BIOLOGICAL RESEARCH IN CANADA
I.T.Co. Canadaq Ltd. have an arrangement with Dr.
J.P.W. Gilman
(Ontario Veterinary College, Guelph# Ontario) and
Dr. G.F. Wright
(University of Toronto) whereby Dr. Gilman carries
out biological tests
on smoke condensates produced in the laboratories
of I.T.Co. Canada, Ltd.
Where possible this 13 financed by grants from the
National Cancer Institute,
but where the N.C.I. may not be willing to provide
financial support,
I.T.Co. Canada would defray the cost. Under these
arrangements, there is
no restriction upon Dr. Gilman publishing his
findings, but with the close
liaison established between him and Messrs.
Laporte and Wads, it Is unlikely
that embarrassment will ever be caused.
The work which has boon carried out to date falls
into three investigational-
(i) A comparison of the biological effects of
condensates from
U.S. cigarettes and Canadian cigarettes.
(ii) A comparison of the effects Of condensate
from cigarettes
made from unextracted and extracted tobaccos.
(iii).An evaluation of Polish cigarettes said to
be manufactured
by a process akin to that advocated by Dr. J.
Beffinger,
namely, fermentation following flue curing.
The biological tests are run in conjunction with
chemical analyses of
the smoke condensates for benzpyrene, etc. made in
the laboratories of
I.T.Co, Canada Ltd.
Types of Biblooical Test
Dr. Gilman employs two types of biological test.
BATCo document for Province of British Columbia 3
November 1999


2 -
(a', Induction 7est (Initiation)
In this smoke condensate is applied to the backs
of mice, a
control group being maintained under similar
conditions. Dosage
levels vary somewhat from experiment to experiment
and where possible,
base-free tar is used In order to avoid toxic
effects due to nicotine.
This test is intended to measure directly the
tumour producing
capacity of condensate.
(b) Promotion Test
A single application of a small quantity of a
proven animal
carcinogen, dimethylbenzanthracent, is followed by
repeated
applications of small quantities of smoke
condensate. Controls are
generally provided consisting of:-
(1) Single application of carcinogen only without
condensate.
(ii) Repeated applications of small quantities of
condensate
without Initiating treatment with carcinogen.
(III) lb treatment. This test is designed to
measure the
co-carcinogenic potency of smoke condensate.
Preparation of Condensate
Smoke condensate produced under standard Canadian
smoking conditions
(a 35 ml. puff of two seconds duration, taken once
a minute) Is collected
in spiral cold traps and is extracted therefrom In
a mixture of methanol
and methylene chloride. The organic solution Is
washed repeatedly with
1 Z hydrochloric acid to remove bases, and after
filtration Is flash
evaporated under rater, pump vacuum at 500C. The
residue Is taken up In
acetone and the concentration adjusted to 10% w/v.
Light mineral oil
Or-
BATCo document for Province of British Columbia 3
November 1999


(10%) Is added to P:event the mouse skin drying up
and peeling after
treatment (a blank treatment of 2D mg. light
mineral oil three times
weekly Is made along the whole dorsal region of
control mice with
nega-tive results). The solution Is then sealed in
glass phials under
nitrogen and stored at -26 + 10C.
It is shipped to Dr. Gilman In polystyrene
containers an cardice
and on receipt by him Is again stored In deep
freeze. The aim Is to
use condensate within four to seven weeks of
preparation.
M Comparison of U.S. and Canadian Ciqarettes
(a) Induction Test
This experiment compared condensate froms-
A. Canadian flue cured cigarettes (one brand).
B. U.S. blended cigarettes (composite of five
standard brands).
C. Tennessee Eastman condensate (from one U.S.
brand
only - PALL MALL).
The mice used were either C3H males and C57 X
C31i-F1 male and female
hybrids. They were shaved in the dorsal region
before treatment was started.
Condensate was applied in repeated doses over a
short time period (e.g. 2DO Mg.
in six doses over thirty-six hours) and then
followed by painting twice a
week with a solution of croton oil (55% in light
mineral oil reducing after
one hundred applications to a 1% solution). The
results of this experiment
are summarised in Table I. Any growth* which
persisted for fourteen days
were scored as -papillomata whether or not they
subsequently regressed.
A second experiment compared different dose
applications of condensate.
BATCo document for Province of BritiSh Columbia 3
November 1999


0
0
CL
0
0
C
0
CD
0
h
W
:r
0
0
z
0
Cr
CD
6601/61001
TABLE I
COMPARISON OF CANADIAN AND U.S. CONDENSATES
INDUCTION TEST
NO. OF SURVIVORS
Condensate At Start At Time of At 600 Time to
Tumour-bearina No. of N6. of No. of
First Tumour days First Tumour Mice (as Aot Tumour
Popillomata Carcinomata
(days) of Survivors at First Bearina
Mice
Canadian cigs. 50 40 9 305 23 9 11 1
A
U.S. cigs. 66 50 10 190 16 8 10 1
B
Tennessee- 40 38 4 210 21 a 10 2
Eastman
C
Control
Croton all only 60


Two groups# each of seventy mice$ were treated.
Group one received a
total dose of 2DO mg. condensate In six doses
spread over thirty-six hours,
and group two received 250 mg. condensate in ton
doses over ninety-six
-hours. Both groups were then painted twice a week
with croton oil.
50% of the mice died during the Initial eight
months without any tumours
appearing. The results are summarlsed in Table H.
W Promotion Test
An initial experiment established that OH males
were more susceptible,
to tumour formation than were the C57 X C3H-F1
hybrids. This was done by
applying 1.5% solution of dimethylbenzanthracens
followed by twice weekly
applications of 5 mg. of a 10% solution of
base-free condensate. The
results (see Table III) demonstrate thatq although
the b45e-free condensate
is only a very week carcinogen, it possesses a
fairly strong co-carcinogenic
effect. (The source of condensate for this test
was Canadian flue cured
cigarettes).
In a subsequent test the promotion activity of
U.S. and Canadian
condensates were compaxedp and it was concluded
that the condensate from
Canadian cigarettes was somewhat less effective in
promoting the production
of tumours. These results are listed in Table IV.
It should be emphasised that the numbers of
animals employed are
probably Insufficiently large to amble statistical
meaning to be attached
to the results. Dr. Gilman himself realises this.
The results vvere submitted for publication In the
British Journal of
Cancer by Dr. Gilman, but were rej*ctod an the
grounds that too much similar
work had already been published.
BATCo document for Province of BritiSh ColUmbia 3
November 1999


>
q
0
0
CL
0
0
C
.9
0
0
W
0
0
Cr
z
0
<
(D
3
Cr
(D
W
10 0 1
TABLE 11
COMPARISON OF MODES OF APPLICATION OF CONDENSATE
Condensate A (CANADIAN)
SURVIVORS TIME TUMOUR BEARING MICE RESPONSE
Group Conan, of At Time At End To Appearance To
end of Survivors Number No. of
Condensate of list Tumour of First Tum*ur of
Experiment at lot Tumour Papillomata
N/v)
10% - .0 440
11 109 23 12 390 26 6 13
Control - 0 - 460 - - -
Condensate C
(Tennessee-
Bastman)
33% 38 1 86 328 13 5 6
10% - 0 330 - -
Croton Oil only - 0 390


TABLE III
CCJdPARISON OF STRAIN SUSCEPTIBILITY OF MICE
AND OF FROMOTIM ACTIOW OF CONDENSATE
Survivors Time (days) Tumour Bearinq Uice
Treatment Start At Time of To First To % of
Survivors no. Favillomata Carcinomata
First Tumour Tumour End at Time of
Ist Tumour
STRAIN C 3H 0
EMBA + Tar A so 55
DMBA only 40 33
Tar A only 20 -
SMIN C 57 X C 3H F1 HY-ILRIDS
DMBA + Tar A 90 72
DMEA only 21 21
Tar A onl y 2D -
259 690 58 32 .45 26
403 62D 15 5 3 4
- 810 - - - -
292 880 15 11 10 3
434 623 14 3 3 0
- 830 - - - -
C=)
BATCo document for Province of BritiSh Columbia 3
November 1999


Im
>
I
0
0
0
0
<
0
h
0
0
ro,
W
z
0
(D
a
Cr
(D
co
TABLE IV
COMPARISON OF CANADIAN AND U.S. CONDENSATES -
PROMOTION TEST
U.S. CONDENSATE
EXP05ure No. of Tumour Total
2 pA Survivors Bearinci Mice Turwurs Carcinomata
No.
I -
(at time Of
(Ist Tu ur~
CANADIAN CONDENSATE
No. of Tumour Total
Survivors Bearina Mice Tumours Carcinomata
I AG--
(at time of)
(1 st Tu-,,T-)
0 75 - - - 60
350 39 3 1 1 0 42 - - - -
400 31 8 3 3 0 38 5 2 2 0
450 27 1e 7 a 0 31 5 2 4 0
500 24 31 12 16 4 28 16 6 11 0
i on


- 5 -
(II) Comparison of Extracted and 1--jon-extracted
Tobaccos
The extracted tobacco was prepared by percolation
using hexane.
The yields of total condensate from cigarettes
made from the two forms
of tobacco were as follows:-
Unextracted tobacco 27.8 mg./cig.
Extracted tobacco 19.9 mg./cig.
~Condensate weights determined by evaporation of a
methanol and
methylene chloride solution by an airstream, while
heating on a hot
plate followed by drying in a desiccator).
(a) Induction Experiments
The mice used were C31i males obtained from the
Jackson Memorial
Laboratory, Bar Harbor, Maine. The condensate
employed was whole tar
rather than base-free condensate, and the final
dose of 40 mg. tar three
times a week was built up slowly starting with 10
mg. twice a week and
rising after one month to 20 mg. three times a
week. When the activity
of condensate from extracted and unextracted
tobaccos was compared on an
equal weight of condensate basis, no difference in
biological activity
between the two groups could be detected.
(b) Promotion Experiment
A solution of dimethylbenzanthracene (0.25% in
acetone) was used as
initiator and a variety of treatments were
studied.
D.M.B.A. Application Condensate Treatment Number
of Mice
1.5 c.c. 10 mg. Tar A 60
Twice weekly
1.0 C.c. 10 mg. Tar A 30
Twice weekly
BATCo document for Province of BritiSh Columbia 3
November 1999


6 -
D.14.3.A. Apolication Condensate Treatment Number
of Mice
10 mg. Tar A 30
Nice weekly
1.5 c c. 10 mg. Tar B 60
Twice weekly
1.0 C.c. 10 mg. Tar B 30
Twice weekly
10 mg. Tar B 30
Twice weekly
1.5 c.c. 3D
1.0 C.c. 20
Tar A from extracted cigarettes
Tar 9 non-extracted cigarettes
Although both tars A and 3 had promoting activity,
there was no difference
in statistical terms between them. At the levels
at which the tars were
applied, only papillomata were found.
Further Analyses of Smoke
During the period of biological testing, the
I-T-Co., Canada smoke
laboratory carried out further comparisons of the
smoke composition of
cigarettes made from extracted and unextracted
cigarettes in order to
substantiate and confirm previous results. In this
series of experiments,
estimations w-re made of 3i4-benzpyrene, and
although a reduction in the
amount of this compound per cigarette was found,
when this was expressed
in parts per million of tar, no reduction was
found. The following
table illustrates-the findingsa-
C=)
BATCO document for Province of BritiSh Columbia 3
November 1999


7
Cigarettes Made Run Tar 3:4-benzpyrene
From
cig.
Reduc-
.,it./cig. educ-
% R
PPM-
M.g. tion Ugm. tion
Unextracted 1 27.7 2.6 0.94
tobacco 2 3c.. 6 3.4 1.1
Extracted 1 24.2 12.6 2.4 7.7 0.99
tobacco 2 26.6 13.1 3.o 11.7 1.1
(III) An Evaluation of Polish Fermented Flue Cured
Ciqarettes
This study was initiated to Investigate the theory
propounded by
Dr. J. Beffinger that fermentation following flue
curing (and in place of
re-drying, which causes a "pasteurization") leads
to a smoke which is
notably less biologically active. As a result of
the initiative of
Dr. Wright, special cigarettes were made by the
Polish ,.onopoly in rhich
the tobacco was processed by a chamber
fermentation following flue curing.
This forced enzymatic fe=entation is the process
which Dr. Beffinger
claims to have developed in 1,932, and which is
said to be used in Poland,
and later in the U.S.S.R, to process all cigarette
tobacco leaves. Details
of the process are given in a memorandum dated 2nd
September, 1960,
"Polish Fermentation Procedures" prepared by R.S.
Wade.
Chemical Investiqations
Details of the chemical investigations of the
Polish cigarettes, and of
a typical Canadian flue cured blend are given In
Table V.
Several interesting facts emerging from this
comparison:-
(I) The nicotine content of the Polish tobacco is
very low, yet
the nicotine content of the smoke derived
therefrom in
c::>
BATCo document for Province of British Columbia 3
November 1999


TABLE V
Analytical comparison of Polish Fermented Flue
Cured and Canadian
Flue Cured Tobaccos and Smoke
TUBACCD A114LYSIS POLISH CANADIAN
Nicotine content o.6 1.8 2.2
Sugar content (%) 22.4 - 21.7 19.5 22
pH (2 cigs in 50 ml. HP) 5.00 5.25
CIGARETTES
Length (mm.)
Butt length (mm.)
Length smoked (mm.)
Av. weight/cig: (Mg.
Av. weight (mg /mm-)
No. of puffs to smoke above length
63
10
53
986
15.6
14.8 - 14.9
74
23
51
1125
15.2
10.3
SMOKE A,'TALYSIS
Whole tar (mg./cig.)
Snell
Cold Trap
Base-free tar (mg./cig.)
(% of whole tar)
Neutral tar (mg./cig.)
(% of whole tar)
Nicotine (mg./cig.)
Phenols (mg./cig.)
(% of whole tar)
pH (I cig. in 75 ml.)
Benzpyrene (pg./100 cig.)
ppm. of whole tar
Pyrene (pg./100 cig.)
ppm. of whole tar
24.6
25.2
15
60
4.98
19.8
1.39; 1.42
1. 2B
5.1
4.7; 4.7; 4.65; 4.75
4.2
1.6
12.6
4.7
28
34.1; 36.0
16
55
7.90; 7.67; 6.81
2 D - 21.9
2.4
1.54; 1.40; 1.24
3.6 - 4.3
5.2; 5.25
2.9 '3.8 - 64 mm. smoked
0.91,10-B91 10 mm. butt
7.6
2.4
CD
C__
__j
BATCo document for Province of BritiSh Columbia 3
November 1999


RESEARCH & DEVELOPMENT ESTABLISHMENT.
British-American Tobacco Co.. Ltd.. Southampton.
- 8 -
comparison to that from Canadian tobacco, is not
so greatly
reduced. It would seem, therefore, that the Polish
tobacco
has a much higher nicotine transfer co-efficient.
(ii) Despite the fermentation process undergone by
the Folish
tobacco, the sugar content of this material is
unusually high
(a ppro x. 22,46) .
(iii) The acidity of the Polish tobacco is greatcr
than that of the
Canadian tobacco (5.00 as compared rith 5.25).
This again
is not characteristic of fermented tobaccos
gonerally (cigar
leaf has a PH of 6.8 and Burley 5.95 to 6.0).
(iV) Although previous expcrience would suggest a
fermented tobacco
would give an alkaline smoke, the iolish fermented
tobacco
yields a smoke more acid than that of Canadian
tobaccos.
(v) The tar yield from the Folish cigarettes is
les5 than that from
Canadian cigarettes, especially when it is noted
that the Polish
c.garettes were smoked to a much shcrter butt
length, since
they were 63 mm. long to begin with. The
percentage of base-
free tar is about the same as that from Canadian
cigarettes,
and the neutral fraction is also unexceptional.
The Folish
cigarettes yield somewhat more benzpyrene an:J
considerably more
pyrene, eithr-,r on a per cigarette or parts per
million condensate
basis. Again this may be a reflection of the
different length
of cigarette to start with.
Indeed, from the results found, one would not have
concluded that the
Polish tobacco had undergone fermentation to the
extent known from past
experience.
CD
BATCo document for Province of BritiSh Columbia 3
November 1999


TABLE VI
COMPARMN OF POLISH (FJLLY FEW[ENTED) FLUE CURED,
CANADIAN FLUE CURED AND U. S. BLEIZED TC3ACC0 WOKE
MICENSATES
Induction Test
Tar
Polish fermented
Canadian Flue
Cured
U. S. T31 ended
3 ase-free tar (20 mg. - 3 x weekly) C3H nice
Respon se at 500 days
No. Survivors At 1110. of Tumour % Mice" Total
Flice at 350 Days Fnd Days Total Bearinq Vlith
Tumours
Exposure Mice Tumours
No.
40 12 8 493 O(l) O(Bi) O(l)
60 42 28 500 6 14 11
75 39 24 500 12 31 16
NOTES X Calculated from survivors at 350 days. The
effective number of
test animals for the Polish tar is only twelve. By
comparing
the total number of tumours in the other two
groups with the
relative numbers of survivors, one can predict an
expected four
tumours in the remaining twelve mice exposed to
Polish tar.
The difference between Canadian and U.S. tar is
not significant.
C=)
'D
BATCo document for Province of BritiSh Columbia 3
November 1999


TABLE VII
GLIVARISON OF POLISH (FULLY FERMENTED) FLUE CURED,
CANADIAN FLUE CURED AND U.S. BLENDED TOBACOD SWKE
CONDENSATE
Promotion Test C 311 mice given a single exposure
to 0.75 mg. of 0M]BA as 1.5% solution in
mineral oil, followed by 5 mg. base-free
condensate applied twice weekl'y.
z
0
<
CD
3
Cr
(D
(0
Treatment
Control DMBA only
5 mg. Polish base-free
tar 2 X weekly
Canadian base-free tar
(fresh)
U.S. base-free tar (old)
Control Polish tar only
01 0 it
No. of No. of
_
- 140. of Tumour Total Total Tumours
mice aE. survivors Bearinq Ace Tumours Carcinomata
Tumour Bearinq Mce
No. %
43 493 2 2 4.6 2 7 1
40 493 1 5 12.5 5 1 1
30 493 1 10 33.3 13 6 1.3
30 493 2 7 23.3 11 4 1.6
40 493 12 0 0 0 0


'o-ac,-o Co.. -tc.. joutriampton.
Bioloqical Tests
The condensate from Pcl-,sh cigarettes ras tested
by Dr. Gilman in
two ways:-
(a) Induction test using base-free condensate
applied three times
weekly at the rate of 20 mg. per application. The
condensate
was compared sim6ltaneously with condensate from
Canadian
flue cured and American blended cigarettes. The
results are
summarised In Table VI, which includes results up
to August 18th.
Recently one mouse, receiving Polish condensate,
has developed
a pa.nilloma.
(b) Promotion test. Mice were exposed to a single
dose of 0.75 Mg.
dimethylbenzanthracene applied as a 1.55*1'
solution in mineral
oil. This was followed by twice-weekly
applications of 5 mg.
of base-free ccndensates. The results are listed
in Table VII,
which was compiled at the completion of the
experiment.
Corw.ent s
Dr. Gilman realises that, because of the
relatively small number of mice
used (these numbers were limited by the arnount of
tar available) the results
are probably not statistically significant, but he
is very Impressed by the
lack of tumours in both experiments using Polish
condensate. It is
noteworthy that the Polish condensate appears to
be considerably more toxic
to mice than either Canadian or American tar, and
this fact may conceal an
otherwise equal biological activity ( i.e. mice
rece iving Polish condensate
may have died before they could develop tumours).
In view of these findings, however, it has been
decided to pursue the
matter further. Dr. Wright has been asked to
approach the Polish authorities
CD
BATCO document for Province of BritiSh Columbia 3
November 1999


R=-SEARC~ a D=-V=-'-ODMENT ESTABLISHMENT.
British-Americin Tobacro Co.. Ltd.. Southampton.
- 10 -
,;,-ith a request for a further 35,XG cigarettes
made from fully fermented
flue cured leaf. It was intended also to ferment a
quantity of C-inadian
flue cured tobacco and to compare Canadian
fermented and unfermented
tobacco. I suggested that when Dr. ",'right
approached the Polish Authorities,
he sh~i-uld request samples also of unfermented
Polish cigarettes, since there
is no guarantee that the fermentation process, as
carried out In Canada,
will duplicata that In Poland exactly and,
moreover, there is at present no
basis for saying that it is the fermentation
process per se responsible
for the results. It could be that Polish tobacco,
wheth3r feirented or
unfermented, may yield equally interesting
results. t!essrs. Laporte and
Wade accepted this suggestion, and I understood
that Dr. 11right's letter
would be modified accordingly.
Dr. 14'right has been persuaded that it is not the
right tire to inform.
Dr. Beffinger of these results and the draft
letter to the Folish authorities,
which I had the opportunity of reading, made it
clear in pleasant, but certain
terns,that Dr. '-"aright is unhappy about
Beffinger's scientific integrity.
The letter car:--,ed .-.,ith .-It an a,-~F:aol ~o
the Pol---, not to cc.-r-municate with
Beffinger until the present interesting, but
statistically inconclusive,
results had been verific-i ',y a sccond 5crics of
experimcnts. This research
was supported by the N-C.I. and Dr. r?right felt
that Dr. Gilman should
publish it. Gilman, on the other hand, feels that
until he has obtained
statistically significant results, he does not
i-,ish to publish. it vould
therefore, seem that apart from the Polish
authorities, this information will
not be made generally available.
I asked Mr. Laporte v&ether I might inform
T.1-1.S.C. of the results.
Fie replied that he had no objection tc telling
Sir Charles Ellis, but that
C:D
C~:)
BATCo document for Province of BritiSh Columbia 3
November 1999


RESEARCH & DEVELOPMENT ESTABLISHMENT,
British-American Tobacco Co., Ltd.. Southampton
hi would prefer that I did not inform T.M.S.C.
Later he modified this
view and said he left it to Sir Charles and myself
to decide whether
we-inform 1.11A.S-C_ I agreed that if we did so
inform them, I would
immediately write to Mr. Laporte and tell him of
this action.
CZD
BATCO document for Province of BritiSh Columbia 3
November 1999