A novel cultivation technique for long-term maintenance of bloodstream form trypanosomes in vitro

ELSEVIER Molecular and Biochemical Parasitology 70 (1995) 157-166

MOLECULAR

EHEMICAL PARASITOLOGY

A novel cultivation technique for long-term maintenance of bloodstream form trypanosomes in vitro

Friedemann Hesse, Paul M. Selzer ‘, Kerstin Miihlsttidt, Michael Duszenko *

Physiologisch-chemisches Institut der Unillersitiit, Tiibingen, Hoppe-Seyler-Str. 4. 72076 Tiibingen, Germany

Received 15 August 1994; accepted 10 January 1995

Abstract

We used an axenic cultivation system to grow African trypanosomes in vitro. Long-term cultivation for more than 60 days has been achieved by replacing the culture medium at regular intervals between 6 and 48 h. In contrast to a control culture without medium replacement, increasing amounts of maximum cell concentrations have been obtained, ranging from 5 X lo6 to 2 X 10’ trypanosomes ml-‘, whereas the generation doubling time remained constant (about 6 h). Higher cell

concentrations have only been obtained by total medium replacement; neither addition of fresh medium nor serum led to a higher cell yield, suggesting that a trypanosome-derived factor or metabolite accumulated in the medium rather than medium

was depleted of an essential nutrient. Most interestingly, however, successive waves have been obtained which eventually led to a damped oscillation curve with a constant high population density after about 40 days of cultivation. Cultures were started with a homogeneous population of the long-slender form. As judged by light microscopy, cells showed a stumpy morphology during the declining phase and became slender again in the following growth phase. At later time points, when

cells remained in a stationary phase at high population density, many different morphological stages have been observed, similar to those described by early authors as intermediate forms [Ormerod, W.E. (1979) In: Biology of the Kinetoplastida, Vol. 2, pp. 340-3931, although many dividing forms are still present at that time. In contrast, identically treated procyclic

cultures were unable to produce cyclic growth waves. Based on these results, a novel concept considering a possible differentiation mechanism is discussed.

Keywords: Trypanosoma brucei; Axenic cultivation; Differentiation

1. Introduction

Abbreviations: FCS, foetal calf serum; FITC, fluorescein isothio-

cyanate; MEM, minimum essential medium, cMEM, conditioned

minimum essential medium; VSG, variant surface glycoprotein.

* Corresponding author. Tel.: (49-7071) 29-3343; Fax: (49-

7071) 29-6390; e-mail: [email protected]

’ Present address: University of California, San Francisco Lab-

oratory of Molecular Pathology, 4150 Clement Street 113B, San

Francisco, CA 94121, USA.

Various subspecies of Trypanosoma brucei cause sleeping sickness and livestock diseases in tropical areas of Africa. A natural infection is typically char- acterized by an oscillating parasitaemia with a cycli- tally increasing and decreasing infection rate [l]. The declining of the number of parasites is caused

0166-6851/95/$09.50 0 1995 Elsevier Science B.V. All rights reserved

SSDI 0166-6851(95)00027-5

158 F. Hesse et al. /Molecular and Biochemical Parasitology 70 (1995) 157-166

by specific antibodies raised against the variant surface glycoprotein (VSG); due to antigenic variation, however, eradication of trypanosomes is never com-

plete 121. Trypanosomes undergo morphological alterations during the course of infection, ranging from the rapidly dividing long-slender to the non-dividing

short-stumpy form [3], which is accompanied by

marked biochemical changes [4,5]. Parasite strains able to cause cyclical infections are termed pleomor-

phic, in contrast to monomorphic strains. The latter

strains are adapted to rats and mice and are often used for convenience in the laboratory, because of

the fulminant infection, which leads to a high parasitaemia (up to 1 X lo9 trypanosomes ml-‘) within 2 to 4 days. They are also best suited for axenic cultivation in vitro, although the maximum cell yield obtained in culture is still limited to about 5 X lo6

trypanosomes ml-‘. We have investigated the cause

of growth limitation and present data showing that

a b

the low cell yield is not caused by a limited supply of substrate but rather by a metabolite or factor produced by the parasite during the course of cultivation. This factor accumulates within the culture medium and forces the cells to differentiate to a non-dividing form of altered morphology and bio-

chemical properties. Interestingly, replacing the culture medium at regular intervals (ranging from 6 to

48 h) leads to a prolonged logarithmic growth phase and higher cell densities (up to 2 X lo7 try-

panosomes ml- ’ > but will not prevent cells from

entering a declining phase. Much to our surprise, however, trypanosomes will enter into a dividing phase again with a 3-4-day delay and subsequently

show a damped oscillation growth curve similar to a natural infection, although antibodies are not present in this system. From these results we would like to suggest that differentiation is controlled by the para-

site itself via a secreted factor, resembling the cy-

IO’ 1-I 10’ I-

O 2 4 6 8 10 0 2 4 6 8 10

Time [days]

C

Fig. 1. Effect of medium exchange on growth curves of axenically cultivated trypanosomes. (a) The medium was replaced every 48 h (01,

every 24 h (A ) or not at all (0). (b) The medium was replaced every 12 h ( 0 1, every 6 h (V ) or not at all (01. (c) In order to check if

substrate depletion is responsible for cell differentiation, trypanosomes were grown for 48 h identical to the control cells, i.e., with one

medium exchange after 24 h. After 48 h, however, either 10% FCS ( A) or 10% serum-free MEM (V) was added in 2-h intervals, whereas

the complete medium was exchanged in the control cells. See text for further details. The graphs show representative growth curves of four

independent experiments with virtually identical results.

F. Hesse et al. /Molecular and Biochemical Parasitology 70 (1995) 157-166 159

tokine system of mammalian organisms. This is also

supported by our finding that procyclic trypanosomes, cultivated as described before, does not

show a cyclical growth curve, although a high cell density is obtained.

from infected mice under sterile conditions or from frozen stabilates. The latter were prepared aseptically from infected rat blood and cryopreserved using a standard protocol [81, except that glycerol (20%) and 2 I.U. ml-’ sodium heparin (Promota, Hamburg, Germany) were used.

2. Materials and methods

2.1. Trypanosomes

Trypanosoma brucei clone MITat 1.2 (Molten0 Institute Trypanozoon antigen type), derived from

the T. brucei strain 427, was kept as frozen stabilates in liquid nitrogen. This clone is monomorphic and leads to a high parasitaemia (> lo9 cells ml-‘) in

mice and rats. Where necessary, trypanosomes have been propagated in NMRI mice or Sprague-Dawley

rats (Interfauna, Tuttlingen, Germany), which were intraperitoneally infected with 2 X lo6 parasites

ml-’ (taken from frozen stabilates). Animals have

been sacrificed at an infection rate between 5 X 1Oa and 1 X lo9 cells ml-‘, monitored by tailblood counting. Infected blood was collected by cardiac

puncture under sterile conditions and parasites have

been isolated as previously described [6].

To start a cultivation experiment, stabilates were rapidly thawed and immediately diluted with a lo-

fold volume of culture medium. Cells were pelleted

(1500 X g, 4°C 5 min) and resuspended in 1 ml fresh culture medium. Following cell counting, try-

panosomes were seeded at 1 X lo5 cells ml-’ in preincubated culture medium and propagated in a CO,-incubator at 37°C.

For most experiments the culture medium was exchanged after various time intervals by spinning the cells down (1500 X g, room temperature, 5 min),

removing the supematant and resuspending the cell pellet within the same volume of medium which had been preincubated at 37°C in a CO,-incubator.

Cell densities were determined using a special

Neubauer haemocytometer (depth 0.02 mm) or a cell counter (Scharfe System, Reutlingen, Germany), as previously described [7]. Only intact motile cells

have been counted.

Procyclic trypanosomes were derived from bloodstream forms of the T. brucei clone MITat 1.4 by

transformation in vitro as previously described 171.

2.4. Cultivation of procyclic trypanosomes

2.2. Culture media

Bloodstream form trypanosomes were cultivated in a modified minimum essential medium (with

Earle’s salts) [7], supplemented with 0.25 mM cysteine/O.Ol mM bathocuproine disulfonic acid (to prevent copper-catalyzed oxidation of cysteine;

Serva, Heidelberg, Germany)/lS% (v/v) heat-inactivated foetal calf serum (FCS).

Procyclic trypanosomes were taken from cryopreserved stabilates and propagated in procyclic form

medium in T 25 culture flasks (Falcon) at 27°C. Culture medium was exchanged once a day as de-

scribed above.

2.5. Immunofluorescence

Procyclic trypanosomes were cultivated in MEM without cysteine and bathocuproine sulfonate but containing 5.2 mM praline/0.012 mM haemin/lO% (v/v> heat-inactivated FCS.

To check for VSG-type homogeneity, trypanosomes were fixed with Hepes-buffered saline (50 mM, pH 7.4, containing 1% formaldehyde), and stained with polyclonal antibodies against VSG-type MITat 1.2 and FITC-coupled anti-IgG antibodies. Trypanosomes from the clone MITat 1.4 were used as a negative control.

2.3. Cultivation of bloodstream form trypanosomes 2.6. Preparation of conditioned media

Cultivation experiments were initiated using bloodstream form parasites either freshly isolated

Cell-free conditioned media (cMEM) were prepared from bloodstream form cultures without

160 F. Hesse et al./Molecular and Biochemical Parasitology 70 (1995) 157-166

medium exchange after 144 h. Cells were removed by centrifugation (3000 X g, 10 min, 4”C), and the supernatant was used following sterile filtration. cMEM was stable at 4°C for more than 4 weeks.

2.7. Molecular mass estimation of the growth inhibitory factor

To determine the molecular mass range of the growth inhibitory factor, cMEM was dialysed through a 3 kDa cut-off membrane, using either a dialysis tubing or Centricon Concentrators (Amicon, Witten, Germany). Dialysates or retentates, respectively, were lyophilized and redissolved in double distilled water to yield the original volume. These cMEM samples were mixed with an equal amount of fresh MEM and supplemented as described above, thereby neglecting any remnants of supplements left over in the respec- tive cMEM fraction. Finally, the media were tested in a cultivation experiment for their growth inhibitory activity.

3. Results

3.1. Effect of medium exchange at different time intervals on axenic bloodstream form cultures

Axenic cultivation of bloodstream form trypanosomes became possible in 1985, when Duszenko et al. [9] and Baltz et al. [lo] showed independently that supplementation with cysteine, 2-mercapto- ethanol or monothioglycerol renders co-cultivation of fibroblasts as feeder cells unnecessary. However, all cultivation techniques available so far are limited in the maximum cell concentration, which is up to 5 X lo6 parasites per ml, while lOO-200-fold higher parasitemia are readily obtained in vivo.

In order to reach higher cell concentrations in vitro, we investigated the effect of completely replacing the culture medium on trypanosome growth in axenic culture. For this purpose, culture media were exchanged at different time intervals (every 6, 12, 24 or 48 h, respectively). Cell growth was monitored by determination of the cell concentration at least once a day. The resulting growth curves are shown in Fig. 1. All curves are identical for the early exponential growth phase and are indistinguishable from the

control experiment (conventional cultivation without medium exchange) up to a cell concentration of about 3 X lo6 trypanosomes ml-‘. During this ini- tial exponential growth phase, medium exchange had no effect on the population doubling time, which was about 6-8 h, and thus as high as in animal infections.

At higher cell concentrations, however, medium exchange significantly improved cultivation conditions and led to both a prolonged exponential growth phase and higher peak concentrations. The effect obviously depends on the frequency of medium exchange, i.e., the more often the medium is replaced, the higher is the maximum cell yield (Fig. la,b).

In all cases the end of the exponential growth phase was reached between day 2 and 3 of cultivation and cell concentrations decreased thereafter leading to a population minimum between day 7 and 8 of cultivation. During this declining phase, medium exchange could not prevent death of most of the trypanosomes. If the medium was not exchanged during this time, all parasites died rapidly between day 4 and 5 of cultivation (Fig. la). However, when the medium was replaced every 24 h, parasites started to grow again after 7 to 8 days and entered into a new exponential growth phase (Fig. la,b).

3.2. Effect of serum or medium addition during cultivation

In order to control for substrate-depletion causing parasites to enter the declining phase, either 10% MEM or 10% serum was added to growing cultures. For this purpose, only 10% of the cell suspension was removed. Cells of these aliquots were pelleted, resuspended in foetal calf serum or MEM, respectively, and added back to the corresponding culture vessel. In this way, the original cell concentration as well as the culture volume were readjusted, 90% of the putative differentiation factor was kept in place, and fresh substrates from serum or MEM were supplemented. This procedure was performed at 2 h (Fig. lc) or 6 h (data not shown) intervals with equal results: in contrast to a control culture, where medium was completely replaced, cells could not grow further but entered into a stationary phase, comparable to a culture without medium replacement (Fig. la,b).


3.3. Trypanosomes can be maintained for more than 60 days in menic culture if the medium is replaced in daily interuals

To investigate how long bloodstream form trypanosomes can be propagated in axenic culture if medium is exchanged regularly, we started a long- term cultivation experiment replacing the culture media at daily intervals. Fig. 2 shows a typical cultiva-

tion experiment during the first 45 days. As in the

experiments described above, trypanosomes first grew exponentially up to a cell concentration of

about (6-7) X lo6 parasites ml-‘. Regardless of medium exchange, cell concentration declined there-

after until the population reached a minimum concentration about day 8. This peak was followed by several others, which were similar to the first one, showing the same maximum cell concentration and

similar growth and declining phases. The minimum cell concentrations reached at the end of the declin-

ing phases increased in successive waves, so that every new wave had a higher starting population

than the former one. Hence the amplitude of succes-

sive waves decreased constantly, leading to a virtually constant cell concentration of (4-7) X lo6 parasites ml-’ after 4 to 6 marked waves.

Interestingly, at least in the beginning of these

cultures the oscillating waves were similar to those observed in natural infections caused by pleomorphic

strains, and regarding the different morphologies appearing in these cultures, we also found a similar development as in vivo. The culture was started with a pure long-slender population and no other morphologies appeared during the exponential growth

phase. In the declining phase, however, more and more intermediate and stumpy forms have been ob-

served and almost no dividing forms were present.

Slender forms appeared again in the late declining phase, and when the minimum cell concentration

was reached, a nearly pure long slender population was found, leading to a new exponential growth

phase. During the course of cultivation, the amount of intermediate and stumpy forms increased signifi-

cantly and all different morphological stages could be seen at the same time after about 4 weeks, when

the population density remained high (data not shown). During this time, the VSG type remained essentially unchanged as judged by immunofluores-

cence using type-specific antibodies and FITC- labelled secondary antibodies, i.e., unlabelled para-

sites have only been found occasionally (data not shown).

0 2 4 6 8 10 12 14 16 16 20 22 24 26 26 30 32 34 36 36 40 42 44

Time [days]

Fig. 2. Long-tern cultivation of bloodstream forms. The entire culture medium was replaced daily as described in Materials and methods.

The growth curve is representative for 4 independently performed experiments with comparable results.


3.4. Trypanosomes taken from long-term culture showed a different growth characteristic 108

Trypanosomes were taken after 40 days, adjusted to 1 X lo5 cells ml-’ and used for long-term cultivation, replacing the medium at daily intervals. In these experiments, cells started to grow exponentially after ;

a lag phase of about 1 day. They grew up to a cell 2

concentration of 7 x lo6 trypanosomes ml-’ and : 3

could be kept at a concentration of about 4 X lo6 cells ml- ’ thereafter (Fig. 3). No marked declining

z 10’ 2 :

phase appeared during the following weeks and all morphological stages were present at the same time.

r

i t I I I / I1 I I I

0 1 2 3 4 5 6 7 6 9 10

Time [days]

Fig. 3. Trypanosomes taken after 40 days from long-term cultures

showed a different growth characteristic compared with try-

panosomes taken from stabilates. (01, trypanosomes taken from a

long-term culture at day 40, (01, trypanosomes taken from a

culture stabilate. Culture media were replaced every 24 h. No

marked declining phases appeared when cells were taken from

long-term cultures, but a stationary phase was reached and cell

concentration maintained at about 4 X lo6 cells ml-‘.

r

1 2 3 4 5 6 7 6 9 10 11 12713

Time [days]

Fig. 4. Effect of medium exchange on cultures of procyclic

trypanosomes. Culture medium was replaced every 24 h. A maxi-

mum cell concentration of about 7 x lo7 trypanosomes ml-’ was

reached and cells could be propagated at population densities

between 3 and 7 X lo7 parasites ml-’ for more than 8 days.

3.5. Effect of medium exchange on procyclic cultures

The effect of medium exchange was also tested on procyclic cultures. In this case, replacing the culture medium every 24 h also led to a prolonged exponential growth phase and a higher maximum cell concentration of about 7 X lo7 trypanosomes ml-’ (Fig. 4). In contrast to bloodstream forms, however, procyclic insect forms entered a stationary phase at maximum cell concentration, and oscillating waves, as seen in bloodstream form cultures, were not observed. Nevertheless, medium exchange at regular intervals improved cultivation conditions significantly and led to a 4-fold increase of cell concentration compared to the conventional cultivation technique.


3.6. Accumulation of a growth inhibitory factor in conditioned media

Cell-free cMEM was mixed with fresh MEM and following addition of supplements used to start cultivation experiments. As shown in Fig. 5, the ability of cMEM to inhibit cell growth increased during the course of cultivation as long as viable cells were present. Cell growth was inhibited in accordance to the amount of cMEM, leading to a 50-75% inhibition (depending on the maximum cell concentration reached during the preceding cultivation) if cMEM prepared from the declining phase was used in a 1:l ratio with fresh medium. In contrast, completely supplemented cell-free media, stored for up to 150 h in the CO,-incubator, supported cell growth even in the absence of fresh media (data not shown).

In a preliminary approach to determine the molecular size of the putative inhibitory factor, conditioned media were dialyzed using membranes of different cut-off sizes. As shown in Fig. 5b, using CentriconTM tubes with a cut-off of 3 kDa, more than 95% of the

a

0 24 48 72 96 120 144”

Time [h]

inhibitory component appears in the dialysate, whereas only negligible amounts remained in the retentate.

4. Discussion

Nearly 10 years ago we developed an axenic cultivation technique to grow bloodstream form trypanosomes in vitro, showing that cysteine is an essential growth factor for this parasite [6,7,9]. Ever since, we and others extensively used this system to grow trypanosomes and to investigate the ultimate cause of growth limitation, which is restricted to about 5 X lo6 trypanosomes ml-’ in all cultivation systems available to date, although the same strain yields about 1 X 10’ cells ml - ’ in laboratory animals and procyclic forms grow up to about 5 X lo7 cells ml-’ in culture. In an earlier study [7] we have shown, that: firstly, bloodstream form parasites can only be grown constantly if subculturing occurs during the logarithmic growth phase; secondly, essential

b

Fig. 5. Accumulation of a growth inhibitory factor in conditioned ,media. All experiments were performed in duplicate. (a) Trypanosomes

were grown without medium exchange and cMEM were prepared from the corresponding cultures at various time points ( 0 ). These cMEM

were mixed in a 1:l ratio with MEM and, after addition of all supplements, used to grow long-slender parasites. Growth inhibition (B) is

expressed as a percentage of control cultures. The results are the means of 3 independent experiments. Standard deviations are within the

range of + 5%. (b) cMEM prepared from culture medium after 144 h (see panel a) was fractionated as described in Materials and methods.

Fractions were mixed with fresh medium and assayed as above: 1, untreated cMEM; 2, dialysate; 3, retentate. Error bars show the standard

deviation for 3 independent experiments.


substrates like glucose, amino acids, or serum com- ponents are not limited during the course of cultivation; and thirdly, the parasites of the stationary phase

resembled short-stumpy-like cells as judged by morphological alterations at the light microscopical and electron-microscopical level, as well as by the expression of specific enzymes like proline oxidase and

cytochromes. Based on these results we concluded that, in analogy to a natural infection, the parasite

may differentiate from the long-slender to the short- stumpy form under cultivation conditions, although

this particular strain is monomorphic and will not undergo differentiation in infected laboratory animals. In the light of our new data, we would like to discuss whether or not the results obtained in vitro

may simulate a natural infection.

4.1. Can monomorphic strains be used to investigate differentiation in vitro?

Monomorphic strains are highly adapted to labo-

ratory animals (rats and mice) and grow logarithmi- cally until the animal dies. The generation doubling time is about 6-8 h, and thus a population density

between 5 X lo* and 1 X lo9 trypanosomes ml-’ is reproducibly reached after 2 to 4 days, starting with an inoculum of about 1 X lo6 parasites ml-‘. Obvi-

ously, there is no effective immune response and a cyclic parasitaemia, as seen in natural infections, will not occur. Accordingly, parasites of short-stumpy

morphology are not observed in this case. In contrast, pleomorphic strains show a decreased genera-

tion doubling time, reach peak densities at lower cell concentrations and show a typical oscillating growth curve [ll]. Since similar differences are obtained in vitro, monomorphic strains grow more readily and

more reproducibly in culture than pleomorphic strains. Surprisingly, however, a rather low peak cell density is obtained in culture (up to 5 X lo6 cells ml-’ with monomorphic strains and even less with

pleomorphic ones) and experimental evidence for the appearance of short-stumpy forms during the course of cultivation was found earlier [7]. Entering the stationary phase seemed not to be due to substrate limitation, because addition of fresh medium or serum

could not prolong cell growth. In this paper we present data suggesting that a cyclic infection may be simulated in culture if medium is replaced at

regular intervals. In addition to our earlier observa- tion, where we demonstrated that stationary form parasites in culture resemble short-stumpy forms

morphologically and biochemically [7], we show here that transition from dividing to stationary phase forms is inducible by transferring long-slender parasites into cMEM. Moreover, activity of S-adeno-

sylmethionine decarboxylase and omithine decar-

boxylase, two key enzymes of the polyamine metabolism, is lost during this transition with a t,,, of 7 or 10 h, respectively, also indicating a controlled mechanism (Selzer and Duszenko, unpub- lished results). To us these results seem to be most readily explained by assuming a trypanosome-de-

rived factor which, reaching a threshold concentration in the medium, induces transition to a non-dividing population. Since experimental evidence sug-

gested that the latter form may represent the short- stumpy form, we questioned if cultivation conditions may cause differentiation of a monomorphic strain. It

is interesting to note, by the way, that observations have been published some years ago, indicating that

monomorphic strains show pleomorphism if grown in a cow instead of laboratory animals [12].

4.2. What causes differentiation from the long-slender to the short-stumpy form?

To date, there is no satisfactory explanation about the molecular mechanism which leads to the forma-

tion of short-stumpy forms. Based on our results, we would like to propose that trypanosomes may con-

stantly produce and, above a certain threshold concentration, respond to a differentiation factor, similar to earlier proposals by other authors [13-161. Con- centration of such a factor would be low as long as

the population density is low, especially because this factor would be diluted within the blood and lym-

phatic system and could also be metabolically degraded. Factor concentration would increase, however, as the parasites grow. In order to respond properly, trypanosomes would require a receptor and ligand-receptor interaction which would limit the maximum population density by the induction of non-dividing parasites. Such a mechanism, like the cytokine system in other cells, would ensure commu- nication among the parasites and, teleologically


speaking, would avoid a life-threatening para-

sitaemia, which would render trypanosomiasis epi- demiologically irrelevant. Differences in factor pro-

duction, receptor expression, or ligand-receptor interaction could easily account for the differences in maximum cell yield and growth characteristics ob-

served in different strains. Assuming a reduced factor response in monomorphic strains, the factor concentration would remain too low to induce differentiation in infected laboratory animals, whereas in culture, where the factor can neither be diluted nor

degraded, differentiation would occur. This mecha-

nism could also account for the different population densities obtained with monomorphic and pleomorphic strains in vivo and in vitro. Reducing the factor

concentration by replacing the medium or by dialysis using trans-well plates, should lead to the observed higher cell densities without increasing the popula-

tion doubling rate. However, since population density and thus factor production increases exponentially, maximum cell yield is still limited, although a

4-fold increase was measured (2 X lo7 instead of 5 X lo6 cells per ml).

4.3. Does the damped oscillation growth curve obtained in uitro resemble a cyclical infection?

Trypanosomes undergo antigenic variation, and the characteristic ups and downs of a natural infection are usually explained by parasite killing due to

the appearance of VSG-specific antibodies and the successive growth of parasites expressing a different antigenic variant, not yet recognized by antibodies [2,17]. Interestingly, however, in mice infected with pleomorphic strains, a decline of the population is

frequently observed within 4 days after infection,

before specific antibodies have been formed, which takes about 7 days [18]. Moreover, a growth curve, similar to our results in cultures (Fig. 31, have been observed in immunosuppressed mice infected with a pleomorphic strain [14]. Both results are in favour of a differentiation process induced by the parasite and

not by the host. Taking these results into considera- tion, we interpret our data as follows: We start with monomorphic trypanosomes, i.e., with a homogeneous population of long-slender morphology. As the cell density increases, a factor or metabolite accumulates within the medium, which eventually induces

profound metabolic changes, including degradation

or inactivation of key enzymes of polyamine metabolism and expression of proline oxidase and cytochromes [7]. This differentiation process leads to

non-dividing parasites and, due to cell death, to a decline of the cell number. If the medium is not replaced, the process becomes irreversible and the parasites die out. If the medium is replaced at regular

intervals, however, parasites again enter the cell division cycle after some days and population density increases. At the moment, we cannot disregard

the possibility that during the declining phase some parasites remain long-slender and become the start-

ing point of the newly appearing population. How- ever, extrapolation of the second growth curve to the

time point of the first peak (after 48 h), would lead to less than one dividing parasite. We thus favour the idea that transition to a non-dividing form is re- versible. Obviously, once transition is initiated, it

cannot be reverted immediately, no matter how often the medium is exchanged. Instead, there seems to be a programmed schedule which has to be passed before cell division becomes possible again. As culti-

vation proceeds, the starting population of a new cycle becomes less homogeneous, leading to a damped oscillation curve. Accordingly, try-

panosomes taken from the late stage (after 40 days), diluted to 1 X lo5 cells ml-’ and grown in fresh

medium, showed a growth characteristic similar to pleomorphic strains grown in immunosuppressed

mice [14]. The immune response of an infected animal leads to the extinction of recognized variants and thus to marked incisions between the population waves. This process, however, does not induce dif-

ferentiation, which, if our hypothesis is correct, would be an independent mechanism to control para-

sitaemia. Accordingly, the regular switching rate of VSG expression (about 10m6 per generation) [19] should not be changed during cultivation and expression of different VSG antigens in successive waves

in culture is not to be expected and was indeed not observed by immunofluorescence (data not shown).

This proposed differentiation mechanism is one possible attempt to explain the often puzzling experimental data and is thus open for discussion. Efforts to isolate and characterize the corresponding factor from conditioned media are in progress in our laboratory.

166 F. Hesse et al./Molecular and Biochemical Parasitology 70 (1995) 157-166

Acknowledgements [91

We thank R. Hiimke, V. Kolb, T. LiSfler, and U. Wille for technical assistance and discussion. This study was supported by a grant from the Deutsche Forschungsgemeinschaft.

DO1

1111

References

[II

El

[31

[41

[51

161

[71

@I

Ross, R. and Thomson, D. (1910) A case of sleeping sick-

ness studied by precise enumerative methods: regular period-

ical increase of the parasites disclosed. Proc. R. Sot. Series B

82, 411-415.

Borst, P. (1986) Discontinuous transcription and antigenic

variation in trypanosomes. Annu. Rev. Biochem. 5.5, 701-

732.

Ormerod, W.E. (1979) Development of Trypanosoma brucei in the mammalian host. In: Biology of the Kinetoplastida

(Lumsden, W.H.R. and Evans, D.A., eds.), Vol. 2, pp.

340-393.

Vickerman, K. (1985) Developmental cycles and biology of

pathogenic trypanosomes. Br. Med. Bull. 41, 105-114.

Bienen, E.J., Hammadi, E. and Hill, G.C. (1981) Try- panosoma brucei: Biochemical and morphological changes

during in vitro transformation of bloodstream to procyclic

trypomastigotes. Exp. Parasitol. 51, 408-417.

Duszenko, M., MtihlstBdt, K. and Broder, A. (1992) Cysteine

is an essential growth factor for Trypanosoma brucei blood-

stream forms. Mol. Biochem. Parasitol. 50, 269-274.

Hamm, B., Schindler, A., Mecke, D. and Duszenko, M.

(1990) Differentiation of Trypanosoma brucei bloodstream

trypomastigotes from long slender to short stumpy-like forms

in axenic culture. Mol. Biochem. Parasitol. 40, 13-22.

Cross, G.A.M. (197.5) Identification, purification and proper-

ties of clone-specific glycoprotein antigens constituting the

surface coat of Trypanosomo brucei. Parasitology 71, 393-

417.

WI

r131

[141

[I51

(161

[171

[I81

[191

Duszenko, M., Ferguson, M.A.J., Lament, G.S., Rifkin, M.R.

and Cross, G.A.M. (1985) Cysteine eliminates the feeder cell

requirement for cultivation of Trypanosoma brucei blood-

stream forms in vitro. J. Exp. Med. 162, 1256-1263.

Baltz, T., Baltz, D., Giroud, C.B. and Crockett, J. (1985)

Cultivation in a semi-defined medium of animal infective

forms of Trypanosoma brucei, T. equiperdum, T. euansi, T.

rhodesiense and T. gambiense. EMBO J. 4, 1273-1277.

Vickerman, K. (196.5) Polymorphism and mitochondrial ac-

tivity in sleeping sickness trypanosomes. Nature 208, 762-

766.

Black, S.J., Jack, R.M. and Morrison, W.I. (1983) Host-

parasite interactions which influence the virulence of Try-

panosoma (Trypanozoon) brucei brucei organisms. Acta

Trop. 40, 11-18.

Giffin, B.F. and McCann, P.P. (1989) Physiological activa-

tion of the mitochondrion and the transformation capacity of

DFMO induced intermediate and short stumpy bloodstream

form trypanosomes. Am. J. Trop. Med. Hyg. 40, 487-493.

Seed, JR. and Sechelski, J.B. (1989) Mechanism of long

slender (LS) to short stumpy (SS) transformation in the

African trypanosomes. J. Protozool. 36, 572-577

Clayton, C.E. (1992) Developmental regulation of nuclear

gene expression in Trypanosoma brucei. Progr. Nucleic Acids

Res. Mol. Biol. 43, 37-66.

Pays, E., Hanocq-Quertier, J., Hanocq, F., Van Assel, S.,

Nolan, D. and Rohn, S. (1993) Abrupt RNA changes precede

the first cell division during the differentiation of Try-

panosoma brucei bloodstream forms into procyclic forms in

vitro. Mol. Biochem. Parasitol. 61, 107-114.

Cross, G.A.M. (1990) Cellular and genetic aspects of anti-

genie variation in trypanosomes. Annu. Rev. Immunol. 8,

83-110.

Newson, J., Mahan, SM. and Black, S.J. (1990) Synthesis

and secretion of immunoglobulin by spleen cells from resis-

tant and susceptible mice infected with Tvpanosoma brucei

brucei GUTat 3.1. Parasitol. Immunol. 12, 125-139.

Lament, G.S., Tucker, R. and Cross, G.A.M. (1986) Analysis

of antigen switching rates in Trypanosoma brucei. Parasitol-

ogy 92, 355-367.

Documents

A novel cultivation technique for long-term maintenance of bloodstream form trypanosomes in vitro