Integrative proteomics and bioinformatic prediction enable a high-confidence apicoplast proteome in malaria parasites

Abstract

Malaria parasites (Plasmodium spp.) and related apicomplexan pathogens contain a nonphotosynthetic plastid called the apicoplast. Derived from an unusual secondary eukaryote-eukaryote endosymbiosis, the apicoplast is a fascinating organelle whose function and biogenesis rely on a complex amalgamation of bacterial and algal pathways. Because these pathways are distinct from the human host, the apicoplast is an excellent source of novel antimalarial targets. Despite its biomedical importance and evolutionary significance, the absence of a reliable apicoplast proteome has limited most studies to the handful of pathways identified by homology to bacteria or primary chloroplasts, precluding our ability to study the most novel apicoplast pathways. Here, we combine proximity biotinylation-based proteomics (BioID) and a new machine learning algorithm to generate a high-confidence apicoplast proteome consisting of 346 proteins. Critically, the high accuracy of this proteome significantly outperforms previous prediction-based methods and extends beyond other BioID studies of unique parasite compartments. Half of identified proteins have unknown function, and 77% are predicted to be important for normal blood-stage growth. We validate the apicoplast localization of a subset of novel proteins and show that an ATP-binding cassette protein ABCF1 is essential for blood-stage survival and plays a previously unknown role in apicoplast biogenesis. These findings indicate critical organellar functions for newly discovered apicoplast proteins. The apicoplast proteome will be an important resource for elucidating unique pathways derived from secondary endosymbiosis and prioritizing antimalarial drug targets.

Fig 1. The promiscuous biotin ligase BirA* biotinylates proteins in the P. falciparum apicoplast and ER.

(A) Schematic (not to scale) of constructs for apicoplast- and ER-targeting of GFP–BirA*. (B) Fixed-cell imaging of BioID–Ap and BioID–ER parasites stained with antibodies raised against the apicoplast marker ACP or the ER marker BiP, respectively. Scale bars, 5 μm. (C) Western blot of untreated and biotin-labeled Dd2^attB, BioID–Ap, and BioID–ER parasites. (D) Fixed-cell imaging of biotinylated proteins in biotin-labeled BioID–Ap and BioID–ER parasites. Scale bars, 5 μm. ACP, acyl carrier protein; ACP_L, ACP leader sequence; BirA*, promiscuous E. coli biotin ligase; BioID, proximity-dependent biotin identification; BiP, binding immunoglobulin protein; ER, endoplasmic reticulum; GFP, green fluorescent protein; SDEL, ER-retention motif; SP, signal peptide.

Fig 2. Accurate, unbiased identification of apicoplast proteins using BioID.

(A) Abundances of 728 proteins identified by mass spectrometry in BioID–Ap and BioID–ER parasites. Protein abundances were calculated by summing the total MS1 area of all matched peptides for a given protein and normalized by the total summed intensity of all P. falciparum peptides matched. Dotted line represents 5-fold apicoplast:ER enrichment. (B) ROC curve used to identify the apicoplast:ER enrichment that maximized true positives while minimizing false positives. Dotted lines denote the sensitivity and false positive rate of the 5-fold cutoff used. False positive rates for hypothetical 2-fold and 1-fold enrichments are shown for reference. (C) Sensitivities of BioID, PATS, PlasmoAP, and ApicoAP based on identification of 96 known apicoplast proteins. (D) PPV of BioID, PATS, PlasmoAP, ApicoAP, and a data set consisting of proteins predicted to localize to the apicoplast by all 3 bioinformatic algorithms. Calculated as the number of true positives divided by the total number of true positives and false positives. Error bars in (C) and (D) represent 95% confidence intervals. Tabulated data are available in S1 Data. ApicoAP, Apicomplexan Apicoplast Proteins algorithm; BioID, proximity-dependent biotin identification; ER, endoplasmic reticulum; ND, not detected; PlasmoAP, Plasmodium falciparum Apicoplast-targeted Proteins algorithm; PATS, Predict Apicoplast-Targeted Sequences algorithm; PPV, positive predictive value; ROC, receiver operating characteristic.

Fig 3. Diversity of protein labeling by apicoplast BioID.

(A) Fraction of proteins identified by apicoplast BioID that are predicted to localize to a membrane. Proteins were considered “membrane” if they had at least 1 transmembrane domain annotated in PlasmoDB ending >80 amino acids from the annotated N-terminus. (B) Number of lumenal and nonlumenal positive controls identified. Percentages above bars indicate the percentage of known proteins from each category identified. (C) Number of proteins from established apicoplast pathways identified. Percentages above bars indicate the percentage of known proteins from each pathway identified. Tabulated data are available in S1 Data. BioID, proximity-dependent biotin identification.

Fig 4. Improved prediction of apicoplast proteins using the PlastNN algorithm.

(A) Schematic of the PlastNN algorithm. For each signal peptide-containing protein, a region of 50 amino acids immediately following the signal peptide cleavage site was selected, and the frequencies of the 20 canonical amino acids in this region were calculated, resulting in a vector of length 20. Scaled RNA levels of the gene encoding the protein at 8 time points were added, resulting in a 28-dimensional vector representing each protein. This was used as input to train a neural network with 3 hidden layers, resulting in a prediction of whether the protein is targeted to the apicoplast or not. (B) Table showing the performance of the 6 models in PlastNN. Each model was trained on five-sixths of the training set and cross-validated on the remaining one-sixth. Values shown are accuracy, sensitivity, specificity, NPV, and PPV on the cross-validation set. The final values reported are the average and standard deviation over all 6 models. (C) Comparison of accuracy, sensitivity, specificity, NPV, and PPV for 3 previous algorithms and PlastNN. NPV, negative predictive value; PlastNN, Apicoplast Neural Network; PPV, positive predictive value.

Fig 5. Apicoplast BioID identifies novel and essential proteins.

(A) Percentage of proteins identified that have 1) annotated gene products but unknown function, 2) gene products annotated explicitly with “unknown function,” or 3) annotated gene products and function in a known cellular pathway. (B) Percentage of proteins identified that are Plasmodium- or Apicomplexa-specific based on OrthoMCL-DB. (C) Percentage of proteins identified that are essential, cause slow growth when deleted, or are dispensable based on PlasmoGEM essentiality data of P. berghei orthologs [41]. (D) Percentage of proteins identified that were classified as mutable or nonmutable based on genome-scale transposon mutagenesis in P. falciparum [42]. In each panel, absolute numbers of proteins are indicated within bars. Tabulated data are available in S1 Data. BioID, proximity-dependent biotin identification; OrthoMCL-DB, Ortholog Groups of Protein Sequences database; PlasmoGEM, Plasmodium Genetic Modification project.

Fig 6. Localization of candidate apicoplast proteins identified by BioID.

(A) Transit peptide processing assay for C-terminally GFP-tagged candidates. Ring-stage parasites were either untreated or treated with 10 μM actinonin/200 μM IPP for 3 days and protein processing was assessed by western blot. (B) Fixed-cell imaging of GFP-tagged candidates in parasites stained with an antibody raised against the apicoplast marker ACP. ROM7–GFP-expressing parasites were also stained with anti-GFP antibody due to low signal from intrinsic GFP fluorescence in fixed cells. Arrowheads indicate regions where PF3D7_0721100-GFP puncta appear adjacent to as opposed to colocalizing with ACP. Scale bars, 5 μm. ACP, acyl carrier protein; BioID, proximity-dependent biotin identification; GFP, green fluorescent protein; IPP, isopentenyl pyrophosphate; ROM, rhomboid protease homolog.

Fig 7. Localization of candidate apicoplast proteins identified by PlastNN.

(A) Transit peptide processing assay for C-terminally GFP-tagged candidates. Ring-stage parasites were either untreated or treated with 10 μM actinonin/200 μM IPP for 3 days and protein processing was assessed by western blot. (B) Fixed-cell imaging of GFP-tagged candidates in parasites stained with an antibody raised against the apicoplast marker ACP. Scale bars, 5 μm. ACP, acyl carrier protein; GFP, green fluorescent protein; IPP, isopentenyl pyrophosphate; PlastNN, Apicoplast Neural Network.

Fig 8. ABCF1 is an essential apicoplast protein required for organelle biogenesis.

(A) Fixed-cell imaging of ABCF1-3xHA knockdown parasites stained with antibodies raised against the HA tag and the apicoplast marker ACP. Scale bar, 5 μm. (B-F) ABCF1-3xHA knockdown parasites were grown in the presence of ATc (+ATc), the absence of ATc (-ATc), or the absence of ATc with IPP supplementation (-ATc/+IPP) for 4 growth cycles. (B) Western blot of ABCF1-3xHA expression. (C) Parasite growth. At each time point, data are normalized to the untreated (+ATc) control. Error bars represent standard deviation of the mean of 2 biological replicates. *P < 0.05, ****P < 0.0001 compared to untreated control, ††††P < 0.0001 compared to -ATc condition, repeated measures two-way ANOVA with Tukey’s multiple comparisons test. (D) Relative apicoplast:nuclear genome ratio as determined by quantitative PCR. At each time point, data are normalized to the untreated (+ATc) control. Error bars represent standard deviation of the mean of 2 biological replicates, each performed in technical triplicate. ***P < 0.001, ****P < 0.0001, repeated measures two-way ANOVA with Sidak’s multiple comparisons test. (E) Western blot of ClpP processing. (F) Fixed-cell imaging showing ACP localization after 2 cycles of knockdown. Scale bars, 5 μm. Tabulated data for (C) and (D) are available in S1 Data. ABC, ATP-binding cassette; ACP, acyl carrier protein; ATc, anhydrotetracycline; ANOVA, analysis of variance; ClpP, caseinolytic protease subunit P; HA, hemagglutinin; IPP, isopentenyl pyrophosphate.