more
independent organism" (see page 3 of your letter). You will appreciate,
therefore, thet I wes Celighted to read the last paragraph of your letter of
the 5th January, where you discussed the results of the recent work on phege
lambda by irs. Lederberg and yourself.

I feel on more familier ground on turning to the various points
relating to serology and antigenic enalysis, which you heve raised in your
letters.

1). Reversion from 0 variant to 0+ H verient in strain 0 901. - It
was certainly a surprise to me to learn thet the cultures of strain 0 901
which you received from Keuffmann and Boulgakov readily revert to the 0 + H
form, It is true that in my own laboratory O veriants of any organism
hendled, ere, es a rule, maintained on ary agar slopes plates, free from
water of condensation, to avoid encouragement of development of H entigen.
Eowever, since this strain has been first recommended for use in the
"Qualitative Serum Diagnosis of Enteric Fevers" (Lancet, 1930, 1, 505), it
has been distributed to many workers throughout the world by the Netional
Collection of Type Cultures or by myself, end I have not heard before of its
reverting to the parent form H 901.

I enclose a reprint of a peper on the "'Stendardization of Diagnostic
Agglutination Tests", published in the Bull.World Hlth Org., 1950, 2, page
643, which I had not sent you before. On page 645, line 1, you will find
the following statement:

"(2) in most laboratories throughout the world the suspensions for
the estimation of TO end BO agglutinins are now mede from cultures of the
permenent O variants of the two special strains recommended for the purpose
by Felix (1930)."

In this country the suspensions distributed by the Oxford Standards

//

 
Laboretory were made for more than ten years from broth cultures of these two
strains; nevertheless, reversion to the 0 + H variant hes not been observed
(see also Felix & Gardner, Bull.Hlth. Org. Lo.N., 1937, &, on page 234).

I sent you today by ordinary airmail a parcel containing duplicate
Lemeo stab cultures of the two streins H 901 end 0 901. The cultures were

specially prepared from the oldest stock cultures evailable and ere labelled
as follows:

0 901 No,l derived from old Lemco stab culture dated 18.9.1935
0 901 No.2 derived from old Lemco stab culture dated 1.2.1938
H 901 No.1 derived from dried culture dated 56.1936

H 901 No.2 derived from old Lemco stab culture dated 24.2,.1937

I took the precaution of going back to the oldest aveilable stock
cultures because you are employing these strains in experiments with various
bacteriophages, and I wanted to make sure that the cultures have not been
accidentally exposed to any bacteriophages in my laboratory, where bacterio-
phage work dié not start until 1940. I shall be very interested to hear
whether these two cultures of 0 901 can readily be made to revert to the
O+ H variant.

2). Anti-O phages - I am afraid you will find the information I am going
to give you on this subject rather disappointing. The source of this
disappointment is Kauffmann's wrong teaching. <At-first he neglected the
existence of/humerous partial 0 antigens in the complex 0 entigens of the
various Selnonellae; subsequently, he and his followers attempted to carry
the differentiation of the various O-antigen complexes into minute details,
without realizing or admitting the limitations of this technique. Consequently,
the information given in the Keuffmann-White scheme under the heading
"0 antigen" has always been misleading. In the early years it was an over-
simplification; later on the tables conveyed the impression of a degree of
accuracy which, in fact, they did not possess. Undue reliance on the slide-
agglutination technique was one of the reasons for those mistakes.

You mey have seen my recent pepers in the December number of the Journal
of Hygiene, 1952, 50, pages 515-579. I mey refer you to pages 525-526, and
again to psge 546, where the question of the complexity of the Salmonelle 0
antigens is briefly discussed. The reprints of the four papers heve just
arrived and I am posting them today under separate cover by ordinary mail.

Table 1 of the Oxford Symposium, to which you referred in your letter
of the 23rd January, was primarily intended to show in « general way the
remarkable parallelism between phage action and antigenic structure. At the
same time the table showed that the lytic action of Anti-Vi end Anti-H phseges
is "relatively specific", whereas those of anti-O and anti-R phages are
"widely over-lapping". The anti-0 phages referred to in the table are
exemplified by the three pheges (Numbers 1, 2 and 3) I sent you recently,
whose lytic spectrum extends over the great majority of Salmonelle species
belonging to all the different O groups in the Kauffmann-White scheme. I
thought thet the "widely over-lapping" i.e. relatively non-specific action
of these phages was adequately indiceted in the table.

In the original paper (Brit.med.J., 1943, 2, 127) these anti-O pheges
were described as follows: "It was soon found that most of the phages
obtained from these sources were anti-O phages thet acted on paratyphoid B
bacilli as well as on Bact. typhi-murium becilli and many other members of the
Salmonella group. Even minor O-antigenic components thet are common to the
various Salmonella species, though they are not listed in the Kauffmann-White
diagnostic scheme, provide en adequate point of attack for anti-O bacterio-
phages." (See page 2, lines 9-15 in the reprint of the 1943 paper which I
sent you last year). On the same page, line 32, you will find that these
pheges were referred to as tthe non-specific O phages" as contrasted with
“the specific Vi phages".

It is obvious from these quotations that I never expected to be able to
identify, by serological methods, a presumed common receptor of these O phages.

{/

 
~-3-.

I know, of course, that other O phages have a much more restricted lytic
spectrum, Nevertheless, I would be much surprised if it were possible, even
in those instances, to associate the phage receptor with e serologically
identifiable QO-antigen fraction.
Vi-negative

Broedly speaking, I would define as O phege any phage acting on a
bacterial cell that still contains some "smooth" O antigen, demonstrable by
serological methods. The quantity of antigen present may be very small indeed,
and mey not be demonstrable by any test in vitro, only by immunization of
rabbits. A pertielly rough varient, which is salt-agglutinable and shows
other criteria of "roughness", may nevértheless contain a considerable amount
of "smooth" O antigen,

In this connection I would like to refer you to the paper published
with Anderson in the Journal of Hygiene, 1951, 49, page 349, dealing with a
strain of Salm.typhi that contained the TVi antigen in such a small quantity
that the cultures might easily be passed as Vi-negative forms, when tested by
the customary routine technique. These cultures were, nevertheless, fully
susceptible to Vi phages.

3). Anti-R phages. - This may be the place to mention anti-R phages.
You wrote in your letter of January 23rd: "I wonder whether a second phege,
possibly rough-specific, mey not be responsible", in order to explain Boyd's
seemingly contradictory finding that "Al" lyses S.bovis-morbificans, In the
light of whet I said in the preceding paragraphs the assumption of a second
phage does not appear to be necessary. I should have mentioned before that
each of the three 0 phages I sent you had, of course, been purified by
repeated single-plaque isolation, so that the question of a second phage could
not arise,

The anti-R phage listed in Table 1 of the Oxford Symposium lyses
"rough" variants of all the different Salmonella species that heve been tested
against it, but does not ect on the corresponding "smooth" veriants. This is
in good agreement with Bruce White's finding about the serological, and
presumably chemical, identity of the R antigen throughout the whole Salmonella
group. To be exact, I ought to mention thet in ell the bacteriophage tests
I have been discussing only lysis has been considered, not absorption of the
pheges.

4.) Anti-H phage. - You may be right in what you wrote ebout insufficient
study of this phage. I have had no personal experience of it and included this
phege in Table 1 on the euthority of the authors quoted in the text, most of
whom I hed good reason for eccepting as competent workers. After the death
of my old friend Schiff I tried to obtain the anti-H phage from New York, but
it was no longer evailable.

& few weeks ego Dr. Stocker sent me Boulgakov's phages (Phage VIII-113
propagated on 377 and Phage VIII-113 propagated on 372) the first of which he
had received from you. I have not used these phage preperations so far.

Sertic left the Paris laboratory more than ten years ago, and since it is a
commercial laboratory I would be reluctant in accepting Boulgakov's phage as
en authentic derivative of the phage originally described by Sertic and
Boulgakov (1936) and later employed by Schiff end his co-workers.

5). Booy and Wolff's paper. - I had some difficulty in tracing this
paper, which hed escapéd my notice. However, a few days ago Professor Dinger,
the Director of the Institute at which Booy and Wolff are working, wrote to
mé, and this gave me an opportunity of telling him ny opinion of the paper.
Insteed of re-writing it I enclose s copy of the letter. Naturally, this is
intended exclusively for your personel information.

6). Professor Creze has not written to me end I have not read anything
published by him, I am not quite clear about what you meant by "recommending
his address" to me, So far, I do not know Professor Creze's address.

7). Although this letter has now become very long I would Like to add
one more remark. You wrote that the O phages I sent you "will not be directly
usable in transduction experiments as they leave very few survivors from most
of the Salmonella cultures exposed to them". In experiments with these phages
some years ago I never succeeded in sterilizing cultures of e« number of

different strains of Selm.typhi. Neither ean O phage nor a Vi phage could
I

 
accomplish thet, but a mixture of the two produced complete sterilization.
This is mentioned briefly in the Proceedings of the 4th Internationel Congress
for Microbiology, Copenhagen, 1947, on pege 363. On the basis of this
experience I assumed thet most of the Salmonelle cultures exposed to these

Q phages would leave surviving resistant mutemts in numbers suitable for your
experiments.

I hope this time the small parcel will be treated with less suspicion
by your customs authorities, I would like to add that, for some obscure

reason, these Lemco cultures survive better et room-temperature than at
2° to 4%.

Yours sincerely,

Dr. Joshua Lederberg,
bepartment of Genetics,
‘ University of Wisconsin,
College of Agriculture,
4gricultural Hall,
Madison 6, Wisconsin.