Apoptosis stalks Plasmodium falciparum maintained in continuous culture condition

The development and propagation of malaria parasites in their vertebrate host is complex and involves various host and parasite factors. In absence of specific anti- parasitic immune responses such as in vitro culture conditions, it is reasonable to assume that the parasite growth rate reflects a balance between cell proliferation and cell loss. This assumption is strengthened by the following observations: 1) in vitro, the parasitaemia of Plasmodium falciparum cultures increases by a factor of 3-8 within 48 hours whereas theoretically it should increase 16 fold [1], and 2) in vivo, the evolution of parasitaemia in fast growing strains seems to be quelled by parasite density even when the RBC supply is not limiting [2]. It is tempting to speculate that the disparity between expected and observed parasite density is a function of self-regulation.

One of the physiological mechanisms of reducing cell number is by apoptosis [3], a process by which self inflicted morphological and biochemical changes result in cell death [4, 5]. At the cellular level, apoptosis is characterized by controlled auto-digestion through the activation of endogenous proteases that result in cytoskeletal disruption, cell shrinkage, and membrane blebbing [6]Due to activation of endonucleases, the nucleus undergoes condensation as DNA is fragmented into oligonucleosomes. Formation of oligonucleosomes can be detected using terminal deoxynucleotidyl transferase-mediated desoxyuridine triphosphate (dUTP) nick-end labeling (TUNEL). [7] Even before DNA damage is evident, mitochondria show collapse in the electrochemical gradient as measured by change in the mitochondrial membrane potential (ΔΨ_m). Changes in ΔΨ_m lead to the insertion of proapoptotic proteins into the membrane resulting in pore formation and subsequent release of cytochrome c into the cytoplasm. [6] Loss of ΔΨ_m can be measured by a cell-permeant, cationic, red-orange fluorescent dye Tetra methyl Rhodamine ethyl ester (TMRE) that is readily sequestered by active mitochondria [8] or cationic dye 5,5,6,6’-tetrachloro-1, 1,3,3’-tetraethylbenzainidazolocarbocyanin iodide, commonly known as JC-1 [9].

For a long time, apoptosis was assumed to be confined to metazoan organism but recent findings indicate that unicellular organisms also undergo apoptosis and features typical of mammalian programmed cell death have been described in several species including the kinetoplastids: Trypanosoma brucei rhodesiense, Trypanosoma cruzi and Leishmania major[7, 10‐15]. Features typical of apoptosis including condensation of chromatin, DNA fragmentation, the appearance of phosphatidyl serine (PS) on the outer leaflet of the plasma membrane and eventual formation of apoptotic bodies have been demonstrated in Plasmodium berghei[16]. Picot et al[17] previously reported oligonucleosomal DNA fragmentation in the erythrocytic stages of chloroquine treated Plasmodium. Recently, DNA fragmentation and mitochondrial membrane potential disruption have been shown in erythrocytic stages of 3D7 and 7G8 isolates of P. falciparum treated with apoptosis- inducer etoposide or anti-malarial chloroquine [17, 18]. Although clear homologues of metazoan caspases have not been identified in Plasmodium, two cysteine proteases have been annotated (PF13-0289 and PF- 0363) in the P. falciparum genome database [19]. The two proteases have 24% sequence identity and have the conserved catalytic dyad histidine and cysteine required for catalysis activity and contains a putative caspase recruitment domain in the N-terminal amino acid sequence [18]. Presence of caspase-like proteins in Plasmodium was also suggested by use of caspase inhibitors (Z-VAD.fmk (Benzyloxycarbonyl-Val-Ala-Asp flouromethylketone) and Z.DEVD.FMK (Benzyloxycarbonyl-Asp-Glu-Val-Asp- flouromethylketone)) that inhibited Plasmodium apoptosis leading to suggestion that caspase-like activity and aspartate specificity may be crucial for apoptosis in Plasmodium[16]. In addition to metacaspase gene, an apoptosis related gene (PfARP) has been purified and characterized in P. falciparum although the role of this protein in growth, multiplication and stage progression has not been identified [20].

Existence of apoptosis in protozoa suggests that even these unicellular organisms live as communities where communication occurs between individual cells. This interaction between protozoa is similar to the quorum sensing (QS) and biofilm formation in bacteria [21, 22]. QS as is known in bacteria, refers to regulation of gene expression in response to fluctuations in cell population density and is mediated by signaling molecules released into the environment. In vitro studies and epidemiological data provide evidence that malaria parasites exhibit QS [23]. The process relies on production of low molecular mass-signaling molecules called auto-inducers, the extra-cellullar concentration of which is related to the population density [23].

Although P. falciparum could regulate parasite density by varying the rate of cell division or varying the rate of invasion of RBC, its been suggested that varying the rate of cell death is the most effective of these mechanisms because it slows down the increase of the parasitaemia while at the same time decreasing the percentage of infected RBCs independent of the host’s immune system [1].

In this study, density dependent processes that limit expansion of the malaria parasites are described. It is possible that these processes may one day be exploitable to provide new strategies of curtailing disease severity associated with parasite density.

Like in multicellular organisms where apoptosis is essential for maintaining order and harmony among cells that must live together, protozoan parasites growing and differentiating in culture, in insect vector or in mammalian host display a structured organization that suggest that they ‘talk’ to each other and influence how they interact with elements of their environment [27]. Some of the factors used for “talking” have been identified as derivatives of homoserine lactone (HSL), which on secretion and at a threshold concentration induces a common genetic response in the population. [28] Such factors have been identified in protozoan organisms such as Dictyostelium[29], Plasmodium[30] and Trypanosoma brucei[12].

Studies of free-living bacteria and single-celled organisms have shown they are capable of a form of quorum sensing whereby all individual cells secrete a signal factor. Such factors are derivatives of homoserine lactone which have been shown to induce a common genetic response in the population once a threshold concentration is reached [28].

In this study, in order to determine whether apoptosis triggered the events that were observed in the growth curve of the Dd2 parasite populations used in this study, a set of experiments were conducted to examine the use of apoptotic machinery at various stages of the parasite growth. In this experimental set up, parasitaemia increased by a factor of 1.5 to 2 every 48 hours, while the theoretical growth has been estimated to be by 16-32 folds [26]. Thereafter, the culture crashed when the nutrients or red blood cells were not limiting (Figure 1). Early in the growth curve, the morphological features reviewed by microscopy were those of healthy malaria parasites [32, 33]. Parasites with compromised morphological features were seen later as parasitaemia increased (Figure 1).

An important molecular indicator of apoptosis is DNA fragmentation whereby the DNA is degraded at the internucleosomal regions by endogenous DNAses into fragments of 180-200 base pairs [34]. The degraded DNA can be revealed by electrophoretic separation of these fragments in agarose gel or by incorporating labeled nucleotides to the exposed 3’-OH terminal of nicked DNA ends. Using the latter method, evidence of DNA fragmentation was discernible albeit at very low level by flow cytometry and by fluorescent microscopy at the seeding parasitaemia and increased dramatically by the time of parasite culture crash (Figure 2, panel A and B). Most of the DNA degradation was evident at the trophozoites and schizont stages, a finding that probably reflects increased cellular DNA content in these two stages [35]. These findings represent the first report of density dependent DNA fragmentation in asexual stages of P. falciparum. Previously reports showed DNA fragmentation following chloroquine treatement [17] and in the ookinates of P. berghei.[16].

During apoptosis, death stimuli induce mitochondrial membrane permeabilization that results in loss of mitochondria membrane potential. This loss can be revealed as decline of TMRE staining of the mitochodria. As shown in Figure 2, panel C, the proportion of cells with intact mitochondria declined as parasite density increased and was maximal at a day before the crash. The uptake of TMRE in the mitochondria varied among the stages, with the ring showing the greatest mitochondrial response (Figure 2, panel C). This difference in uptake of TMRE is probably as a result of differing level of metabolic activities in the three asexual blood stages [36], although further biochemical evidence will be required to validate this observation. From MedLine literature search, this is the first report in on the use of TMRE to study depolarization of mitochondria in P. falciparum.

The next evaluation examined the expression level of P. falciparum metacaspase gene and/or protein in relation to changes in parasite density. The metacaspase gene was expressed in all the asexual stages (Figure 3, panel A). Large variations in levels of gene expression in the different parasite stages were observed at different parasite densities (Fig 3, Panel A). It was confirmed that the seeming differential expression of this gene was not due to variation in quantity or quality of RNA by parallel analyses of expressing level of a house keeping gene, the seryl tRNA gene. A reverse transcription control was also used to confirm that amplification was not from genomic DNA. For ring stages, metacaspase was not expressed until the parasite density of 4% was attained (Figure 3, panel A, lane 8). While the expression of metacaspase gene was only evident in the trophozoites at 0.5% and 4% parasitaemia, in the schizonts, metacaspase was present at all parasitaemia levels (Figure 3, panel A). We are not sure what the cause of this large variation in the transcripts is, but could reflect changes in mRNA stability as the parasites proceed through different development stages.

Since gene transcriptional activity does not necessarily translate into protein expression, [37]changes in metacaspase protein expression as parasite density changed were also determined. The proteolytic activation of caspases is highly conserved in higher eukaryotes and depends on prodomain removal and cleavage of protein fragments [38]. Using anti- caspase 3 and 7 polyclonal antibodies to probe for P. falciparum metacaspase, two protein fragments of 45 kDa and 28 kDa were detected in all the asexual stages (Figure 3, B) as has been described before [16]. The expression level of metacaspase protein increased with rise in parasite density (Figure 3, B, lanes 8-12). A low level of protein of approximately 28 kDa was present in uninfected erythrocytes (lane 13) and has been reported before in other negative control samples [16]. Therefore this protein is from un-infected RBCs.

The four key indicators of apoptosis, namely DNA fragmentation, collapse of mitochondrial potential, up-regulation of metacaspase gene and protein transcripts accompanied collapse of malaria in vitro culture that was maintained under conditions that were not limiting for nutrients or red blood cells. These findings and others fro earlier studies are indicative of presence of a density dependent process for limiting uncontrolled growth of malaria parasites. These findings offer important insights that may in future be exploitable as chemotherapeutic targets.