Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Most eukaryotic cells have stringent regulatory mechanisms to coordinate DNA duplication with chromosome segregation followed by cytokinesis so that daughter cells are formed with exact copies of their parental genetic material. In mitotic cells, the genome is copied just once during the S phase [1]. DNA-repair mechanisms ensure that errors generated during synthesis are corrected. The cell then prepares to segregate the two copies of the genome (M phase). Duplicate copies of the genome are aligned on a bipolar microtubular spindle and separated into two daughter cells [2].

Our studies during the past two decades showed that this paradigm was not followed in trophozoites of Entamoeba spp. We focused most of our studies on Entamoeba histolytica HM1:IMSS strain and also compared the data with other strains such as 200:NIH and more recent isolates that were maintained in axenic culture [3]. Analyzing the progression of the cell cycle in E. histolytica posed several challenges because routine microbiological methods and genetic manipulations could not be carried out in this organism. Thus, our understanding of the cell-cycle processes has largely relied on the use of multiparameter flow cytometry, immunofluorescence, scanning cytometry, and other molecular biological techniques.

1.1 The Axenic Cell Cycle of E. histolytica

In contrast to other eukaryotic cells, the different phases of the cell division cycle (G1, S, G2/M) were not clearly demarcated in flow cytometric analysis of the axenic cultures of E. histolytica [4, 5]. Mathematical modeling analyzed the G1, S, and G2/M phases of the E. histolytica cell cycle as a series of overlapping Gaussian curves, differing from the discrete peaks of these phases in other organisms (Fig. 16.1a, b) [4]. These results suggested that E. histolytica cultures were made up of cells containing heterogeneous amounts of DNA. Cells with polyploid nuclei were found in axenically growing populations of HMI:IMSS cells (Fig. 16.1c). Thus, heterogeneity of DNA content could be caused by differences in nuclear DNA content or the number of nuclei or both. The average DNA content of cells and the number of cells with DNA content >2n was found to increase with time in axenic culture [6], which indicates that E. histolytica trophozoites possibly reduplicated their genome several times without nuclear division or cytokinesis. We confirmed that this heterogeneity is not specific to the laboratory strain E. histolytica HMI:IMSS and that other isolates of E. histolytica showed a similar phenotype [3]. Thus, typical checkpoint mechanisms that ensure the progression of chromosome segregation and cytokinesis (immediately after the completion of genome duplication) were either absent or altered in this parasite.

Fig. 16.1
figure 1

Comparison of the cell-cycle phases as analyzed in a flow cytometer. a Flow cytometric analysis of a typical eukaryotic cell cycle (adapted from Louise Russel, Cutaneous Research, Institute of Cell and Molecular Science, Queens Mary College, London) showing distinct G1, S, and G2/M phases. b Flow cytometric analysis of the Entamoeba histolytica cell cycle (adapted from Gangopadhyay et al. [5]) shows the absence of discrete cell-cycle phases. c A multinucleated E. histolytica cell in an axenically growing culture. DNA has been stained with DAPI.Bar 20 μM. d The DNA content of Entamoeba invadens cyst and trophozoites (~40 Mb) shown in the logarithmic scale. 4× and 40× indicate the average genome content corresponding to the major peaks in each cell type [3]

Full size image

It has been difficult to synchronize E. histolytica cells in any one phase of the cell cycle. Mitotic blockers such as colchicine, nocodazole, and thiabenzadole were not active against E. histolytica cells. Our best results were obtained by using serum starvation for 12 h followed by addition of serum [5, 6]. During serum starvation, cells were arrested at different phases. Using this method in our subsequent studies we could obtain synchronization for one mitotic cycle [7, 8]. Using BrdU (5′-bromo-2′-deoxyuridine) incorporation, we showed that after addition of serum, DNA synthesis was initiated after a lag phase of 2 h. More than 80 % of the cells showed a uniform DNA content at this time [6]. Cells continued to accumulate multiple genome contents subsequently in the next several hours [3, 6]. These data were confirmed both by BrdU incorporation and by scanning cytometry, clearly showing that nuclear and cell division are not temporally linked to genome duplication and segregation. In addition, the number of duplication cycles was not fixed as the nuclei accumulated heterogeneous amounts of DNA [3, 6]. Thus, genome segregation did not necessarily occur immediately after genome duplication. In serum synchronized cells, we observed that the number of microtubular assemblies increased 4 h after addition of serum and were highest after 8 h of serum addition [8]. The number of binucleated cells was highest at 10 h after addition of serum. These observations suggested that the majority of cells in this synchronized population underwent several rounds of genome multiplication followed by chromosome segregation, possibly between 4 and 8 h after initiation of DNA synthesis, followed by nuclear division between 8 and 10 h [8].

2 The E. histolytica Genome Segregates on Atypical Microtubular Assemblies

Chromosome segregation is carried out on the mitotic spindle in eukaryotic cells. The sister chromatids are pulled apart on these spindle fibers: complete sets of chromosomes are moved to each pole of the cell where they are packaged into daughter nuclei. The mitotic spindle is composed of dynamic microtubular fibers that are polymers of α- and β-tubulin subunits. These subunits are nucleated at the microtubule organizing center (MTOC) [9]. One of the key proteins of the MTOC is the γ-tubulin that regulates the nucleation of microtubules in higher eukaryotes [10]. The homologues of α-, β-, and γ-proteins have been identified in E. histolytica [11–13]. The amino acid sequences of Ehαβγ-tubulins are significantly divergent from their eukaryotic homologues, and this difference has been attributed to their resistance to antimitotic drugs such as colchicines and benomyl [11, 13], although sensitivity to the microtubule stabilizing drug, taxol was predicted from the conserved taxol-binding site in the modeled tertiary structure of Eh γ tubulin [14].

Despite the presence of essential spindle and MTOC forming proteins i.e alpha, beta and gamma subunits [12, 15], metaphase-like equitorial alignment of condensed chromosomes could not be identified in E.histolytica cells [16, 17]. Anaphase and telophase were identified on the basis of nuclear shape [18]. Moreover, nuclear microtubular assemblies with fibers radiating from a central region in most E. histolytica cells were shown by indirect immunofluorescence [15].

In one of our studies, serum synchronized cells were fixed and stained with anti-β tubulin antibody, and a time-course study was done to observe the microtubular structures in E. histolytica cells. Several novel microtubular (MT) assemblies including monopolar, bipolar, and multipolar spindles for the segregation of chromosomal DNA were identified [8]. These unusual MT structures suggested that genome segregation occurred on different kinds of MT structures in E. histolytica cells, likely required for the heterogeneous number of genome copies in a single nucleus. A model was based on these observations(Fig. 16.2) from Mukherjee et al. [8]. This model highlights the possible modes of genome segregation on MT assemblies ranging from radial, bipolar, or fan-shaped structures to bundles of multiple MTs in assorted shapes. Real-time data with fluorescence-tagged MTs to validate this exciting hypothesis remains to be validated with live cell imaging using fluorescently labelled microtubules.

Fig. 16.2
figure 2

Model showing chromosome segregation on different types of microtubular assemblies seen in Entamoeba histolytica. The schematic diagrams are adapted from the microscopic images shown alongside (on right) [8]. The cells were stained with anti-beta tubulin antibody and DAPI for staining the microtubules and DNA, respectively. Images were taken in a Zeiss LSM 510 META confocal microscope under 63× oil objective. The different MT assemblies suggest multiple modes of genome segregation

Full size image

3 Cytokinesis in Entamoeba histolytica

Cytokinesis is the process of physical separation of a mother cell that gives rise to two daughter cells. One of the earliest modes of cell division used might have been the motility and consequent mechanical force driven by actin polymerization in a polymorphic cell [19]. Cytokinesis in E. histolytica can occur in three different ways. In the first mode, the intercellular bridge between dividing E. histolytica trophozoites is formed at random sites. The extension and rupture of this cytoplasmic bridge leads to formation of two daughter cells. In the second mode of cytokinesis the severing of the intercellular bridge is assisted by helper cells. In such cases the helper cells migrate to the intercellular bridge and rupture the bridge either mechanically or by unidentified mechanisms. Helper cell-assisted cytokinesis was estimated to occur in 45 % of the cases (Fig. 16.3a). Interestingly, the microtubular structure was found to extend through the entire length of the intercellular bridge (Fig. 16.3b). Finally, failure of cytokinesis is quite common where (Fig. 16.3c) the two halves of a cell are separated by an extremely long bridge that ultimately failed to separate and cytokinesis was aborted in approximately 20 % of the cells [8]. This event may well lead to cells with multiple nuclei in which the mitotic cycle continues for many rounds without successful cell division. Unequal or aberrant division was frequent and gave rise to “anucleate” cells [8]. Altogether, the data suggested that the heterogeneous modes of cytokinesis contributed to the genetic heterogeneity in the population of E. histolytica cells.

Fig. 16.3
figure 3

Various events leading to cytokinesis in Entamoeba histolytica. a Helper cell (red arrow) approaches the intercellular bridge (black arrow) between two daughter E. histolytica cells. b The microtubular assembly stretches across the intercellular bridge, a part of which was seen on top of a helper cell as the latter moves beneath the bridge. The cells were stained with anti-β-tubulin antibody and DAPI to visualize microtubules and DNA, respectively. c A still from live-cell imaging of cytokinesis in E. histolytica cells shows a long and thin intercellular bridge (black arrow) connecting the two daughter cells

Full size image

4 Checkpoint Genes Are Absent in Entamoeba histolytica

The regulation of a typical eukaryotic cell division cycle depends upon a set of proteins known as the “checkpoint” proteins [20–22]. Cell-cycle progression is closely monitored by these checkpoint proteins and occurs only if the preceding phase has been completed correctly. Analysis of the E. histolytica genome shows that most of the conserved checkpoint proteins are absent. A large amount of information regarding the cell-cycle regulation in eukaryotes has been obtained from genetic and biochemical studies on the yeasts Saccharomyces cerevisiae and Schizosaccharomyces pombe [23]. Additionally, it was seen that proteins controlling the cell cycle in yeasts were well conserved in other eukaryotes [24]. The complete genome sequence of E. histolytica, published in 2005, consists of about 8,201 genes with an average size of 1.17 kb. No homologues could be identified for one-third of the predicted proteins (32 %) from public databases. Even before the completion of the genome sequence few cell-cycle-related genes had been identified, such as p34Cdc2 [25], Mcm2-3-5 [26], α-, β-, and γ-tubulin [11–13], and Diaphanous1 [27] in E. histolytica. An in silico analysis of the E. histolytica genome for homologues of cell-cycle genes in S. cerevisiae was conducted for a better understanding of the amoeba cell cycle.

  1. 1.

    Genes involved in DNA replication initiation and entry into S phase.

    Replication initiation in eukaryotes is characterized by the formation of a pre-replication complex (Pre-RC) with subsequent firing of the replication origins. In S. cerevisiae Orc1-6 (origin recognition complex) proteins bind at the pre-RC site and subsequently Cdc6, Cdt1, and Mcm2-7 are sequentially recruited at the site to form the pre-RC [28]. The helicase complex of Mcm2-7 remains inactive until it is activated by Cdc45 protein. The kinases Cdc7 and Cdk2 are involved in the loading of Cdc45 on origins. Cdc7 phosphorylates the subunits of Mcm2-7, which, through an unknown mechanism, changes the conformation of Mcm2-7 to facilitate the loading of subsequent factors such as Mcm10 and Cdc45. Our studies showed that sequence homologues of several proteins required for DNA replication initiation such as Mcm2-7, Cdc45, and subunits of GINS complex were present in the E. histolytica genome (Table 16.1). Except for a single protein that shared homology with both Cdc6 and Orc1p, none of the Orc2–6 proteins were identified in this organism. Studies with Eh Cdc6/Orc1 are currently in progress to understand the role of this protein in regulating DNA synthesis initiation.

    Table 16.1 Entamoeba histolytica genes encoding homologues of Saccharomyces cerevisiae genes involved in DNA replication initiation
    Full size table
  2. 2.

    G1–S-phase checkpoint proteins in E. histolytica.

    Regulation of DNA replication in S. cerevisiae involves four checkpoint proteins: Mec1, Mrc1, Tof1, and Dpb11. No homologs of these 4 proteins are found in E. histolytica. Several crucial genes required for the G1–S transitions in yeast and humans such as the homologues of p21, p27, p53, and retinoblastoma (RB) genes are absent in the genome of E. histolytica. The conserved intra-S-phase checkpoint proteins Chk1 and Chk2 (which act downstream of ATM and ATR kinases) are also absent in the E. histolytica genome; however, a homologue of human Chk2 has been identified in this organism [29] (Table 16.2).

    Table 16.2 E. histolytica genes encoding homologues of S. cerevisiae checkpoint genes for G1–S phase transition
    Full size table
  3. 3.

    Chromosome segregation and spindle checkpoint proteins in E. histolytica.

    Chromosome segregation occurs in the M phase and is the process of separating the two sister chromatids formed as a result of DNA replication in the S phase. The key players regulating this process are the kinetochore proteins. An assembly of the kinetochore proteins is organized around the centromeric nucleosomes. The outer, central, and inner kinetochore assemblies form the bridge between DNA and microtubule in budding yeast [30]. Strikingly, except five kinesin-like proteins (Eh KlpA1–5), no other components of the kinetochore are found in E. histolytica (Table 16.3). One of these Eh Klps (Eh KlpA1/A2) is homologous to CENP-A. Eh Klp5, a BimC kinesin homologue, was localized with both radial and bipolar spindle assemblies and required for regulating genome content in E. histolytica [31]. However, the absence of other sequence homologues required for kinetochore formation suggests a typical kinetochore may not be present in E. histolytica.

    Table 16.3 E. histolytica genes encoding homologues of S. cerevisiae genes required for chromosome segregation
    Full size table
  4. 4.

    Several G2/M or spindle checkpoint proteins are absent from E. histolytica.

    Spindle checkpoint proteins are activated when there is a perturbation in the alignment of chromosomes to the spindle. Monopolar, bipolar, and multipolar spindles have been observed in E. histolytica [8, 32]. However, the mechanism of chromosome alignment and partitioning on either of these spindles is not well understood. Sequence analysis shows absence of several spindle checkpoint proteins. Interestingly, homologues of two checkpoint proteins, Bub2, which blocks the mitotic exit by inhibiting Cdc14 in response to checkpoint activation, and Mps2, a dual specificity protein kinase required for spindle pole body duplication and spindle checkpoint activation, were found to be present in E. histolytica (Tables 16.4, 16.5, and 16.6). E. histolytica accumulates multinucleated cells, suggesting uncoupling of nuclear and cell division. Sequence analyses revealed that the homologues of all proteins required for cytokinesis in budding yeast are present in this organism (Table 16.7).

    Table 16.4 E. histolytica genes encoding homologs of S. cerevisiae checkpoint genes required for spindle formation
    Full size table
    Table 16.5 E. histolytica genes encoding homologues of S. cerevisiae genes involved in mitosis
    Full size table
    Table 16.6 E. histolytica genes encoding homologues of S. cerevisiae genes involved in exit from mitosis
    Full size table
    Table 16.7 E. histolytica genes encoding homologues of S. cerevisiae genes involved in cytokinesis
    Full size table

5 Novel Proteins That Regulate the Cell Cycle of Entamoeba histolytica

Analysis of the genome of E. histolytica showed that a large number of proteins involved in the cell division process of this parasite were either absent or significantly divergent. Given the fact that a large number of proteins encoded by the parasite are hypothetical opens up a possibility the E. histolytica may contain novel regulatory mechanisms to ensure that the cell cycle proceeds even in the absence of the conventional regulatory proteins. Functional analysis of a kinesin-like protein, E. histolytica Klp5 (EhKlp5), showed that increased expression of this protein, whereas promoting microtubular spindles leads to homogenization of the average DNA content in growing cells [31]. In addition, EhKlp2 was also found to alter the frequency of bipolar spindles and genome content in this parasite. EhKlp2-4 were found to associate with both microtubules and actin cytoskeletal networks although none of these proteins has any actin-binding domains [33]. Yeast two-hybrid analysis with the nonmotor domains of these proteins identified several actin-binding proteins as interactors (Grewal and Lohia, unpublished observations). This observation suggests that these kinesin proteins might associate with the actin cytoskeleton with the help of these interactors and might be involved in regulating motility and cell division.

A novel group of formin proteins was also discovered in E. histolytica of which EhFormin-1 and -2 led to delay in cell division [34]. EhFormins-5 and -8 were phylogenetically distinct from their other eukaryotic counterparts [34]. The characteristic FH1 domain, which is crucial in initiating actin polymerization, is absent in these formin proteins. An InterProScan (EMBL-EBI) of EhFormin-5 and -8 identified a GTPase-binding domain at the N-terminal of these proteins,indicating that this region was unique and may respond to unknown signaling mechanisms or protein networks that subsequently may effect actin remodeling. EhFormin-5 was a nucleocytoplasmic protein, which localized as a ring inside the nucleus of E. histolytica. Furthermore, EhFormin-8 also colocalized on microtubular structures, suggesting that both these proteins might be involved in the regulation of essential nuclear function such as assembly and disassembly of microtubules. Ectopically expressed EhFormin-5 and -8 decreased the genome content in the stable transformants, suggesting that both these proteins are involved in the process of chromosome segregation. To identify the downstream targets of EhFormin-5 and -8, the N-terminal regions of these two formins used as a bait to screen the Eh cDNA library by yeast-two-hybrid assay. EhFormin-5 interacted with a p21 Ras family GTPase, EhRas family GTPase, and five hypothetical proteins. The identification of EhRas family GTPase as an interactor of EhFormin-5 was a significant and novel finding, because, among the members of the p21 Ras family GTPases, only Rho and Rac proteins have been implicated in binding to formin proteins and assisting actin polymerization in other eukaryotic systems. EhFormin-8 interacted with a zinc-finger domain-containing protein, a helicase domain-containing protein, and two hypothetical proteins. Our data suggest that EhFormin proteins have evolved to perform cell cycle-specific functions in addition to their role as actin nucleators Grewal and Lohia unpublished observations. Previous studies have shown that expression of a constitutively active mutant of RacGEh (RacGGly12Val Eh), a dominant-negative mutant of RabA, and the kinase domain of PAK2Eh led to cytokinetic defects and consequently to the accumulation of multinucleated cells [35–37]. Taken together, these observations suggest that cell division in E. histolytica requires the combined activity of several signaling molecules and proteins.

Calcium signaling plays a major role in the cell cycle of eukaryotes. E. histolytica encodes a large repertoire of novel multi-EF hand CaBPs [38]. The calcium-binding proteins characterized thus far have been implicated in having a crucial role in phagocytosis and cell proliferation [39, 40]. EhCaBP6 on the other hand was found to localize at the end of microtubular structures and on the intracellular bridge during cytokinesis [41]. Immunolocalization data suggest that EhCaBP6 is functionally similar to mitotic CaM proteins in other organisms. A large number of cellular processes are regulated by protein–protein interactions. Most proteins require physical interactions with other proteins to execute their biological function. In most eukaryotic cells, calmodulin initiates various signaling cascades by binding to target proteins. Immunofluorescence studies show that EhCaBP6 in mitotic cells may be interacting with the microtubules or with microtubule-associated proteins. In an effort to identify the proteins interacting with EhCaBP6, the latter was used as the bait in a yeast two-hybrid genetic screen against the E. histolytica cDNA library. Among the interactors obtained from the screen EhCaBP6 was also found to interact with a zinc-finger domain protein and ribosomal protein P2. The zinc-finger protein obtained in this screen was the same as obtained as an interactor for EhFormin-8. The interaction of both EhFormin-8 and CaBP6 with a zinc domain protein suggests that EhFormin-8 and EhCaBP6 are part of the same protein complex that affects genome segregation in E. histolytica. Thus, novel proteins such as EhFormins, kinesins, and EhCaBP6 have evolved to regulate crucial processes of microtubule assembly and chromosome segregation in the protozoan parasite E. histolytica.

6 Genome Content of Cysts and Trophozoites

Axenic cultures of E. histolytica have been used since 1961 after the introduction of the TYI-S-33 medium and are indispensable for molecular biology and cell biology studies of this parasite. It was observed that heterogeneity of DNA content was a common feature of all the E. histolytica strains growing under axenic conditions. E. histolytica normally grows in the presence of the microbial flora of the large intestine. We compared the DNA content of two recent isolates, E. histolytica 2592100 and DS4-868, that could be cultured both under xenic and axenic growth conditions. E. histolytica HM1:IMSS has been cultured axenically for more than 40 years and is difficult to revert to xenic culture. The nuclear DNA content of both strains was found to be tenfold lower when grown under xenic conditions compared the corresponding axenically grown cultures. Furthermore, a comparison of the size of the nuclei of these two populations of E. histolytica revealed that the nuclear size in the xenically growing E. histolytica cells was less than that of axenically growing cells. Additionally, the number of multinucleated cells was also greater in the axenic cultures [3]. These results were corroborated by histological sections obtained from patient’s large intestine [3]. In addition to xenic and axenic growth we compared the DNA content of trophozoites and cysts. Entamoeba cells are found in two major forms in nature, that is, cysts and trophozoites. Because in vitro encystation is difficult for E. histolytica, we analyzed the genome content of the cyst and trophozoites in E. invadens. During excystation a cyst gives rise to single amoeba. Scanning cytometry showed that excysted trophozoites had 40 times more DNA content than the cyst nucleus (Fig. 16.1d). Molecular data [3] supported our interpretation that several rounds of whole genome duplication occur during the conversion of a cyst into a viable trophozoite.

Concluding Remarks

Taken together, the data suggest that the Entamoeba genome is able to lose or gain multiple copies that are partitioned into cysts or daughter cells which are actively multiplying. Dynamic plasticity of the genome allows the cells to adapt to different growth conditions both inside and outside the human host. Absence of conserved regulatory mechanisms is therefore a likely necessity for these parasites to constantly adapt to challenging environmental changes. Over and above all these mechanistic differences, genome duplication, segregation, cell division, and conversion to different forms of the protist are unique among all eukaryotes, reminding us again that there is always a deviation from paradigms and rules.