| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Institute of Cell Biology and
Institute of Parasitology, University of Bern, Bern, Switzerland
2Correspondence: Institute of Cell Biology, Baltzerstrasse 4, CH-3012 Bern, Switzerland. E-mail: thomas.seebeck{at}izb.unibe.ch
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: African sleeping sickness • PDE inhibitor • cytokinesis • life cycle stages • paraflagellar rod structure
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, starvation (2)
, local changes of NO or carbon dioxide concentrations (3
4
5)
, hyperosmotic or heat shock (6
, 7)
, oxidative stress (8)
, or the control of cell differentiation (9)
and tumor formation (10)
. Cyclic nucleotide specific phosphodiesterases (PDEs) constitute the only means by which eukaryotic cells can rapidly dispose of the cyclic nucleotides generated as a response to such signals and thus can control shape, amplitude, and duration of cyclic nucleotide signals (11)
. Such signals can activate or modulate a variety of downstream effector proteins, including protein kinases A or G, guanine nucleotide exchange factors (12)
, cyclic nucleotide gated channels (13)
, or the activity of certain adenylyl cyclases (14)
and PDEs (15)
. Some PDEs can also mediate crosstalk between cAMP and cGMP signaling pathways, as exemplified by human PDE2 where cAMP hydrolysis can be stimulated by cGMP binding to the regulatory GAF domain (16)
. PDEs can also integrate signals from different protein kinase signaling cascades since they might be phosphorylated and their activity modulated by kinases that belong to different pathways (17)
.
In view of these complex and highly interactive functions of the PDEs, it has become increasingly clear that the catalytic properties of PDEs are not the only parameters that matter for their intracellular function. Rather, functionality is determined to a large extent by their precise subcellular localization (16
, 18
, 19)
.
Anchoring of PDEs brings them into close physical proximity to the cyclases that generate the cyclic nucleotides, to proteins such as kinases that modulate their activity, and to the downstream effector proteins that take up the cyclic nucleotide signal (20)
. Thus, the PDEs not only exert tight control over cyclic nucleotide steady-state levels, but they also act as gauges that define how much of a signal actually reaches downstream effectors, and they restrict the space within which the cyclic nucleotides may diffuse. PDEs can thus generate well-defined microdomains of cellular space where cyclic nucleotide concentrations can reach high levels over short times (16
, 20)
. Only effector proteins that are able to sample such microdomains will become activated. This architecture allows a tight spatial and temporal control of each signal. The importance of PDE association with protein scaffolds for proper function has been further emphasized by evidence obtained from genetic diseases (21)
. Binding of human PDE4 to the scaffolding protein DISC appears to be crucial for correct cAMP signaling in the brain. A genetic disruption of the DISC protein is correlated with schizophrenia and depression.
In most cases, pharmacological inhibition or genetic ablation of PDE activity does not produce a lethal phenotype in individual cells. This is exemplified by the generation of yeast mutants where the disruption of both PDE genes produced no phenotype under standard growth conditions. Only when these cells are exposed to stress does a phenoytype become apparent (7
, 8)
. The disruption of PDE genes or the pharmacological inhibition of PDE activities does not grossly disrupt the overall functioning of multicellular organisms. A number of mouse strains deficient for individual PDE genes have been constructed and are viable, although they exhibit lesser symptoms such as reduced growth, increased prenatal mortality (22)
, or complete female infertility due to a block in oocyte development (23)
. These and many similar findings suggest that cyclic nucleotide signaling mostly exerts a modulatory role on cellular functions. This paradigm has enabled the development of numerous therapeutic compounds that are based on PDE inhibitors. All of these exert specific actions in distinct organs but do not lead to cell death or irreversible changes in target cell physiology.
The human pathogen Trypanosoma brucei is a pathogenic protozoon of the order of the kinetoplastida, and it is the causative agent of human sleeping sickness in Africa (24)
. T. brucei undergoes a complex life cycle, including several proliferation and differentiation stages both in the mammalian host and in the tsetse fly vector. The regulatory role of cAMP in these processes is still poorly understood. cAMP signaling may be involved in the removal of antibody-bound cell surface glycoproteins triggering (25)
, as well as in triggering differentiation of the proliferative "long slender" forms into the nonproliferative, fly-infectious "short stumpy" forms in the mammalian bloodstream (26)
. The genome of T. brucei codes for 
50 adenylyl cyclases, all with the same overall architecture consisting of a large, N-terminal extracellular domain, a single transmembrane helix, and a conserved catalytic domain (27)
. Although many of these genes are constitutively expressed, a distinct subgroup is coexpressed only with individual variable surface glycoprotein genes from specific subtelomeric expression sites (28)
. Three genes for catalytic subunits of protein kinase A (PKA) were identified in T. brucei (29)
, as well as single gene for a PKA regulatory subunit (30)
. However, despite its sequence conservation with PKA of higher organisms, T. brucei PKA is not regulated by cAMP in vitro (30)
.
The genome of T. brucei contains five genes that code for class 1 PDEs (31)
and none for class 2 or metallo-ß-lactamase PDEs, and a unified nomenclature has been proposed (32
, 33)
. TbrPDEB1 and TbrPDEB2 code for two closely similar, cAMP-specific PDEs. The two genes are tandemly arranged on chromosome 9, with their open reading frames separated by 2379 base pairs (see Fig. 1
A). The first gene of the tandem (TbrPDEB1) was described earlier as TbPDE2C (34
; NCBI acc. nr AY028446, GeneDB number Tb09.160.3590), while the second gene (TbrPDEB2) was initially published as TbPDE2B (35
; GeneDB number Tb09.160.3630). Syntenically arranged, highly conserved homologues of TbrPDEB1 and TbrPDEB2 are also found in the genomes of other kinetoplastid pathogens, including T. vivax, T. cruzi, L. infantum, and L. major (32
, 36
37
38)
.
|
The open reading frames of TbrPDEB1 (2793 bp) and TbrPDEB2 (2778 bp) code for polypeptides of 930 and 925 amino acids with a similar overall sequence organization. Their N termini contain two tandemly arranged GAF domains, followed by a C-terminal, highly conserved catalytic domain. The overall sequence identity between TbrPDEB1 and TbrPDEB2 is only 29.5% between the N-terminal 212 amino acids but is 88.5% throughout the remainder of the polypeptides, including the GAF domains and the catalytic domains. The catalytic domains are completely identical, with the exception of a single stretch of 33 amino acids that is dissimilar between TbrPDEB1 and TbrPDEB2 (only 9 of these 33 amino acids are identical). Interestingly, similar stretches of divergence are also found, at identical positions, in the respective homologues of T. cruzi and L. major (36
, 37)
. This stretch of divergence spans the predicted helices 12 and 13, a part of the PDE catalytic domain that is also diverged between all human PDEs, both in terms of sequence and length (39)
. The GAF-A domain of TbrPDEB2 binds cAMP with an affinity of 16 nM (40)
. cAMP binding to this region of the protein lowers the apparent Km of the catalytic domain, suggesting that the GAF domains may modulate the activity of the catalytic domains. The GAF-A domains of TbrPDEB1 (Oberholzer, unpublished observations) and of TbrPDEB2 (40)
remain the only experimentally verified cAMP-binding domains in T. brucei so far.
Interestingly, in T. brucei many of the proteins predicted to be involved in cAMP signaling were found to be located in the flagellum and within the flagellum often tightly associated with the paraflagellar rod structure (PFR; Oberholzer et al., unpublished observations). In T. cruzi, the direct physical interaction of an adenylyl cyclase with a PFR protein was demonstrated (41)
, and the flagellar localization of the cAMP-specific PDE TcrPDEB2 was shown by immunofluorescence (36)
.
Earlier studies from this laboratory have indicated that RNAi against TbrPDEB1 and TbrPDEB2 interferes with cell proliferation and leads to cell death. At the same time, intracellular levels of cAMP were raised significantly (34)
. Although these results clearly suggested an important role of cAMP in cell cycle control and proliferation, the RNAi vectors available at that time were too leaky to allow more detailed analyses.
The current study now confirms and extends these early results both in cell culture and in vivo, and it demonstrates that TbrPDEB1 and TbrPDEB2 exhibit distinct subcellular localizations. TbrPDEB1 is localized exclusively in the PFR of the flagellum (42)
to which it is tightly associated. In contrast, TbrPDEB2 is mostly found in the cytoplasm and can be solubilized by nonionic detergents. A minor fraction of TbrPDEB2 colocalizes with TbrPDEB1 in the PFR. In the procyclic form of T. brucei (corresponding to the form of the parasite that proliferates in the midgut of the tsetse fly vector), simultaneous RNAi against the two genes produces no discernible phenotype. In stark contrast, the bloodstream form of T. brucei is exquisitely sensitive to RNAi against TbrPDEB1 and TbrPDEB2 if both genes are targeted simultaneously. Such a double-RNAi against TbrPDEB1 and TbrPDE2 is also capable of completely preventing or eliminating trypanosome infections in animals, suggesting that these PDEs might be interesting targets for developing specific PDE inhibitors as a novel class of trypanocidal drugs.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
RNAi against the TbrPDEB family in procyclic forms
DNA fragments specific for TbrPDEB1 (bp 3–349) and for TbrPDEB2 (bp 158–520), and a fragment common to both (bp 1094–1444 of TbrPDEB1 and TbrPDEB2) were polymerase chain reaction (PCR) amplified from genomic DNA with the high fidelity polymerase system (Roche, Mannheim, Germany) with the following primers (underlined XhoI and BamHI restriction sites): B1-RNAi-f (ATCTCGAGGTTCATGAACAAGCCCTTTGGCA); B1-RNAi-r (ATGGATCCTCCCCGGGCACCCGGTACAC); B2-RNAi-f (TACTCGAGCCCCTGGTTTGCCCACCCAC); B2-RNAi-r (ATGGATCCAACGGCATCCCCTCGGAGAT); B1,2-RNAi-f (TACTCGAGCCTTCTCTATGTTTGCTGGCGCCT); and B1,2-RNAi-r (GCGGATCCTGGTACGCGTCCTGAATA).
DNA fragments were verified by sequencing and ligated in two opposite directions into the pSLcomp1 hairpin vector (a derivative of pNA8; ref 43
). Restriction sites used were BclI, SalI, XhoI, and BamHI. The constructs were linearized with NotI, ethanol precipitated, and transformed into the NYDM procyclic double-marker cell line (44)
. This procyclic (insect form) cell line was cultivated in SDM-79 medium (45)
, supplemented with 5% FCS and containing 15 µg/ml G418 and 25 µg/ml hygromycin. Selection for the RNAi constructs was done with 1 µg/ml phleomycin. RNAi was induced with 1 µg/ml tetracycline.
RNAi against the TbrPDEB family in bloodstream forms
DNA fragments specific for TbrPDEB1 (bp 150–428), for TbrPDEB2 (bp 314–542), and a fragment common to both (bp 1965–2201 of TbrPDEB1 and TbrPDEB2) were PCR amplified from genomic DNA with the high fidelity polymerase system with the following primers (underlined HindIII, XbaI, XhoI, and BamHI restriction sites): B1-f (CAGCAAGCTTGGATCCGGCAGCCCTTATCAAACGTA); B1-r (CCTCTAGACTCGAGTTTCATCAATCAATGCTGCC); B2-f (CAGCAAGCTTGGATCCACGAGCATTTGTCCCTTGTC); B2-r (CGGATCTAGACTCGAGCCACAAGGACTCAGTACGCA); B1,2-f (CAGCAAGCTTGGATCCCATTCTGCAATGCCGTAAGA); and B1,2-r (CGGATCTAGACTCGAGGCGCTGGAAAGAATACCAAG).
DNA fragments were verified by sequencing and ligated in two opposite directions into the pMP10 RNAi vector plasmid (a derivative pLEW100; ref 44
). Restriction sites used were HindIII, XbaI and XhoI, and BamHI. The construct was linearized with NotI, ethanol precipitated, and transfected into the NYSM single marker cell line (44)
. This line is a derivative of bloodstream form MiTat 1.2(221) that expresses the VSG variant 221. The parental NYSM line was cultivated in HMI-9 medium (46)
supplemented with 10% FCS and containing 1 µg/ml G418. Selection of the RNAi constructs was done with 0.1 µg/ml puromycin. RNAi was induced with 1 µg/ml tetracycline.
In situ tagging of TbPDEB1 in the 427 procyclic strain
One allele of TbrPDEB1 was C-terminally tagged with a triple hemagglutinin (HA) tag, using our one-step PCR strategy (47)
. Transfectants were selected with 15 µg/ml G418, and clones were isolated by limiting dilution and were tested by Southern blotting for the correct integration of the tagging cassette and by Western blotting for the expression of the tagged protein. Clone 2CA5 was used for further experiments.
In situ tagging of TbPDEB2
To obtain a cell line in which one allele of each PDE is differentially tagged, clone 2CA5 was used to C terminally tag one allele of TbrPDEB2 with a triple c-Myc tag (47)
. The last 1327 bp of the open reading frame of TbrPDEB2 (without the stop codon; bp 1448–2775) and the first 1006 bp of the 3'UTR were amplified, using the following primers (KpnI, XhoI, BamHI, and XbaI sites underlined): B2-tag-ORF-f (ATGGTACCTCAACCGTGAGGTTGACAAACA); B2-tag-ORF-r (TACTCGAGAGACGAAGCCCCTGTACTCC); B2-tag-3'-f (ATGGATCCGCGGAAGGTTAATAATCGTGAG); and B2-tag-3'-r (ATTCTAGACAGCAGCGCCTATAAAGGCCA).
The PCR products were finally ligated into the vector pMOTag43M. The final construct was digested with KpnI and NotI, ethanol precipitated, and transformed into 2CA5 cell line. Transfectants were selected with 25 µg/ml hygromycin and were verified by Southern blotting and expression analysis. One of the doubly-tagged clones, 2CA5(2B-4M)1, was used for further analysis.
Triton-X100 fractionation
Trypanosomes (1x108) were washed once in PBS and lysed on ice for 5 min in PBS + 0.5% Triton-X100 supplemented with protease inhibitor (Roche complete mini, EDTA-free). The cell lysate was centrifuged 10 min at 13,000 rpm (4°C), and the supernatant and pellet fractions were analyzed by Western blotting.
Immunofluorescence microscopy
Immunofluorescence microscopy was done as described previously (47)
. The following primary antibodies were used: rabbit anti-Bip (staining of the endoplasmic reticulum; gift of Jay Bangs, University of Wisconsin, Madison; dilution 1:1000); rabbit antialdolase (visualization of glycosomes; gift of Paul Michels, University of Louvain, Belgium; dilution 1:1000); rabbit anti-TbV-H+-PPase (visualization of acidocalcisomes; gift of Norbert Bakalara, University of Bordeaux II, France; dilution 1:500); rat anti-PFR Ab (42)
; dilution 1:1000), preabsorbed rabbit anti-hemagglutinin Y-11 (Santa Cruz Biotechnology, Santa Cruz, CA, USA; dilution 1:250; preabsorbed as described previously (47)
; and mouse antic-Myc 9E10 (Santa Cruz; dilution 1:200). Secondary antibodies were Alexa Fluor 488 or 594 conjugated goat anti-rabbit, anti-mouse, or anti-rat polyclonal antibodies (Molecular Probes, Eugene, OR; highly cross-absorbed, dilution 1:750). 4',6'-diamidino-2-phenylidole (DAPI) staining was done with Vectashield mounting medium with DAPI (Vector Laboratories, Burlingame, CA, USA).
For immunfluorescence microscopy of cytoskeletons, cells were extracted once with cold MME + 0.5% Triton X-100 and once with MME + 0.1% Triton X-100 (48)
before fixation.
Immunoelectron microscopy
For immunogoldlabeling and electron microscopy (EM), parasites were Triton-extracted in suspension as for immunofluorescence and were fixed in PBS containing 3% paraformaledehyde for 30 min at room temperature. During fixation, cytoskeletons were allowed to adsorb to formvar-carbon-coated nickel grids. After three washes in PBS, nonspecific binding sites were blocked in PBS containing 1%BSA and 50 mM glycine for 1 h and were incubated with rabbit anti-HA antibodies at a dilution of 1:100 for 1 h at room temperature. Specimens were washed three times for 5 min in PBS and were subsequently incubated with colloidal gold (10 nm) conjugated goat-anti-rabbit-IgG (Amersham Biosciences, Otelfingen, Switzerland) at a dilution of 1:6 for 1 h. The grids were then washed five times for 5 min in PBS, and samples were postfixed in 1% glutaraldehyde in PBS and air dried. They were finally contrasted in 1% uranyle acetate for 1 min and air dried. Specimens were viewed on a Philips EM400 transmission electron microscope operating at 80 kV.
Intracellular cAMP concentration determination
The intracellular cAMP concentrations were determined using the cyclic AMP PLUS kit (format A) enzyme immunoassay (EIA) of Biomol International (Plymouth Meeting, PA, USA; Cat Nr AK-215) according to the instructions of the supplier.
Infection experiments
Young adult female NMRI mice (Charles River Laboratories, Wilmington, MA, USA) were infected intraperitoneally with 1 x 105 cultured trypanosomes. Parasitemia was monitored daily by collecting 1 µl of tail blood into 100 µl of 0.85% NH4Cl, 10 mM TrisHCl, pH 7.5 on ice, followed by counting in a Neubauer chamber. For transfering of trypanosomes from blood into cultures, the tail tips were thoroughly cleaned with 70% ethanol, and a 5 µl drop of blood was collected with a sterile pipette tip and transferred directly into 1 ml of growth medium. Cells were daily diluted 1:2 with fresh growth medium until the culture was fully adapted to the medium.
Determination of the toxicity of cAMP analogs
Stock solutions of cAMP analogues were prepared by dissolving 8-bromo-cAMP (Cat No 203800) and 8-CPT-cAMP (Cat No 116812; Calbiochem, San Diego, CA, USA) in DMSO to 400 mM. AlamarBlue assays of drug dilutions were performed exactly as described previously (34)
.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The correct integration of the tagging construct was verified by Southern blot analysis of genomic DNA, and the expression of both tagged proteins was confirmed by Western blotting (Fig. 1B
). Immunofluorescence imaging demonstrated that despite their structural similarity, the two PDEs exhibit distinct subcellular localizations (Fig. 1C
). TbrPDEB1 is located predominantly along the flagellum. A closer inspection shows that it is present in the flagellum only once the latter exits the cell body. This localization immediately suggested an association of TbrPDEB1 with the PFR of the flagellum (49
, 50)
. In contrast, most of TbrPDEB2 is found in dot-like structures that are distributed throughout the cytoplasm. However, a minor fraction of TbrPDEB2 colocalizes with TbrPDEB1 in the flagellum (Fig. 1C
).
Extraction of trypanosomes with nonionic detergents such as Triton X-100 results in detergent-resistant cytoskeletons that completely maintain the cellular architecture, consisting of the submembraneous cage of microtubules, the microtubular axoneme of the flagellum, and the PFR (48)
. Triton X-100 extraction of trypanosomes expressing the tagged PDEs demonstrated that TbrPDEB1 is fully resistant to detergent extraction (Fig. 2
A). Under the same conditions, the PFR structure stays intact, as demonstrated by the observation that the structural protein PFR2 is not solubilized by Triton X-100 (Fig. 2A
). TbrPDEB1 becomes solubilized only in the presence of stronger detergents (e.g., 0.1% SDS), conditions under which also the PFR structure starts to disintegrate. In contrast to TbrPDEB1, and in agreement with the findings from immunofluorescence microscopy, TbrPDEB2 is mostly but not completely solubilized by the Triton X-100 extraction. A minor but significant proportion of TbrPDEB2 remains associated with the Triton-insoluble cytoskeleton (Fig. 2A
).
|
Immunofluorescence microscopy of the Triton-extracted cytoskeletons demonstrated that both remaining PDEs colocalize along the flagellum (Fig. 2B
). In contrast to intact cells, no TbrPDEB2 was visible within the cell body of Triton-extracted cytoskeletons. This is compatible with the view that the cytoplasmic TbrPDEB2 is extractable by Triton X-100, whereas TbrPDEB2 that is located in the flagellum is similarly resistant to the detergent as is TbrPDEB1, with which it colocalizes.
Immunoelectron microscopy finally served to confirm the localization of TbrPDEB1 within, or in close association with, the PFR structure (Fig. 2C
). Colocalization of TbrPDEB1 with the PFR structure is in good agreement with the observations from immunofluorescence microscopy (see Fig. 1C
) that TbrPDEB1 staining along the flagellum is observed only from the point where the flagellum exits the cell body (50)
.
To exclude the possibility that the tags used in our experiments interfere with correct localization and function of TbrPDEB1 and TbrPDEB2, cell lines were constructed where either of the two genes were C-terminal tagged with yellow fluorescent protein (YFP), followed by a triple HA tag. Triton X-100 fractionation and immunofluorescence microscopy studies showed that the proteins carrying this much larger tag show an identical subcellular localization as was observed with the small triple HA or triple c-Myc tags. The subcellular localization of untagged TbrPDEB1 was also verified by using a TbrPDEB1-specific polyclonal antibody (data not shown). Immunoprecipitated TbrPDEB2 carrying a triple c-Myc tag shows strong activity in PDE assays, suggesting that the protein activity is not altered due to the C-terminal tags (L. Wentzinger, personal communication).
Association of TbrPDEB1 and TbrPDEB2 with the PFR structure
The potential association of TbrPDEB1 and TbrPDEB2 with the PFR structure of the flagellum was further explored using the procyclic snl-2 mutant strain of T. brucei. This strain contains an inducible RNAi construct that is targeted against one of the major PFR structural proteins (PFR2; ref 51
). On induction of RNAi, snl-2 cells remain fully viable and proliferation competent, but their flagella lack a PFR structure and are paralyzed. Proteins destined to become incorporated into the PFR structure accumulate in a droplet-like structure at the tip of the flagellum (52)
. TbrPDEB1 and TbrPDEB2 were C-terminally tagged (triple HA tag) in the snl-2 strain, and their intracellular distribution was observed in the presence or absence of tetracycline-induced RNAi against the PFR structure. As shown in Fig. 3
A (panels 1 and 3), TbrPDEB1 locates in the flagellum of snl-2 just as it does in wild-type (WT) cells. On induction of RNAi against the PFR structure, TbrPDEB1 accumulates in the droplet-like structure at the tip of the flagellum, as expected for integral PFR proteins (Fig. 3B
, panels 4 and 6). TbrPDEB2 behaves very similarly as far as its flagellum-associated subpopulation is concerned. Induction of RNAi against the PFR structure causes the flagellar TbrPDEB2 to accumulate in the distal droplet-like structure (Fig. 3B
) in a manner identical to what happens with TbrPDEB1. In contrast, the distribution of the cytoplasmic TbrPDEB2 subpopulation remains entirely unaffected by RNAi (Fig. 3B
). Control stainings with anti-Bip Ab that visualizes the endoplasmic reticulum further demonstrate that the induction of RNAi against PFR2 does not affect the distribution of proteins in the cytoplasmic compartment (Fig. 3A, B
, panels 2 und 5).
|
Procyclic trypanosomes are not sensitive to RNAi against TbrPDEB1 and TbrPDEB2
With the subcellular localization of TbrPDEB1 and TbrPDEB2 already established, tetracycline-inducible RNAi was applied to study the effect of ablating these enzymes. The mRNAs of both PDEs were first targeted in the procyclic forms by appropriate RNAi constructs. The two mRNAs were either targeted individually, or they were targeted jointly by a double-RNAi construct that contained a conserved nucleotide sequence present in both mRNAs. For all RNAi experiments, a stem-loop type of RNAi vector was used where RNA synthesis is driven by a single procyclin promotor that is under the control of a tetracycline repressor. In our hands, these constructs proved to be much more tightly controlled than the constructs with opposing T7 promotors used in earlier RNAi experiments (34)
. The effectiveness of RNAi-mediated reduction of target mRNA levels was rather variable between individual clones of a given construct. Clones exhibiting strong RNAi effects were obtained for all constructs and were selected for further analyses.
Northern blot analysis of the resulting strains grown in the presence or absence of 1 µg/ml tetracycline demonstrated that the RNAi constructs were functional, that the level of target mRNA was strongly reduced, and that the RNAi constructs were indeed target-specific (Fig. 4
A, B). The data also demonstrate that TbrPDEB1 and TbrPDEB2 do not mutually regulate their mRNA levels since the ablation of TbrPDEB1 does not affect the mRNA level of TbrPDEB2 and vice versa (Fig. 4A, B
). Induction of the double-RNAi construct leads to the reduction of the steady-state levels of both target mRNAs (Fig. 4C
). This reduction does not lead to a compensatory up-regulation of the mRNA levels of the other trypanosomal PDEs (32
; Fig. 4C
).
|
None of the RNAi clones, including the ones that carry the double-RNAi construct, showed any discernable phenotype, neither on the level of individual cell morphology, nor at the level of population growth. Their generation times are identical both in the presence or absence of tetracycline, and they are very similar to that of the parental NYDM strain (Fig. 4D
). Taken together with the results from Northern blotting, these results indicate that proliferation of procyclic forms is not affected by the reduction of mRNAs for TbrPDEB1 and TbrPDEB2.
TbrPDEB1 and TbrPDEB2 are essential enzymes in the bloodstream form
Previous experiments from this laboratory had indicated that ablation of TbrPDEB1 and TbrPDEB2 might be lethal for bloodstream form trypanosomes (34)
. These experiments were difficult to interpret since the cells could not be cultivated and no analysis of vector integration or mRNA levels could be done. To reconfirm these earlier findings, bloodstream forms were transfected with a newer RNAi vector the expression of which can be tightly controlled. Clones were selected that carry RNAi constructs directed either against TbrPDEB1 or TbrPDEB2 individually or that confer double-RNAi against both PDEs simultaneously. When clones carrying the individual RNAi constructs were induced with tetracycline for 18 h, the levels of the respective target mRNAs were strongly reduced (Figs. 5A and B
). As observed in procyclics, the ablation of TbrPDEB1 does not affect the mRNA level of TbrPDEB2, and vice versa (Fig. 5A, B
). Both strains showed no discernible phenotype on induction of RNAi, neither at the level of cellular morphology nor in terms of population growth (data not shown).
|
When bloodstream clones were induced that carry the double-RNAi construct, the levels of both TbrPDEB1 and TbrPDEB2 mRNAs were strongly reduced. At the same time, the mRNA levels of the other three PDEs remained unchanged (Fig. 5
C).
In striking contrast to the situation seen with the individual RNAi constructs or seen in procyclic forms, induction of the double-RNAi construct produced a dramatic phenotype. The cells start rounding up (Fig. 5E
), become multinuclear, and contain numerous kinetoplasts and multiple flagella (Fig. 5F
). The ratio between nuclei and kinetoplasts is maintained at approximately one, suggesting that the cell division cycle proceeds in an orderly manner but that cytokinesis is inhibited. The resulting giant cells still contain an intact endoplasmic reticulum and well-defined acidocalcisomes and glycosomes (Fig. 5E
). Nevertheless, they no longer proliferate, and all will eventually lyse (Fig. 5D
) after 50 h after induction of RNAi.
Ablation of TbrPDEB1 and TbrPDEB2 raises intracellular cAMP both in procyclic and in bloodstream forms
cAMP levels were determined in cell extracts from procyclic and from bloodstream clones expressing the double-RNAi construct. Uninduced cells, as well as the parental strains in the presence or absence of tetracycline, exhibited the expected low intracellular cAMP concentrations (0.1–0.4 pmol/3x106 cells). Induction of RNAi by tetracycline for 48 h (procyclics) or 18 h (bloodstream forms) resulted in a dramatic increase in intracellular cAMP in both cell types in good agreement with our earlier findings (34)
. In procyclics, the cAMP level reached a maximum of 4 picomol/3 x 106 cells after 48 h of RNAi induction, i.e., an 10-fold increase over resting cAMP levels. In bloodstream forms, induction of the double-RNAi construct resulted in cAMP levels of 50 picomol/3 x 106 cells after 18 h induction, an 100-fold increase over resting levels. Induction of RNAi against either of the two PDEs individually did not measurably change cAMP levels.
The marked difference between the cAMP levels in procyclic and in bloodstream forms was not unexpected, since bloodstream form trypanosomes exhibit a much higher cAMP flux than do procyclic forms (Marti and Seebeck, unpublished observations). The results obtained from these experiments indicate that the difference in sensitivity of procyclic and bloodstream forms toward RNAi-mediated ablation of TbrPDEB1 and TbrPDEB2 reflects a higher tolerance of procyclic forms against sustained high cAMP levels. This is further supported by the finding that the membrane-permeable cAMP analogues 8-CPT-cAMP and 8-Br-cAMP are toxic to bloodstream forms (IC50 of 1.2x10–6±4.5x10–7 M (n=3) and 3.7x10–5 M, respectively), whereas procyclic forms are not affected at concentrations up to at least 1 mM (data not shown).
TbrPDE1 and TbrPDEB2 are essential for establishing an infection in vivo
After it was demonstrated that TbrPDEB1 and TbrPDEB2 are essential enzymes for bloodstream trypanosomes in culture, their role during in vivo infections was explored. Mice were infected with the RNAi strains directed separately against TbrPDEB1 or TbrPDEB2 and with the double-RNAi strain. Control infections were done with the parental NYSM strain and with three trypanosome strains containing RNAi constructs against irrelevant mRNAs. All strains were similarly virulent, reaching parasitemias of >108/ml by 3–4 days after infection (Fig. 6
A). When animals received 1 mg/ml doxycyline in the drinking water from 2–3 days before infection, the course of infection with the parental NYSM strain and the RNAi control strains was similar to that in untreated control animals. The same lack of an infectivity phenotype was also observed with the RNAi strains directed individually against TbrPDEB1 or TbrPDEB2, in agreement with what had been observed in culture. In striking contrast, animals infected with the double-RNAi strain remained entirely trypanosome-free (3 separate experiments with groups of 3–5 animals each, one representative experiment is given in Fig. 6A
) over the entire 5 day period in which the infection took its course in all controls. When such aparasitemic animals were transferred to drinking water without doxycycline after 5 days, no parasites could be detected for the entire follow-up period of 35 days (data not shown).
|
The double-RNAi strain was also unable to infect animals in the presence of doxyxcycline when the animals had been immunocompromised by pretreatment with either cyclophosphamide (a single dose of 200 mg/kg ip 6 h before infection) or rolipram (10 mg/kg ip every 2nd day from day 2 before infection) to suppress the immune system of the host (Fig. 6B
). Rolipram is an inhibitor of human PDE4 that elicits strong immunosuppression but does not inhibit TbrPDEB1 or TbrPDEB2 (34
, 35)
and does not affect trypanosome proliferation in culture.
TbrPDE1 and TbrPDEB2 are essential enzymes for maintaining an infection
To explore if induction of RNAi was also sufficient to clear an ongoing infection, animals infected with the double-RNAi strain were transferred to doxycyline-containing drinking water only after parasitemia had reached high levels. For animals with near-fatal levels of parasitemia (>108 trypanosomes/ml), the onset of RNAi was too slow and the animals succumbed. However, in animals with parasitemias of <108/ml, trypanosomes were cleared within 24 h, and the animals remained trypanosome-free for several days (Fig. 6C
). Eventually, parasitemaia relapsed in the majority of the animals, despite the continuous presence of doxyxcycline. When trypanosomes were collected and cultured from such relapsing animals, they proved to be no longer sensitive to tetracycline in culture (data not shown). These findings indicate that the relapse is caused by trypanosomes that had acquired spontaneous mutations to tetracycline resistance and hence were no longer able to maintain RNAi in the presence of the inducer.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 33)
are coded for by two tandemly arranged, closely linked genes on chromosome 9. Both genes and their products have been described earlier (TbrPDEB1, formerly TbPDE2C; ref 34
), GenBank accession number AY028446; and TbrPDEB2, formerly TbPDE2B (35)
, GenBank accession number AF192755). Subsequent sequence analyses have shown that the latter sequence might contain an assembly error, so that the following reference sequences from the T. brucei database, release 4, were used for this work: TbrPDEB1: Tb09.160.3590, and TbrPDEB2: Tb09.160.3630. Genes homologous to TbrPDEB1 and TbrPDEB2 are also found in syntenic arrangements and in the genomes of T. cruzi, T. vivax, and L. infantum. They are also present in the genome of L. major (37)
although the current version of the database (release Jan 11, 2006) still contains an assembly error that masks the presence of one of the genes. Similar assembly problems may also exist in the current versions of the T. congolense and L. braziliensis genome databases.
In situ C-terminal tagging of TbrPDEB1 and TbrPDEB2 allowed the intracellular localization of the respective gene products. Despite the overall similarity of the two PDEs, their subcellular localization is quite distinct. TbrPDEB1 is strictly confined to the flagellum. Several lines of evidence have established that it forms an integral part of the PFR structure, from which it only can be dissociated by conditions that also disrupt the PFR structure. In contrast, a large proportion of TbrPDEB2 is localized in the cytoplasm and can be extracted with Triton X-100, while a minor part colocalizes with TbrPDEB1 in the flagellum. These data confirm a recent analysis of the PFR proteome that demonstrated the presence of both PDEs in the high-salt extracted core structure of the PFR (53)
. They are also in excellent agreement with a study describing the presence of a T. cruzi homologue of TbrPDEB2 in the flagellum of these organisms (36)
.
The finding that the two PDEs are located in the flagellum and are tightly associated with the PFR adds to an emerging picture of the trypanosome flagellum as an important site of cAMP signaling (Oberholzer et al., unpublished observations). Besides the two PDEs, adenylyl cyclases were also shown to be located in the flagellum and to be associated with the PFR (36
, 54
; Bregy and Seebeck, unpublished observations), and the PFR proteome contains several putative cAMP-binding proteins (53; Oberholzer et al., unpublished observations).
RNAi against TbrPDEB1 and TbrPDEB2 is lethal for bloodstream forms but produces no discernible phenotype in procyclic forms
Earlier experiments in this laboratory demonstrated that the PDEs are essential enzymes in bloodstream forms. RNAi in bloodstream forms was not possible due to the marked leakiness of the vector construct employed at the time. RNAi in procylic forms produced clones with often distorted cell morphology but without a dramatic population growth phenotype. Intracellular cAMP in these clones was strongly elevated (34)
. In the present study, we have reconfirmed and extended these early observations by using a much more tightly controllable RNAi system, which is crucial when analyzing bloodstream form trypanosomes. In overall agreement with the earlier findings, RNAi against both PDEs resulted in a lethal phenotype in bloodstream forms and produced essentially no phenotype in procyclics. In view of the specific and distinct subcellular locations of TbrPDEB1 and TbrPDEB2, the latter observation was unexpected. Northern blotting experiments demonstrated that the reduction of TbrPDEB1 and TbrPDEB2 mRNAs did not result in a compensatory up-regulation of other PDE mRNAs. This finding is further supported by the observation that steady-state intracellular cAMP levels are increased 10-fold on double-RNAi induction. The lack of a phenotype in procyclic forms may suggest that procyclic trypanosomes are inherently less sensitive to elevated cAMP levels than are bloodstream forms. This has already been suggested in our earlier work where the cAMP level in procyclic formed after RNAi was even higher than in the present study, but the effect on cell proliferation was only slight, although an increase in odd-shaped cells was observed (34)
. These observations are also in line with data showing that the proliferation of procyclic forms is not affected by up to 1 mM of membrane-permeable cAMP analogs.
In striking contrast to the situation in procyclics, bloodstream from trypanosomes is exquisitely sensitive to simultaneous RNAi against TbrPDEB1 and TbrPDEB2. Interference with each PDE individually produced no phenotype, suggesting that the two enzymes can compensate for each other, despite their different subcellular localizations. This is also supported by the observation that RNAi against the individual PDEs does not measurably change the intracellular cAMP steady-state levels. In contrast, targeting both PDEs simultaneously leads to cell swelling, multinuclearity, and cell lysis within 50 h. During this period, cells undergo several rounds of apparently normal nuclear and kinetoplast duplication, and they produce several flagella. Intracellular cAMP levels raise 100-fold over the resting level within 18 h of RNAi induction. All evidence currently available suggests that elevated cAMP blocks cytokinesis in bloodstream form trypanosomes.
Two aspects of the markedly different phenotypes exhibited by bloodstream forms and procyclics are noteworthy. Induction of double-RNAi induces an increase in "steady-state" cAMP levels in both forms. However, these levels are consistently (over a large number of different experiments) 10-fold lower in procyclics. This difference is not only seen on RNAi induction against the PDEs but also when these are inhibited by highly specific inhibitors (unpublished observations). These findings, combined with our observation that at least some adenylyl cyclases are expressed at much higher levels in bloodstream forms (Bregy and Seebeck, unpublished observations), indicate that the cAMP flux is much higher in bloodstream forms than it is in procyclics, despite the very similar resting steady-state cAMP levels in the two forms. In addition to a less elevated level of intracellular cAMP on RNAi, procyclic forms might be inherently less sensitive to excess cAMP. This view is supported by the observation that procyclics are oblivious to high concentrations (at least up to 1 mM) of various cell-permeable cAMP derivatives. Although we cannot at present rule out other causes for this observed cAMP resistance, the most straightforward interpretation would be an inherent insensitivity to high cAMP levels.
Prevention/elimination of parasitemia
The sensitivity of cultured bloodstream form trypanosomes against the inactivation of TbrPDEB1 and TbrPDEB2 highlights their potential as candidate targets for trypanocidal PDE inhibitors. Infection studies with the RNAi strains demonstrated that in vivo induction of RNAi by the orally applicable tetracycline analog doxycycline effectively prevents infection and is also able to quench an ongoing infection. This effect is independent of the immunocompetence of the animal, suggesting that ablation of the two PDE activities is sufficient to eliminate the parasite. These findings clearly validate the TbrPDEB1 and TbrPDEB2 as potential targets for the development of a new generation of trypanocidal PDE inhibitors.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication July 26, 2006. Accepted for publication October 11, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
-ß tubulin intergenic region; G418, Hyg: antibiotic resistance to G418 and hygromycin, respectively). B) Control of expression of tagged PDEs by immunoblotting. M: MW markers; 1: WT cells; 2: strain 2CA5, TbrPDEB1 tagged with triple HA; 3: strain 2CA5(2B-4M)1, TbrPDEB1 tagged with triple HA and TbrPDEB2 tagged with triple c-Myc. Bottom panel: control staining with Ab against a major structural protein of the PFR. Each lane corresponds to 8 x 106 cells. C) Immunofluorescence detection of tagged PDEs. TbrPDEB1: green; TbrPDEB2: red. Nuclei and kinetoplasts visualized by DAPI staining (blue). D) Control staining of WT cells.
