STUDIES ON THE IN VIVO AND IN VITRO
GROWTH OF TRYPANOSOMA CRUZI
JACQUELINE ADAMS HYNES
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
ACITI01 LED3GI ;TS
I rish to thank Dr. E. C. Bovce for initially suugestinC a por-
tion of this problem and for his constant aid and cncourzceccnt during
the course of the investigation. I am very gratefull for the opportunity
of sc:.'vi as his research assistant on work supported by national Insti-
tutes of Health Grant E-l158.
I am grateful to Doctors Levis Berncr, R. II. DeUitt, R. P. Hussey,
J. R. Redoond, T. 1. Stearns, and H. II. Uallbrunn for helpful suges-
tions concerning ry research, for the loan of equip~ncnt, and for read-
ing this dissertation. I especially wish to tank Mr. William C. Sloan
for the valuable ideas which he contributed while discussinG with no
problems which arose during this investigation.
Some equipment used in this research vas purchased irith funds
Created by the Florida Acadeji/ of Science.
TABLE OF CONTETS
IIRODUCTION .. . . ..... . . . 1
MATERIAIAIS AIMD METHOD ............ . ........ 5
Organifsms and Their Maintenance . . . . . . 5
In vivo Experiments ....... ............ 7
In vitro Experimentz .................... 14
RE ULT . . . . . . . . . . . . 20
In vivo LcEx riments .... ................ 20
In vitro Experiments ........ . . . . 44
DISCLUSIG0 ............... . ........
SIt6IARY ADID COICLUIIO3S. . . . . . ... 57
LITUIATURE CITED . . . . . . . . . . . 58
BIOGRAPHY . . . . . . . .......... 63
LIST j0 TABLE3
1 Mortality of 21-Day-Old Mice Injected with
Trypanozoncs and/or Cortisone. ... . . .. 8
2 Bazol Kediua . . . . .. . . . . . . 17
3 diu 3 . . . . . . . . . . . .. 19
4 Mortality of 17-Day-01d Mice Injected with
Trypanosomes and/or Cortisone. . . . . . . . 22
5 Mortality of 17-Day-Old lice Injected with
Trypanosomes, Cortisone, and Adenosinc Triphosphate.. . 27
6 Mortality of 17-Day-Old Mice Injected with Trypanosomes
and Cortisone, Treated with Thiamine . . . . .. 30
7 Mortality of 17-Day-Old Mice Injected with Trypanosomns,
Cortisone, Thinaine, Mi esiur and Adenosine Triphosphate 34
8 Mortality of 17-Day-Old Mice Injected with Trypanosomes,
Cortisone, Adenosine Diphosphate, and Inorganic Phosphate 38
9 Paraplegia in Infected ce. . . ... . . . .. 43
10 Growth in the Partially Defined Medium of Citri and
Grossovicz (1955a) . . . . . . . . . 45
11 Grawth of Costa Rican Strain in Experimental Media . . 47
LLT 0? FIGURlE
1 Additive Growth Rates of 17-Day-Old Mice Injected
with Trypano3ones and/or Cortisone . . . . . 24
2 Additive Growth Rates of 17-Day-Old Mice Injected
vith Trypanosomes, Cortisone, and Adenosine Triphosphate 28
3 Additive Growth Rates of 17-Day-Old Mice Injected with
Trypanosomes and Cortisono, Treated rith Thioaine. . . 32
4 Additive Growth Rates of 17-Day-Old icec Injected with
Trypanosoaes, Cortisone, Thiamine, Magnesium, and
Adenosine Triphosphate . . . . . . . ... 35
5 Additive Growrth Rates of 17-Day-Old Mice Injected with
Trypanosomes, Cortisone, Adenosine Diphosphate, and
Inorganic Phosphate. ................... 39
Trypanosoma cruzi vas first described and identified as the
causative cent of American trypanosomiasis or Chags' disease by Chagas
in 1909. The disease, once believed limited to the state of Minas
Gerais, Brazil, is now known to affect millions of persons from Mexico
to Argentina (Dias and Laranja, 1948). Naturally infected triatomids
and mammals have frequently been found in Texas, California, and Arizona
(Hall, 1953). The first indigenous case of human Chagas' disease in
this country was reported in 1955 from Corpus Christi, Texas (Woody and
I. ile information concerning its distribution continues to increase,
knowledge of other aspects of the disease remains scanty. There is still
no treatment for the infection (Goble, 1956; Hawking, 1953). Nor is the
manner in which the parasite harms the host known. The idea that the
tryponosome injures its host through the consumption of large amounts of
blood glucose has been discarded (von Brand et al., 1949). Evidence con-
cerning the production of endotoxins by T. cruzi is highly contradictory
(von Brand, 1951). Cameron (1956) believes that the effects of this
parasite are due chiefly to mechanical injury to host tissues. It is un-
likely that solutions to these problems of treatment and causation will
be found until much more basic information on the metabolism and nutri-
tional requirements of T. cruzi has been gained. The present study was
undertaken in an effort to gain some further insight into the physiology
of this organism.
Several excellent reviews of our present day knowledge of the
physiology of T. cru.i have recently appeared (von Brand, 1951, 1952;
LDoff, 1951; Hutner and Provasoli, 1955). It is well established that
culture forms of T. cruzi, which correspond to the form found in the in-
vertebrate host, utilize glucose (von Brand et al., 1949). Ryley (1956)
demonstrated clucose consumption by blood stream forms. Baornstein and
Rees (1952) found aldolase in culture forms of the parasite, and Baern-
stein (1953b) demonstrated tie presence of isomerase and triosephosphate
dehydrogenase. From this, he concluded that T. cruzi probably has the
classical Embden-~'jyrhof-Parnas gljcolytic scheme.
Considerable evidence for the presence in the parasite of a func-
tional tricarboxylic acid cycle has been accumulated. Succinic dehydro-
cenase activity has been shown by Seaman (1953), Agosin and von Brand
(1955), and Ryley (1956). Ialic dehydrogenase and fumarase were demon-
strated by Bacrnstein (1953a), and isocitric dehydrogenase by Agocin and
Weinbach (1956). Von Brand and Agosin (1955) found that the respiration
of T. cruzi was stimulated by the addition of various Krebs cycle inter-
rcediates to the medium. Ryley (1956) detected small amounts of citric
acid in media containing T. cruzi. He suggested that the tricarboxylic
acid cycle was operating in these organisms and that traces of this key
metabolite leaked from the cells. Ryley found no significant differences
in glucose metabolism between blood stream and culture forms. Therefore,
the above findings on culture types of T. cruzi are probably applicable
to the blood stream forms also, and suggest that aerobic metabolism of
the organism is carried out via the tricarboxylic acid cycle.
In view of these data, the report that injection of adenosine tri-
phosphate reduced the mortality of cortisone-injected mice infected with
T. cruzi (Adans, 1954) appeared to be highly significant.
Different strains of T. cruzi vary widely in their pathoenicity
for mice, ranging from a complete avirulcnce to an .etrcem virulence
(HEasch!a, 1947). Since 1951 when Jarpa et al. found that the adtLinis-
tration of cortisone acetate to mice infected with T. cruzi increased
the parasitemia and the mortality rate, this compound has been used to
enhance the effect of the less virulent strains on host organisms.
Cortisone has this same effect on various other spontaneous and experi-
mentally induced infections, including tuberculosis, poliomyelitis, and
certain kinds of malaria. The mechanics of this action is not definitely
!-norn, but there seems to be rather general agreement that cortisone
depresses host resistance to infection rather than having any direct ef-
fect on the infectious agent. It ha been found that the hormone affects
metabolism through its action on various enzymes and that it acts on
lymnpoid tissue, reducing capillary permeability and inhibiting phago-
cytosis, antibody formation, and reticulo-endothelial activity. Any or
all of these effects may be related to its infection-enhancing ability.
This subject has been reviewed in a series of papers edited by Shwartzman
The nutritional requirements of T. cruzi are still unknown. The
chief obstacle to the determination of these grort-h needs and to the
study of their metabolism has been the need for a defined synthetic
medium in which to grow the organism. Until quite recently it was be-
lived that T. cruzi could not be grorn in the absence of erythrocytes
(Little and Olecon, 1951) or scrum (Liuoff, 1951). Citri end Grossowicz
(1954) then deviscd a liquid medium in which cr stalline serum aolbuin,
hcomtin, and tomato juice co.mpletely3 replaced blood. They refined the
nodiu. further by szubstituting ikojom gyoirth factors for the tomato
juice (1955a). This modiuu supported satisfactory growth of the Cul-
burtson strain of T. cruzi. Sea an (1957) suc&ested that their medium
might be rather selective for this strain, since several other strains
did not grow wcll in it.
The aims of the present study rsrc: (1) to repeat Adams' (1954)
experiments vrith adenosine triphosphate, (2) to extend his workl to in-
clude the study of the effects of other compounds involved in carbohy-
drate etabolism on trypanosome-infected mice, (3) to test the ability
of Citri and Grossovics's medium (1955a) to support crorth of T. cruzi
strains other than the Culbertson strain, (4) to define further the
partially defined medium of Citri and Grossowicz.
ATER IALS AND MTTDS
Organisms and Their Maintenance
The "Brazil" and "Costa Rica" strains of Trpanosoma crusi were
used for in vivo studios. In addition to these two strains the Culbert-
son strain ~ras utilized in the culture work for comparative purposes,
since it u s for this strain that Citri and Grossorics (1954, 1955a)
devised their media. Axenic cultures were used throughout this study.
The Brazilian strain was obtained from Dr. R. G. Yeager (Tulane
Iledical Center) who had maintained it in a diphasic medium consisting
of Difco brain-heart infusion agar plus 25 per cent defibrinated rabbit
blood overlaid with 2 to 3 ml. of sterile Locke's solution. This
strain was originally isolated from a Brazilian patient in 1942 (Goble,
1951) and has been maintained in several laboratories in this country
since that time.
The Costa Rican strain was secured in Costa Rica by Dr. Herbert
Johnstone of the University of California IMedical Center (Noble et al.,
1953) and was given to me by Dr. E. R. Noble (University of California,
Santa Barbara College). The latter grew the organisms in a medium com-
posed of glucose, NaC1, and peptone to which was added red cell coagulum
or hemoglobin coagulum (McRary et al., 1953).
Dr. N. Grossoricz supplied a culture of the Culbertson strain
which had been maintained at Hebrew Lkiversity, Jerusalem, on Adler's
medium (Adler and Theodor, 1926) since 1949. The early history of this
strain is incomplete (Hauschka, 1947).
All three strains grov tvll in a modification of the medium used
by Younger. This modification is prepared as follows:
Difco brain-heart infusion broth 3.7 ga,
Bacto-agar 1.5 gm.
Pyrex-distilled water 100 ml.
Thio was autoclaved for 15 minutes at 15 p.o.i., ccoled
to approximately 460 C., and to it was added:
Whole blood 10 ml.
Citrated (ACD) blood discarded by the J. H. Thomas Memorial Blood Bank,
Gainesville, Florida, was used and found to produce highly satisfactory
Growth. Each slant of this medium was overlaid with 5 ml. of sterile
Locke's solution (NaC1, 8.0 go.; KC1, 0.2 gm.; CtC12, 0.2 gm.; KH2PO4,
0.3 gn.; dextrose, 2.5 gm.; Pyrex-distilled water, 1000 ml.) to which
were added penicillin (60 ag./ml.) and streptomycin (100 ag./ml.) to
decrease the risk of bacterial contamination. Stock cultures of all
strains were grown in this medium at 270 C., and subcultures were made
every 2 to 3 weeks.
All of the mice used in this study were bred in the laboratory
from a small initial stock of white mice, CF Ho. 1 strain, obtained from
Carworth Farms, Inc., New City, Naw York. Although this procedure
limited the number of mice available for experimentation, it did as-
sure that the nice used were in good health and that they were of th=
desired age. The mice were permitted to feed ad libitun on Purina
Laboratory Chow, and a constant supply of water was provided. Young
mice were weaned at 28 days of age.
In vivo Experiments
1. Injection of mice. Mice to be infected were injected with
both cortisone acetate and trypanosomes, since the Brazilian and Costa
Rican strains of T. crusi are only moderately virulent. Five g. of
cortisone acetate (saline suspension, Cortone Merch) wer injected
subcutaneously. One-half ml. of trypanosome-containing overlay from
two- to three-ueek-old stocl- cultures was injected subcutancously or
intraperitoneally. The split litter technique, with consideration for
sex, was used in all experiments.
At the beginning of this study, it was felt that it night be de-
sirable to inject the mice after they had been weaned. The earliest age
suggested fo weaning is 21 da:s, and 28 days is the recommended time
(Snell, 1941). Culbertson and Kessler (1942) found that when mice under
25 days of age were inoculated with T. cru.i (strain not indicated)
they acquired an infection of greater intensity than that exhibited by
older mice and Generally died, while mice above this age usually sur-
vived. Therefore, three litters of weaned 21-day-old mice were injected
to determine whether infections could be produced at that age. Each
litter was divided into the following treatment groups: (1) no injec-
tion (no inj); (2) cortisone (cort); (3) trypanosomes, Brazilian (Br) or
Costa Rican (CR) strains; (4) trypanosomes and cortisone. At the end of
60 days only two mice (and one of these not infected) had died (see
Table 1). With such a low mortality rate, it would be difficult to
determine rwether injections of adenosine triphosphate or other compounds
MORTALITY OF 21-DAY-OLD MICE INJECTED UIT TRYPAH'OSOMS AThD/OR CORTISOME
Br + court
CR + court
io. days between
injection and death
reduced the number of deaths resulting from the disease. For this
reason, 17-day-old mice, which are more susceptible to infection with
T. cruzi, were used for all of the experiments in this study, following
the practice of Adams (19514).
IIaeoocytomctr counts of trypanosomes used for injection were
made by diluting the culture with a 1:10 dilution of neutral formalin
(3641 formaldehyde) in distilled water. or i were counted in
80 squares (as recommended for crythrocyte counts) with a magnification
of 430X. Duplicate counts were made in each case and the results of the
two cveraced to Give the number of organisms per ml. of sample.
Other injections were administered as indicated in the description
of the individual experiments.
LExprimental mice wore weighed on the day of injection and eery
other day thereafter for 1 month. Veifits were taken every fourth day
for a second month.
2. IIistoloical techniques. All dead mice were autopsied, and
macroscopically visible changes in internal organs noted, as soon after
death as the animals were discovered. Tissues for histological study
were irr-ediately fixed in Bouin's fluid, and 8, paraffin sections were
stained in Harris' haematoxylin and counterstained in a saturated solu-
tion of orange G in 95 per cent ethanol. Spinal and brain smears were
fixed in methanol and stained with Wright's blood stain. Fresh tissues
to be examined w~re teased or macerated in a drop of Neff's (1955) basic
salt solution (NaC1, 120 nZ.; .IC12.6120, 3 mg.; CaCl2, 3 rag.; FcS04.7T 0,
3 ma.; 0.1.1 KirP04 and 0.2O1I IIa'q, 20 ril. to adjust pH to 6.5; Pyrex-
distilled water to 1000 ml.) and were examined under variable phase con-
trast at a magnification of 400X with an AO-Bakcr Interference licroocope.
3. Trypanosomcs and/or cortisone. A series of experircnts was
performed to determine what percentage of injected mice died, the number
of days clapsing between time of injection and death,and the Growth pat-
tern (based on weight) of injected nice. These data provide a basis
cainst which to compare the results of later series of experiments in
which the mice =erc, in addition, injected with nucleotides or thiamine,
and also serve for comparing the effects of the Brazilian and Costa
Rican strains of T. crazi.
Ten litters of 17-day-old mice were used in these experiments.
rEac litter was divided into groups, usually four, each of which was
treated in one of the following ways: (1) no injection, (2) cortisone,
(3) trypanosomes, (4) trypanosomes and cortisone. Occasionally, large
litters wero divided into more groups in order to test the two strains
of trypanoscmes on members of the same litter.
4. Adenosine triphosphate (ATP). Each litter was divided into
groups which were injected as follows: (1) trypanosones and cortisone,
(2) trypanosomes, cortisone, and ATP, (3) cortisone and ATP. Later,
same of the litters were further subdivided to test the effects of dif-
ferent counts of ATP on members of the rsme litter. The trypanosomes
and cortisone were injected as previously described, and 8 hours later
the ATP solution was injected subcutaneously. Nine litters of mice were
used in these experiments.
The disodium salt of ATP was dissolved in sterile Locke's solu-
tion to give the desired concentration. The solution vas prepared im-
mediately before it was to be injected, for ATP bre-ks down rapidly in
To repeat Adams' (1954) worh, 5.0 mg. of ATP were given to each
mouse. When it was found that more of the infected mice treated in this
way were dying than those untreated, the amount was reduced to 2.5 mg.
5. Tliamine. Several mice in the preceding experimental series
7hich had been injected with Brazilian strain and cortisone displayed
symptoms similar to those of thiamine deficiency (Morris, 1947; Woolley
and White, 1943). They were unable to retain their balance and rolled
over and over; they went into convulsions when held by the tail; and
spasticity of the legs and feet was apparent. These observations, coupled
with the reports that culture forms of T. cruzi may produce appreciable
counts of pyruvic acid (Chang, 1948) and that blood stream forrs pro-
duce at least small amounts of it (Ryley, 1956), suggested that a thi-
amine deficiency might be involved in the course of the infection, since
thinmine is required for the metabolism of pyruvate via the Krebs tri-
carboxylic acid cycle. It seemed possible that: (1) the trypanosomes
were using thiamine in amounts that caused the host to become deficient,
(2) the trypanosomes ere producing pyruvic acid to such an extent that
the mouse vas unable to metabolize it, and so the accumulated pyruvate
proved toxic (Robinson, 1951; Grant and Fulton, 1957), or (3) an inde-
perndntly occurring thiamine deficiency might exist in scae mice, per-
haps making them more susceptible to infection. Mice were, therefore,
treated with thiamine to see what effect it might have on the infection.
Each litter was divided into three treatment groups: (1) try-
panocomes end cortisone, (2) t-rpanosomes, cortisone, and thiamine, (3)
thiamine. Ilice from two litters were given 60 "g. of thiamine hydro-
chloride (amount adopted from Woolley and MUhite, 1943) orally on the day
following injection and every other day thereafter for 4 xec!e, giving
a total of fourteen doses. Mice from five litters were injected sub-
cutancously with 25 rg. or 50 *U.. of thiamine hydrochloride on the day
following the trypanosome injection and every other day thereafter for 4
weeks. The thiamine solution was prepared by dissolving thiamine hydro-
chloride in Locke's solution and was sterilized by filtration.
6. Thiaminc, mapgnesiin, and adenosine triphosphate. In the con-
version of pyruvate to acetyl coenzyme A, thiamine is in the form of
thiamine pyrophosphate or cocarboxylase. It has been 1nown for some time
that ATP is necessary for the phosphorylation of thiamine in vitro (Glasa,
1951), and it has recently been demonstrated (Rossi-Fanelli et al., 1954)
that this reaction also occurs in the living animal. Within 1 hour of in-
jecting thiamine and ATP into intact rats, phosphoric groups are trans-
ferred to thiamine, producing di- and triphosphothiamine. Mecnesium ions
ar2 probably necessary for the phosphorylation of thiamine (Glass, 1951;
lieilands and Stumpf, 1955).
Therefore, in conjunction with the thiamine experiments, four
litters of mice were treated with thiamine, I4++*, and AMP. The injec-
tion solution ias prepared to give a 1:1:1 molar ratio of the components.
The desired amounts of thiamine hydrochloride aid MzCl2.6 12 vere dis-
solved in Pyrex-distilled water and sterilized by filtration. The ATP
was dissolved in this solution just before the mice ircro injected. Each
litter was divided into these treatment groups: (1) trypanosomes and
cortisone, (2) try panoso!mes, cortisone, thiamine., M ++ and ATP, (3)
thiamine, MC++, and ATP. On the day following injection of the trypano-
somes and every other day thereafter for 4 weeks, the nice were given sub-
cutaneous injections of thiamine hydrochloride, 200 pg.; MC12.6i20,
120.6 .; and ATP (disodium salt), 370 ag.
7. Adenosine diphosphate and inorganic phosphate. Under physio-
logical conditions the coupling of oxidation and phosphorylation appears
to be obligatory, and both reactions are dependent upon the supply of
adenosine diphosphate (ADP) and inorganic phosphate (iP) (Krebs, 1957).
In other irords, phosphate and/or phosphate acceptors are the rate-limit-
ing factors in oxidation under different conditions u-rich have been
studied (Lardy, 1956), ad I2rcbs (1957) has sucgcsted that both are
probably important in vivo. Since, in most tissues the available con-
centrations of ADP and iP are below; the critical level, the rate of oxygen
consumption depends upon the rate at which ATP splits to form ADP and iP
In vicrr of these facts, it seemed possible that whatever effect
ATP had upon the trypanosorm-infectcd mice might be hastened and/or
enhanced by supplying ADP and iP rather than ATP. Five litters of mice
were divided into those treatment groups: (I) trypanosomes and corti-
sone, (2) trypanoscms, cortisone, DP, cand iP, (3) cortisone, ADP, and
iP. The injection solution was prepared so that an amount of ADP equiv-
olent in moles to 2.5 M3. of ATP would be administered and with the
molar ratio of ADP to iP 1:1. The desired amount of I:aHPO4.12H20 wva
dissolved in Pyrex-distilled uater and the solution sterilized at 15
p.a.i. for 20 minutes. The ADP (sodium salt) was dissolved in the
phosphtc solution immediately before injection. Eicht hours after
receiving trypanosome injections, the mice received 2.4 m-. of ADP and
1.8 mg. of Ia2HPO4.12 1120 ubcutaneously.
In vitro Experiments
1. Inoculation of cultures. To establish initial cultures of
trypanosomes in the partially defined medium of Citri and GrossoVicz
(1955a), several stock cultures of each strain were pooled and centri-
fuied at 1400 rpm (r = 15.0 cm.) for 10 minutes. The organisms were
washed twice in 0.85 per cent NaCl solution and finally suspended in 5
Ml. of this. This suspension was used to inoculate tubes of Citri's
medium, and samples were taken for haenocytometer counts.
Other experimental media were inoculated with washed organisms
grown in Citri's medium.
Cultures were grown in 15x125 mm. Pyrex culture tubes, each con-
taining 5 ml. of medium. These were incubated at 270 C. in a slanted
position. Both cxperinental and stock cultures rere routinely checked
for bacteria, using nutrient agar pour-plates.
Population densities were determined by making direct haemocyto-
meter counts as previously described. The counts from duplicate tubes
were averaged to give a measure of growth for each experiment. All ex-
perimnnts were run at least in triplicate.
2. Preparation of media. water rodistilled in a Pyrex glass
still s used in compounding all media. Vitamins, amino acids, and
henin were obtained from Nutritional Biochemicals Corporation. Concen-
trated stock solutions of vitamins and metal mixes were prepared and
stored in the refrigerator. Because the activity of folic acid deteri-
orates during storage (Tracer, 1957), solutions of it were used within
2 months of preparation. The solution was prepare- by dissolving the
folic acid in a few drops of 0.01U IIaOH, then adding water and heating
Gently. Heoin was dissolved in a mall amount of 511 HI0H, then made
up to volume with Pyrex-distilled vater. Solutions of glucose, henin,
cerun albumin fraction V, thiamine, and riboflavin were sterilized by
filtration through a Pyrex brand fitted glass filter disc of ultra-
fine porosity and added aseptically to the autoclaved medium.
After ex-perimnting vith various modifications on Citri's medium,
it vas found that a Iknom mixture of amino acids, the composition of
which is based on an analysis of p-lactoglobulin (Tristram, 1953), could
be substituted for the casein hydrolysate.' Serum albumin fraction V was
omitted from the medium, various metals were added, and slight changes
;ere made in the concentrations of some other ingredients. The growth
factors and their concentrations terc not altered. The composition of
the medium is shoun in Table 2.
The amino acids were dissolved in three-fifths of the final volume
of boiling water, then the salts were added, and the mixture cooled.
TJeen 80 (polyoxycthylene sorbitan monooleate), cytidylic acid, creatine,
creatinine, and ribonucleic acid were added to this solution, and 1.011
IHaOll was used to increase the pH before the vitamins were added. The pH
was finally adjusted to 8.0, the volume made up with redistilled water,
and the medium autoclaved at 15 p.s.i. for 20 minutes. The remaining in-
gredients, as indicated in the table, were added aseptically to the
cooled, sterile medium.
In Medium 1 the amount of hemin was increased to 2.5 ma./100 ml.
of mndium. Medium 2 was identical to Medium 1 except that the hemin was
added before autoclavinG.
Medium 3 differed from the basal medium in several ways. IHemin
polymerizec in al-maline solution so that part of it may not be avail-
able for use by the organisms; therefore, the heain for this medium was
dissolved in triethanolamino (5 nm. hemin/ml. 50 per cent (v/v) aqueous
triethanolamine) as suSgested by Corrperthwaite et al. (1953). The
amount of triethenolamine added to the medium with the hemin was only
half the total amount to be included; the remainder was added as the 50
per cent (w/v) aqueous solution. The triethanolamine also served as a
buffer in the medium. The concentrations of CaC12, WMSO4.7H20, and
IMtal mi B*-* +
Paxa-cainobnzoic acd 0.01
Folic acid 0.25
Pyridoxinc. Hl 0.2
Iyridoxamine. 2HC1 0.2
S-toci solution sterilized by filtering throuGh ultra-fine sintered
glass filter and added Cseptically to the autocl3vod medium.
i*Stoc;- solution autocl2ved at 15 p.s.i. for 15 min. and added aseptically
to the autoclaved medium.
-l ml. contains: II BO,, 0.057 mr.; CoS04.7T1,0, 0.238 =r;.; CuSO .5HiO,
0.394 G.; Fe So04.7 0 5.0 me.; MnSO.H20, 1.54 r~.; ZnSO.71H, JI.0
zC. (Iatthan and Covpurtlhaite, 1955).
glucose were chn a~d, and ethylencdicnmine-tctra-acctic acid (EDTA) ad
FcS04 .1120 vere added to tie mEdiut.i. Netal mix A vws used instead of
me.tal mix B. These chanZgs are surnmmized in Table 3.
I EDIU" 3
BaEal Ilcdiun vith Follo ing Changes and Additions
Before autoclvving add:
LMetal nix A+
After autoclavinZ add:
*Stock solution sterilized by filtering through
ultra-fine sintered glass filter and added
aseptically to the autoclaved medium.
+1 ml. contains: H3BO3, 0.023 rg.; CoSO4.7T20,
0.095 mi.; Cu SO*.51Lo, 0.039 ag.; MnSO*.1O20,
3.076 mr.; ZnSO4.7I20, 4.398 mg.; Nal, 0.001 mg.;
EDTA, 1.0 Lg. (Nathan and Cowperthrmitc, 1954).
In vivo Experiments
Before the results of these experlucnts could be cumraxzcd, it
:.as necessary to dctor-iiinc hcethcr there Iercr statistically significant
differences aong2 litters ~hiich would ha'e an effect upon c::perimental
results. For this pu-ipos, an analy-sis of variaice i~; carried out on
all the litters (14) which contained at least threc mnice injected with
Lrazilian strain can cortisonc. The per cent rcicht change fromn the
day of injection to the trclfth day after injection in Dr + cort-in-
jccted mice was the chLracter selected for analysis.
Analysis of Variance
Sourcc of CLu of
v2Xil nce D.f. squares Meaan square F ratio
Ectw.een litters 13 60,460.89 4,650.C 4 F = 4650.84/2780.616
Within litters 23 77,857.25 2,780.616 F (13, 28) is
between 2.12 and
Total 41 138,318.14
Therefore, we accept the null hypothesis: the litter means are not cic-
nificantly different. Litter differences, with respect to a basic physi-
ological characteristic, growth rate, are not such as to affect experi-
mental results significantly, and data from similar tests on different
litters may be pooled.
For each series of e:peri.i-'nts a mortality table covering the
O--lday period following injection is presented. A graph portraying ad-
ditive grorbth rate curves of each experitrental group for this period is
also includc-1. Each point on a graph represents the me~ per cent
chenSe in weight on a given day- fro: that ?: days earlier,. ;ith the
weight at the time of injection taken as 100 per cent. Only the weights
of mice whiich rsurvivcd for at least 60 day-s oere used in ccmputing these
curves. The mortality tc-olcs may then be considered to show the effect
of acute trypanosomiazsi, and the growth cures, that of the chronic
Statistical iaalyscs of results presented in the mortality tables
rrec performed using Fisher's exact treatment of the 2x2 table. Selected
points (those fich appeared most lilhly to show differences) on growth
curves emre analyzed using the t-test for the comparison of 2 means
(:;ith pooln- var-iances). The 0.05 level of significance was selected.
1. Trypaniosarmes and/or cortisone. The results of these experi-
rnnts are presented in Table 4.
There was no statistically significant difference in mortalities
between the follo.rin treatment Groups: lo inj vs. cort (P = 0.22);
cort vs. Br -- cort (0.25); cort vs. CR + court (0.15); Br vs. Br + court
(0.13); CR vs. CR + court (0.63). The probability indicated is that of
obtaining values at least this widely separated by chance alone. The
differencesin mortalities between the following groups were statisti-
cally significant: no inj vs. Br. + court (0.0478) and no inj vs.
CR + court (0.028); but that between Br + cort vs. CR + cort (0.82)was
MORTALITY OF 17-DAY-OLD MICE I~ JECTED
ITTH TRYPAIUSO;IES AID/OR CORTISONE
Br + court
C0 + court
I~~~o. nPc Dah
3 .' .
not. Vc therefore reject the null h2 othesis of no effect. for injec-
tion of the mice with tr;panosoics and cortisone does affect their
sur-vival. It may elso be observed in Tibll 4 that the mean numbers of
days cl31sin; betucon injection anl death for the latter two Cgoups
are vor- close.
T'e additive co;th r'.tes for each e::porin::ntal group are shown
in FiCCre 1. The siril.aritics of the cro.tlh patterns shou'n are evident.
Tnceo are especially intcres'tin, since some of the mice represented in
the d~at of Graphs b, e, and f of FiC're 1 had chronic trypTanosomniasi
as cvi.denced by their subsequent deaths ans the appearance of hce,-ily
parasitis. oranms upon cutopsy.
On day 4 only, statistically significant differences exist be-
treen the ric'n per cent vcicht changes of the following groups: no
inj vs. Br + court (0.05 ,P 0.02); Br vs. Br + court (0.01>P 0.001);
no inj v,. CR + court (0.01 P 0.001); CR vs. CR + court (0.01 >P >0.001).
The probability given is that of the differences bctrrcn the means be-
in3 this great by chance alone. By day 8 this difference is made up and
is no longer statistically significant. Differences on subsequent days
are not statistically siCnificant. CR + cort and Br + cort groups dif-
fer significantly only on day 12 (0.02 >P >0.01). The mean values for
other Croups showed such slight differences upon inspection that analyses
ecre not carried out on them. Actual heights of some of the above Croups
were also compared (t-test for comparison of two means) for the days on
thich they rwre most different; no statistically significant differences
Figure 1. Additive Groirth rates of 17-day-old nice injected with
trypanosomez and/or cortisone.
a. ITo injection
b. Brazilian strain + cortisone
c. Brazilian strain
e. Costa Rican strain + cortisone
f. Cocta Rican strain
iS 32 48 44
1I 32 41 44
1I 32 46 44
0AYS AFTER INJECTION
2. Adenosine triphosnhate. The results of these experiments
are summarized in Table 5.
The difference in mortalities between groups of nice injected
with cort + 5.0 mg. ATP and court + 2.5 mr. ATP iwn not statistically
significant (P = 0.49). The mortality difference between Br + cort-
and Br + court + 2.5 ag. ATP-treated mice is significant (0.016), as is
that for Br + cort + 2.5 mg. ATP- and Br + court + 5.0 ms. ATP-treated
ones (0.000); while that between Br + cort- and Br + court + 5.0 ag. ATP-
treated ones is probably biologically significant, although it is not
significant statistically (0.052). No statistically significant dif-
ference exists between mortality rates of CR + cort- and CR + cort +
2.5 me. ATP-treated groups (0.78).
Additive growth rate curves for these groups are shown in Figure
2. Again, statistically significant differences in mean per cent weight
gains occur only on single isolated days. The differences between the
means for the court + 2.5 rag. ATP- and court + 5.0 mg. ATP-treated groups
are significantly different on day 24 (0.01 P > 0.001), as are those
for Br + cort- and Br + court + 2.5 mg. ATP-injected ones on day 32
(0.05 > P > 0.02).
3. Thiamine. Results of these experiments are shown in Table 6.
Because the numbers of mice in some of the experimental groups
were small, the mortality values for all mice treated with thiamine were
combined, as were all those for mice treated with BR + cort + thiamine
(differences in mortality rates among the pooled groups were not
MORTALITY OF 17-DAY-OLD MICE INJECTED
WITH TRYPAVTOSOMES, CORTISOH E,D D ADEHOSEIIE TRIPHOSPIPAE
--o ic I II
Cort + 5.0 g. ATP
Cort + 2.5 m3g. ATP
Br + court
Br + court + 5.0 mg. ATP
Dr + court + 2.5 mn. ATP
CR + court
CR + cort + 2.5 riG. ATP
Ho. days between
injection and death
Me an Range
9.0 7 -11
12.0 3 21
11.5 6 30
10.8 5 39
Figure 2. Additive growth rates of 17-day-old mice injected with
trypanosco-es cortisone, and adenosine triphosphate.
a. Brazilian strain + cortisone + 5.0 rag. ATP
b. Brazilian strain + cortisone
c. Brazilian strain + cortisone + 2.5 mg. ATP
d. Cortisone + ATP
S_ = 2.5 ms. ATP)
x.-.,----x = 5.0 me. ATP
c. Costa Rican strain + cortisone
f. Costa Rican strain + cortisone + 2.5 rG. ATP
DAYS AFTER INJECTION
MiOtTALITM OF 17-DAY-OLD I. ITCT -TECTD ITHI
TRYPfAOS0!tOS AND CORTISOT, TREATED WITTH THIAIMS
lio. da, s bet rccn
INo. nice Dc.ths injection and death
Src- atcnt jeinjcctd 1o. 1ln TRa
Br + court
Lr + cor + tiliiaingel
60 rcg. oral
Er + cort + thlioino,
Er + cort + thicminc,
50 C., inj
CR + court
CR + court + thiiminc,
50 .G- ianj
5 0 0
statistically significant) for statistical analyses. lo statistically
significant differences in mortality rates were found between the cxperi-
=cntal groups j this series: no inj vs. Br + cort (P = 0.32); Br + court
vs. Br + court + thiamine (0.34); no inj vs. CR + cort (0.45); CR + court
vs. CR + court + thimnine (0.O0). There is an indication that cortisone-
injected mice infected iith the Costa Rican strain ma react differently
to thiamine treatnt than do those infected with the Brazilian strain.
Grorth curves for this series of experiments are presented in
Figur' 3. Values for all mice treated wirth thiamine and for those given
Br + court + thiamine were also pooled here. Moan per cent weight change
values for Br + court + thiecine-treated mice are never significantly
different from those for Br + cort-injected ones, nor are those for the
CR + cort + thiamine and CR + court groups. The difference in means on
day 12 of the no inj and thimine-treated groups is statistically sig-
nificant (0.02 p P 0.01).
4. Thiamine, naGnesium, and adonosine triphosphate. The results
of these e:.xriments are shoim in Table 7, and the grmoth curves in
Figure 4. The number of litters used in this set of experiments and in
the next series is so small that it was not felt that extensive statis-
tical analyses would be meaningful. The differences betuean mortali-
ties in Br + cort- and Br + court + thiamine, Mg+ ATP-treated mice
(P = 0.45), and in CR + cort- and CR + court + thiamine, M14++, ATP-
treated ones (0.27) are not statistically significant. Growth rate
curves for treated and untreated animals in these two groups also appear
to be very similar.
Figurc 3. Additive crcnrth rates of 17-day-old nice injected with
trypanosozoa and cortisone, treated irith thiamine.
a. io injection
b. Brazilian strain + cortlcone
c. Drazirilan strain + cortisone + thiamine
c. Costa Rican strain + cortisone
f. Costa Rican strain + cortisone + thiemine
1 10 130
110 \ 10
10 "\ /
t o no
16 32 4t0 4
0 t st 40onC
I I StFTE 48 84 CTI
OAVS AFTEN INIECTION
TOCRTALITY OF 17-DAl-0LD IICE IIJECTED 1UIT1
TRYPAIIOSOIES, CORTISOIS0, THiiLAiiE, I .AGI-IESIUi'I, AID ADQIOSEIE TRIPHOSPHATE
llo. days between
Ilo. mic2 Dcaths injection and death
Treatrannt injectcad Ho. i Mean Ran-e
Tlcnimic, M-++H, ATP 10 0 0
B + court 5 1 20.0 3.0
Dr + cori + thiamine, 5 0 0
CR + court 6 1 16.7 24.0
CR + court + thicainc,
M3-, ATP 6 3 50.0 38.3 11 56
Trijze 4. Additive c-orLth rates of 17-day-old mice injoctcd with
trypanos cec, cortisone, thi'annc, magnesium, and adenosine triphosphate.
a. Tiaisine, n;ium, ad ATP
b. EBr'ilian strain + cortisone
c. Brarilic. strain + cortisone + thiamine, rnagcsiun, and ATP
d. Costa Rican strain cortisone
e. Costa Rica~ strain + cortisone + thiamine, magnesiiu, and ATP
90 t 3t 46 64
I. 1- 130
,o -s--------- 90 me I s t 41 4
"To AFTER HUISTON
5. Ader.osi c diphosphato and Lnorganic phosphatc. The rcuults
of this e:xperinmntal treatnJnt are presented in Table 8 and Ficure 5.
There was no statistically significant difference in mortalities of
treated and untreated CR + cort-injected mice, but the difference be-
turcen the two Br + cort-injected ones was close to the significance
level (P = 0.077). Here again there appears to be a difference in
reaction to treatment between mice infected with the Costa Rican and
Brazilian strains. Gr!.th curves for the treated and untreated groups
are quite csiilar.
If the mortality rates of all the mice injected with Br + court
and those injected with CR + court from all experiments are compared, the
difference between them is not statistically significant (P = 0.24).
Trenty-nine per cent of the mice injected rith Br + court died within
60 days after injection, as compared to 21 per cent of the CR + cort-
When the numbers of days elapsing between injection and death for
these same two groups are compared, it is found that the two variances
are significantly different (P e 0.02 of obtaining an F-value as large
as that found by chance alone). The means cannot, therefore, be
analysed using the usual t-test with pooled variances. However, when
the mcans are compared by the approximate method of Cochran and Cox
(Snedecor, 1946) with no hypothesis about the variances, it is found
that the means probably are not significantly different (P>0.05 of ob-
taining the t-value calculated). The mean number of days elapsing be-
tween injection and death of Br + cort-injected nice is 10.7 (3 30);
MIORALITL OF 17-DAY-OLD MICT: I-JECTED WITH
TRYPANOS(~OES CORTISOIE, ADEMOSNE DIPIIOSPIIATE, AIMD INORCGAJIC PHOSPHATE
Cort + ADP, iP
r'r + court
Br + cort + ADP, iP
cn + court
CR + court + ADP, iP
i;o. dv, betwccn
injection and death
4.3 3 6
19.0 6 31
Figure 5. Aviditiva CrovTh ratoe of 17-2:v.-old nice injected with
tryp3aosomns, cortisone, otdenosino diphosphate, and inorgcnic phosphate.
a. Cortisone + ADP nd iP
b. irazilion strain + cortisone
Brazilian train + cortisone + ADP and iP
d. Costa Rican strain + cortisone
e. Costa Rican strain + cortisone + ADP and iP
110 46 6
o16 3 4W 64
110, / \.
46 4 .LC
0AVS AFTER mt.GIoImN
30 2 46 64
for CR + cort-!njectei ones, 13.2;. (4 51).
r. JHistolo-y. ITi.stoloLical ex-mination of tissues from mice
infected with both Brazilian sand Costa Rican strains showed the follow-
ing organs to be parazitized: esophjausz stomach, sEa=ll intestine, heart,
liver, lung, spleen, urinary bladder, brain, and spinal cord. Leishmin
bodies 'wre not observed in the Iidney or pancreas. These findings
compare with those of other workers (HauschI-a, 1947; von Brand et al.,
1949) using the Brczilian and different strains of T. cruzi.
In about half of the nice dying of the acute disease the stomachs
trere enormously distended. They were filled with partially digested
food, and the material at the pyloric end of the stomach soaetimcs con-
tained blood. Gray to blacl spots were frequently visible macroscopi-
cally on tie external stomach wall. Microscopic examination of sec-
tions through these regions and of the fresh tissue macerated in a
drop of saline solution showed these spots to be dense concentrations
of leishnan bodies. In several cases the stomachs were actually per-
forated at these points.
This observation of stomach infections confirms the findings of
Lesser and Luleman (1957) who have presented the only previous report
of stomach involvement. Their studies were done with the Pilcher strain
of T. crusi.
7. Priapism. Tro cases of priapi m ere observed, one in a
nouse injected with CR + court + thianine, b,++, and ATP, the other in
one injected with Br + cort + thiamine. The condition in each lasted
for five days and uac follovwd by paraplegia in one of the anirzals.
This phcnoacnon has been reported in young rats suffering from
a cortisone-vLcravated thimaine deficiency (Uilrcrth and Mcites, 1953).
hcn those authors treated the a~nials vith subonptial mounts of thi-
aoino, the priapisM disappccrol although cortisone injections were
8. Paraplegia. Ten nice wrc afflicted wirth paraplegia. In-
formation concerning these cases is suanarized in Table 9. The mice
seemed to fall rather definitely into two groups: these which died
very soon after the onset of paralysis, and those which lived for a
long time afterward. In the latter group the condition of two of the
mice alternately improved and regressed, but in no case did a mouse
completely recover from the paralysis. Even when some improvement was
shown here, the mice remained thin. The female infected with Brazilian
strain also developed visual troubles several weeks before its death,
one eye finally becoming completely opaque.
Spinal smears of all the paraplegic mice which died showed leish-
man bodies, but never in great numbers. The parasites rwcr also ob-
served in fresh material from the spinal cord which was macerated in a
drop of saline solution. Brain smears were prepared for two mice, and
the brains of four others were sectioned. One brain smear was positive
for leishman bodies, and throe of the sectioned brains showed then.
9. Chronic infections. During the course of this study 15
deaths resulted from chronic trypanosomiasis. These mice died from
PARAPLEGIA IN flFCTID MICE
M CR + cort
M.I CR + court
M CR + court
M2 CR + court
F CR + court
II CR + court
M CR + cort +
M CR + court +
II CR + court +
Io. da..s after injection
of psralysio Died
Lic -manu bodies
5Gtill living 124 days after injection.
**Still living 114 days after injection.
---------- ------------------ --------------
73 45h days (mrn-n = 216) after injection. During thin; period no un-
injected mice of the name eges died. Chronic infections vcre found in
mice urith the following injections: (1) Br, (2) Br + court, (3)
Br + court + ATP, (4) Br + cort + thica.ine, (5) CR, (6) CR + court, (7)
CR + cort + thiamine, uith the larCest number of casos (five) occurring
in the CR + cort-injected group. Since not all of the mice used in ex-
periments could be hept indefinitely, no statement can be made conccrn-
ing the total number of deaths which might have occurred.
Upon histological examination the same organs mentioned previ-
ously were found to be parasitized. Hearts, spleens, and livers were
frequently onlarcgd. Stomachs wcre not distended rlith food as in the
acute cases although leishmen bodies were frequently present in the
tissues. Liver, lung, and urinary bladder, in about half of these cases,
contained unusually large numbers of leislman bodies.
In vitro Expcriments
1. Growth in Citri's pnrtiall.y defined medium. It has been
found that the Erazilian and Costa Rican strains of T. cruzi Goir as
well as, or better than, the Culbertson (C) strain in Citri and Gros-
sovicz's partially defined medium and that this growth is steadily main-
tained through at least eight subcultures. Counts from a representative
series of samples are presented in Table 10.
These population densities arc just slightly lower than those of
the same strains in the diphasic medium. For the Bra.ilian strain the
mean of counts from 26 4-week--old cultrros in diphasic medium was
GROWTHi El THE PARTIALLY DIIfED (1DIU.I
OF CTRI ALID IGOSSOWICZ (1955a)
Organisms per ml. (x 106)
train At tine of Alter After After
inloculattion 2 wcol:z 3 vcecs :-. eehl
C 0.23 1.7 (6x) 11.0 (39.:) 19.4 (69x)
CR 0.06 5.7 (95x) 20.9 (348x) 29.6 (493x)
Br 0.03 7.9 (263x) 19.2 (640o:) 28.2 (940x)
31.6 x 10i organism per ml.; for the Costa Rican strain, the mean from
19 cultures of the same age ran 42.7 x 10 .
2. Growth in experimental media. A synthetic medium completely
defined except for ribonucleic acid, has bee devised. The Costa Rican
strain crew as vell in each of the modifications of this medium as it
did in control tubes of Citri's medium. The results of those experiments
arc summarized in Table 11.
CR(OTH OF COSTA RICAN STRAIN III EXPERIM"PTAL MEDIA
Organisms per nl. (x 106)
At time of
tediumm inoculation After 5 eaL-ek
Basal medium 0.9 1.2 (1.33x)
Citri's medium 0.9 0.9 (1.00x)
Kediu 1 0.75 1.4 (1.87:)
Citri's medium 0.75 0.9 (1.20x)
Medium 2 0.75 1.3 (1.73x)
Citri' medium 0.75 1.4 (1.87x)
Media 3 0.9 1.5 (1.67x)
Citri's medium 0.9 1.75 (1.94x)
Incomplete oxidation of carbohydrates is characteristic of the
metabolism of all the trypanosomes studied to date. For example, the
pathogenic African trypanosomes break down glucose only as far as pyru-
vato. Both blood stream and culture forms of T. cruzi are able to
oxidize glucose more completely, and about half of the glucose molecule
is converted to carbon dioxide, the remainder being degraded to succinic,
acetic, and lactic acids (Ryley, 1956). Chang (1948) and von Brand et
al. (1949) both demonstrated glucose utilization by culture forms of T.
cruzi, but Glucose consumption was not shown for blood stream forms (von
Brand et al., 1949). Ryley (1956) reported that both forms use glucose
and that they do not differ markedly in their metabolism of it.
Ryley's analyses for organic acids were carried out on two to two
and one-half hour incubates in Ringer-bicarbonate-glucose medium. Chang
(1948), in analyzing 1- to 2-week-old diphasic cultures of T. cruzi,
found appreciable amounts of succinic, pyruvic, and lactic acids and
small amounts of carbon dioxide and formic acid, but no acetic acid.
Unfortunately, no metabolic studies have been carried out on tissue
forms of T. cruzi.
Pyruvate is the only one of these metabolic end-products for which
host tissues have been analyzed. Coleman and von Brand (1957) studied
blood pyruvate levels in rats infected with various trypanosomes. They
found that in all cases, whether the host was infected with T. cruzi or
with one of the trypanosomes producing nuch larger amounts of pyruvic
acid, the host animals were able to maintain normal blood pyruvate
levels. However, they made no mention in their report of how long the
host animals had been infected. In T. cruzi infections there is fre-
qucntly extensive damage to liver and A muscle tissues. It is well known
that with severe liver disorders in man there is a decreased utiliza-
tion, and consequent accumulation in the blood, of pyruvate (Parida and
Karl, 1956). Therefore, maintenance of normal pyruvate levels may very
.Sll depend upon the extent of tissue, especially liver, damage in the
host. As tissue damage progressed it is quite likely that pyruvate
levels would rise, eventually to concentrations toxic to the host.
Under the conditions of the present experiments the treatment of
T. cruzi-infected mice with thiamine or with thiamine, magnesium, nd
ATP apparently had no effect. If, as suggested above, the mice are able
to rmtabolize pyruvate satisfactorily until tissue damage becomes too ex-
tensive, it can be seen why thicmine did not affect the course of the
disease. However, if an animal were suffering from a slight pre-existent
thiamine deficiency, the increased amount of thiamine necessary to
._.tabolize the pyruvate produced during a trypanosome infection might
be enough to meke the deficiency pronounced. In such a case, thiamine
therapy might be effective, at least temporarily. It should be borne
in mind also that even if a thiamine deficiency did occur during a
trypanosome infection and were treated, it is very likely that the
disease has other effects which miGht completely mask any results of
the thiamine therapy.
Grant and Fulton (1957) found abnormally high levels of pyruvate
in the blood of rats infected with T. rhodesiense and showed a positive
correlation to exist b between the pyruvate level and the dejee of para-
citemia. Upon treating the rats with a trypanocidal drug, they noted
that parasites had disappeared from the blood 90 minutes after treat-
ment, and at 120 minutes the lcto-acid levels in these rats were almost
identical to those of control animals. Here there was apparently no ex-
tensive tissue damage. The failure to maintain normal blood pyruvate
levels in this case may have been due to a limiting mount of some co-
factor rcqaired for pyruvate metabolism. The possibility that the para-
site is in some way able to inhibit this metabolic reaction should not
be ruled out.
These ideas presented concerning pyruvate metabolism are probably
applicable also to the utilization by the host of the other organic acids
produced by T. cruzi. The host should be able to maintain normal levels
of cuccinate, lactate, and acetate until tissue damage becomes extensive.
Perhaps in early infections light pre-existing deficiencies of vitamins
or other compounds required for the oxidation of these organic acids may
be aggravated. In this case, treatment of the host with the needed factor
would prevent the accumulation of these acids to toxic levels, staving off
death from this cause and giving the animal time to fiCht off, or adjust
to, the infection. Such deficiencies miCht also occur in chronic cases
of trypanosomiasis. Deficiencies of thiamine, a-lipoic acid, pantothenic
acid, riboflavin, and nicotinamide are among those which might be ex-
pected to occur.
The one factor c oon to the utilization of all these metabolic
end-products of the trypancsome is ATP or ADP and iP. As pointed out
carlier, the rate of cbstra.te dc'raddation is dependent upon the rate
of o::idativc phosphorylation, which is in turn controlled by the rate
at .rhic AiTP is split to ADP and iP.
The results of the present experim~L ts confirm Ada-ms' (1954-) find-
ing that ATP treatmCat reduces the mortality of mice infected lrith the
Brazilian strain of T. cruzi. Failure of the treatment to have the sei
effect on mice infected with the Costa Ricea strain may be attributed
to: (1) the smaller number of experimental animals involved, (2) the
greater variability of the Costa Rican strain; or (3) a real difference
in reaction to the trcatmont btreecn the two strains.
Tlc finding that 2.5 r,. of A TP reduce the mortality of infected
nice ~lile 5.0 mS. increase it was unexpected. It was the latter amount
liichl Admns used in his experiments. The larger amount of ATP was not in
itself toxic, for whcn Given to cortisone-injected mice it van not fatal.
Only ihen the rice had previously been injected with trypanosomes did the
5.0 mr. dosage have an adverse effect. I ha no explanation for the re-
action to this dosage difference, if indeed it is a real thing. It is in-
teresting, haoever, to imrnine that the host may be able to claim for its
own use certain amounts of ATP, but that above a certain level the ATP
is not irrsdiately utilized and becomes available to the trypanosomes.
The manner in which ATP acts to reduce the mortality of trypano-
some-infected mice remains hypothetical. Adams (195)) suggested that it
was used in coRbating a hypoglycemia produced in the host by the trypano-
somes. Since it has been sh .on that animals infected with T. crusi
usually maintain normal blood sucGr levels (von Brand et el., 1949),
thi i~oulI not seen to be the c.planalition. The fact that the ATP Ias
administered only once, very early, in the infection, precludes the
possibility that its enerjr ras available for repair of dama-ed tissues.
It is possible however 'hlat it could be used for synthesis in the produc-
tion of antibodies or pha ocytic cells. The ATP injections rcre adminis-
tered at about the time that the metacyclic trypanosomes injected are
becinn:ing to change to leishman forms (Pere:-Rcyes, 1953), the forms
which reproduce wirthi the vertebrate host. It seems more likely, how-
cver, that the ATP becomes involved directly in the metabolism of
rouse or the trypanosome or both. It may very well be used to provide
phosphate and/or phosphate acceptors neccssry for the oxidation of the
nmtabolic end-products of the trypanosomos, Giving the host tine to ad-
just its metabolic pattern to acco. nodate this new situation.
Surprisingly, hardly any studies have been made of the effect of
ATP administration on intact animals. From the few experiments per-
formed it has been learned that there is a direct transfer of phosphate
groups from ATP to thiamine and incorporation of the resultant di- and
triphosrhothiamines into the liver within 1 hour of injecting rats rith
ATP32 and thia-ine (Rossi-Fanolli et al., 1954). Hascitelli-Coriandoli
and Boldrini (1957) found th.t the injection of ATP raises the levels of
ATP-phosphorus, creatine phosphate-phosphioruc, end inorganic phosphorus
to almost normal levels in the heart muscle of rats treated vith thyrox-
inc. Until more is IIown concerning the fate of injected ATP in animals,
and until the nutritional requir!emnts of T. cruzi and the effect of
this parasite upon its host are better understood, no final c=planation
for the effect of ATP on tr'pio l.zis cona be given.
The treatrmant of trypar.nosonc-infectcI uic. with UAP and iP did not
reduce uc e at ortclity rat:e. In view. of the sail number of niec used in
this set of c::pri ments, thisec results cannot be considered to be con-
clusive. With a lage enoui sample, the results of this treat-xct
should be siilr to those obtainid with ATP injection.
The high degree of adaptation bcticen T. cruzi and its host is well
c naplified in the very close similaiaity of growth rates of infected and
uninfected juice. Further evidence of this adaptation is found in the
relatively larc number of infected mice which survived the acute infec-
tion end continued to live rith chronic infections. At least a partial
eplanation for this host-parasite relationship lies in the very nice
nLtabolic exchange between the two. The parasite takes glucoce from the
Iiost, but does not completely nctobolize it, returning at least half of
the moleculc to the host in the form of organic acids uhich the host is
then able to utilize. In this waj both organisCi obtain encercv from the
The occurrence of chronic trypanosomiasis in laboratory ;ii'ials
is rarely mentioned in the literature. Johnson (1933) reported that of
19 dos c::perincntally infected with T. crusi, 9 of then developed acute
infections and 11, chronic infections. Coble (1951) und Haucschia (1947)
licce both noted the occurrence of the chronic disease in nice. Since
clhonic trypaosermisis is the moDt frequent form encountered in humans,
it is suggested that nice writh clironic infections should lend themselves
':cll to ezpcrim2ntcal .;'orl concerned itlh the cfiect' of the chronic
disoasc on the host an1 '.with therapy of the disease.
Paraplegia in nice infected with T. cruzi Las been reported, to rLy
1:ncr iZlea, only once before. Do Souse C,.j-os (1925) state that .- of 12
infected nice shzowd parcaplcgia. Thi condition is fairly ccnan. xa in in-
fected dogs (Goble, 1952; Villela, 192n) cnd rabbit.o (do Sousa Corpo;,
1724). De Sousa Cmipos (1925) reported that lesians of the central
nervous system wcre not obscrvlcd in all casos of paraplegia and s gCrcted
that this might depend on the duration of teo infection. The aflaocllated
forao were found in the central nervous system of all tie other para-
plegic cases rcported, although they were never numerous. Inflamiatory
centers formed by infiltration of macrophages .'cre encountercl more fre-
quently during microscopic c aination of brain and spinal cord sections
then iore lciscLG -n bodies. Goble (1952) found Uthat dogs may die within
a few days of becoming paraplegic or may continue to live for scme tine
with varying dcreo of wrcaLness and paralysis. The neurotropic form of
tryprnosomiasis appears to be very similar in dogs land in the mice ob-
sr7red in the present c.prcrimnts. Goble, however, easo reported the
recovery of one paraplegic dog. In view of those reports where leish-
ian bodies could not be demonstrated in the central nervous system and
where there was a complete recovery from paraplegia, paralysis in try-
panoscme-infected mice should not be haphazardly attributed to the
direct effect of the trypanosomc. It might be pointed out that a phos-
phorus deficiency will produce paralysis of the hind legs (Footer ct al.,
1949) in mice. A thiamine deficiency also affects the lces of nice,
-.i'ini':; them so v ncx: thntt thly ',e un:i'le to support t'ih body (Uooll.y:
and M2hite. 1943). Thierfore, it is quite possible that paraplecia in
trypraosa-n-infeetod ice szou.ld not be wholly attributed to a direct
effect of tha pcrasitcs in the central nervous system.
It is no- ::n.ro that the partially defined synthetic nodiu-i of
Citri and Grosrovics (1955a) will support grourth of three strains of
T. crtzi, .s orhwn by the rcesTlto presented earlier in this paper, and
31 different strains of Loishme.nia tropica (Citri and Grossowicz, 1955b).
The latter muthors foundr that the medium did not support gro rth of L.
infentui. L. donovan-. L. braciliensis, ran- L. mae, hou'ver. This
cstuesztz that the Growth requireC-cnts of L. tropical and T. cr--i ere
norc i:211ar than are those of L. tropical cnd the other leis'- .mias of
D1uin1 the tine that gro.wt Y- :oriments in redia containing !mo.-.m
\nino acidc w:cr bein3 conducted, the cultures in this laboratory ~;or
undcrgoinu, one of the une::plained Grcwth depressions ~Iich occasionaCly7
affect tr;'paioso o cultures (Balamuth; 1957). It is for thick reason
that the popu.ltions attained in the excprimental media cn: in the Citri'3
mediun controls were rchi lo-1.r thaen V :ould be extpcto1.. 'Th relative
c-o unt of Crowth obtained in the enGermantal medin as conparcd to that
in the control medii appccars to be soiCificcnt, vnd there is no reason
to believe thcV this rill chCen.ce under improved CroI.th conditions. No
differences in rocrth supporting ability Lerre observed cmoen the dif-
ferent e:pcrirmnntal modia. Future e:;periicnts ere plaUmed in ihich
prince and pyrinidincs or nuclcotidcs irill be substitute for the
ribonuclcic acid in the icdia now bcing usec. If this substitution is
accomrplished, the nodium for T. cruzi will be completely defined.
SUMMARY AID CONCLUSIONS
Injection with adenosine triphosphate reduced the mortality of
mice infected with Trypanosoma cruzi. Treatment of infected mice with
thiamine or with thiamine, magnesium, and adenosine triphosphate appar-
ently had no effect on the course of the disease. With a small number
of mice, adenosine diphosphate and inorganic phosphate appeared to pro-
duce no effect on the infection.
Chronic trypanosomiasis resulted frequently when mice were injected
with trypanosomeo and cortisone. Crowth rates of normal, uninfected mice
wrre not siGnificantly different from those of mice with the chronic
Leishman bodies were observed in heart, spleen, liver, lung, and
other orGans commonly parasitized by T. cruzi. In addition, the aflagel-
lated forms were found in stomach tissues of about one-half of all mice
dying of acute trypanosomiasis, confirming a single previous report of
Ten infected mice developed paraplegia. LeishmGan bodies were
present in the central nervous systems of these animals.
The partially defined medium of Citri and Grossowicz (1955a) was
found to support growth of the Brazilian and Costa Rican strains of T.
cruzi as well as it did that of the Culbertson strain for which it was
devised. Modifications of the medium rere made, and ribonucleic acid is
now the only undefined component. A known mixture of amino acids com-
pletely replaced the protein source. Growth of T. cruzi (Costa Rican
strain) was as good in amino acid-substituted media as in the original
AdM-s, B. II. (1954).
Lethal infections of Trypcnosoma cruzi in cortisone
adenosine triphozphate-treat d white nice. Unpub-
of Science thesis, University of Houston, Houston,
Adlcr, S., and rTeodor, 0. (1926).
and IIerptomonas papataLii.
The identity of Leismani:. tropical
Ann. Trop. Ied. Paraitol., 20, 355-
AC-osin, M., and von Brand, T. (1955). Characterization and intracellu-
lar distribution of the succinic dehydrosenase of Trypanosoma
cruzi. Exp. Parasitol., 4, 548-563.
ACsoin, M., and Weinbach, E. C. (1956). Partial purification and charac-
tcrization of the ioocitric dchydro gnasc from Trypanosoa cruzi.
Biochin. et Biophys. Acta, 21, 117-126.
Baornmtein, 11. D. (1953a). Malic dchydrocenase and related enz:mes in
the culture form of Trypanosma cruzi. E:p. Parasitol., 2,
S_(1953b). The enzyme systems of the culture form of Tryanosoma
cruci. Ann. 11. Y. Acad. Sci., 56, 982-994.
Bacrnstcin, IH. D., and Rees, C. W. (1952). Aldolase in the culture form
of Trypanosoma cruzi. Exp. Parasitol., 1, 215-228.
Balanuth U. (1957). Personal comnunication to E. C. Bovec.
von Brand, T. (1951). IMetabolism of Tirpanosomidae and Bodonidae, in
L:roff, A., Biochemistry and lPysiolo.-y of Protozoa, Vol. 1,
Academic Prccs, Inc., New Yorl: 177-234.
(1952). Chemical Paysiolo,7y of Endoparasitic Animals, Acadciic
Press, Inc., ~kr York.
von Brand, T., and Asosin, M. (1955). The utilization of Krebs cycle
intermediates by the culture forms of Trypanosoma cruzi and
Leishmania tropical. J. Infectious Diseeees, 97, 274-279.
von Brand, T., Tobie, E. J., Iissling, R. E., and Adamc, G. (1949).
Fnysiological and pathological observations on four strains of
Trypan)~ oao cruzi. J. Infectious Disesees, 85, 5-16.
Cameron, T. W. W.. (1956). Par-sitco and Parasitisn, Iethuon and Co.,
Chagas, C. (1909). Ueber eine noue Trypanosomiasis des Menschen. Mm.
Inst. OswQldo Cruz, 1, 159-218.
Chang, S. I. (1943). Studies on heaoflaGelatoc V.IV. Observations
concerning some biochemical activities in culture and respira-
tion of three cpeciez of loiChmbanias and Tryrnascan cruzi. J.
Infectious Diseoasco, 8, 109-118.
Citri, II., and Grosso-icz, N. (1954). A liquid modi~ for the cultiva-
tion of Trypaoso.a czuzi. Nature, 173, 1100-1102.
(1955a). A partially defined culture medium for Tryjanosama
cruzi and some other hcemoflagellates. J. Gen. Microbiol., 3,
(1955b). Grawth requirements of Lcishmania troDica and other
loishmanias. Trans. Roy. Soc. Trop. gd. 49, 603-604.
Colemca, R. II., and von Brand, T. (1957). Blood pyruvate levels of
rats during hemoprotozoan infections. J. Parasitol., 43, 263-270.
Cowperthwaite, J., Weber, M. M., Pacl:er, L., and Hutner, S. H. (1953).
INutrition of Hcrpotcmonas (Strigomonas) culicidarum. Ann. U. Y.
Acad.. Sci. 972-981.
Culbertson, J. T., and Kessler, W. R. (1942). Age resistance of mice
to Trypanosoma cruzi. J. Parasitol., 23, 155-158.
Dias, E., and Laranja, F. S. (1948). Chagas disease and its control.
Proc. Intern. Congr. Trop. Med. Malaria, 4th Cong'., Washington,
D. C., 2, 1159-1170.
Foster, C., Jones, J. H., Henlc, U., and Brenner, S. A. (1949). Iutri-
tion and poliomyelitis. The effects of deficiencies of phos-
phorus, calcium, and vitamin D on the response of mice to the
Lansing strain of poliomyelitis virus. J. Infectious Diseases,
Glass, B. (1951). A suom~ary of the symposium, in McElroy, W. D., and
Glaso, B., Phosphorus Metabolism, Vol. 1, Johns Hopkins Press,
Goble, F. C. (1951). Studies on experimental Chagas' disease in mice
in relation to chemotherapeutic testing. J. Parasitol., 37,
(1952). Observation on experimcntal Chagas' disease in
dogs. Am. J. Trap. Med. L.-, 1, 189-204.
(1956). American trypanosoniasis. J. An. Med. Assoc., 161,
Grant, P. T., and Fulton, J. D. (1957). The catabolism of glucose by
strains of Trypan)aoma rhodcsicnMc. Biochin. J., 66, 242-250.
Hall, R. P. (1953). Protocoology, Prentice-Hall, Inc., Ne- York.
Hauschka, T. S. (1947). Sex of host as a factor iu Chagas' disease.
J. Farasitol., ~, 399-404.
Hawking, F. (1953). Recent advances in the chemotherapy of protozoal
infections. Intern. Congr. Micrdbiol., 6th Congr., Rom., 4,
Hutner, S. H. and Provezoli, L. (1955). Comparative biochemistry of
flagellates, in Htnaer, S. I., and twoff, A., Diocheistry and
Physiology of Protozoa, Vol. 2, Academic Preso, Inc., 1ow York,
Jarpe, A., Agoein, M., Ciristcn, R., and Atias, A. V. (1951). Ensavos
de quimioterapia de la enfermedad de Chagas e-perimental. VII.
Cortisona y fosfato de pentaquina. Bol. Inforn. Parasitol.
Chilenes, 6, 25-27.
Johnson, C. M. (1938). Cardiac changes in dogs experimentally infected
with Trypanosoma cruzi. Am. J. Trop. Med., 18 197-206.
Krebs, 1. A. (1956). The effects of extraneous agents on cell metabol-
ism, in Walstenholme, G. E. W. and O'Connor, C. M., Ionizin Radi-
ations and Cell Metabolism. Ciba Foundation Symposium, Little
Brown and Company, Boston, 92-103.
(1957). Control of metabolic processes. Endeavour, 16,
Lardy, H. A. (1956). Energetic coupling and the regulation of metabolic
rates. Proc. Intern. Coagr. Biochem., 3rd Congr., Brussels, 1955,
Lesser, E., and Lukeman, J. M. (1957). Stcmach infections with
Trygpnoaom cruzi. J. Parasitol., 4_, 65.
Little, P. A., and Oleson, J. J. (1951). The cultivation of Trypaosome
cruui. J. Bacteriol., 61, 709-714.
L:roff., i. (1951). The nutrition of poaasitic flagellates (Trypanoso.-i-
dac, Trichoonondinae), in Lwoff, A., Biochcmistry and PBsiology
of Protozoa, Vol. 1, Academic Pressc, Inc., iUew Yor'!, l29-1~..
Mascitelli-Corio~doli E., and Boldrini, R. (1957). IEfect of injection
of organic phosphates on coe pho.sphorus fractions in the heart
muscle of rats treated irith tlhy-oziine. Iature, 179, 196-1197.
MicRo ry, U. L., Dcavor, E. L., aid 11oble, E. R. (1953). In vitro of-
fects of ProdiGiosin and other antibiotics on Trypanoson cruzi.
E__. Parasitol.; 2, 125-128.
Morris, H. P. (1947). VitLain requri:snts of the moue. Vitamins
and IIorm-oncs: 5, 175-195.
Ilathan,. H. A.; and Covpcithwaita J. (1954). Use of the tryponosomid
flUaellate, Crithiidia fasciculata, for evaluating critinalarials.
Proc. Soc. 22p. Biol. MId., 55, 3-17-119.
(1955). "Crithidia factor"--a ne:r mcribc of the folic acid
group of vitamins. J. Protozool., 2, 37-42.
h'cff, R. J. (1955). Piycioloyo of aroeba: vrith special reference to
nuclear physiology and intermediary carbohydrates mctaboliam in
soil Ecocbae. Terminal report on Public Health Rcscarch Grant
l-623 (I and G),; :.4S o., 2-3.
Ileilands, J. 3., and Stuppfl, P. K. (1955). Outlines of Fazjrie
Chaiistry, John! Wiley and Sons, Inc., licI.r Yor--.
Iloblc, E. R., tcRary; U. L., and Bcavcr, E. T. (1953). Cell division
in trypaiososac. Trans. Am. Ilicroscop. Soc., 72, 236-243.
Parida. I ., and ?:aa:, R. 1.. (1956). Blood pyruvic acid, serum gamma
Clobulin, and other tests of hepatic function. J. Lab. Clin.
aod., 47, 42-50.
Pcez-Reyes, R. (1953). La cvolucion de Schizotrypanum cruzi in ratones
blanco. Cioncia (xic.), 13, 20,'-225.
Robinson, F. A. (1951). The Vit:nin E. Corplex, John Wiley and Sons,
Inc., liNe Yor:.
Rossi-Fonelli, A., Silipria di, iI., Fasclla, P., Siliprandi, D.; and
Salvetti, II. (1954). On the phosphorylation of thiamine in
the living atinal. Expcrientia, 10, 73-74.
Rylcy, J. F. (1956). Studies on the metabolism of the protozoa. 7.
Coiparative carbohydrate nctabolism of -eleven spociez of t1ypcno-
soees. Biochcm. J., 62, 215-222.
Senan., G. R. (1953).
The succinic dchydrogcnase of Trypanosoma cruzi.
(1957). Personal co=aunication.
Shwartzman G. (1953). The Effect of ACTi and Cortisone upon Infection
and Resistance, Col -Ibia Univelrsity Pross, HIow York.
Snodecor, G. W. (1946). Staitistical Mthod- 4th edition, Iowa State
College Press, Amcs, 83.
Sncll, G. D. (1941). Biology of the Laboratory Iouse, Blckiston Co.,
do Souza Caopos, E. (1924). Sur la paraplegic dcs cnlnaaux infected
ee:perintalm.ient par Ic Trypanosoma cruzi (Chagas 1909). Compt.
rend. soc. biol., 91, 934-395.
. (1925). Sur la paralysis des animaux (chien, souris)
experimentalement avec les cultures de Trypanosoma cruzi.
rand. soc. biol.; 93, 40-42.
Tracer, W. (1957). nutrition of a hemoflcellate (Lcihomania tareutolac)
having an interchangeable requirement for choline or plrido::al. J.
Protozool., 4, 269-276.
Tristra, G. R. (1953). The Eaiino acid composition of proteins, in
IIeurath, H., and Bailey, K., The Proteins, Vol. 1, Part A,
Academic Press, Inc., Nov Yorl, 151-233.
Villela, E. (1924). Paralysie experimental chez la chin par le
Trypanosom cruzi. Conpt. rend. soc. biol., 91, 979-983.
Wilverth, A. M., cand Ilitcs, J. (1953).
mine-deficient young rats. Proc.
Effects of cortisone on thia-
Soc. Exp. Biol. Med., 83,
Woody, N. C., and Woody, I. B. (1955). American trypcnosomiasic (Chagas'
disease). First indigenous case in the United States. J. Am. Med.
Assoc., 159, 676-677.
Woolley, D. W., and Mlhite, A. G. C. (1943). Production of thiamine de-
ficiency disease by the feeding of a pyridine analogue of thiiiine.
J. Biol. Chem., 149, 235-239.
Jacqueline Adam. was born in Ashland, Kentucky, on June 3, 1930.
In 1952 she married Dennis Hynos. She received the Bachelor of Arts
Decree in June, 1952, from MIacalester College, St. PIal, Minnooota, and
the Icastcr of Science Dcgree from the University of IlichiCan in Febru-
ary, 1954. From 1954 to 1956 she iras employed as Assistant in Research
(Instructor) in the Department of Civil Engineering, University of
Florida. She rcCumn-d her graducto studies at the University of Florida
in 195' and while thero held a crEduate acsistentship, tie Dudley
Beaumont Menorial Followmhip, and a research assistantsnip.
Sis dissertation vas prepared under the direction of the
chairman of the camidldato's supervisory committee and has been ap-
proved by all menmbes of that comnittcc. It was submitted to the
Dean of the Collse of Arts and Sciences and to the Graduate Council,
a'd vac approved as partial fulfi.l ent of the requiraem ~ts for the
doercc of Doctor of Phlilosophy.
A--ust 9, 1958 D3'______
Dean, ColloGe of Art eand Sciences
Dean, Graduate School
SUPaTVIS CY COMMITTEE: