Extracellular release of two peptidases dominates generation of the trypanosome quorum-sensing signal

Abstract

Trypanosomes causing African sleeping sickness use quorum-sensing (QS) to generate transmission-competent stumpy forms in mammalian hosts. This density-dependent process is signalled by oligopeptides that stimulate the signal transduction pathway leading to stumpy formation. Here, using mass spectrometry analysis, we identify peptidases released by trypanosomes and, for 12 peptidases, confirm their extracellular delivery. Thereafter, we determine the contribution of each peptidase to QS signal production using systematic inducible overexpression in vivo, and confirm this activity operates through the physiological QS signalling pathway. Gene knockout of the QS-active peptidases identifies two enzymes, oligopeptidase B and metallocarboxypeptidase 1, that significantly reduce QS when ablated individually. Further, combinatorial gene knockout of both peptidases confirms their dominance in the generation of the QS signal, with peptidase release of oligopeptidase B mediated via an unconventional protein secretion pathway. This work identifies how the QS signal driving trypanosome virulence and transmission is generated in mammalian hosts.

Fig. 1. Mass spectrometry analysis of released trypanosome proteins.

a Schematic representation of the samples used for the analysis of the secreted proteome from bloodstream slender forms, stumpy forms or stumpy form parasites undergoing differentiation to procyclic forms. b Scatter plots showing the reproducibility of three biological replicates of secreted material released from slender forms, stumpy forms or cells 3 h or 6 h into differentiation to procyclic forms. c Volcano plots of proteins detected as released from either slender or stumpy forms (left-hand panel) or during early differentiation (right-hand panel) highlighting released peptidases. Log2 of the fold change (FC) is plotted on the x-axis and the −log10 of the p value (ANOVA) on the y-axis. The vertical dotted lines indicate a fold change >1.5 and the horizontal dotted line a p value < 0.05. Red dots represent peptidases detected in the dataset; significance thresholds are indicated by dotted lines. SL is slender, ST is stumpy and PCF is procyclic. Source data are provided as a source data file.

Fig. 2. Peptidases identified by mass spectrometry are released from intact trypanosomes.

a Release of identified peptidases by bloodstream form parasites. In each case, the identified peptidase was epitope tagged using constructs targeting the endogenous gene locus to ensure close to physiological expression. Resultant cell lines were assayed in serum-free Creek’s medium for 2 h and then the cell pellet (P) and supernatant (S) isolated by centrifugation. Isolated proteins were monitored for the distribution of each peptidase using the BB2 antibody detecting the Ty1 epitope tag, with the abundant cytosolic protein EF1-alpha used to monitor non-specific escape of cytosolic proteins. Molecular weight markers in kDa. b Inducible ectopic expression of each peptidase identified as released by parasites. In each case, N-terminally epitope-tagged genes were cloned into a doxycycline-inducible expression vector and then the expression and extracellular release of the expressed proteins were monitored using the epitope tag-specific antibody BB2, with the cytosolic protein EF1-alpha used as a control. Released and cell-associated proteins were isolated after incubation of parasites in serum-free Creek’s medium for 2 h and centrifugation. Molecular weight markers in kDa. Source data are provided as a source data file.

Fig. 3. Ectopic expression of three peptidases enhances QS.

a Ectopic overexpression phenotypes of identified peptidases in vivo. 1 × 10⁴ of each parasite line were inoculated into MF1 mice and progression of the parasitaemia monitored after induction of the ectopic peptidase expression from the day of infection. For each peptidase, mice were infected, being either induced or uninduced to express the peptidase via provision of doxycycline in 5% sucrose (induced; red) or 5% sucrose (uninduced; blue) to the drinking water of infected animals (n = 2 biologically independent analyses, with n = 3 where effects were seen in trial studies). The parasitaemia of the parental T. brucei AnTat 1.1 90:13 cells was also assayed (yellow) and is shown in panel Tb927.11.12850 to illustrate its equivalent progression to uninduced parasites. Values are mean ± SEM. Two-way Anova followed by Tukey’s or Šídák’s multiple comparison test, p < 0.05 (*); p < 0.002 (**), p < 0.0002 (***) p < 0.0001 (****). Tb927.11.6590; p = 0.0092 and 0.0295 respectively, Tb927.8.7020; p = 0.0070, 0.0207 and 0.0004 respectively, Tb927.11.12850; p = 0.0004 and 0.0177 respectively, Tb927.11.2500; p = 0.0085 and 0.0038 respectively. b Relative expression and extracellular release of each identified peptidase. In each case, cell pellet and supernatant were isolated after 2 h incubation in Creek’s medium and the epitope-tagged peptidase was detected using the Ty1-specific BB2 antibody. Loading is indicated by the cytosolic protein Trypanothione synthetase (TyrS). The expression level and extracellular release of the respective peptidases did not correlate with phenotypic effects observed either in vitro or in vivo. Source data are provided as a source data file.

Fig. 4. Accelerated differentiation activated by the expression of three peptidases is mediated via the QS signalling pathway.

a Infection profile for mice infected with parasite lines ectopically expressing each peptidase (Tb927.11.2500, Tb927.11.12850 or Tb927.8.7020) under doxycycline regulation with the QS signalling pathway disrupted (ΔRBP7A/RBP7B). In each case, six mice were infected (biologically independent samples), with three being provided with doxycycline to induce ectopic peptidase expression (uninduced, blue; induced, red). In all cases, the parasitaemia (mean ± SEM) progressed unchecked due to the lack of transduction of the QS signal caused by deletion of RBP7A/B. b Confirmation of the inducible expression and extracellular release of each peptidase in an ΔRBPA/B background. Parasites were harvested from the infections shown in a, with three mice being induced for ectopic peptidase expression and three being uninduced as biologically independent samples. Detection of the stumpy marker PAD1 demonstrated equivalent stumpy generation in the parasites regardless of the inducible expression of the peptidase. Stringent doxycycline-regulated expression is observed in each case. The loading control for the cell pellet was EF1-alpha. Parasites were not successfully purified from one of the mice in the +Dox group at the end of the experiment due to a failure with the DEAE resin. c Quantitation of PAD1 expression on day 6 post infection when parastaemias exceeded 1 × 10⁹/ml. PAD1 immunofluorescence assays were based on three biologically independent infection profiles represented in a and analysed by western blotting represented in b. Source data are provided as a source data file.

Fig. 5. Null mutants for two released peptidases show reduced quorum sensing.

a PCR analysis of genomic DNA derived from null mutants of Tb927.8.7020, Tb927.11.2500 and Tb927.11.12850 analysed using gene-specific primers. Several null mutant clones were analysed in each case, together with wild-type parental parasites. For each target gene no amplicon was detected with the null mutant clones, whereas a band of the expected size was detected using gDNA from wild-type parasites. Integrity of the gDNA in each case was validated using primers targeting a control gene (lower panels). b Parasitaemias (mean ± SEM) of infections initiated with null mutant line (red) or parental line (blue) for each target gene. Biologically independent replicate infections are shown (n = 2), with Tb927.11.2500 and Tb927.11.12850 showing elevated parasitaemia. Note that the scales differ on the y axis between graphs. c Upper panel- PCR confirmation of the successful add back of a copy of Tb927.11.12850 to the null mutant. Lower panel- parasitaemias (mean ± SEM) of biologically independent infections initiated with parental T. brucei AnTat 1.1 parasites (yellow, n = 2), a Tb927.11.12850 null mutant (red, n = 3), or the Tb927.11.12850 null mutant containing an add back for the gene (blue, n = 2). The null mutant exhibited elevated parasitaemia reflecting reduced differentiation, whereas the add back of the Tb927.11.12850 gene restored the growth profile to levels similar to the parental line. d PAD1 expression of parasites at day 5, 6 and 7 post infections. The null mutant exhibited reduced PAD1 expression at peak parasitaemia compared to the parental cells or add back line in biologically independent analyses (respectively n = 3, n = 2, n = 2). Beyond day 5, PAD1 levels were similar between lines, reflecting that the null mutant exhibited delayed but not abrogated stumpy formation, with stumpy parasites further enriched due to their enhanced tolerance of the developing immune response beyond day 5. Source data are provided as a source data file.

Fig. 6. Two released peptidases contribute combinatorially to the generation of the quorum sensing response.

a biologically independent infection profiles (mean ± SEM) of parental T. brucei AnTat 1.1 parasites (yellow, n = 2), Tb927.11.12850 null mutants (blue, n = 3) and Tb927.11.12850 plus Tb927.11.2500 double null mutants (red, n = 3) in vivo. The Two-way ANOVA analysis followed by Tukey’s comparison test was performed for statistical significance between Tb927.11.12850KO vs Tb927.11.12850 plus Tb927.11.2500 double null mutant (Supplementary Table 1). The double peptidase KO was most virulent, the single peptidase KO was less virulent but more virulent than the parental line. Two-way ANOVA analysis followed by Tukey’s comparison test was performed, P < 0.0001 (****). b PAD1 expression (mean ± SEM) of T. brucei AnTat 1.1 parasites, Tb927.11.12850 null mutants and Tb927.11.12850 plus Tb927.11.2500 double null mutants on day 5, 6 and 7 of infection (respectively, n = 2, n = 3, n = 3 in biologically independent samples). At peak parasitaemia on day 5 both the single peptidase and double peptidase infections exhibited less PAD1 expression than parental cells. Beyond day 5, PAD1 levels were similar between lines, reflecting that the null mutants exhibited delayed but not fully abrogated stumpy formation, with stumpy parasites further enriched due to their enhanced tolerance of the developing immune response beyond day 5. c biologically independent infection profiles (mean ± SEM) of parental T. brucei AnTat 1.1 parasites (yellow, n = 2), Tb927.11.12850 plus Tb927.11.2500 double null mutants (red, n = 3), and Tb927.11.12850 plus Tb927.11.2500 double null mutants with a restored Tb927.11.12850 gene (blue, n = 3) in vivo. The statistical significance between Tb927.11.12850 plus Tb927.11.2500 double null mutant and the restored Tb927.11.12850 cell line is shown (p < 0.0001). The double peptidase KO was most virulent, whereas the add back less virulent than the double KO but more virulent than the parental cell line. Two-way ANOVA analysis followed by Tukey’s comparison test was performed, P < 0.0001 (****) (Supplementary Table 1). d PAD1 expression (mean ± SEM) of T. brucei AnTat 1.1 parasites, Tb927.11.2500::Tb927.11.2500 double null mutants and Tb927.11.2500::Tb927.11.2500 double null mutants with restored Tb927.11.12850 add back on day 5, 6 and 7 of infection (respectively, n = 2, n = 3, n = 3 in biologically independent samples). At peak parasitaemia on day 5 both the double peptidase null mutants and the add back mutant infections exhibited less PAD1 expression than parental cells. Beyond day 5, PAD1 levels were similar between parasite lines, reflecting that the null mutants and add back line exhibited delayed but not fully abrogated stumpy formation, with stumpy parasites further enriched due to their enhanced tolerance of the developing immune response beyond day 5. Source data are provided as a source data file.

Fig. 7. Oligopeptidase B exhibits unconventional protein secretion.

a Growth in vitro (mean ± SEM) of T. brucei AnTat1.1 90:13 constitutively expressing epitope tagged Tb927.11.12850 (oligopeptidase B), induced to express dominant-negative SAR1 (upper panel) or wild type SAR1 (lower panel)(n = 2 in biologically independent analyses). Expression of the dominant-negative SAR1 induced growth cessation within 24 h. b Extracellular release of epitope tagged Tb927.11.12850 from parasites induced to express dominant-negative SAR1 (upper panel) or wild type SAR1 (lower panel). Tb927.11.12850 was released at equivalent levels in each case (with an additional degradation product in the pellet fractions only), whereas the cytosolic control EF1-alpha remained cell associated. c Growth in vitro (mean ± SEM) of T. brucei AnTat 1.1 90-:13 constitutively expressing epitope tagged Tb927.11.12850, with induced RNAi targeting Rab11 (n = 2 in biologically independent analyses). RNAi against Rab11 induced growth cessation within 24 h. d Extracellular release of epitope tagged Tb927.11.12850 from parasites induced to deplete Rab11 by RNAi. Extracellular release of Tb927.11.12850 was sustained at equivalent levels, whereas the cytosolic control EF1-alpha remained cell associated. Source data are provided as a source data file.

Fig. 8. A model for the quorum-sensing signalling in trypanosomes.

During trypanosome infection, the parasites release peptidases into the bloodstream of the mammalian hosts. These hydrolyse host proteins, generating oligopeptides which are transported by GPR89, a surface transporter protein found only on the surface of the slender forms of trypanosome parasites. * peptidases identified in this study, # peptidases shown to enhance stumpy differentiation by ectopic overexpression in ref. . TbOPB is trypanosome oligopeptidase B (Tb927.11.12850), TbMCP1 is metallocarboxypeptidase 1 (Tb927.11.2500), Peptidase 1 is Tb927.8.7020, TbPOP is prolyl oligopeptidase (Tb927.10.8020) and TbPGP is pyroglutamyl peptidase (Tb927.4.2670).