Impact of Continuous Axenic Cultivation in Leishmania

infantum Virulence
Diana Moreira1., Nuno Santarém1., Inês Loureiro1, Joana Tavares1,2, Ana Marta Silva1, Ana Marina
Amorim1, Ali Ouaissi1,3, Anabela Cordeiro-da-Silva1,2", Ricardo Silvestre1*"
1 Parasite Disease Group, IBMC - Instituto de Biologia Molecular e Celular, Universidade do Porto, Porto, Portugal, 2 Departamento de Ciências Biológicas, Faculdade de
Farmácia, Universidade do Porto, Porto, Portugal, 3 INSERM, UMR, CNRS 5235, Université Montpellier II, Montpellier, France

Abstract
Experimental infections with visceral Leishmania spp. are frequently performed referring to stationary parasite cultures that
are comprised of a mixture of metacyclic and non-metacyclic parasites often with little regard to time of culture and
metacyclic purification. This may lead to misleading or irreproducible experimental data. It is known that the maintenance
of Leishmania spp. in vitro results in a progressive loss of virulence that can be reverted by passage in a mammalian host. In
the present study, we aimed to characterize the loss of virulence in culture comparing the in vitro and in vivo infection and
immunological profile of L. infantum stationary promastigotes submitted to successive periods of in vitro cultivation. To
evaluate the effect of axenic in vitro culture in parasite virulence, we submitted L. infantum promastigotes to 4, 21 or 31
successive in vitro passages. Our results demonstrated a rapid and significant loss of parasite virulence when parasites are
sustained in axenic culture. Strikingly, the parasite capacity to modulate macrophage activation decreased significantly with
the augmentation of the number of in vitro passages. We validated these in vitro observations using an experimental murine
model of infection. A significant correlation was found between higher parasite burdens and lower number of in vitro
passages in infected Balb/c mice. Furthermore, we have demonstrated that the virulence deficit caused by successive in
vitro passages results from an inadequate capacity to differentiate into amastigote forms. In conclusion, our data
demonstrated that the use of parasites with distinct periods of axenic in vitro culture induce distinct infection rates and
immunological responses and correlated this phenotype with a rapid loss of promastigote differentiation capacity. These
results highlight the need for a standard operating protocol (SOP) when studying Leishmania species.

Citation: Moreira D, Santarém N, Loureiro I, Tavares J, Silva AM, et al. (2012) Impact of Continuous Axenic Cultivation in Leishmania infantum Virulence. PLoS Negl
Trop Dis 6(1): e1469. doi:10.1371/journal.pntd.0001469
Editor: Shaden Kamhawi, National Institutes of Health, United States of America
Received April 22, 2011; Accepted November 20, 2011; Published January 24, 2012
Copyright: ß 2012 Moreira et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by Fundação para a Ciência e Tecnologia (FCT) and FEDER Ciência 2010, project numbers PTDC/SAU-FCF/101017/2008, PTDC/
SAU-FCF/100749/2008, program Ciência 2008 and FCT grants SFRH/BDP/48340/2008, SFRH/BD/64528/2009, SFRH/BD/37352/2007. The funders had no role in
study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: [email protected]
. These authors contributed equally to this work.
" These authors also contributed equally to this work.

Introduction without limitations, it has been demonstrated that continuous

culture over time induces loss of virulence. In fact, long-term in
Protozoan parasites of the genus Leishmania undergo several vitro culture of promastigotes was one of the first empirical
developmental transitions during their life cycle. Ingestion of approaches to efficiently identify parasite virulence genes leading
infected macrophages during a blood meal by the sandfly vector to the experimental development of attenuated strains [4].
leads to the release of intracellular amastigotes into the vector’s Similarly, long-term in vitro growth of drug-resistant parasites
midgut. This abrupt change in environment induces the was suggested to mediate a loss of the resistance phenotype [5].
transformation into extracellular procyclic promastigotes. The This can be due to either loss of virulence factors induced by the
procyclic form within the vector midgut replicates and ultimately lack of a survival pressure or due to disadvantageous adaptations
transforms into virulent metacyclic promastigotes, in a complex to the media resulting in phenomena similar to clonal selection [6].
process that encompass parasite migration towards the upper gut Either way, alterations in the physiology of the parasite that are
of sandfly vector [1]. In laboratory conditions, it is possible to induced by long-term growth in these media may lead to
achieve indefinite promastigote growth outside the sandfly using misinterpretation and contradictory results. Thus, one must
several established media; the procyclic forms correspond to carefully consider the influence of the different laboratorial factors
promastigotes in exponential phase of growth that will eventually in order to minimize these variables.
pass into a stationary phase, a fraction of these stationary parasites The current study is based upon the hypothesis that
differentiates into the metacyclic form, with properties resembling maintenance of Leishmania spp. in axenic in vitro culture results in
those of sand fly promastigotes [2,3]. Although stationary-phase a progressive loss of virulence quickly generating a significant bias
promastigotes with undefined in vitro passages are commonly used towards the experimental data. We have compared the in vitro and

www.plosntds.org 1 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

Author Summary cultures were initiated at 106 parasites/ml and passed each 5 days.
Promastigote to amastigote differentiation was achieved by cul-
Protozoan of the genus Leishmania undergo several turing 107 stationary phase promastigotes/ml at 37uC in a cell free
developmental transitions during its life cycle. Leishmania culture medium (MAA20) [7]. Amastigote to promastigote dif-
alternates between two morphologically distinct forms, ferentiation was performed by culturing 107 axenic amastigotes/
promastigotes (insect stage) and amastigotes (vertebrate ml in complete RPMI medium for 4 days at 27uC. In alternative,
stage). Most of the available information about Leishmania spleens of infected Balb/c mice were placed in similar culture
spp. has been obtained from studying in vitro cultured conditions for 7 days.
promastigotes, an excellent experimental model for the
different developmental stages present in the insect
vector. Although promastigotes are grown in a controlled
Ficoll density purification assay
environment, the maintenance of long term culture results Metacyclic promastigotes were purified from cultures with 3, 5
in loss of virulence, which can lead to a misinterpretation or 9 days or from 5-day cultures with 4, 21 and 31 (P4, P21 and
and often contradictory experimental results. It is then of P31) in vitro passages by Ficoll density gradient, as previously
great interest to unravel the defects arising from sustained described [8]. Briefly, 6 ml of 40% Ficoll was overlaid by 6 ml of
axenic parasite culture in laboratory settings. The authors 10% Ficoll in RPMI base. Then, 6 ml of PBS containing 1.26109
demonstrate a correlation between the maintenance of parasites was placed at the top of the Ficoll gradient. The step
parasite culture with a growing defect of the promastigote gradient was centrifuged for 10 minutes at 370 g at room tem-
form to differentiate in the mammalian amastigote form. perature without brake. The metacyclics promastigotes were re-
This research provides a biological explanation for the loss covered from the layer between 0% and 10% Ficoll solution.
of virulence due to sustained parasite culture and Metacyclic promastigotes, identified by morphological criteria, i.e.,
discusses the impact for all experimental work done with short and slender with a long flagellum twice the body length using
visceral Leishmania species. phase contrast on a Nikon Eclipse 80i.

qPCR analysis
in vivo infections and focused on the influence that axenic parasite Total RNA was isolated from cells with the TrizolH reagent
growth and long-term maintenance can have on in vitro infections (Invitrogen, Barcelona, Spain), according to the manufacturer’s
outcome. Our results demonstrate, for the first time, that the loss instructions. Briefly, parasites were washed with ice-cold phos-
of virulence caused by the maintenance of axenic promastigotes in phate-buffered saline (PBS), harvested and homogenized in 800 ml
culture can be the result of a growing inability to differentiate into of Trizol by pipetting vigorously. After addition of 160 ml of
amastigote forms. Moreover, the induction of differentiation from chloroform, the samples were vortexed, incubated for 2 min at
promastigote to amastigote and then back to promastigote forms room temperature and centrifuged at 12.000 g, for 15 min, at
both in vitro and in vivo was capable to restore parasite virulence. 4uC. The aqueous phase containing RNA was transferred to a new
Overall, our study demonstrated the need of a standard operating tube and RNA precipitated with 400 ml of isopropanol for at least
protocol (SOP) to study visceral Leishmania spp. highlighting the 10 min at room temperature. Following a 10 min centrifugation at
crucial importance for proper control of parasite cultures in studies 12.000 g, the pellet was washed with 1 ml of 75% ethanol and
focusing on the mammalian stage, such as drug development or resuspended in 10 ml of 60uC heated RNase free water. The RNA
vaccine trials. concentration was determined by using a Nanodrop spectropho-
tometer (Wilmington, DE, USA) and quality was inspected for
Materials and Methods absence of RNA degradation or genomic DNA contamination,
using the Experion RNA StdSens Chips in the ExperionTM
Animals and parasites automated microfluidic electrophoresis system (BioRad Hercules,
Ten to twelve-week-old female Balb/c mice were obtained from CA, USA). RNA was stored at 280uC until use. RT was per-
Instituto de Biologia Molecular e Celular (IBMC; Porto, Portugal) formed with equal amounts of total extracted RNA (1 mg) obtained
animal facilities. Under laboratory conditions, the animals were from parasites recovered from different experimental conditions
maintained in sterile cabinets and allowed food and water ad by using Superscript II RT (Gibco BRL) and random primers
libitum. Animal care and procedures were in accordance with (Stratagene). Real-Time quantitative PCR (qPCR) reactions were
institutional guidelines. All conducted experiments were done in run in duplicate for each sample on a Bio-Rad My Cycler iQ5
accordance with the IBMC.INEB Animal Ethics Committee and (BioRad, Hercules, CA, USA). Primers sequences were obtained
the Portuguese Veterinary Director General guidelines. RS has an from Stabvida (Portugal) and thoroughly tested. qPCR was per-
accreditation for animal research given from Portuguese Veteri- formed in a 20 ml volume containing 5 ml of complementary
nary Direction (Ministerial Directive 1005/92). A cloned line of cDNA (50 ng), 10 ml of 26 Syber Green Supermix (BioRad,
virulent L. infantum (MHOM/MA/67/ITMAP-263) was grown at Hercules, CA, USA), 2 ml of each primer (250 nM) and 1 ml H2O
26uC in RPMI 1640 medium (Lonza, Swtzerland) supplemented PCR grade. Specific primers for histone H4 (forward: 59
with 10% heat-inactivated Fetal Bovine Serum - FBS (Lonza, ACACCGAGTATGCG -39; reverse: 59- TAGCCGTAGAG-
Switzerland), 2 mM L-glutamine, 100 U/ml penicillin, 100 mg/ GATG-39; LinJ35.1400 histone H4: Gene ID 5073031), Small
ml streptomycin and 20 mM HEPES buffer. The MHOM/MA/ Hydrophilic Endoplasmic Reticulum-associated Protein (SHERP)
67/ITMAP-263 clone (zymodeme MON-1) was originally isolated (forward: 59 CAATGCGCACAACAAGAT -39; reverse: 59-
from the bone marrow of a human patient in Morocco and cloned TACGAGCCGCCGCTTA-39; LinJ23.1190 SHERP: Gene ID
by micromanipulation. In some experiments, a previously uncha- 5069222) and rRNA45 (forward: 59CCTACCATGCCGTG-
racterized field attenuated L. infantum strain was used (species TCCTTCTA -39; reverse: 59- AACGACCCCTGCAGCAATAC
confirmed by pteridine reductase 1 sequencing and currently -39) [9] were used for amplification. After amplification, a
under ongoing characterization in our laboratory). To minimize threshold was set for each gene and cycle threshold-values (Ct-
the possibility of clonal bias, we have performed three independent values) were calculated for all samples. Gene expression changes
recoveries of parasite from Balb/c mice for these experiments. All were analyzed using the built-in iQ5 Optical system software v2.1

www.plosntds.org 2 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

(Bio-Rad laboratories, Inc). The results were normalized using as suspended in sterile PBS. A volume of 200 ml of PBS containing
reference gene the rRNA45 rRNA sequence [9]. 108 parasites was injected intraperitoneally. Mice of each group
were sacrificed at 56 days post-infection. The parasite burden in
Viability analysis the spleen and liver was determined by limiting dilution as
Purified and non-purified promastigotes at a density of 105/ml previously described [10].
were washed and suspended in Annexin V binding buffer.
Parasites were incubated at room temperature for 15 minutes Statistical analysis
with AnnexinV-Cy5 (BD Pharmingen, San Diego, CA) and 7- The data was analyzed using the non-parametric Kruskal-
AAD (Sigma). Parasites subjected to Ultra Violet light during Wallis test followed by Dunn posttest for multiple comparisons
30 minutes and kept in culture for 4 hours were used as positive when necessary.
control. In amastigote differentiation, 26106 cells with 1 mM of
propidium iodide (PI) were used. Data were collected in a BD Results
FACScalibur cytometer (20.000 gated events) and analyzed by
FlowJo software (Ashland, OR). The maintenance of L. infantum promastigotes in axenic
cultures results in diminished virulence
Promastigote CFSE-labeling We started by clarifying our in vitro model of L. infantum infection
7
Purified and non-purified promastigotes (6610 /ml) were in relation to the parasite development stage. L. infantum parasites
washed two times, suspended in PBS containing 5 mM of car- recovered from the spleen of infected Balb/c mice were used to
boxyfluorescein succinimidyl ester (CFSE) (Invitrogen Molecular start axenic cultures at a 106 parasites/ml. The first task was to
probes, Eugene, Oregon) and incubated at 37uC for 10 minutes. clearly define the culture time frame in which we can recover
Labeled parasites were washed, incubated at 4uC for 5 minutes. stationary parasites. Performing basic cell cycle analysis we
Parasites were washed again to remove the exceed CFSE dye and excluded the use of the parasites until 2 days of culture because
suspended in culture medium before proceeding to macrophage there was still significant active division (Fig. S1A). In order to
infections. For promastigote to amastigote differentiation and evaluate the infectivity of stationary L. infantum, we used CFSE-
proliferation analysis, 107 CFSE-labeled promastigotes were placed labeled stationary promastigotes recovered at 3, 5 and 9 days of in
on 1 ml of MAA20. Each day, 100 ml of culture added with 1 mM of vitro growth and BMMø as infection cellular target. Our data
PI was analyzed by flow cytometry. Axenic amastigotes, identified demonstrate that 3rd culture-day L. infantum promastigotes were
by the absence of visible flagella and oval shape body, were observed significantly less infectious when compared with the 5th and 9th
in phase contrast on a Nikon Eclipse 80i. days of culture (Fig. S1B). These differences were already observed
at 4 hours post-infection indicating a deficient parasite uptake with
3rd culture-day L. infantum promastigotes. Intraphagolysosomal
In vitro macrophage infection
adaptation mechanisms do not appear to be involved in the
Cell suspension of bone marrow was obtained by flushing the
infection differences since similar infection percentages reductions
femurs of susceptible Balb/c mice. The cell suspension was cultured
were found between 4 and 24 hours (15.660.4 for 3rd culture-day;
in Dulbecco’s modified Eagle’s medium (DMEM) (Lonza, Switzer-
15.061.2 for 5th culture-day; 18.161.1 for 9th culture-day, when
land), supplemented with 10% heat-inactivated FBS (Lonza, comparing 4 with 24 hours post-infection). Many reports now
Switzerland), 2 mM L-glutamine, 100 U/ml penistreptomycin relate virulence with parasites culture viability [11,12]. In order to
and 1 mM sodium pyruvate. After overnight incubation at 37uC, lay down the hypothesis that differences in infectivity could be
non-adherent cells were recovered (300 g for 10 min, at room attributed to non-viable parasites, we evaluated the percentages of
temperature) and cultured in 24-well culture dishes at 26105cells/ apoptotic or necrotic parasites by AnnV/7AAD labeling [13].
ml in supplemented DMEM. For bone-marrow derived macro- Nevertheless, no significant differences were found between all
phages (BMMø) differentiation 10% L-929 cell conditioned culture days (data not shown).
medium (LCCM) was added at days 0 and 4. At day 7 of culture, Several groups have already reported that long-term in vitro
CFSE labeled promastigotes were incubated with the BMMø at a cultivation (more than 12 months) of Leishmania spp. leads to a
10:1 ratio. After 4 hours, infection was stopped the infection rates totally avirulent promastigote population [6,14]. According to
were determined at 4, 24 and 48 hours post-infection by a BD these findings, we decided to evaluate if the sustained maintenance
FACScalibur cytometer and analyzed by FlowJo software. In some of L. infantum promastigotes in axenic culture at shorter time
experiments, BMMø were infected and submitted to lipopolysac- periods lead to distinct BMMø in vitro infection rates with
charide (LPS) stimulus. Briefly, four hours after infection, infection distinctive immunologic phenotypes. In order to accomplish this,
was stopped and 1 mg/ml of LPS (Sigma) added. Twenty-four hours we maintained L. infantum promastigotes recovered from the spleen
post-infection, BMMø culture supernatants were collected for of infected mice for 4, 21 and 31 passages, which are equivalent to
cytokine quantification by Enzyme-Linked Immunosorbent Assay - 20, 105 and 155 division events, considering simple exponential
ELISA (TNF-a, IL-12p40, IL-6 and IL-10), using commercial growth. The long-term maintenance of L. infantum in culture did
sandwich immunoassay kits (Biolegend and BD, San Diego, CA). not modify the promastigote growth behavior (Fig. 1A) neither
Also, BMMø were recovered at 24 hours post-infection for surface their viability that was always superior to 90% (data not shown).
co-stimulatory markers analysis. Thus, BMMø were stained with Taking into account the distinct infection profiles depicted in Fig.
CD40-PE and MHCII-APC at 4uC, during 30 minutes in the dark. S1B, we chose 5-day culture promastigotes to compare infectivity.
The cells were then washed in PBS and suspended in 200 ml of PBS- When non-purified parasite cultures were used to in vitro infect
2% FBS. Data were collected by a BD FACScalibur cytometer and BMMø, a marginal but significant loss of infectivity at 48 hours for
analyzed by FlowJo software. P21 and P31 when compared to P4 was observed (Fig. 1B).
Metacyclic forms have been understood to be the most infective
Animal experiments and parasite quantification parasite form [8,15]. Therefore, we enriched the promastigote
Promastigotes recovered from stationary culture with 4, 21 and culture recovered from P4, P21 and P31 in metacyclics recovered
31 in vitro passages stationary-culture were collected, washed and by Ficoll density gradient, herein referred as Ficoll-purified

www.plosntds.org 3 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

promastigotes [16], and analyzed their infectivity in primary the differences observed in parasite infection rate and co-
BMMø cells (Fig. 1C). When Ficoll-purified metacyclic promas- stimulatory markers found with stationary-phase promastigotes
tigotes from these cultures were used, differences were abrogated in different passages, we further investigated whether the cytokine
irrespective of the passage used (Fig. 1C). To confirm the en- bias was similarly altered. Indeed, the use of Ficoll-purified meta-
richment of metacyclic promastigotes in this fraction and as an cyclic enriched promastigotes, whatever their source, shifted the
internal control of our experimental conditions, we analyzed the cytokine environment towards an anti-inflammatory ratio. Al-
expression of two genes, SHERP and histone H4 that can be used though some statistical differences were found between Ficoll-
to evaluate metacyclogenesis. SHERP gene is found to be up- purified promastigotes from different passages, all displayed lower
regulated in infective metacyclic promastigotes [17]. On the other pro-inflammatory/IL-10 ratios when compared with LPS stimu-
hand, higher expression of histone H4 is associated with ex- lation (Table 1). Overall, these data suggests that when Leishmania-
ponential phase promastigotes [18]. Indeed, the qPCR analysis infected BMMø are faced with an inflammatory stimulus, there is
demonstrated a significant increase in the SHERP mRNA tran- a specific overall loss of modulatory capacity that seems to be
scripts in Ficoll-purified promastigotes when compared with non- related to the highly immunoregulatory population of metacyclic
purified parasite cultures (Fig. S2). These results suggested that the parasites.
maintenance of L. infantum in axenic cultures resulted in a viru-
lence leakage affecting their infectivity probably by the loss of Sustained culture of L. infantum promastigotes results in
metacyclic parasites. Nevertheless, no significant differences were
an in vivo loss of virulence
found between the percentage of Ficoll-purified promastigote
We have above demonstrated that sustained axenic parasite
recovery from each culture (P4: 6.660.1%; P21: 3.461.6%; P31:
culture results in a rapid loss of in vitro virulence. Previous reports
4.060.9%).
demonstrating an in vivo loss of virulence were based on long-term,
usually more than 1 year, parasite maintenance [6,14]. However,
Maintenance of long-term axenic L. infantum cultures we have observed a clear in vitro defect after only 21 passages.
decrease parasite capacity to modulate host cell Therefore, we decided to validate the observed phenotype by
functions in an inflammatory context performing in vivo infections using the susceptible Balb/c mice
We have above demonstrated that the BMMø infection by L. model. Six weeks after the infectious challenge with non-purified
infantum promastigotes depends not only upon the days of culture stationary-phase promastigotes recovered from P4, P21 or P31
but is significantly modulated by their axenic culture period. cultures, a significant difference was found between P4 and high
However, we were unable to detect any major changes on passage number parasite infections in the liver (Fig. 3A). Similarly,
macrophage activation status when submitted to L. infantum we have observed a significant lower parasite burden in the spleen
infection (data not shown), which can be explained by the of P31 infected mice that P4 infections (Fig. 3B). These results
Leishmania silent entry mechanism [19]. Therefore, we hypothe- confirm the observed in vitro loss of virulence with parasite culture
sized that, when facing an inflammatory stimulus, axenic cultures maintenance.
with high passage number should be less successful in subverting
macrophage effector functions being less capable of promoting
Loss of virulence originates from inadequate capacity to
infection. In order to investigate this hypothesis, we incubated
BMMø cells with Ficoll-purified or non-purified parasites from differentiate into amastigote forms
distinct culture periods, which were 4 hours later submitted to LPS To elucidate the biological mechanisms that account for the loss
stimulation. As before, we observed a decrease of the infection rate of virulence due to long term parasite culture, we started by
with the augmentation of parasite in vitro passages (Fig. 2A). This hypothesizing two major potential reasons: decrease number of
difference was minimized if Ficoll-purified promastigotes were metacyclic promastigotes or inadequate differentiation into
used instead, although it was still statistically significant at amastigote forms. The quantification of Ficoll-purified promasti-
48 hours post-infection (Fig. 2B). LPS stimulation rapidly induces gotes described above did not show any significant differences
a surface up-regulation of MHCII molecules and co-stimulatory among the passages suggesting similar metacyclic quantities.
marker CD40. The analysis of these markers demonstrated that However, the Ficoll density gradient assay is not a specific and
high passage number parasites had lower capacities to counteract sensible test for the quantification of metacyclic promastigotes in a
the LPS activation stimulus (Fig. 2C). Once again, these culture but rather a method for its enrichment. Thus, to evaluate
differences were abrogated when metacyclic-enriched populations the hypothetical deficit on the generation of metacyclic promas-
were used (Fig. 2C). We have also evaluated the levels of secreted tigotes, we have performed in vitro infections using BMMø as
IL-6, IL-12p40 and TNF-a and of the anti-inflammatory IL-10 targets, where we substitute 5% or 10% of non-purified stationary-
cytokine. We found that the capacity to control LPS-induced phase promastigotes from each passage with similar percentages of
cytokines was variable depending on the number of parasite Ficoll-purified fractions of P4 cultures to increase the total
passages, likely reflecting its distinct virulence. While significant percentage of metacyclic promastigote. As a positive control, we
differences were found with P31 parasites in the BMMø secretion used a naturally attenuated L. infantum from which we were unable
levels of IL-6 and TNF-a (Fig. S3A and S3B, respectively), the to recover metacyclic promastigotes. Indeed, although the
major modifications were observed at the IL-12p40 and IL-10 promastigotes of this L. infantum strain presents a similar axenic
levels (Fig. S3C and S3D, respectively). Indeed, P4 parasites were growth curve (Fig. S4), we were always unable to recover by Ficoll
more capable to down-regulate IL-12p40 secretion induced by density gradient relevant number of promastigotes (lower that
LPS stimulus, while increasing IL-10 cytokine secretion. This 0.1% of initial culture) from stationary-phase cultures. The
demonstrates that high passage parasites failed to counteract the quantification of CFSE-positive BMMø demonstrated that
secretion of pro-inflammatory cytokines induced by LPS in a increasing the percentage of Ficoll-purified promastigotes did not
similar manner. Moreover, if a pro-inflammatory/IL-10 ratio is significantly enhance, at any time point, the percentage of infected
constructed, a strong correlation was observed between shorter BMMø for P4 (Fig. 4A), P21 (Fig. 4B) or P31 (Fig. 4C)
axenic culture maintenance periods and lower pro-inflammatory/ promastigotes. However, the opposite was observed with the
IL-10 ratios (Table 1). Since the metacyclic enrichment diminished naturally attenuated strain, where a significant increase of infected

www.plosntds.org 4 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

www.plosntds.org 5 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

Figure 1. Long-term in vitro maintenance of L. infantum promastigotes does not alter parasite growth. L. infantum growth curves were
performed by neubauer chamber counting (A). BMMø were infected at a 1:10 (cell/parasite) ratio with non-purified (B) and Ficoll-purified (C)
promastigotes labeled with CFSE. Data were acquired by FACScalibur cytometer and analyzed by FlowJo software. Three independent experiments
were performed; one representative experiment is shown. The mean and standard deviation are shown. *P,0,05, **P,0,01, ***P,0,001 statistical
significance relatively to P4.
doi:10.1371/journal.pntd.0001469.g001

www.plosntds.org 6 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

Figure 2. Long-term in vitro maintenance of L. infantum promastigotes results in loss of virulence in vitro. BMMø were submitted to LPS
stimulation 4 hours after infection with non-purified (A) or Ficoll-purified (B) CFSE-labeled promastigotes at a 1:10 (cell/parasite) ratio. The percentage of
infected cells was determined by the number of CFSE-positive cells in a FACScalibur cytometer. Expression of MHCII and co-stimulatory molecule CD40 at
24 hours post-infection. Thick line – P31, dotted line - P4, thin line - LPS and shadded hystogram - isotype control (C). Two independent experiments were
performed; one representative experiment is shown. The mean and standard deviation are shown. *P,0,05; **P,0,01 statistical significance relatively to P4.
doi:10.1371/journal.pntd.0001469.g002

macrophages was observed at 48 hours post-infection (Fig. 4D). tigotes to amastigotes forms would select the most virulent
These results demonstrated that the lack of virulence originated parasites in a heterogeneous culture assuring the continuity of
from sustained parasite culture cannot be reverted by the addition competent and adapted parasites. Therefore, to explore this
of enriched-metacyclic fractions. This excludes a defect in the assumption we have differentiated promastigotes from each
capacity to generate metacyclic promastigotes as the inherent passage number in amastigotes both in axenic and in vivo
biological cause for the virulence loss. Therefore, we investigated conditions. Axenic amastigotes were obtained by differentiating
the potential role of inadequate capacity to differentiate in the promastigotes in MAA20 medium [7] for a period of 3 days. The
amastigote form. CFSE-labeled stationary-phase promastigotes viable axenic amastigotes were maintained axenically in culture for
recovered from each passage were placed on MAA20 [7] and 10 days, after which were re-differentiated to promastigotes forms.
followed for three days. We evaluated the promastigote differen- In alternative, we recovered L. infantum parasites from the spleen of
tiation by light microscopy, axenic amastigotes proliferation by infected Balb/c mice by allowing amastigote to promastigote
CFSE labeling and overall viability by PI staining. All cultures differentiation for a period of 7 days. All these promastigotes were
presented axenic amastigotes-like cells after 3 days of differenti- sub-cultured for 4 passages and used to infect BMMø. Remark-
ation (data not shown). However, high passage number promas- ably, we did not observe any difference between infections
tigotes displayed a striking decrease of differentiated cells. To whatever the initial parasite source used (Fig. 6A and B). Again,
quantify these differences, we have assessed the progressive we used as a control the naturally attenuated L. infantum strain.
diminution in the intensity of CFSE staining after the differenti- Although an increase of virulence was observed after the
ation process. In fact, while P4 promastigotes progressively differentiation protocol (Fig. 6A), when compared to non-
diminished CFSE fluorescence (Fig. 5A and B), high passage differentiated parasites (Fig. 4D), we observed a general lower
number promastigotes exhibit a severe defect to proliferate as infection percentage with the exception of 4 hours post-infection.
amastigotes forms, as observed in both histogram curves (Fig. 5A) Overall, these results demonstrate that the defect on virulence due
and quantification of mean fluorescence CFSE intensity (Fig. 5B). to sustained parasite maintenance can be recovered either by in
Moreover, this defect was not correlated with a difference on cell vitro or in vivo full differentiation to amastigote and back to
death, since similar percentages of viable parasites were found for promastigotes forms.
all cultures during the differentiation process (Fig. 5C). Interest-
ingly, the naturally attenuated strain did not display any significant
change to differentiate when compared with P4 promastigotes, Discussion
suggesting a distinct mechanism of loss of virulence that is not Visceral Leishmania infections studies have been the center of
related with the capacity to generate axenic amastigotes. some controversy which can occasionally be traced back to the use
of distinct in vitro promastigotes culture conditions. The plasticity of
L. infantum virulence is restored after full differentiation the Leishmania genome [20] is an important variable to consider
to amastigote forms when axenic promastigotes are used for in vitro or in vivo studies.
It is a current empirical methodology to pass Leishmania spp. Thus, parasite phenotypic plasticity allows it to adapt to the
promastigotes in experimental models to maintain virulence. We environment generating discrepancies between studies in different
have hypothesized that the differentiation process from promas- laboratories even when using the same Leishmania strain.

Table 1. Ratio of several pro-inflammatory cytokines/IL-10 quantified by ELISA on cell supernatants of 24 hours infected BMMø.

IL-12p40/IL-10 IL-6/IL-10 TNF-a/IL-10

Non - purified LPS 1,15 6 0,13 3,46 6 0,14 6,74 6 0,40

LPS P4 0,806 0,01a,b 0,40 6 0,003*** a,b
0,30 6 0,016*** a,b

LPS P21 2,41 6 0,04* 1,39 6 0,10* 1,09 6 0,06*

LPS P31 1,95 6 0,72 1,15 6 0,13* 0,84 60,07*
c,d c,d c,d
Ficoll - purified LPS P4 0,18 6 0,00*** 0,15 6 0,01*** 0,09 6 0,01***
LPS P21 0,45 6 0,02*** 0,39 6 0,01** 0,19 6 0,03**
LPS P31 0,40 6 0,01*** 0,40 6 0,01** 0,18 6 0,01**

The values are expressed in arbitrary units. One representative experiment out of two is shown. The mean and standard deviation are shown.
*P,0.05,
**P,0.01,
***P,0.001 statistical significant relatively to LPS and between passages of culture presented.
a
*P , 0.05 between LPS P4 and LPS P21,
b
*P , 0.05 between LPS P4 and LPS P31,
c
*P , 0.05 between LPS P4 and LPS P21 and,
d
*P , 0.05 between LPS P4 and LPS P31. All Ficoll-purified ratios are significantly different (at least *P , 0.05) from the related non-purified population.
doi:10.1371/journal.pntd.0001469.t001

www.plosntds.org 7 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

In the current work, we have started by investigating in our in

vitro model of L. infantum infection the relation between the parasite
development stage and its infectivity. The first step was to discard
logarithmic parasites because they are not ultimately responsible
for the infection [3], so we used a basic cell cycle analysis to
discard multiplying parasites (less than 10% of total population are
in S/G2 phase after the third day of culture). This data correlated
clearly with basic morphological visualization. Stationary L.
infantum cultures in day 5 and 9 induced higher BMMø infection
rates than day 3 parasites. This difference in infectivity might
translate the time frame required for becoming truly metacyclic
parasites [21], which was also corroborated by the less amount of
metacyclic recovered (data not shown). Since the presence of
apoptotic parasites is essential for a virulent inoculum of Leishmania
promastigotes [22], we decided to quantify the percentage of
apoptotic and dead parasites in each case to remove this possible
bias from our analysis. In fact, for all time frames tested, the
differences in infectivity were not related to apoptotic or dead
parasites in the non-purified or Ficoll-purified populations, with
culture viability always higher than 90%.
Some authors described that in vitro maintenance for long
periods constitute an important factor for the loss of virulence in L.
infantum [14] and L. major [6] promastigotes. Nonetheless, this loss
of virulence is a reversible phenomenon, since serial passages on
susceptible mice allow the parasite to recover a virulence
phenotype [23]. In the present study, we complemented the
previous observations by comparing the impact of continuous in
vitro culture on Leishmania promastigote virulence and also into the
capacity of host macrophage manipulation. Our data clearly
demonstrated a loss of L. infantum virulence related to the
Figure 3. Long-term in vitro maintenance of L. infantum augmentation of in vitro culture periods although no modification
promastigotes results in loss of virulence in vivo. Balb/c mice was observed in the axenic promastigote growth behavior. This
were infected with stationary phase promastigotes submitted to 4, 21 or
31 successive in vitro passages. After 6 weeks post-infection, the parasite
significant loss of infectivity was observed as soon as 105 days of
load was determined in liver (A) and spleen (B) by limiting dilution. The successive (21 passages) culture and worsened with parasite
mean and standard deviation are shown. *P,0,05; **P,0,01; ***P,0,001. maintenance in culture. In fact, 20 passages was the soonest time
doi:10.1371/journal.pntd.0001469.g003 point where we could have a significant reproducible loss of

Figure 4. Decreased capacity to infect is not correlated with a loss of metacyclic promastigotes. BMMø were infected at a 1:10 (cell/
parasite) ratio with non-purified promastigotes submitted to 4 (A), 21 (B) or 31 (C) successive in vitro passages with 5% or 10% of Ficoll-purified
parasites or without (Mock). As a control, BMMø were infected with a naturally attenuated strain in the same conditions (D). Data were acquired by
FACScalibur cytometer and analyzed by FlowJo software. Two independent experiments were performed; one representative experiment is shown.
The mean and standard deviation are shown. **P,0,01.
doi:10.1371/journal.pntd.0001469.g004

www.plosntds.org 8 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

www.plosntds.org 9 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

Figure 5. Diminished capacity of long-term cultured L. infantum promastigotes to differentiate and proliferate as axenic
amastigotes. CFSE-labeled non-purified promastigotes submitted to 4, 21 or 31 successive in vitro passages or from a field recovered naturally
attenuated strain (Aten) were cultured in MAA20 for 3 days to induce differentiation into the amastigote form. (A) Parasite multiplication was
followed by FACScalibur quantification of CFSE fluorescence. (B) For each time, the mean fluorescence intensity (MFI) was calculated. The mean and
standard deviation are shown. **P,0,01; ***P,0,001 (C) At the same time, parasite viability was followed by propidium iodide (PI) incorporation.
doi:10.1371/journal.pntd.0001469.g005

infection. There is a grey area between passage 9 and passage 20 at 48 hours post-infection when facing an inflammatory stimulus,
where we can have variation of infection in a manner that showing that the phenomenon of loss of virulence, although less
probably reflects the initial parasite inoculum recovered from the prominent in the metacyclic-enriched parasites, was not restricted
mammal. This was also observed after in vivo infection where we to the unpurified culture. Since the differences at the infection
had a significant decrease in parasite burden after 21 and 31 level were significant, we examined if there was a potential effect
passages, confirming the in vitro observations. on the macrophage activation status. In the presence of a strong
The percentage of metacyclic in a heterogeneous stationary- inflammatory stimulus, L. infantum is able to suppress certain LPS-
phase is an important factor in the parasite infectivity since they derived pro-inflammatory cytokine responses in an active parasite-
are significantly more infective than the non-purified population. specific process while it augments the production of some anti-
We used a Ficoll density gradient methodology to enrich the inflammatory cytokines (Silvestre et al, unpublished data). Indeed,
percentage of metacyclic promastigotes [24]. Beyond the mor- the addition of LPS to Leishmania spp. infected cells was
phological changes, during the Leishmania spp. differentiation demonstrated to synergistically induce the secretion of the anti-
process, modifications also occur in gene expression and in the inflammatory IL-10 cytokine in monocytes [25] and in macro-
composition of parasite surface that help to characterize phages [26]. The functional polarization of macrophages into IL-
metacyclic promastigotes. Thus, we have evaluated the enrich- 10 producers characterized as M2 cells [27] has been long
ment of metacyclic promastigotes in the Ficoll-purified fraction by understood to play a crucial role in the success of parasite infection
microscopy (data not shown) and through qPCR analysis of the process [28]. Our results demonstrated a growing defect of high
SHERP and histone H4 gene expression. The augmentation of passage parasites to modulate the LPS stimulatory effect.
SHERP gene expression in Ficoll-purification supported an Furthermore, it is clear from the inflammatory profile depicted
enriched metacyclic population. The use of Ficoll-purified in Table 1 that metacyclic enriched fractions are always
promastigotes abolished the differences found among the different significantly more effective in abrogating a macrophage response
passages. Yet, significant differences were found for P21 and P31, to the inflammatory stimuli than their non-purified counterparts

Figure 6. Virulence is recovered after in vitro or in vivo promastigote-amastigote differentiation process. BMMø were infected at a 1:10
(cell/parasite) ratio with non-purified promastigotes differentiated from in vitro axenic amastigotes (A) or ex-vivo intracellular amastigotes (B). As a
control, a naturally attenuated strain (Aten) was submitted to the same experimental conditions. Data were acquired by FACScalibur cytometer and
analyzed by FlowJo software. Three independent experiments were performed; one representative experiment is shown. The mean and standard
deviation are shown. *P,0,05; **P,0,01.
doi:10.1371/journal.pntd.0001469.g006

www.plosntds.org 10 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

revealing the metacyclic parasites as a highly immunomodulatory which may probably be originated from the absence of a complete
population with a distinct profile from the non-purified popula- life cycle. Therefore, special care must be taken when performing
tion. There is a distinct and significant loss of immunomodulatory experiments with axenic Leishmania promastigotes. The systematic
properties from P4 to P21 that becomes stable after P21. This loss and rigorous control of Leishmania culture conditions should be
of immunomodulatory properties seems reminiscent of phenom- considered as a keystone for each experimental protocol. The
enon of transient gene expression similar to what happens under differences found in infectivity accompanied by disparate effects at
drug pressure [29], being lost upon the terminus of immunological the macrophage activation levels point to significant differences at
pressure. Indeed, this might be happening in just a few passages of biochemical and structural level, enlarging the effects of careless
axenic culture. In an attempt to explain the loss of virulence parasite maintenance to other experimental fields. This informa-
mechanism, some authors referred to a reduction of metallo and tion is extremely relevant especially for those developing new drug
cysteine peptidases activity, important for virulence, in L. and vaccine approaches. In such cases the immune response to the
braziliensis [30,31] and in L. amazonensis [32] or mitochondrial parasite is the essence of the experimental procedure.
defects [33] during long-term culture. However, others have been
unable to detect any differences in the parasite enzymatic profile Supporting Information
with long in vitro periods of cultivation [34,35].
One can speculate that the overall loss of immunomodulatory Figure S1 Cell cycle analysis and in vitro virulence of L.
properties over time, and in consequence loss of virulence might infantum recovered in distinct culture days. L. infantum
reflect a diminution of the number of metacyclic parasites in the promastigotes were cultured at a 106/ml. Each day, 26106
population. Although the percentage of Ficoll-recovered promas- promastigotes were recovered and the cell cycle analyzed by PI
tigotes was quite similar among the three tested passages, these staining (A). BMMø were incubated with non-purified CFSE
fractions do not constitute a pure metacyclic population, so we labeled L. infantum promastigotes at a ratio of 1:10 (cell/parasite).
decided to complement older passage parasites with Ficoll The percentage of infected cells was obtained by quantifying the
recovered metacyclic promastigotes to access if the loss of number of CFSE-positive cells (B). Data were acquired at 4 and
virulence could be reverted by exogenous addition of metacyclic 24 hours post-infection in a FACScalibur cytometer and analysed
parasites from an early passage. No improvement in the overall by FlowJo software. Three independent experiments were
infection was observed, although, when these same metacyclic performed; one representative experiment is shown. The mean
parasites were added to an avirulent field strain, from which we and standard deviation are shown. *P,0,05, **P,0,01 statistical
were unable to recover metacyclics, there was an improvement on significance relatively to 3rd day of parasite growth.
the infection. (TIF)
Another possibility to explain the virulence loss was the
Figure S2 Indirect quantification of metacyclic promas-
possibility of a defective promastigote to amastigote differentiation.
tigotes in heterogenous and Ficoll-purified cultures by
Our data clearly demonstrated that high passage number
gene transcription analysis. Transcription profile of SHERP
promastigotes displayed decrease capacity in differentiating, which
and Histone H4 genes obtained by qPCR, for non-purified and
was not correlated with decreased cellular viability. This
Ficoll-purified promastigotes. Normalizations were made against
incapacity translates into fewer parasites able to differentiate
leading to a less capable population to face host cell response. the reference gene rRNA45. Three independent experiments were
Ultimately, this results in lower parasite burdens in vitro and in vivo. performed, each performed in duplicate; one representative
Moreover, these axenic amastigotes recovered after differentiation experiment is shown. The mean and standard deviation are
were morphologically indistinguishable and retained similar shown. *P,0,05.
growth capacity (data not shown). To ultimately state and confirm (TIF)
the importance of promastigote to amastigote differentiation as a Figure S3 Long-term cultured L. infantum promasti-
driving selective force for virulence, we showed that parasites gotes show decrease capacity to modulate an inflamma-
passed through the amastigote stage, either in vitro or in vivo, revert tory stimulus in vitro. BMMø were submitted to LPS
the loss of virulence. This fact in conjunction with the remarkable stimulation 4 hours after infection with non-purified promasti-
loss of immunomodulatory properties leads to the early loss of gotes. The levels of IL-6 (A), TNF-a (B), IL-12p40 (C) and IL-10
virulence detected in our model. (D) were quantified 24 hours post-infection on BMMø superna-
We do not rule out metacyclogenesis related defects as a driving tants by ELISA. Three independent experiments were performed;
force for a virulence. In relation to metacyclogenesis it has been one representative experiment is shown. The mean and standard
argued that a successful and complete differentiation is dependent deviation are shown. *P,0,05; **P,0,01; ***P,0,001 statistical
on the presence of large amount of metacyclic promastigotes [36]. differences relative to LPS unless depicted by a bar.
However, it is still not clear whether this process is an essential step (TIF)
in the differentiation in vitro, since procyclic promastigotes appear
to differentiate with equal efficiency as metacyclics [37–40]. Figure S4 Naturally attenuated L. infantum strain has a
Indeed, our results with the naturally attenuated strain support this similar axenic growth in comparison to WT strain. L.
notion. Our data do not rule out that P21 or P31 metacyclic infantum growth curves were performed by neubauer chamber
promastigotes could not display any sort of biochemical or protein counting. The mean and standard deviation are shown.
expression defect that may impact the differentiation process. (TIF)
Similarly, we cannot reject the idea of longer periods of sustained
cultured originating defective metacyclic cultures. However, Author Contributions
during our study, the loss of virulence was related to a specific Conceived and designed the experiments: NS RS. Performed the
defect on promastigote to amastigote differentiation. experiments: DM NS AMA RS. Analyzed the data: DM NS AO RS
Overall, our data demonstrated that the loss of virulence is ACdS. Contributed reagents/materials/analysis tools: IL JT AMS. Wrote
linked with decreased capacity to differentiate in amastigote forms, the paper: DM NS RS.

www.plosntds.org 11 January 2012 | Volume 6 | Issue 1 | e1469

 Virulence Loss in Long-Term Axenic Promastigotes

References
1. Bates PA (2007) Transmission of Leishmania metacyclic promastigotes by protozoan parasite Leishmania major in its sand fly vector, Phlebotomus papatasi.
phlebotomine sand flies. Int J Parasitol 37: 1097–1106. Cell Microbiol 12: 1765–1779.
2. Sacks DL, Perkins PV (1984) Identification of an infective stage of Leishmania 22. van Zandbergen G, Bollinger A, Wenzel A, Kamhawi S, Voll R, et al. (2006)
promastigotes. Science 223: 1417–1419. Leishmania disease development depends on the presence of apoptotic
3. Sacks DL, Perkins PV (1985) Development of infective stage Leishmania promastigotes in the virulent inoculum. Proc Natl Acad Sci U S A 103:
promastigotes within phlebotomine sand flies. Am J Trop Med Hyg 34: 13837–13842.
456–459. 23. Katakura K, Kobayashi A (1985) Enhancement of infectivity of Leishmania
4. Mitchell GF, Handman E, Spithill TW (1984) Vaccination against cutaneous donovani promastigotes by serial mouse passages. J Parasitol 71: 393–394.
leishmaniasis in mice using nonpathogenic cloned promastigotes of Leishmania 24. Saraiva EM, Pinto-da-Silva LH, Wanderley JL, Bonomo AC, Barcinski MA,
major and importance of route of injection. Aust J Exp Biol Med Sci 62(Pt 2): et al. (2005) Flow cytometric assessment of Leishmania spp metacyclic
145–153. differentiation: validation by morphological features and specific markers. Exp
5. Hadighi R, Mohebali M, Boucher P, Hajjaran H, Khamesipour A, et al. (2006) Parasitol 110: 39–47.
Unresponsiveness to Glucantime treatment in Iranian cutaneous leishmaniasis 25. Meddeb-Garnaoui A, Zrelli H, Dellagi K (2009) Effects of tropism and virulence
due to drug-resistant Leishmania tropica parasites. PLoS Med 3: e162. of Leishmania parasites on cytokine production by infected human monocytes.
6. Segovia M, Artero JM, Mellado E, Chance ML (1992) Effects of long-term in Clin Exp Immunol 155: 199–206.
vitro cultivation on the virulence of cloned lines of Leishmania major 26. Lapara NJ, 3rd, Kelly BL (2010) Suppression of LPS-induced inflammatory
promastigotes. Ann Trop Med Parasitol 86: 347–354. responses in macrophages infected with Leishmania. J Inflamm (Lond) 7: 8.
7. Sereno D, Lemesre JL (1997) Axenically cultured amastigote forms as an in vitro 27. Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, et al. (2004) The
model for investigation of antileishmanial agents. Antimicrob Agents Chemother chemokine system in diverse forms of macrophage activation and polarization.
41: 972–976. Trends Immunol 25: 677–686.
8. Yao C, Chen Y, Sudan B, Donelson JE, Wilson ME (2008) Leishmania chagasi: 28. Cunningham AC (2002) Parasitic adaptive mechanisms in infection by leishmania.
homogenous metacyclic promastigotes isolated by buoyant density are highly Exp Mol Pathol 72: 132–141.
virulent in a mouse model. Exp Parasitol 118: 129–133. 29. Ubeda JM, Legare D, Raymond F, Ouameur AA, Boisvert S, et al. (2008)
9. Ouakad M, Bahi-Jaber N, Chenik M, Dellagi K, Louzir H (2007) Selection of Modulation of gene expression in drug resistant Leishmania is associated with
endogenous reference genes for gene expression analysis in Leishmania major gene amplification, gene deletion and chromosome aneuploidy. Genome Biol 9:
developmental stages. Parasitol Res 101: 473–477. R115.
30. Bates PA, Robertson CD, Coombs GH (1994) Expression of cysteine proteinases
10. Silvestre R, Cordeiro-Da-Silva A, Santarem N, Vergnes B, Sereno D, et al.
by metacyclic promastigotes of Leishmania mexicana. J Eukaryot Microbiol 41:
(2007) SIR2-deficient Leishmania infantum induces a defined IFN-gamma/IL-10
199–203.
pattern that correlates with protection. J Immunol 179: 3161–3170.
31. Lima AK, Elias CG, Souza JE, Santos AL, Dutra PM (2009) Dissimilar
11. Wanderley JL, Moreira ME, Benjamin A, Bonomo AC, Barcinski MA (2006)
peptidase production by avirulent and virulent promastigotes of Leishmania
Mimicry of apoptotic cells by exposing phosphatidylserine participates in the
braziliensis: inference on the parasite proliferation and interaction with
establishment of amastigotes of Leishmania (L) amazonensis in mammalian hosts.
macrophages. Parasitology 136: 1179–1191.
J Immunol 176: 1834–1839.
32. Chaudhuri G, Chang KP (1988) Acid protease activity of a major surface
12. Wanderley JL, Pinto da Silva LH, Deolindo P, Soong L, Borges VM, et al. membrane glycoprotein (gp63) from Leishmania mexicana promastigotes. Mol
(2009) Cooperation between apoptotic and viable metacyclics enhances the
Biochem Parasitol 27: 43–52.
pathogenesis of Leishmaniasis. PLoS One 4: e5733.
33. Nasyrova RM, Kallinikova VD, Vafakulov S, Nasyrov F (1993) [The virulence
13. El-Fadili AK, Zangger H, Desponds C, Gonzalez IJ, Zalila H, et al. (2010) and cytochemical properties of Leishmania major during long-term cultivation].
Cathepsin B-like and cell death in the unicellular human pathogen Leishmania. Parazitologiia 27: 233–241.
Cell Death Dis 1: e71. 34. Cuervo P, Santos AL, Alves CR, Menezes GC, Silva BA, et al. (2008) Cellular
14. Grimm F, Brun R, Jenni L (1991) Promastigote infectivity in Leishmania infantum. localization and expression of gp63 homologous metalloproteases in Leishmania
Parasitol Res 77: 185–191. (Viannia) braziliensis strains. Acta Trop 106: 143–148.
15. Louassini M, Adroher FJ, Foulquie MR, Benitez R (1998) Investigations on the 35. Soares RP, Macedo ME, Ropert C, Gontijo NF, Almeida IC, et al. (2002)
in vitro metacyclogenesis of a visceral and a cutaneous human strain of Leishmania chagasi: lipophosphoglycan characterization and binding to the midgut
Leishmania infantum. Acta Trop 70: 355–368. of the sand fly vector Lutzomyia longipalpis. Mol Biochem Parasitol 121:
16. Spath GF, Beverley SM (2001) A lipophosphoglycan-independent method for 213–224.
isolation of infective Leishmania metacyclic promastigotes by density gradient 36. Cysne-Finkelstein L, Temporal RM, Alves FA, Leon LL (1998) Leishmania
centrifugation. Exp Parasitol 99: 97–103. amazonensis: long-term cultivation of axenic amastigotes is associated to
17. Knuepfer E, Stierhof YD, McKean PG, Smith DF (2001) Characterization of a metacyclogenesis of promastigotes. Exp Parasitol 89: 58–62.
differentially expressed protein that shows an unusual localization to intracellular 37. Barak E, Amin-Spector S, Gerliak E, Goyard S, Holland N, et al. (2005)
membranes in Leishmania major. Biochem J 356: 335–344. Differentiation of Leishmania donovani in host-free system: analysis of signal
18. Soto M, Quijada L, Alonso C, Requena JM (1997) Molecular cloning and perception and response. Mol Biochem Parasitol 141: 99–108.
analysis of expression of the Leishmania infantum histone H4 genes. Mol Biochem 38. Goyard S, Segawa H, Gordon J, Showalter M, Duncan R, et al. (2003) An in
Parasitol 90: 439–447. vitro system for developmental and genetic studies of Leishmania donovani
19. Silvestre R, Santarem N, Tavares J, Silva AM, Cordeiro-Da-Silva A (2009) phosphoglycans. Mol Biochem Parasitol 130: 31–42.
Recognition of Leishmania parasites by innate immunity. Immun, Endoc & 39. Debrabant A, Joshi MB, Pimenta PF, Dwyer DM (2004) Generation of
Metab Agents in Med Chem 9: 106–127. Leishmania donovani axenic amastigotes: their growth and biological characteris-
20. Bastien P, Blaineau C, Pages M (1992) Leishmania: sex, lies and karyotype. tics. Int J Parasitol 34: 205–217.
Parasitol Today 8: 174–177. 40. Saar Y, Ransford A, Waldman E, Mazareb S, Amin-Spector S, et al. (1998)
21. Sadlova J, Price HP, Smith BA, Votypka J, Volf P, et al. (2010) The stage- Characterization of developmentally-regulated activities in axenic amastigotes of
regulated HASPB and SHERP proteins are essential for differentiation of the Leishmania donovani. Mol Biochem Parasitol 95: 9–20.

www.plosntds.org 12 January 2012 | Volume 6 | Issue 1 | e1469