Plasmodium APC3 mediates chromosome condensation and cytokinesis during atypical mitosis in male gametogenesis

The anaphase promoting complex/cyclosome (APC/C) is a highly conserved multi-subunit E3 ubiquitin ligase that controls mitotic division in eukaryotic cells by tagging cell cycle regulators for proteolysis. APC3 is a key component that contributes to APC/C function. Plasmodium, the causative agent of malaria, undergoes atypical mitotic division during its life cycle. Only a small subset of APC/C components has been identified in Plasmodium and their involvement in atypical cell division is not well understood. Here, using reverse genetics we examined the localisation and function of APC3 in Plasmodium berghei. APC3 was observed as a single focus that co-localised with the centriolar plaque during asexual cell division in schizonts, whereas it appeared as multiple foci in male gametocytes. Functional studies using gene disruption and conditional knockdown revealed essential roles of APC3 during these mitotic stages with loss resulting in a lack of chromosome condensation, abnormal cytokinesis and absence of microgamete formation. Overall, our data suggest that Plasmodium utilises unique cell cycle machinery to coordinate various processes during endomitosis, and this warrants further investigation in future studies.


Supplementary Figures
. Structural analysis of APC proteins, generation of APC3-GFP parasites and identification of proteins immunoprecipitated with CDC20-GFP.
(f, g) Analysis of CDC20-GFP by mass spectrometry following immunoprecipitation from a parasite lysate using gametocytes activated for 7 min. Venn diagrams display mean totals of shared and unique proteins compared with WT-GFP and WT-ANKA lines (labelled on graph as WT-ANKA). The table lists unique proteins found only in CDC20-GFP samples. Full results can be found in Supplementary Table S2. (a) Yeast two hybrid assay. Budding yeast transformants expressing the indicated gene products were spotted in serial dilutions on minimal medium SD-2 (-Leu, -Trp) or SD-3 (-Leu, -Trp, -His). As a positive control, plasmids expressing T antigen and p53 were used.
(b) Pb APC10 forms a dimer. Using a Y2H strain expressing both HA-tagged APC10 and myc-tagged APC10, APC10 was immunoprecipitated with either anti-myc or anti-HA antibody and then bound proteins were detected with anti-HA or anti-myc antibody, respectively.
(c) Cell extracts from Y2H strains used were immunoblotted with anti-HA or anti-myc antibody. Plasmid pGADT7 expresses Gal4 activation domain (AD) with a HA tag whereas plasmid pGBKT7 expresses Gal4 DNA binding domain (BD) with a myc tag. AD-or BD-APC10, APC11 and CDC20 bands in blots are highlighted with an orange dot. Neither ADnor BD-APC3 was clearly detected at the expected molecular weight (MW) although some signals were observed at higher MW area (marked with red dots). APC3 may be aggregated or poorly expressed even using a 'yeast' codon-optimized gene.  (c) Southern blot analysis of WT and P ama1 apc3 parasite (cl.2 and cl.5) genomic DNA following PacI digestion. A probe specific for the apc3 3'UTR bound to a 4.7 kb band in WT and to a 7.9 kb band in P ama1 apc3 parasite DNA.  (i) centrin, (j) γ-tubulin and (k) α-tubulin localisation with specific antibodies in P ama1 apc3 or WT-GFP microgametocytes activated for 8 or 15 min. DAPI DNA staining and the merged images are also shown. Scale bar = 5 µm. Supplementary Table S1. List of main protein hits in the APC3-GFP immunoprecipitation experiments.

Gene Products
Gene

Generation of transgenic parasites
All of the oligonucleotides used to genetically confirm the mutant parasite lines can be found in Table S3. For GFP-tagging of apc3 by single homologous recombination, a region of apc3 (PBANKA_1101500) downstream of the ATG start codon, amplified using primers T1141 and T1142, was used to generate the construct as described previously 5 . The gene deletion targeting vectors for ∆apc3, ∆apc10 (PBANKA_1433200) and ∆apc11 (PBANKA_1123400) were constructed using the pBS-DHFR plasmid, which contains polylinker sites flanking a T.
gondii dhfr/ts expression cassette conveying resistance to pyrimethamine, as described previously 4 . The three constructs were transfected in at least 4 independent experiments (a positive control of the APC3-GFP construct was also used). The conditional knockdown construct (P ama1 apc3) was derived from P ama1 (pSS368) where APC3 was placed under the control of ama1 promoter as described previously 10 . PCR primers PTD0031 and PTD0032 were used to generate a 609 bp fragment of 5′ sequence upstream of apc3, directly before the start codon. A 531 bp fragment generated with primers PTD0033 and PTD0034 from the 3′ flanking region of apc3 was then inserted downstream of the ama1 promoter region. P.
berghei ANKA line 2.34 (for GFP-tagging) or ANKA line 507cl1 (for gene deletion and promoter trap) parasites were transfected by electroporation. Six-to-eight week old female Tuck-Ordinary (TO) (Harlan) outbred mice were used for all experiments.

Genotypic analysis of mutants
For the C-terminal fusion GFP tagged parasites, diagnostic PCR confirmed correct integration and Western blot confirmed GFP expression and the correct protein size. For the gene deletion and promoter swap parasites, a diagnostic PCR reaction and Southern blot were used. These approaches are similar to those described previously 10,11 .

Yeast two hybrid assay
Budding yeast strain Y2HGold was transformed with the 2-hybrid vectors pGADT7 (AD) and pGBKT7 (BD) containing codon optimised Plasmodium apc3, apc10, apc11, cdc20 or Tantigen and p53. As each vector contains a nutritional marker, LEU2 and TRP1, respectively, transformants were selected on SD medium lacking leucine and tryptophan (SD-2: -Leu, -Trp). In order to examine protein interactions, the transformants were spotted out on SD medium lacking leucine, tryptophan and histidine (SD-3: -Leu, -Trp, -His) in serial dilutions and incubated at 30˚C for 3 to 4 days. Each plate contained a positive control (T antigen/p53) and a negative control (pGADT7 vector/pGBKT7 vector).

Purification of schizonts, gametocytes and ookinetes
Purification of gametocytes was achieved using a modified version of a previous protocol 12 .
Mice were treated by intraperitoneal injection of 0. penicillin and streptomycin; per 0.5 ml blood) for 24 h at 37°C for schizont development (with rotation at 100 rpm) and at 20°C for ookinete production as described above. Schizonts and ookinetes were purified on 60% and 63% NycoDenz gradients, respectively and harvested from the interface and washed.

Immunoprecipitation and Mass Spectrometry Analysis
Purified

Live imaging and Immunofluorescence assay
Gametocytes were purified and activated in ookinete medium then fixed and processed for IFA with different markers. P. berghei schizonts and gametocytes were collected from parasite-infected blood (as described above

Nuclear DNA content analysis
The nuclear DNA content was analysed by FACS as described previously 5

Quantitative RT-PCR
Total RNA was isolated from purified parasites using an RNeasy purification kit (Qiagen).
cDNA was synthesised using an RNA-to-cDNA kit (Applied Biosystems) allowing quantification from 150 ng of total RNA. qRT-PCR reactions used SYBR green fast master mix (Applied Biosystems) and were analysed using an Applied Biosystems 7500 fast machine with the following cycling conditions: 95 °C for 20 sec followed by 40 cycles of 95 °C for 3 sec; 60 °C for 30 sec. Gene expression in wild type parasites was determined using the Pfaffl method 13 . Relative quantification in the mutant line was normalised against wildtype expression using the ∆∆Ct method. Both experiments used hsp70 and arginine-tRNA synthetase as reference genes. Three biological replicates were used for each stage (each with two technical replicates). The primers used are described in Supplementary Table S3.
Statistical analyses were performed using Excel and GraphPad Prism (GraphPad Software).
For wild type expression an unpaired Student's t-test was performed between apc3 NAG