Developing Drugs From Bugs

Many parasites, including schistosomes, have evolved to produce molecules which manipulate host biology in order to promote propagation of the parasites’ life cycle. An excellent example of one such molecule is the Interleukin-4-inducing Principle from Schistosoma mansoni Eggs (IPSE, Schramm et al.), originally named alpha-1 (McLaren et al.), and later also known as the S. mansoni chemokine-binding protein (smCKBP), as termed by Smith et al.
Click image to open 3D viewer. Click and drag on this crystal structure of IPSE to view it from different angles .
Click image to open 3D viewer. Click and drag on this crystal structure of IPSE to view it from different angles .

IPSE is only expressed by schistosome eggs (Schramm et al.), is secreted into the environment (Mathieson and Wilson and Smith et al.), and has multiple host modulatory properties. First, as its name suggests, IPSE within soluble egg antigen can induce IL-4 release from basophils of schistosomiasis-naïve humans (Falcone et al.) and mice in an IgE-dependent fashion (Schramm et al.). In vitro work indicates IPSE activates NF-AT in IgE-bearing basophils, which is consistent with IPSE’s ability to trigger IL-4 release from these cells:

From Wan et al. Activation of RS-ATL8 reporter system by IgE binding factor IPSE/alpha-1 expressed in wheat germ (white bars) and E. coli (black bars).

All cells (except IgE only) were sensitized for 16 hours with pooled serum from P. Judaica-allergic individuals diluted 1∶50 (v/v). The experiment included positive controls: sensitized with 1 µg/mL IgE and stimulated with the indicated amounts of polyclonal anti-human IgE (hatched bars); stimulated with Par j 2 (100 pg/ml; grey bar), as well as the negative controls (grey bas): serum sensitized/unstimulated cells (serum only). Data are mean ±SD of the readings of three separate wells. Multiplicity adjusted p-values were obtained by ANOVA followed by Dunnett’s test for each condition compared with serum only control ****: p<0.0001; ***: p<0.001; n.s.: not significant.

Other work demonstrated that IPSE was a pepsin-sensitive, heat stable, 30-40 kDa glycoprotein (Wuhrer et al., Meevissen et al., and Meevissen et al.) that binds to IgE (Haisch et al. and Xie et al.), possibly through a crystallin fold (Meyer et al. and Meyer et al.). IPSE has also been reported to potentially induce IL-10 secretion by Bregs (B regulatory cells) by binding antibodies on the surface of these cells (Haeberlein et al.).

Another host modulatory property of IPSE is its capacity to bind to and neutralize the activity of multiple chemokines (Smith et al.). It is possible that this activity of IPSE allows it to “mold” schistosome egg granuloma progression, since induction or administration of anti-IPSE antibodies leads to increased granuloma size in the mouse liver (Fahel et al.) and lung (Smith et al.).

Perhaps the most astonishing ability of IPSE is its ability to enter host cell nuclei through the action of a nuclear localization sequence (Kaur et al.). It is possible that once in host cell nuclei, IPSE induces changes in gene transcription.

The importance of IPSE in schistosomiasis-associated immune responses was underscored by Oliveira et al., who reported that individuals living in schistosomiasis-endemic areas and yet resistant to infection had higher titers of anti-IPSE IgE antibodies.


IPSE’s profound immunomodulatory properties may have implications that extend beyond schistosomiasis. Li et al. observed that pig fibroblasts engineered to produce IPSE and injected into mice shifted the xenogeneic immune response to a type 2 immunity-associated profile. These workers speculate that IPSE could be used to prevent rejection of xenogeneic transplants. Doenhoff et al. and Igetei et al. demonstrated that antibodies to peanut allergens cross-react with antibodies to IPSE, which they conjecture is consistent with the hygiene hypothesis (the theory that allergies are becoming more common in the developed world because our immune systems are no longer “primed” by parasites and other infectious agents present in less clean environments):


From Doenhoff et al. Western blot immunoreactivity of acid-eluted antibodies and control rabbit antisera monospecific for S. mansoni egg antigens; IPSE/alpha-1 and kappa-5 against S. mansoni egg antigens. M = molecular size markers. Primary antibodies: 1 = anti-SmSEA; 2 = anti-SmSEA antibodies eluted from the 43 kDa latex molecule; 3 = rabbit antiserum specific for S. mansoni egg antigen IPSE/alpha-1; 4 = rabbit antiserum specific for S. mansoni egg antigen kappa-5
We have cloned the Schistosoma haematobium orthologs of IPSE (hereafter referred to as H-IPSE, Pennington et al.):


Although there are multiple H-IPSE orthologs predicted to be present in the S. haematobium genome, they cluster into two main clades exemplified by H03-IPSE and H06-IPSE (panel A). These orthologs exhibit >60% identity to M-IPSE (S. mansoni ortholog of IPSE) and 86% to each other, and feature a signal peptide, multiple disulfide bonds, and a nuclear localization sequence like M-IPSE (panels A and B). H-IPSE orthologs are only found in egg-secreted protein and soluble egg antigen (panel C), and mRNA is only expressed in eggs and adult female worms (which carry eggs):


Stage -specific expression of H-IPSE mRNA. RT-PCR results for H-IPSE obtained from cDNAs, prepared by reverse transcription of DNAse-treated RNA of various life stages of S. haematobium. Ladder: 100 basepair (Bp) DNA ladder; egg: S. haematobium egg cDNA; mir: miracidial cDNA; cer: cercarial cDNA; som: in vitro  mechanically transformed schistosomula cDNA; Ad, F, M mixed cDNA from female, male or mixed adult worms, respectively. ShTub: control housekeeping gene, S. haematobium tubulin.

We expressed M-IPSE and H-IPSE in mammalian HEK293-6A cells, which allows for glycosylation of the recombinant proteins:


Expression of M-IPSE and H-IPSE in HEK293-6A cells. A) Schematic diagram of pTT5 H03/06 -IPSE expression cassette. eCMV=Cytomegalovirus (CMV) 801 enhancer sequence; pCMV=CMV promoter; TPL=tripartite leader sequence from adenovirus; eMLP=enhancer element from the adenovirus major late promoter (MLP); hVEGF=human vascular endothelial growth factor signal sequence; 8xHis=octahistidine tag; TEV=Tobacco Etch Virus protease cleavage site; STOP: stop codon; pA: β -globin poly adenylation signal. B) Coomassie -stained 4 -20% SDS – PAGE gradient gel and C) Western Blotting of recombinant H03 -IPSE (and M -IPSE, used as comparison) expressed in HEK 293SF -3F6 cells and purified by IMAC from serum -free culture supernatant, and run under non -reducing (NR) or reducing (R) conditions.

We then incubated H03-IPSE and H06-IPSE with HTB-9 urothelial cells and confirmed that these proteins localized to host cell nuclei:


<Prev    Next>

Recombinant H03 -IPSE is taken up by HTB -9 host cells and translocates to the nucleus. HTB -9 cells, incubated for 24 hours with 0.40 nM recombinant H03 – IPSE, were stained with 5 μM DRAQ5 nuclear stain for 15 minutes at room temperature, followed by staining with a mouse anti -His antibody and Alexa Fluor®  555 conjugated Goat anti -Mouse IgG (H+L) as secondary antibody. The right column shows the overlay of the two channels. The uptake in HTB -9 cells was visualized by confocal microscopy. The primary anti -His antibody was omitted in the control lane.


Fluorescence microscopy of HTB -9 cells incubated with recombinant H03 – IPSE (NLS: SKRRRKY), H06 -IPSE (NLS: SKRGRKY) or H03 -IPSE mutant (NLS: 820 SKAAAKY). HTB -9 cells were stained with Hoechst 33342 nuclear stain for 15  minutes at room temperature, followed by staining with a mouse anti -His antibody and Alexa Fluor® 555 -conjugated Goat anti-Mouse IgG (H+L) as secondary antibody. The right column shows the overlay of the two channels. The primary anti –  His antibody was omitted in the control lane. Bar size is 100 μm.

The nuclear localization sequences of H-IPSE were even able to direct a large, tetrameric GFP construct to host cell nuclei, but this was dependent on an intact sequence:



Effect of multiple amino acid substitutions on NLS in H -IPSE. A) The nucleotides encoding the H0 3/H0 6 -IPSE nuclear localization sequence (SKRRRKY and SKRGRKY, respectively) were inserted into the pTetra -EGFP construct (2,3). pTetra -EGFP encodes a tetrameric EGFP construct resulting in the expression of a  fluorescent protein which due to its size ( >100 kDa) is excluded from the nucleus in the absence of a functional NLS (Tetra -EGFP) or imported into the nucleus in the presence of a functional NLS (canonical SV40 NLS, H0 3/H0 6 -IPSE NLS). Nuclei were stained with DAPI and green fluorescence measured with the GFP light cube on an EVOS fl microscope , 24 hours after transfection. Bar is 100 μm. B) Comparison of wild -type H06 -IPSE, H03 -IPSE and H03 -IPSE mutant NLS effect on nuclear localisation of Tetra -EGFP fusion protein. One hundred transfected HTB9 cells were evaluated under the EVOS fl microscope for each transfection and the percentage of cells displaying exclusive nuclear fluorescence, as opposed to cytosolic only or mixed cytosolic/nuclear localization, recorded. Positive control: Sv40 canonical NLS sequence; negative control: unmodified Tetra -EGFP vector (Tetra -EGFP).

In summary, the S. haematobium genome possesses multiple IPSE orthologs, which are expressed in vivo and feature nuclear localization properties akin to M-IPSE. Given the potent immunomodulatory properties of M-IPSE, we are currently exploring the ability of H-IPSE proteins to favorably modulate inflammation and pain in multiple bladder disease models. Some of this work is presented in the next section.


Prior work by Macedo et al. indicated that administration of recombinant IL-4 can alleviate ifosfamide-induced hemorrhagic cystitis. This study is important because hemorrhagic cystitis caused by ifosfamide and other alkylating chemotherapy agents can be refractory to MESNA, the current standard of care prophylactic. Giving recombinant IL-4 to patients to prevent chemotherapy-induced hemorrhagic cystitis is unrealistic; high doses of this cytokine have been shown to induce cytokine storm symptoms such as capillary leak syndrome and hypotension.


The interleukin-4-inducing properties of IPSE led us to hypothesize that IPSE may have therapeutic effects in ifosfamide-induced hemorrhagic cystitis. Indeed, a single intravenous dose of H06 H-IPSE was highly efficacious in reducing ifosfamide-induced spontaneous pain behaviors:


H-IPSE reduces spontaneous pain but not evoked tactile sensitivity in ifosfamide-treated mice. Mice were administered recombinant IL-4, H06 H-IPSE, H06 H-IPSE with neutralizing anti-IL-4 antibody (a-IL-4), MESNA, or saline before ifosfamide challenge. A) Animals were then placed individually in fresh cages with new bedding. Mice were evaluated and spontaneous pain scored (in a blinded fashion). Higher scores indicate more pain. The effect of H06 was nuclear localization sequence (NLS)-dependent, since ifosfamide-challenged mice given the NLS mutant of H06 did not have a reduction in spontaneous pain behaviors. B) Mice were subjected to tactile allodynia assessment using von Frey monofilaments. The change in withdrawal threshold is the difference between the 50% withdrawal threshold at baseline and 10-h post-ifosfamide injection. As a higher 50% threshold indicates less pain, positive values for changes in withdrawal threshold indicate less pain. The reduced pain observed in the H-IPSE- and IL-4-treated groups was not significant compared with the mice treated with only ifosfamide. The change in 50% threshold was significantly lower in the H-IPSE and H-IPSE NLS mutant groups when co-administered neutralizing anti-IL-4 antibody. The plotted data are pooled from 3 repeat experiments. Horizontal bars represent means. *P , 0.05, **P , 0.01, ***P , 0.001. Adapted from Mbanefo et al.

A single dose of H06 H-IPSE was also very efficacious in reducing ifosfamide-induced bladder hemorrhage, but not edema:


H-IPSE reduces bladder hemoglobin levels, but not bladder wet weight, in ifosfamide-treated mice. A) Mice were administered recombinant IL-4, H-IPSE, and the NLS mutant of H-IPSE, with or without neutralizing anti-IL-4 antibody (a-IL4), MESNA, or saline, before ifosfamide challenge. Animals were then euthanized and their bladders excised, weighed to measure bladder wet weight, and then subjected to the Drabkin’s test to measure tissue hemoglobin levels. The H-IPSE-treated group showed significantly reduced bladder hemoglobin levels compared with the ifosfamide-only group, and prior administration of neutralizing anti-IL-4 antibody reversed the effect of H-IPSE, albeit without achieving statistical significance. The NLS mutant-treated group, as well as MESNA-treated mice, did not exhibit a significant decrease in bladder hemoglobin following ifosfamide exposure. Plotted data are pooled from 3 repeat experiments (n = 3–4 each). The bars represent means and SD. B) There were no significant differences in bladder wet weights across treatment groups. Horizontal bars represent means. *P , 0.05, **P , 0.01. Adapted from Mbanefo et al.

In addition, histology confirmed that one dose of H06 H-IPSE lessened ifosfamide-induced hemorrhage, urothelial denudation, and bladder edema:


H-IPSE mitigates ifosfamide-induced histopathological changes in the bladder. Histologic assessment of bladders from mice receiving the following: saline only (A), ifosfamide only (B), MESNA and ifosfamide (C), recombinant IL-4 and ifosfamide (D), H-IPSE and ifosfamide (E), and H-IPSE + neutralizing anti-IL-4 antibody and ifosfamide (F). Bladders from the saline-treated group showed normal urothelium with no sign of edema (ed) or urothelial denudation (ud). Ifosfamide-injected bladders exhibited segmental urothelial denudation, severe edema with effacement of the submucosa and occasional frank hemorrhage. Ifosfamide-injured bladders exposed to HIPSE, IL-4, or MESNA showed intact urothelium and overall preserved bladder architecture with a relative lack of edema. The protective effect of H-IPSE was reversed by prior administration of neutralizing anti-IL-4 antibody. Adapted from Mbanefo et al.

A single dose of H06 also attenuated ifosfamide-induced voiding dysfunction:


H-IPSE reduces abnormal voiding behaviors in ifosfamide-treated mice. Mice were subjected to the voided spot on paper test to measure their voiding frequency over a 4 h period. As compared with the mice exposed only to ifosfamide, mice administered H-IPSE and ifosfamide showed a significant reduction in voiding frequency. Administration of anti-IL-4 mAb seems to reverse the effect of H-IPSE. Plotted data are from 1 of 2 representative experiments. Horizontal bars represent means. *P , 0.05, **P , 0.01. Adapted from Mbanefo et al.

We suspect that the IL-4-dependent effects of H06 are mediated by bladder mast cells:


Intravenous injection of H-IPSE induces IL-4 expression by basophils. A) A higher proportion of basophils from the peripheral blood of H-IPSE-injected mice were IL-4+ compared with PBS-injected mice. After ifosfamide (Ifos) injection, this cell population was similar in both H-IPSE and PBS groups. B) There was no significant difference in IL-4+ peripheral blood mast cells between H-IPSE and PBS groups before and after ifosfamide injection. C) IL-4-expressing basophils were present at higher numbers in the bladders of both H-IPSE and PBS groups after ifosfamide injection. D) There was no difference in IL-4- positive bladder mast cells, with or without ifosfamide, between the H-IPSE and PBS groups. Consistently, BALB/c mice did not express GFP (IL-4), as expected. Plotted data are pooled from 2 repeat experiments (n = 3–7). Horizontal bars represent means. *P , 0.05, **P , 0.01, ***P , 0.001. Adapted from Mbanefo et al.

The therapeutic effects of H06 in ifosfamide-induced hemorrhagic cystitis are associated with increased gene expression of uroplakins (barrier function genes) and less expression of the pro-inflammatory iNOS gene:


H-IPSE restores expression of uroplakin mRNA and reduces proinflammatory mediator mRNA expression in the ifosfamide-injured bladder. Total RNA was purified from the bladders of mice administered ifosfamide alone or ifosfamide in addition to the following: saline, recombinant IL-4, H-IPSE, the NLS mutant of H-IPSE, H-IPSE with neutralizing anti-IL-4 antibody (aIL-4), or MESNA. Real-time PCR was then performed to measure transcription of uroplakin genes [Up1a (A) and Up2 (B)] and a tight junction-associated gene [Cldn8 (C)]. After ifosfamide insult, H-IPSE-injected mice showed significantly increased transcript levels of the uroplakin genes vs. mice exposed to ifosfamide alone. Transcription of proinflammatory mediator genes [Il-1b (D), Tnf-a (E), and iNOS (F)] was also assessed. H-IPSE- and MESNA-injected mice showed significantly decreased transcript levels of iNOS compared with mice exposed only to ifosfamide. Ifosfamide-exposed mice, given the NLS mutant of H-IPSE or H-IPSE in combination with the anti-IL-4 antibody, still had reduced levels of iNOS gene expression compared with mice given ifosfamide alone. H-IPSE-injected mice also exhibited significantly reduced levels of Tnf-a vs. the group treated with the anti-IL-4 antibody but not Il-1b relative to mice exposed to ifosfamide alone. *P , 0.05, **P , 0.01, ***P , 0.001, ****P , 0.0001. Adapted from Mbanefo et al.