Tài liệu Báo cáo khoa học: Identification and characterization of an R-Smad ortholog (SmSmad1B) from Schistosoma mansoni pdf

As determined by yeast three-hybrid assays and pull-down assays, the presence of the wild-type or mutated SmTbRI receptor resulted in a decreased interaction between SmSmad1B and SmSmad4

(SmSmad1B) from Schistosoma mansoni

Joelle M Carlo1*, Ahmed Osman1,2*, Edward G Niles1, Wenjie Wu2, Marcelo R Fantappie2,

Francisco M B Oliveira2and Philip T LoVerde1,2

1 Department of Microbiology and Immunology, School of Medicine and Biomedical Sciences, State University of New York, NY, USA

2 Southwest Foundation for Biomedical Research, San Antonio, TX, USA

The multicellular, dioecious parasite Schistosoma

mansoni has a complex life cycle consisting of both

free-living and host-dependent stages The signaling

mechanisms underlying the growth and development

of S mansoni during these stages have remained

largely undefined In the human host, S mansoni para-sites develop from schistosomules to adults, and can survive in the host mesenteric circulation for years The implication that host molecules may be exploited

by schistosomes to enhance the parasites’ development

Keywords

bone morphogenic protein; Schistosoma

mansoni; Smad; transforming growth

factor-b

Correspondence

P T LoVerde, South-west Foundation for

Biomedical Research, PO Box 7620, NW

Loop 410, San Antonio, TX 78227-5301,

USA

Fax: +1 210 670 3322

Tel: +1 216 258 5892

E-mail: ploverde@sfbr.org

Database

The nucleotide sequence described here is

available in the GenBank database under the

accession number AY666164

*These authors contributed equally to this

work

(Received 14 February 2007, revised 5 June

2007, accepted 11 June 2007)

doi:10.1111/j.1742-4658.2007.05930.x

Smad proteins are the cellular mediators of the transforming growth fac-tor-b superfamily signals Herein, we describe the isolation of a fourth Smad gene from the helminth Schistosoma mansoni, a receptor-regulated Smad (R-Smad) gene termed SmSmad1B The SmSmad1B protein is com-posed of 380 amino acids, and contains conserved MH1 and MH2 domains separated by a short 42 amino acid linker region The SmSmad1B gene (> 10.7 kb) is composed of five exons separated by four introns On the basis of phylogenetic analysis, SmSmad1B demonstrates homology to Smad proteins involved in the bone morphogenetic protein pathway Sm-Smad1B transcript is expressed in all stages of schistosome development, and exhibits the highest expression level in the cercariae stage By immuno-localization experiments, the SmSmad1B protein was detected in the cells

of the parenchyma of adult schistosomes as well as in female reproductive tissues Yeast two-hybrid experiments revealed an interaction between Sm-Smad1B and the common Smad, SmSmad4 As determined by yeast three-hybrid assays and pull-down assays, the presence of the wild-type or mutated SmTbRI receptor resulted in a decreased interaction between SmSmad1B and SmSmad4 These results suggest the presence of a non-functional interaction between SmSmad1B and SmTbRI that does not give rise to the phosphorylation and the release of SmSmad1B to form a het-erodimer with SmSmad4 SmSmad1B, as well as the schistosome bone morphogenetic protein-related Smad SmSmad1 and the transforming growth factor-b-related SmSmad2, interacted with the schistosome coacti-vator proteins SmGCN5 and SmCBP1 in pull-down assays In all, these data suggest the involvement of SmSmad1B in critical biological processes such as schistosome reproductive development

Abbreviations

AP-1, activator protein-1; 3-AT, 3-amino-1,2,4-triazole; BAC, bacterial artificial chromosome; b-gal, b-galactosidase; BMP, bone morphogenetic protein; Co-Smad, common Smad; DPE, downstream promoter element; ERK, extracellular signal-regulated kinase; EST, expressed sequence tag; Gal4AD, Gal4 activation domain; Gal4BD, Gal4 DNA-binding domain; GST, glutathione S-transferase; MBP, maltose-binding protein; MH, Mad homology domain; R-Smad, receptor-regulated Smad; SmGCP, schistosome gynecophoral canal protein; TGFb,

transforming growth factor-b.

and ultimate survival within the host has prompted the

need for better characterization of schistosome

signa-ling networks [1] In recent years, several members of

the transforming growth factor-b (TGFb) superfamily

have been isolated from S mansoni [2–7] The

involve-ment of the TGFb superfamily in critical cellular

pro-cesses such as embryogenesis, differentiation and

apoptosis makes these pathways attractive candidates

for elucidating the growth and development

mecha-nisms employed by S mansoni

The TGFb superfamily comprises a large group of

structurally related, secreted cytokines, including

TGFb, bone morphogenetic protein (BMP), and

acti-vin [8–11] The TGFb signaling cascade is stimulated

through the binding of ligand to a type II receptor, a

transmembranous serine⁄ threonine receptor kinase

The ligand–type II receptor complex recruits another

serine⁄ threonine transmembrane receptor kinase, type I

receptor, which is subsequently phosphorylated and

activated by the type II receptor The activated type I

receptor then interacts with a group of cellular

media-tors called receptor-regulated Smads (R-Smads) The

R-Smads of the TGFb–activin pathway include Smad2

and Smad3 The R-Smads of the BMP pathway

include Smads 1, 5 and 8 The type I receptor

phos-phorylates the R-Smad at its C-terminal MH2 domain,

causing the R-Smad to dissociate from the receptor

The phosphorylation⁄ activation of the R-Smad allows

it to interact with another component of the Smad

family, called a common Smad (Co-Smad) The Smad

complex translocates to the nucleus, where, in concert

with other proteins, it modulates the transcription of

TGFb-responsive genes

In S mansoni, the first TGFb superfamily member

identified was a type I receptor named SmRK1 (later

referred to as SmTbRI) [3] SmTbRI was found to

be expressed in the schistosome tegument, whereas

RT-PCR analysis demonstrated the upregulation of

SmTbRI transcript in S mansoni during stages of

mammalian infection [3] These results, along with

the reported binding of human TGFb to a chimeric

form of SmTBRI [12], followed by the later finding

of the induced association of SmTbRI with the

schistosome type II receptor SmTbRII by human

TGFb [7], suggested a role for TGFb signaling in

host–parasite interactions Two R-Smad genes

(Sm-Smad1 and SmSmad2) and a Co-Smad gene

(Sm-Smad4) were also identified from S mansoni [2,5,6]

It was determined that SmSmad2 acts as a substrate

for receptor activation by SmTbRI, whereas the

acti-vation of SmSmad1 by SmTbRI has not been

dem-onstrated As SmSmad1 resembles R-Smads of the

BMP pathway, it was suggested that both BMP and

TGFb signaling networks may be active in schisto-somes, and that a second type I receptor capable of transmitting BMP-related signals may be present in the genome of S mansoni

Herein, we report the isolation of a new member

of the S mansoni R-Smad family, designated Sm-Smad1B Like SmSmad1, SmSmad1B demonstrates homology to BMP-related R-Smad genes In this study, we report the identification of SmSmad1B cDNA and present its gene structure along with the expression profiles, immunolocalization, and protein interaction properties

Results

Identification of SmSmad1B Through cDNA library screening with the putative SmSmad8⁄ 9 expressed sequence tag (EST) as a probe,

a SmSmad1B cDNA clone was isolated that contained the entire coding region and 3¢-UTR, as well as a par-tial 5¢-UTR sequence (68 bp) (Fig 1A) The 3¢-UTR sequence was determined to be complete by the pres-ence of a polyA tail with the AAUAAA consensus polyadenylation signal [13] located 25 bp upstream of the polyA tail [13] The 5¢-UTR sequence was extended

to its full length by employing 5¢-RACE The organ-ization of the SmSmad1B cDNA comprises a 136 bp 5¢-UTR, a 1143 bp coding region, including a ‘TAA’ stop codon, and a 738 bp 3¢-UTR (Fig 1A)

SmSmad1B protein consists of 380 amino acid resi-dues, and contains all the typical and conserved motifs

of an R-Smad As no sequence or structural properties exist to differentiate the BMP Smads from each other,

a high degree of similarity among the BMP Smads is a common phenomenon SmSmad1B is one of the small-est R-Smads in terms of size, mainly due to the pres-ence of a short linker region comprising only 42 amino acids The N-terminal MH1 domain of SmSmad1B consists of 137 amino acids, and comprises a conserved nuclear localization signal and a b-hairpin structure that serves as a DNA-binding domain (Fig 1A) The

201 amino acid MH2 domain of SmSmad1B contains

a typical C-terminal SSVS phosphorylation motif, which is the site of phosphorylation by type I receptors

of the TGFb superfamily (Fig 1A) The presence of the SSVS phosphorylation site identifies SmSmad1B as

an R-Smad, as this motif is absent in both inhibitory and Smads and Co-Smads [9] Importantly, the L3 loop in the MH2 domain of SmSmad1B resembles that of R-Smads in being specific for transducing BMP signals (i.e amino acid residues H340 and D343) (Fig 1A)

Phylogenetic analysis

Phylogenetic trees were constructed by Bayesian

infer-ence using a mixed protein substitution model with an

inv-gamma distribution of rates between sites using

mrbayesv3.1.1 (Fig 2) Phylogenetic analyses of both the MH1 (Fig 2A) and MH2 sequences (Fig 2B) showed that SmSmad1B clustered within the BMP-related R-Smad group, which includes the homologous proteins Drosophila MAD and the vertebrate Smad1,

Fig 1 Structure of the SmSmad1B gene, cDNA and protein (A) Schematic representations of the SmSmad1B gene, cDNA and protein and the amino acid sequence of SmSmad1B protein Five exons interrupted by four introns constitute the SmSmad1B gene (top) A cDNA of about 2 kb in size is transcribed from the genomic gene (middle) and translated into a 380 amino acid SmSmad1B protein (bottom) Regions encoding MH1, linker and MH2 are in light gray, gray and dark gray, respectively, and the regions representing the 5¢- and 3¢-UTRs are shown in white in the genomic gene and the cDNA Intron size in bp, domain size in bp and domain size in amino acids are indicated at the bottom of each schematic representation of the gene, cDNA, and protein, respectively The schematic representation and the amino acid sequence of SmSmad1B protein show sequence motifs (black boxed) such as nuclear localization signal (NLS), DNA-binding b-hairpin domain (DBD) and the receptor-phosphorylation motif (Pi motif) as well as the amino acid sequence of the peptide region that was used to generate SmSmad1B-specific antibody reagents (Antibody peptide) The L3 loop is also shown (gray box), with R-Smad subtype-specific amino acids

in bold and underlined (B) The promoter region and the 5¢-UTR of the SmSmad1B gene The transcription start site within the Inr is designa-ted by a broken arrow A 50 bp intron that separates exons 1 and 2 is shown (italics, underlined lower-case letters) The promoter region is

in upper-case letters, and the exon sequences are presented in bold upper-case letters Some transcription regulatory elements are listed (boxed): Inr (initiator element); DPE; and AP-1 The underlined ATG is the codon for the translation start methionine.

Smad5 and Smad8 Furthermore, SmSmad1B was

clo-sely related to SmSmad1, another S mansoni Smad

protein previously isolated [2], and to the tapeworm

Echinococcus multilocularis Smad, EmSmadB [14] The phylogenetic data suggest that SmSmad1 and Sm-Smad1B are paralogous genes; they originated from

Fig 2 Bayesian phylogenetic tree of SmSmad1B The dataset was analyzed using a mixed substitution model with an inv-gamma distribu-tion of rates between sites using MRBAYES v3.1.1 The trees were started randomly; four simultaneous Markov chains were run for

3 · 10 6 generations The trees were sampled every 100 generations Bayesian posterior probabilities were calculated using a Markov chain Monte Carlo sampling approach implemented in MRBAYES v3.1.1, with a burn-in value setting at 7500 generations; the values are shown at each branch point (or by arrows) The results suggested that SmSmad1B and SmSmad1 are paralogous genes; they originated from duplica-tion of a common ancestor Smad gene after the split between platyhelminths, arthropods, and vertebrates (A) MH1 tree (B) MH2 tree The GenBank accession numbers of the analyzed sequences were as following: CeSma3 (Caenorhabditis elegans sma-3), U34902; Cem1 (Cae elegans MAD homolog 1), U10327; CeSma4 (Cae elegans SMA-4), U34596, DAD (Drosophila melanogaster DAD), AB004232; dMAD (D melanogaster MAD), U10328; dMedea (D melanogaster Medea), AF057162; dSmad2 (D melanogaster Smad2), AF101386; EmSmadB (E multilocularis SmadB), AJ548428; ckSmad8 (Gallus gallus SMAD8), AY953145; hSmad1 (Homo sapiens Smad1), U59423; hSmad2 (H sapiens Smad2), U65019; hSmad3 (H sapiens Smad3), U76622; hSmad4 (H sapiens Smad4), U44378; hSmad5 (H sapiens Smad5), U73825; hSmad6 (H sapiens Smad6), AF043640; hSmad7 (H sapiens Smad7), AF015261; mSmad2 (Mus musculus Smad2), U60530; rat-Smad8 (Rattus norvegicus rat-Smad8), AF012347; SmSmad1 (S mansoni Smad1), AF215933; SmSmad2 (S mansoni Smad2), AF232025; Sm-Smad4 (S mansoni Sm-Smad4), AY371484; xSmad1 (Xenopus laevis Mad1), L77888; xSmad2 (X laevis Mad2), L77885; xSmad3 (X laevis SMAD3), AJ311059; xSmad8 (X laevis Smad8), AF464927.

the duplication of a common ancestor gene after the

split between the platyhelminths, arthropods, and

ver-tebrates The same results were obtained by maximum

likelihood and neighbor-joining distance analyses

(sup-plementary Figs S1 and S2)

SmSmad1B gene structure and 5¢ upstream

analysis

The location of the exon–intron boundaries were

deter-mined by alignment of cDNA sequence with the

bacterial artificial chromosome (BAC) DNA sequence

(SmBAC1 40G14) The four exon–intron junctions

conform to the eukaryotic consensus GT-AG splice

sites (supplementary Table S1) [15] The locations of

the exon–intron junctions in SmSmad1B are shared by

Smad genes from other species For example, the

loca-tion of the intron within the MH1-encoding region is

conserved in the human Smad4 gene, whereas the

intron within the linker-encoding region of SmSmad1B

is shared by human Smad3 [16] The location of the

intron within the MH2-encoding region is highly

con-served among human Smad9, Smad5, and Smad3,

mouse Smad1, and the E multilocularis SmadB

[14,16,17] PCR amplification of the SmSmad1B

cDNA flanking the linker region did not produce

mul-tiple PCR products (data not shown), indicating the

absence of alternative splicing in this region

The beginning of exon 1 of SmSmad1B was

identi-fied by performing 5¢-RACE Three independent

rounds of 5¢-RACE produced 5¢-UTR fragments that

extended no further than 136 bp upstream from the

translation start site This location was determined to

be the position of the putative transcription start site

(Fig 1B) Analysis of the 5¢ upstream region of exon 1

demonstrated the lack of a conserved TATA box

upstream of the transcription initiation site However,

an AT-rich sequence with a single nucleotide mismatch

from the TATA box consensus is located at position

) 55 ⁄ ) 48 in the SmSmad1B promoter region (GATA

AAAG, as compared to the consensus TATAA⁄

TAAG⁄ A) [18] An initiator element (Inr) is located at

position) 2 ⁄ + 5 that conforms to the mammalian Inr

consensus rather than the Drosophila consensus Inr

(TCA+1AAAC) [19,20] From position + 24⁄ + 29, a

downstream promoter element (DPE; consensus A⁄ G ⁄

T-C⁄ G-A ⁄ T-C ⁄ T-A ⁄ C ⁄ G-C ⁄ T) is also located [21]

The DPE is known to act in conjunction with the Inr

in the initiation of transcription A potential AP-1

(activator protein-1) site is located at position

) 78 ⁄ ) 72 Interestingly, there are three core

Smad-binding elements (GTCT) [22] in the upstream region

(Fig 1B)

Developmental expression of SmSmad1B The expression level of SmSmad1B mRNA was evalu-ated by performing quantitative RT-PCR on cDNA prepared from total RNA isolated from various schistosome developmental stages (Fig 3) The expres-sion levels of SmSmad1B were compared to those of the related R-Smad gene, SmSmad1 The results dem-onstrate that SmSmad1B is expressed in all of the developmental stages examined The SmSmad1B expression pattern closely follows that of SmSmad1, both exhibiting the highest transcript levels in cercariae and lower levels in different developmental stages in the intermediate host, Biomphalaria glabrata snails On the other hand, expression levels show a significant drop in the stages representing different time points in the mammalian host, as early as 3 days postinfection, and there is a gradual decrease thereafter up to 21-day-old schistosomules, which represent the trough of the expression curves of both R-Smads The levels then display a slight increase, reaching maximum levels of expression in the mammalian host in paired adult worms In addition, it appears that the BMP-related Smads, SmSmad1 and SmSmad1B, exhibit relatively lower levels of expression as compared to the TGFb-related SmSmad2 in the late stages of infection (28 day, 35 day and adult worms; SmSmad2 data not shown) Both the SmSmad1 and SmSmad2 expression

Fig 3 Quantitative RT-PCR analysis of schistosome BMP-related R-Smad genes A bar graph comparing the fold expression levels (mean ± SD) of SmSmad1 (black) and SmSmad1B (gray) normal-ized to the levels of Sma-tubulin throughout various stages of schistosome development The following developmental stages were tested: infected Biomphalaria glabrata snails representing daughter sporocysts (inf snail), cercariae, 3-day-old and 7-day-old cultured schistosomules, 15 day, 21 day, 28 day and 35 day para-sites, adult worm pairs, separated adult female and male worms, and eggs cDNA from uninfected B glabrata snails served as a neg-ative control.

levels are consistent with our previously reported

results that showed that SmSmad2 expression levels

exceed SmSmad1 levels by approximately 45% in the

later stages (35 days or older) of mammalian

develop-ment [6]

Identification and immunolocalization of

SmSmad1B protein in adult schistosomes

To detect the native SmSmad1B protein, western blots

were performed using S mansoni adult worm pair

pro-tein extracts (whole and soluble) and the

affinity-puri-fied aSmSmad1B antibody A band was detected

migrating at approximately 53 kDa in the S mansoni

protein extracts when they were probed with the

aSmSmad1B antibody (Fig 4, right panel), which is

higher than the calculated molecular mass of

Sm-Smad1B (43 kDa) The band could not be detected in

the S mansoni protein extracts when they were probed

with a preimmune rabbit IgG antibody (Fig 4A,

mid-dle panel) Furthermore, preincubation of aSmSmad1B

antibody with varied amounts of SmSmad1B linker

peptide resulted in a gradual decrease in intensity of

the 53 kDa band until the native protein was no longer

visualized in the presence of 10 lgÆmL)1 of the peptide

(Fig 4B, right panel, lane 3), indicating that the

53 kDa band is specific The in vitro translation

prod-uct of SmSmad1B runs at approximately 50 kDa (data

not shown) That difference in size could be attributed

to post-translational modifications that occur to the

native protein, such as specific phosphorylation by

type I receptor [23,24], or N-acetylation by p300, CBP,

or P⁄ CAF [25,26] Such modifications may not be seen

in the in vitro translated product

Immunofluorescent staining was performed to

local-ize the expressed SmSmad1B protein in adult

schisto-somes Adult worm cryosections were probed with

affinity-purified aSmSmad1B antibody, and the specific

fluorescence was visualized at 680 nm In female adult

worms, SmSmad1B was prominent in the vitellaria as

well as in the reproductive ducts and subtegumental

tissues (Fig 5) In male adult worms, specific fluores-cence was also visualized in the subtegument but not

in tissues of the reproductive system Rather, a tissue

of undefined origin in the male worms demonstrated consistent, specific SmSmad1B fluorescence The signal was located in the parenchyma within the worm cen-ter, and spanned the entire length of the male worm (Fig 5B)

SmSmad1B protein interactions

To investigate the interaction between SmSmad1B and schistosome TGFb superfamily members, yeast two-hybrid assays were performed When Y190 yeast competent cells were cotransformed with plasmids expressing a SmSmad1B-Gal4AD fusion protein and a SmSmad4-Gal4BD fusion protein, a strong positive interaction was observed, as determined by growth on selective SD media [– Leu, – Trp, – His, + 40 mm 3-amino-1,2,4-triazole (3-AT)] (Fig 6A) and by the development of blue color in a LacZ filter lift assay (Fig 6B) The results of the yeast two-hybrid and the yeast-three hybrid experiments (described below) are summarized in greater detail in Table 1 In comparison

to the SmSmad1B–SmSmad4 interaction, a weakly positive interaction was detected with yeast cotrans-formed with SmSmad1B-Gal4AD and either SmTbRI-0Gal4BD or SmTbRIQD-Gal4BD receptor constructs The protein interactions of the BMP-related SmSmad1 with SmSmad4, SmTbRI and SmTbRIQD were also evaluated, and found to exhibit a relatively comparable interaction pattern to that of SmSmad1B

Yeast three-hybrid assays were performed to evalu-ate the SmSmad1B–SmSmad4 interaction in the pres-ence of SmTbRI receptor constructs Y190 yeast competent cells were cotransformed with the Sm-Smad1B-Gal4AD construct and the pBridge construct that allows for the expression of both SmSmad4-Gal4BD and SmTbRI (or SmTbRIQD), as described

in Experimental procedures In yeast transformants with the pBridge construct, the expression of the

Fig 4 Detection of the SmSmad1B protein in S mansoni worm extract by western blot Composite photograph of SDS gel separation of soluble (S) and whole (W) adult worm extracts (left panel), and membrane strips immunoblotted with: preimmune rabbit IgG (middle panel) and affinity-purified aSmSmad1B IgG (right panel) A competition assay (right panel, lane C) was performed by preincubating the affinity-puri-fied aSmSmad1B IgG with the SmSmad1B linker peptide (10 lgÆmL)1) The molecular size (kDa) of the band is given on the left.

Fig 5 Immunolocalization of SmSmad1B protein in adult schistosomes Immunofluorescent staining of SmSmad1B in adult worm cryosec-tions Column I, phase-contrast images Column II, green autofluorescent images taken with a 522 nm filter Column III, far red immunofluo-rescent images taken with a 680 nm filter (200 · magnification) Worms treated with preimmune rabbit IgG (negative control) are presented

in row A Rows B–E represent worms treated with affinity-purified aSmSmad1B IgG The arrows represent the area of male-specific SmSmad1B fluorescence M, male worm; F, female worm; V, vitellaria; G, gut; ST, subtegument; O, ootype.

receptor should be suppressed in the presence of

methi-onine However, it was determined that the pBridge

construct contains a leaky Met25 promoter that allows

for the expression of the receptor constructs even in

the presence of methionine-containing SD media (data

not shown) Therefore, the pBridge constructs were

only able to be used for three-hybrid analysis when

both the SmSmad4-Gal4BD and the receptor

(wild-type or active mutant) were coexpressed in yeast As

compared to the SmSmad1B–SmSmad4 interaction

observed in the yeast two-hybrid assay, the inclusion

of SmTbRI or SmTbRIQD resulted in decreased

growth of cotransformed yeast on selective SD media

(– Leu, – Trp, – His, – Met, + 40 mm 3-AT) (Fig 7A)

However, little change in blue color intensity was

observed in the filter-lift assay (Fig 7B) Similar results

were observed when SmSmad1B was replaced with SmSmad1 in the three-hybrid experiments

To better examine the effect of the inclusion of TGFb receptor-containing constructs on the Sm-Smad1B–SmSmad4 or SmSmad1–SmSmad4 inter-actions, liquid LacZ assays were performed to quantify induction of b-galactosidase (b-gal) activity (Fig 7C)

In the liquid LacZ assays, the SmSmad1–SmSmad4 interaction produced approximately eight b-gal units, whereas the SmSmad1B–SmSmad4 interaction pro-duced approximately one b-gal unit The presence of the wild-type and constitutively active mutant receptor

in the SmSmad1B–SmSmad4 interaction resulted in statistically significant decreases in SmSmad1B–Sm-Smad4 b-gal induction of 15% and 26%, respectively Decreases in b-gal units of 14% and 38% also resulted

Fig 6 Yeast two-hybrid analysis of SmSmad1B protein interactions (A) Growth of cotransformed Y190 yeast cells on selective SD media (– Leu, – Trp, – His, + 40 m M 3-AT) Numbers 1–8 represent the following yeast cotransformations: 1, p53–pSV40 (positive control); 2, pLamin C–pSV40 (negative control); 3, SmSmad1B-AD–SmSmad4-BD; 4, SmSmad1B-AD–SmTbRI-BD; 5, SmSmad1B-AD–SmTbRIQD-BD; 6, Sm-Smad1-AD–SmSmad4-BD; 7, SmSmad1-AD–SmTbRI-BD; 8, SmSmad1-AD–SmTbRI-QD-BD Cotransformation numbers 4 and 5 were streaked in duplicate (B) LacZ filter-lifts from transformed yeast grown on SD media lacking leucine and tryptophan.

Table 1 Summary of the yeast two-hybrid and yeast-three hybrid experiments showing SmSmad1B protein interactions The following cri-teria were utilized in this table for designating the extent of the protein interactions: the growth rate of cotransformed Y190 yeast colonies (activation of HIS3 reporter); the duration of the development of blue color in the LacZ filter-lift assay (activation of LacZ reporter); and the b-gal units calculated from the liquid LacZ assay (activation of LacZ reporter) + ⁄ –, weak interactions (yeast growth after 9 days of incuba-tion, blue color development on LacZ filter-lift assay after 3 days and ⁄ or < 0.8 b-gal units in liquid LacZ assay); +, moderate interactions (yeast growth after 5 days of incubation, blue color development on LacZ filter-lift assay after 2 days, and 0.9–5.5 b-gal units in liquid LacZ assay; + +, strong interactions (yeast growth after 3 days of incubation, blue color development on LacZ filter-lift assay after 1 day, and 1.0–7.5 b-gal units in liquid LacZ assay); + + +, stronger interactions (yeast growth after 3 days of incubation, blue color development on LacZ filter-lift assay in less than 1 day, and > 7.5 b-gal units in liquid LacZ assay) AD, Gal4 activation domain; BD, Gal4 binding domain.

from the inclusion of wild-type or active receptor in the

SmSmad1–SmSmad4 interaction However, only the

inclusion of SmTbRIQD produced a statistically

signifi-cant decrease in the SmSmad1–SmSmad4 interaction

In the liquid LacZ assays, the extent of the SmSmad1–

SmSmad4 interaction as compared to that of the

Sm-Smad1B–SmSmad4 interaction is more apparent than

what was observed in the filter-lift assay, due to the

quantifiable nature of the liquid assays Also, the

mag-nitude of the decrease in interaction between SmSmad1

and SmSmad4 in the presence of the receptors,

specific-ally SmTbRIQD, is more obvious in the liquid assay

than in the LacZ filter-lift assay

In an attempt to confirm the SmSmad1B protein

interactions in the yeast assays, maltose-binding

protein (MBP) pull-down experiments were performed The resin-bound SmSmad4-MBP fusion protein was incubated with in vitro translated [35S]SmSmad1B in the presence or absence of either in vitro translated SmTbRI, unlabeled SmTbRI or SmTbRIQD MBP-bound resin was used as a negative control to assess nonspecific background binding Similar to the results

of the yeast two-hybrid and three-hybrid protein inter-action assays, SmSmad1B was able to bind SmSmad4

in the pull-down assay (Fig 8A,B) The addition of SmTbRI resulted in a decrease in the interaction strength between SmSmad1B and SmSmad4 by 19%, and the inclusion of SmTbRIQD produced a statisti-cally significant 52% decrease In a previous report, the inclusion of the receptor constructs in an in vitro

Fig 7 Yeast three-hybrid analysis of

SmSmad1B protein interactions (A) Growth

of cotransformed Y190 yeast cells on

select-ive SD media (– Leu, – Trp, – His, – Met,

+ 40 m M 3-AT) Numbers 1–8 represent the

following yeast cotransformations: 1,

p53–pSV40 (positive control); 2, pLamin

C–pSV40 (negative control); 3,

SmSmad1B-AD–SmSmad4-BD; 4, SmSmad1B-AD–

SmSmad4-SmTbRI-pBridge; 5,

SmSmad1B-AD–SmSmad4-SmTbRI-QD-pBridge; 6,

AD–SmSmad4-BD; 7,

SmSmad1-AD–SmSmad4-SmTbRI-pBridge; 8,

SmS-mad1–SmSmad4-SmTbRI-QD-pBridge.

(B) LacZ filter-lifts from transformed yeast

grown on selective SD media lacking

leucine, tryptophan and methionine

cotransformed with plasmids in the same

order as in (A) (C) Liquid LacZ assays.

Induction of b-gal is reported in b-gal units,

where values represent the average of three

independent experiments *Represents

statistically significant value (P ¼ 0.05).

SmSmad1–SmSmad4 interaction assay also had a

neg-ative effect on the strength of the interaction between

SmSmad1 and SmSmad4 [6] MBP pull-down assays

were also employed to investigate the binding of

Smad1B with SmTbRI or SmTbRIQD in vitro

Sm-Smad1B was expressed as an MBP fusion protein and

incubated with either in vitro translated [35

S]methion-ine-labeled SmTbRI or SmTbRIQD, and the bound

proteins were precipitated with amylose resin In

the pull-down assays, SmSmad1B interacted with both SmTbRI and SmTbRIQD, with a slight binding preference for SmTbRIQD (Fig 8B) The preferen-tial binding of SmSmad1B to SmTbRIQD in the pull-down assays, although moderate, could explain the decreased interaction between SmSmad1B and SmSmad4 in the presence of SmTbRIQD (Fig 8A), as the interaction between SmSmad1B and SmTbRIQD made SmSmad1B less available for binding to Sm-Smad4

Pull-down assays were performed to investigate the interaction between the schistosome coactivator proteins SmGCN5 [27] and SmCBP1 [28] and the schistosome R-Smads) SmSmad1, SmSmad2, and SmSmad1B) in the presence or absence of SmSmad4 For the interaction assays with SmGCN5, the schisto-some R-Smads were in vitro translated as glutathione S-transferase (GST)-fusion proteins and incubated with in vitro translated 35S-labeled SmGCN5, in the presence or absence of nonlabeled SmSmad4 GST-bound glutathione Sepharose was used as a negative control to assess nonspecific background binding The results of the pull-down assays in Fig 9A show that the BMP-related R-Smads, SmSmad1 and SmSmad1B, interact at relatively higher levels with SmGCN5 as compared to the level achieved with the TGFb-related SmSmad2 In the meantime, addition of SmSmad4 resulted in decreased levels of interaction with SmGCN5 with all the tested R-Smads (Fig 9A) For SmSmad2, inclusion of SmTbRI-QD in the binding reaction not only significantly increased the interaction

of SmSmad2 with SmGCN5, but also revealed the interaction of SmSmad4 with SmGCN5 and demon-strated its participation in the formation and, prob-ably, the stabilization of the transcriptional protein complex Figure 9B shows that the presence of SmTbRI-QD significantly boosted the interaction with SmGCN5 of the 35S-labeled SmSmad4 (lane 4) and the

35S-labeled SmSmad2 (lane 6) in the presence of either nonlabeled SmSmad2 or SmSmad4, respectively, as compared to the interaction levels attained in the pres-ence of wild-type SmTbRI (lanes 3 and 5)

In a similar approach, a GST pull-down assay was performed to investigate the interaction between the coactivator SmCBP1 and the R-Smads SmSmad1, SmSmad1-B, and SmSmad2 The results in Fig 10A show that GST-SmCBP1 interacted with SmSmad1 and SmSmad2 but not with SmSmad1B, SmSmad4 or the receptor SmTbRI-QD, which served as a negative control Similar to the situation with SmGCN5, when SmSmad4 was included in the reactions, a reduction in interaction level with GST-SmCBP1 was observed with both SmSmad1 and SmSmad2 (Fig 10B), and again,

Fig 8 In vitro interaction between SmSmad1B and schistosome

TGFb superfamily members (A) Evaluation of the SmSmad1B–

SmSmad4 interaction by MBP pull-down experiments; In vitro

translated [ 35 S]SmSmad1B (5 lL) was incubated with

SmSmad4-MBP (2 lg) in the presence or absence of unlabeled in vitro

transla-ted SmTbRI or SmTbRI-QD (10 lL) A graphical representation of

the values obtained from the SmSmad1B–SmSmad4 MBP

pull-downs in the presence or absence of receptor constructs is shown

(bottom panel) *Represents statistically significant value (P ¼

0.05) (B) MBP pull-down experiments demonstrating the

interac-tion between SmSmad1B-MBP and [ 35 S]SmTbRI or [ 35

S]SmTbRI-QD Values represent percentage binding as compared to input,

and are the mean of three independent experiments Background

binding, represented by (–), was accounted for in the calculation of

percentage binding Lanes labeled (I) represents 10% input of

35

S-labeled in vitro translated products.