Improving Protein Pharmacokinetics by Genetic ... - Semantic Scholar

THE JOURNAL OF BIOLOGICAL CHEMISTRY © 2004 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 279, No. 5, Issue of January 30, pp. 3375–3381, 2004 Printed in U.S.A.

Improving Protein Pharmacokinetics by Genetic Fusion to Simple Amino Acid Sequences* Received for publication, October 15, 2003, and in revised form, November 10, 2003 Published, JBC Papers in Press, November 11, 2003, DOI 10.1074/jbc.M311356200

Paula Alvarez‡, Carlos A. Buscaglia‡§, and Oscar Campetella¶ From the Instituto de Investigaciones Biotecnolo´gicas, Universidad Nacional de General San Martıń, B1650WAB San Martıń, Buenos Aires, Argentina

The role of primary amino acid sequences in protein pharmacokinetics, an issue of relevance in both basic knowledge and biotechnology, was addressed here using as a starting point two repetitive antigens from the hemoflagellate Trypanosoma cruzi that are known to stabilize their associated proteins in the bloodstream. A major drawback to their pharmacological application is that these repetitive sequences are highly immunogenic, being therefore the deletion of this characteristic desirable. Based on sequence homology and epitope mapping analyses, an artificial repetitive sequence (PSTAD) was engineered. This motif was tested by genetic fusion to the C terminus of both the trypanosomal trans-sialidase and the rat tyrosine aminotransferase and found to produce a 4.5– 6-fold increase in the half-life of the associated proteins in blood while displaying significantly lower immunogenicity. Residues involved in the stabilizing properties of the novel peptide were mapped by a site-directed mutagenesis approach, allowing us to successfully identify another two motifs. Searching databases for sequences displaying some homology, embedded in proline frameworks and associated to shed virulence factors from unrelated microorganisms, resulted in the identification of four other protein extensions. Remarkably, three of them (from Streptococcus pneumoniae, Actinomyces viscosus, and Escherichia coli) revealed similar pharmacokinetic features, suggesting therefore an analogous evolutionarily acquired mechanism to ensure the biodistribution of their corresponding proteins. Our findings indicate that the insertion of defined motifs into a proline-rich framework constitutes a suitable alternative to construct a chimeric protein with extended half-life in blood. * This work was supported by grants from the Agencia Nacional de Promocioń Cientı´fica y Tecnolo´gica and the Consejo Nacional de Investigaciones Cientı´ficas y Tećnicas (CONICET) from Argentina and the World Bank/UNDP/WHO Special Program for Research and Training in Tropical Diseases. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EBI Data Bank with accession number(s) AY249142 (hydrophilic extension of the TS from T. carassii) and AY243566 (C terminus of the EspF protein from the enteropathogenic E. coli O145:NM strain). ‡ Research Fellow of Consejo Nacional de Investigaciones Cientı´ficas y Tećnicas (CONICET). § Current address: Michael Heidelberg Division of Immunology, Dept. of Pathology, New York University School of Medicine, 550 First Ave., New York, NY 10016. ¶ Researcher of CONICET. To whom all correspondence should be addressed: Instituto de Investigaciones Biotecnolo´gicas, Universidad Nacional de General San Martıń, Predio INTI, Edificio 24, Av. General Paz y Constituyentes, B1650WAB San Martıń, Buenos Aires, Argentina. Tel.: 54-11-45807255; Fax: 54-11-47529639; E-mail: oscar@iib. unsam.edu.ar. This paper is available on line at http://www.jbc.org

Despite its relevance, little is known about the structural requirements that are to be filled by a given protein to remain in blood. This applies even for a long lived blood protein such as the albumin, where detailed knowledge of its structure and function has already been gained (1). The mean residence time in the bloodstream is distinctive for each particular protein and seems to rely on a combination of several mechanisms including proteolytic degradation, hepatic uptake, endothelium permeability, renal excretion, and immunogenic reactions (2). In addition, post-translational modifications such as site-specific carbohydrate attachment (3, 4), sialylation degree (5), and multimerization (6) may modulate this phenomenon. The presence of endogenous receptors for certain molecules also regulates their mean residence time in blood (7–10). Although scarce evidence is at hand, the existence of linear amino acid motifs involved in the bloodstream half-life of proteins might be postulated (11, 12). Understanding the structural requirements that govern the pharmacokinetics of proteins is not only a relevant basic science issue; the need to maintain substances in circulation within their biologic activity ranges is one of the main objectives in therapeutic and diagnostic applications (13). In this regard, several strategies have been described, including mutagenesis, covalent conjugation of molecules to polymers (14 – 18), encapsulation in liposomes or viral envelopes, and physical entrapment in particles (2). Other approaches rely on genetic and chemical techniques to create chimeras between a desired protein and a normal constituent of the mammalian plasma that would eventually act as a carrier (19 –21). This strategy resembles that displayed by several pathogen-derived molecules that recognize and remain attached to circulating proteins such as immunoglobulins, albumin, fibronectin, or fibrinogen through the evolutionary acquisition of specific binding domains (22–24). Trypanosoma cruzi, the protozoan parasite causative of the Chagas’ disease, displays an alternative strategy to improve the pharmacokinetic properties of its trans-sialidase (TS),1 a bloodstream-borne virulence factor (25). The presence of a repetitive domain termed shed acute phase antigen (SAPA) on the TS C terminus, although not involved in catalysis (26), raises its half-life in blood (12, 27). This enhanced serum persistence allows the enzyme to be systemically distributed, thus ensuring the interaction with its target tissues during infection (28, 29). Another repetitive shed antigen from T. cruzi known

1 The abbreviations used are: TS, trans-sialidase; GST, glutathione S-transferase from Schistosoma japonicum; HRM, hybrid repetitive motif; PBS, phosphate-buffered saline; SAPA, shed acute phase antigen; TAT, tyrosine aminotransferase from rat liver; t␤, elimination half-life; TS-HRM, TS with 21 associated HRM repeats; TS-SAPA, TS with 13 associated SAPA repeats; TS-3R, TS with three associated SAPA repeats.

3375

3376

Improving Pharmacokinetics of Proteins

as Tc13 (30), when genetically fused to TS stabilizes the enzyme in blood to a similar extent as SAPA (12). Therefore, both amino acid sequences provide a simple and valuable tool for the elucidation of peptidic motifs underlying the maintenance of proteins in the circulation. Unfortunately, both SAPA and Tc13 are highly immunogenic, being two of the major antibody targets during the acute phase of T. cruzi infection (31). Molecular dissection of both features (i.e. immunogenicity and stabilization capacity) is therefore highly desirable. In the present work, we took advantage of sequence alignments and B-cell epitope mapping information from both T. cruzi antigens to design a novel artificial repetitive motif (termed the hybrid repetitive motif (HRM)). As SAPA and Tc13, this motif carries the ability to stabilize reporter proteins in the circulation but displays significantly lower immunogenicity. Critical amino acid residues involved in the pharmacokinetics properties of HRM were identified by a site-directed mutagenesis approach. Results achieved were extended to peptide motifs present in shed proteins from other pathogenic microorganisms, suggesting an analogous strategy to disseminate their virulence factors inside the vertebrate host. EXPERIMENTAL PROCEDURES

Mice—C3H/HeJ and BALB/cJ 60 –90-day-old female mice bred in our facilities were indistinctly used. All assays were approved by the ethical committee of our institute. Recombinant Proteins—The minimal DNA unit coding for two 5-amino acid-long repeats of the HRM molecule was generated by annealing the oligonucleotide 5⬘-pCGATCCCTCTACGGCCGACCCAAGTACTGC-3⬘ (where p means 5⬘ phosphorylated oligonucleotide) to its complementary sequence. Oligonucleotides (250 pmol each) were mixed, boiled for 5 min, and incubated overnight at 16 °C with 1,000 units of T4 DNA ligase and 75 ␮M ATP. Five units of FspI restriction endonuclease were included in the ligation mixture to release tail-totail ligated products. DNA-modifying enzymes were from New England Biolabs (Beverly, MA), unless otherwise stated. Fragments showing maximal extent of polymerization were purified from agarose gels and subjected to a second round of ligation as described above. DNA fragments showing the desired molecular mass were purified, incubated for 1 h at 72 °C with Taq polymerase in the presence of dATP 0.4 mM, and subsequently cloned into the pGEM T-easy vector (both from Promega (Madison, WI)). The accuracy and the actual number of repeats were determined by DNA sequencing. To obtain the TS-HRM fusion protein, this insert was subcloned into the EcoRI site located at the 3⬘ end of the TS gene in the pTSac plasmid (12) constructed in pTrcHisA vector (Invitrogen). The HRM-encoding fragment was also subcloned into the EcoRI site of pGEX-1␭T vector (Amersham Biosciences) to obtain the GST-HRM protein. A similar strategy was carried up to generate each HRM mutant molecule. The following oligonucleotides annealed to their complementary sequences were used: 5⬘-pAGCACGGCGGATGGCTCAACGGCGGATCCC-3⬘ (where p means 5⬘ phosphorylated oligonucleotide) for the HRM-P construct; 5⬘-pCCGTCAACCGCGAACCCGAGCACCGCGAAT-3⬘ for the HRM-N construct, and 5⬘-pAGCACGGCGAAACCGTCAACGGCGAAACCC-3⬘ for the HRM-K construct. Ligation assays were performed as before but in the presence of SmaI for the HRM-P and HRM-K constructs and SspI for the HRM-N constructs. EcoRI adaptors (5⬘-AATTCTCAGGT-3⬘ and 5⬘-ACCTGAG-3⬘) were annealed and ligated to the purified blunt-ended fragments for their direct cloning into the pTSac. To obtain shorter HRM extensions, the corresponding annealed oligonucleotides were subjected to a single ligation round, and then the pool was ligated to the annealed 5⬘-AATTCACCTCAACGGCCGACCCCTCAACTGC-3⬘ and 5⬘-GCAGTTGAGGGGTCGGCCGTTGAGGTG-3⬘ oligonucleotide pair and cloned in the EcoRI site of the pTSac. A clone containing seven repetitive units in frame with the TS, as determined by DNA sequencing, was selected. The entire rat liver tyrosine aminotransferase (TAT) cDNA (GenBankTM NM 012668) was gently provided by Dr. C. Nowicki in a plasmid constructed in pET24 (Stratagene, La Jolla, CA) and amplified by PCR (primers 5⬘-ACTCGAGAAGGCCAGATGGGACGTG-3⬘ and 5⬘CGGTACCATATTTGTCACACTCCTCCT-3⬘). After XhoI/KpnI digestion (sites underlined in the sequences), the amplicon was inserted upstream of the HRM containing 21 repetitive units encoding sequence

previously subcloned into the EcoRI site of pTrcHisA (Invitrogen). The degenerate repetitive sequence of the sialidase from Actinomyces viscosus strain DSM43798 (32) (GenBankTM X62276/L06898) was obtained by KpnI/SacI digestion of a plasmid containing the whole nanH gene, a kind generous of Drs. R. Schauer and P. Roggentin. The fragment was ligated to EcoRI adaptors constructed with the oligonucleotide pairs 5⬘-AATTCTAATCACGAGCT-3⬘/5⬘-CGTGATTAG-3⬘ and 5⬘AATTCGCCTCAGGGTAC-3⬘/5⬘-CCTGAGGCG-3⬘ and inserted into the pTSac as above. The C terminus of the sialidase A encoded by the nanA gene from Streptococcus pneumoniae (33) (GenBankTM number Q59959) was amplified from genomic DNA using the primers 5⬘-CGAATGGAATGAACGGAA-3⬘ and 5⬘-TATACTCTGCGATTTCAT-3⬘. The PCR product was cloned into the pGEM T-easy vector, released by EcoRI digestion, and subcloned into the pTSac. The same strategy was employed to clone the hydrophilic extension of a TS-like gene from Trypanosoma carassii (34) (GenBankTM number AY249142) (primers 5⬘-ATCCTTCCGTTGTGAAAC-3⬘/5⬘-TCAATCTTGTCCGACTTTAGG3⬘) and the C terminus of the EspF protein from the enteropathogenic Escherichia coli O145:NM strain (35, 36) (GenBankTM number AY243566) (primers 5⬘-TTACTCCCTCTCGTCCG-3⬘/5⬘-TTACCCTTTCTTCGATTG-3⬘). All constructs were subjected to DNA sequencing, and the molecular properties of the encoded proteins were predicted with the aid of the LaserGene software (DNASTAR Inc., Madison, WI). Expression, Purification, and in Vitro Evaluation of Recombinant Proteins—TS and TAT recombinant proteins were expressed in E. coli XL1-Blue (Stratagene) and purified to homogeneity, as judged by Coomassie Blue-stained SDS-PAGE, by immobilized metal affinity chromatography through Ni2⫹-charged Hi-Trap chelating columns using the His tag located at their N termini provided by their respective cloning vectors, followed by ion exchange chromatography through MonoQ columns (both from Amersham Biosciences). GST, GST-SAPA (12), and GST-HRM proteins were purified by affinity chromatography through GSTrap columns (Amersham Biosciences) following manufacturers’ guidelines. Pharmacokinetics Studies—Mice were intravenously injected with 3 ␮g (30 – 40 pmol) of the indicated TS recombinant protein contained in 0.2 ml of PBS by the retro-ocular sinus. Blood samples were taken from the tail at different intervals as required by each experimental design. Sera were assayed for remnant TS activity by measuring the transference of the sialyl residue from sialyl-lactose (Sigma) to [14C]lactose (Amersham Biosciences) as described (37). Results were calculated in cpm/␮l of plasma as enzyme source. Specific TS activity, thermal inactivation at 37 °C, and proteolytic degradation in normal mouse plasma (12, 27) rendered similar values for every chimeric TS generated (data not shown). TAT and TAT-HRM proteins were labeled with 125I (PerkinElmer Life Sciences) using IODO-GEN-coated tubes (Pierce), and unincorporated 125I was removed by Hi-Trap Desalting columns (Amersham Biosciences). About 1.5 ⫻ 106 cpm (specific activities: TAT, 145,000 cpm/␮g; TAT-HRM, 475,000 cpm/␮g) were given intravenously in 0.2 ml of PBS. Fractions of the sera collected at different times postinjection were resolved by SDSPAGE, and labeled proteins were quantified by 1D Image Analysis software (Eastman Kodak Co.) after autoradiography. In all cases, results were expressed relative to values found at the 3-min postinjection time that was taken as 100%. The intravascular half-life values (t of the elimination phase, t␤) were determined from the second part of the decay curve by least squares regression using the slope of a curve fitted to the data points (16, 38). Animal Immunization and Antibody Evaluation—Mice were administered with four intravenous doses (10 ␮g/each) of recombinant TS proteins every 10 days by the retro-ocular route. One week after the last dose, the reactivity of sera toward the SAPA or HRM repetitive motif was evaluated by enzyme-linked immunosorbent assay. Briefly, polystyrene microplates (Maxisorp; NUNC a/s, Roskilde, Denmark) were coated with GST-SAPA or GST-HRM (100 ng/well, about 1.6 pmol of GST-SAPA, or 2.3 pmol of GST-HRM) in PBS overnight. Wells coated with GST were used as negative controls. The amount of GST fusion proteins in each plate was tested assaying three wells/plate with proper dilutions of an anti-GST rabbit serum. Plates were blocked with 5% nonfat milk in PBS for 2 h at room temperature, and then serial serum dilutions (100 ␮l/well) were seeded and incubated for 1 h. Reaction was revealed by the addition of peroxidase-conjugated secondary antibodies (Sigma) followed by 2,2⬘-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (Roche Diagnostics). Each sample was tested in triplicate. Sera obtained from naive animals and assayed in the same plate were taken as negative controls. SPOT Assays—B-cell epitope mapping of antigen Tc13 was achieved by SPOT assay as described (39). Briefly, filters containing peptide


FIG. 1. Designing of a novel HRM motif derived from T. cruzi shed antigens. A, alignment of SAPA and Tc13 antigens. The boxes highlight the core of the SAPA epitope that overlaps with the designed HRM sequence and the single Tc13 epitope disclosed here. Residues deleted to construct the HRM sequence are displayed in boldface type. Lowercase letters correspond to residues of the adjacent repetitive unit. B, B-cell epitope mapping of the T. cruzi Tc13 antigen. Hexameric peptides were synthesized by the SPOT technique following the Tc13 repeat sequence. Membranes were probed with mouse or rabbit antiTc13 sera (1:200 dilution). The boxes highlight the overlapping amino acid residues that determine the disclosed epitopes. spots (Sigma) were treated with casein-based blocking buffer and sera from either rabbits or mice immunized with a GST-Tc13 fusion protein (12) emulsified in Freund’s adjuvant were tested at 1:200 dilution. After extensive washings, secondary antibodies coupled to alkaline phosphatase (DAKO Corp., Carpinteria, CA) were added at 1:1,000 dilution, and a reaction was developed with 5-Br-4-Cl-3-indolyl phosphate/3-(4,5dimethylthiazol-2-yl) 2,5-diphenyltetrazolium bromide (both from Sigma). Statistical Analyses—Student’s t test was used. RESULTS

Design of a Novel Motif Derived from T. cruzi Repetitive Antigens—Both SAPA (SSAHSTPSTPAD) and Tc13 (PKSAE) repetitive antigens from T. cruzi were shown to extend the half-life of associated proteins in blood (12), thus providing a model to search for primary sequences involved in this phenomenon. Sequence alignments indicate that the Tc13 repetitive unit resembles the PSTPAD sequence embedded in the SAPA repetitive unit (Fig. 1A). To design a novel repetitive sequence with potential biotechnological applications, two parameters were considered: the minimum amino acid length required and its potential immunogenicity. In this context, the second SAPA proline (PSTPAD) was precluded from the design, since it plays a central role in SAPA antigenicity (39). To disclose the antigenic organization of the Tc13 repeat, we scanned for its B-cell linear epitopes through the SPOT technique (40). Sera from Tc13 repeat-immunized mice and rabbits were tested against peptides covering the five possible hexamers that can be constructed with the repeat sequence. As shown in Fig. 1B, both species highlight a similar epitope (KSAE for rabbits and KSAEP for mice), indicating the presence of a single antigenic determinant per repeat unit. It can also be concluded from these assays that the lysine residue is involved in Tc13 immunogenicity, whereas the counterpart serine residue in SAPA (PSTPAD) does not participate in the epitope structure of this molecule (39). Taking these results altogether, both the second proline from SAPA and the lysine from Tc13 were excluded, and the 5-amino acid-long motif PSTAD named HRM was designed (Fig. 1A). Sera from rabbits and mice immunized with either GST-SAPA or GST-Tc13 antigens were unable to recognize TS-HRM (not shown), confirm-

3377

ing that the main B-cell epitopes present in both parental sequences were disrupted. The HRM Motif Extends the Bloodstream Half-life of Genetically Fused Proteins—The use of TS as a reporter protein in pharmacokinetic assays was previously validated (12). It also provides some experimental advantages such as its easy enzymatic activity measurement and the absence of specific inhibitors in normal mammalian serum (41). Therefore, 21 HRM units were inserted in frame at the C terminus of TS (TS-HRM; Fig. 2A). This addendum is roughly equivalent in mass to nine SAPA repeats, thus slightly above the threshold of eight SAPA units determined to improve the TS permanence in the circulation (12). TS-HRM protein was injected into mice by the intravenous route, and remnant TS activity was monitored in blood samples collected at different times postinjection. As controls, three different TS proteins previously characterized (12) were used (see Fig. 2A). All of these proteins exhibit similar specific TS activity when assayed in vitro (data not shown) and were thus administered in equimolar amounts (35– 40 pmol/mouse). As shown in Fig. 2B, TS-HRM and TS-SAPA displayed similar half-lives (t␤ of 40 and 39 h, respectively) and can be detected in the bloodstream up to 3 days postinjection. In contrast, TS without any C-terminal extension was absent from circulation, and only 10% of the input of the TS harboring three SAPA repeats (TS-3R) remained after 24 h. In a separate set of experiments, a TS with only seven HRM-associated motifs, roughly equivalent in mass to three SAPA units, was also tested, displaying a similar half-life to that estimated for the TS-3R molecule (t␤ of 24 versus 18 h, respectively), thus suggesting that HRM and SAPA motifs share similar pharmacokinetic characteristics. To assay whether the effect of the HRM motif can be extended to proteins other than TS, the rat liver TAT was chosen as a mammalian protein model. The purified recombinant TAT and TAT-HRM proteins were 125I-labeled and administered intravenously. From the plotting of the autoradiographic signals derived from SDS-PAGE (Fig. 2C), the half-life of the TAT-HRM was extended about 6 times when compared with that of TAT (from 1.3 to 7.9 h). Findings displayed in Fig. 2 indicate that the HRM motif abridges all the molecular features involved in the stabilization properties of both SAPA and Tc13 parental sequences. The HRM Motif Displays Lower Immunogenicity than SAPA—To evaluate the immunogenic properties of the HRM repeat, mice were intravenously administered with four doses either of TS-HRM, TS-3R, or TS-SAPA (10 ␮g/each, Fig. 2A) and the specific antibody response directed against the repetitive domains was assayed by enzyme-linked immunosorbent assay. Although intravenous injection often leads to poor immune responses, high titers of SAPA directed antibodies were detected in TS-SAPA-immunized animals (Fig. 3). Furthermore, the TS-3R protein that contains only three SAPA-repeats and is rapidly cleared from the circulation (Fig. 2B) elicits a similar humoral response as TS-SAPA (Fig. 3). In contrast, the HRM motif displays a significantly reduced immunogenicity although it harbors 21 repetitive units. Reactivity of sera from TS-HRM-immunized animals remained under the cut-off value at the end of the immunization schedule (Fig. 3). Overall, selected amino acid replacements introduced to generate the HRM molecule significantly impair its immunogenic properties, probably by disruption of both SAPA and Tc13 B-cell epitopes (see above and Fig. 1). Therefore, the designed HRM repetitive motif retains the ability to lengthen the permanence of associated proteins in blood while exhibiting a remarkable reduction in its immunogenic properties.

3378


FIG. 3. Immunogenicity of the HRM motif. Animals were administered with four 10-␮g doses (110 –120 pmol) of the indicated protein by the intravenous route. Ten days after the last dose, anti-repeat antibodies were determined by an enzyme-linked immunosorbent assay at the indicated sera dilutions. Data are expressed as the mean ⫾ S.D. (n ⫽ 3 animals). The asterisks denote significant differences (p ⬍ 0.05) from TS-SAPA. Data for normal mouse serum are presented as single columns, since undistinguishable reactivities were recorded against GST, GST-SAPA, and GST-HRM molecules.

FIG. 2. Pharmacokinetics studies of the TS-HRM chimera. A, schematic representation and molecular properties of chimeric TS constructs assayed. The catalytic TS domain and the variable repetitive extension are indicated in gray and black boxes, respectively. Dark gray small boxes at the N termini indicate the His tag. Molecular masses (Mr) are indicated in kDa. The predicted net charge at pH 7 is also shown. N-, the amino terminus of each protein. B, mice were intravenously injected with 3 ␮g (30 – 40 pmol) of the indicated protein, and blood samples taken at different times postinjection were measured for remnant TS activity. Values recorded immediately after injection are indicated as input values and were taken as 100% trans-sialidase activity. Data are expressed as the mean ⫾ S.D. (n ⫽ 5 animals). The asterisks denote significant differences (p ⬍ 0.01) from the TS-SAPA protein. C, mice were intravenously injected with 125I-labeled TAT or TAT-HRM, and serum samples were taken at the indicated times. The autoradiography of the SDS-PAGE and the remaining protein percentage, as estimated by scanning, are shown. All assays were performed at least three times.

Mapping of Residues Involved in HRM Stabilization Properties—Critical amino acid residues involved in the HRM-mediated blood stability were then searched by site-directed mutagenesis. For this purpose, several constructs derived from the HRM sequence were designed and cloned at the TS C terminus. Recombinant proteins having similar mass addenda to TSHRM were assayed (for molecular properties, see Fig. 4A).

Since the repetitive antigens used herein are embedded in a proline framework (Fig. 1A), we first tested the effect of a longer interproline distance in the pharmacokinetics of the HRM molecule. A repeat unit with prolines spaced by 10 residues was built by replacing prolines for glycine in alternating repeats (PSTADGSTAD, termed HRM-P). As shown in Fig. 4B, the TS-HRM-P fusion protein renders a similar bloodstream half-life value as compared with the parental TS-HRM molecule (t␤ of 38 and 40 h, respectively). The second relevant feature displayed by both SAPA and HRM is a biased amino acid composition toward negatively charged residues (Figs. 1A and 2A). Therefore, we replaced the aspartate residue of HRM either for asparagine (PSTAN, termed HRM-N) or lysine (PSTAK, termed HRM-K). The TS-HRM-N recombinant protein displayed the same in vivo performance (t␤ ⫽ 39 h) as the TS-HRM, indicating that this modification imposes minor constraints on its stabilization properties (Fig. 4B). Conversely, replacement of aspartate by lysine significantly drops the mean residence time of the TS-HRM-K molecule in blood (t␤ ⫽ 11 h; Fig. 4C). This value is similar to that recorded for the TS without a C terminus extension and then even below that of the TS-3R molecule (t␤ of 9 and 18 h, respectively) (Fig. 4). Repetitive Units Present in Secreted Virulence Factors from Other Pathogens Also Stabilize TS in Blood—The acquisition of SAPA and Tc13 extensions might represent a novel strategy to disseminate TS and TS-like molecules in the bloodstream. Hence, it is conceivable that other pathogens unrelated to T. cruzi might have developed a similar mechanism to improve the biodistribution of some of their encoded virulence factors. To test this hypothesis, we searched the data bases for sequences displaying the following features: (a) a certain degree of similarity to SAPA and/or Tc13; (b) a proline-rich hydrophilic framework accounting for their extended secondary structure and maximum solvent exposure; and (c) association to shed virulence factors from pathogenic microorganisms. Four sequences matching these criteria were identified and cloned at the TS C terminus to be tested in our pharmacokinetic assays. These sequences span the degenerate repetitive units present at the C termini of the sialidases from A. viscosus (32, 42) and S. pneumoniae (33), the hydrophilic C-terminal extension of the closely related TS from the fish parasite T. carassii (34), and


3379

FIG. 4. Pharmacokinetics studies of TS-HRM mutant proteins. A, schematic representation and molecular properties of mutant TS-HRM molecules assayed. The catalytic TS domain and the variable repetitive extension are indicated by gray and black boxes, respectively. Dark gray small boxes at the N termini indicate the His tag. Molecular masses (Mr), are indicated in kDa. The predicted net charge at pH 7 is also shown. N-, the amino terminus of each protein. B and C, mice were intravenously injected with 3 ␮g of each protein and quantified for remnant trans-sialidase activity in blood as indicated in the legend to Fig. 2. Note that whereas the HRM, HRM-N, and HRM-K repeat unit contains 5 amino acid residues, the HRM-P unit is 10 residues long. Data are expressed as the mean ⫾ S.D. (n ⫽ 5 animals). The asterisks denote significant differences (p ⬍ 0.01) to the TS-SAPA protein. Assays were performed at least three times.

the proline-rich region of the EspF protein from enteropathogenic E. coli (35, 36). The complete sequence of these domains together with the predicted molecular properties of the ensuing chimeric TS proteins is indicated in Fig. 5A. The putative alignment of these sequences with SAPA and/or HRM repetitive units is also displayed. As shown in Fig. 5B, both A. viscosus- and S. pneumoniae-derived sequences increase the persistence of TS in blood to the same extent as SAPA. It is noteworthy that the homologies detected between both molecules and the T. cruzi SAPA repeat are restricted to the PSTPAD motif, which explains the improved pharmacokinetics of the latter molecule (see Figs. 1 and 2). The sequence from E. coli EspF displays an intermediate but significant effect on the stabilization of TS (t␤ ⫽ 24 h), whereas the extension of T. carassii does not show any effect (t␤ ⫽ 13 h). DISCUSSION

The rapid clarification from the bloodstream constitutes a major drawback for many otherwise promising protein pharmaceuticals (17, 43, 44). To overcome this problem, covalent coupling of drugs or proteins to polyethylene glycol has been extensively applied (13). Although polymer conjugation to hormones and enzymes may lead to a reduction in their biological/ enzymatic activities (45– 47). Furthermore, chemical coupling methods usually result in a mixture of heterogeneous molecules displaying different in vivo performances (2, 45– 47). Genetic techniques constitute an interesting alternative, since this method allows the allocation of the desired sequences far from the active site of the target molecule. However, current approaches involve fusion to whole proteins or to entire protein

domains that might produce negative effects in the recombinant protein activity as well (48). A novel approach based on repetitive sequences able per se to retain proteins in the circulation was explored here. Two T. cruzi repetitive antigens (SAPA and Tc13) able to extend the life span of associated proteins in blood have been previously identified (12). Taking into account both structural and antigenic information of these antigens (Fig. 1) (39), an artificial repetitive unit termed HRM was designed. Pharmacokinetic studies showed that the genetic fusion to HRM produced a 4.5– 6-fold increase in the half-life in blood of the associated protein (Fig. 2). Whereas the rapid protein distribution phase from the bloodstream to the extravascular space (termed the ␣ phase) seems not to be substantially modified, most of the improvement achieved by HRM is due to a large increase in the ␤ phase that corresponds to the intravascular elimination of the injected proteins. As previously described for SAPA and Tc13 (12), the HRM extension does not modify the specific activity of the associated TS, probably due to its flexible secondary structure made up of hydrophilic residues interspaced by prolines. More important, the immunogenicity of HRM was significantly reduced as compared with SAPA, overcoming one of the major hurdles for the biotechnological applications of the parental sequences (Fig. 3). The involvement of the net charge of the HRM molecule in its in vivo function was also addressed here. Replacement of aspartate by asparagine did not alter the functionality of the repeat (TS-HRM-N; Fig. 4B), contrasting with the remarkable effect observed when the residue was replaced by lysine (TS-

3380


FIG. 5. Pharmacokinetics studies of TS fused to different peptide motifs isolated from pathogens. A, sequences of the peptide motifs from different virulence factors cloned at the C terminus of the TS. One repetitive unit, when applicable, is highlighted in gray for each protein. Note that in some cases these repeats are not canonical. In the A. viscosus sialidase extension, an additional small repeat can be defined (box). In the extension from T. carassii TS, the minor degenerated repeats are boxed. Homologies to SAPA and HRM motifs are indicated above and below each sequence, respectively. B, the C-terminal extensions from the sialidase of A. viscosus (TS-ACTI), the sialidase A from S. pneumoniae (TS-STRE), the EspF protein from the enteropathogenic E. coli O:145NM (TS-ESCH), and the trans-sialidase from T. carassii (TS-TCAR) were placed at the C terminus of the TS from T. cruzi. Mice were intravenously injected with 3 ␮g of the indicated protein and quantified for remnant trans-sialidase activity in blood as described in the legend to Fig. 2. Data are expressed as the mean ⫾ S.D. (n ⫽ 5 animals). The asterisks denote significant differences (p ⬍ 0.01) from TS-SAPA protein. Assays were performed at least three times.

HRM-K; Fig. 4C). This phenomenon might be ascribed to the positive charge added and is consistent with the fact that positively charged proteins are rapidly depleted from blood mainly due to the negative charges present in the endothelium, liver, and the renal glomerule (49). In view of these results, one striking feature is the presence of one lysine per repeat in the original Tc13 sequence (Fig. 1A). The positive charge provided by this residue, however, seems neutralized in the Tc13 unit (EPKSA) by equimolar amounts of closely located glutamate residues, thus rendering an overall negative net charge. Hence, an overall negative net charge seems to constitute a desired property to design this kind of sequences. The molecular basis underlying the stabilization mediated by SAPA, Tc13, and HRM is still puzzling. Attempts to identify a putative serum component that might act as a carrier molecule, have proven unsuccessful (data not shown and Ref. 12). However, it seems not to be as trivial as to be explained by mass addendum or an increase in the hydrodynamic volume of the reporter protein. In a previous work, we showed that tandem repeats associated to intracellular T. cruzi proteins, although structurally related to SAPA and Tc13, display significantly reduced pharmacokinetic parameters when genetically fused to TS (12). This is also the case for the hydrophilic extension present in the TS protein from T. carassii (Fig. 5). However, a critical amount of the appropriate sequences seems to be necessary, since only a minor enhancement of TS circulation was obtained using three SAPA units in contrast to the major effect achieved when eight or more SAPA units are placed at its C terminus (Fig. 2B) (12). In this sense, HRM seems to be at least as efficient as SAPA, since seven HRM units, roughly equivalent in mass to three SAPA units, displayed t␤ of 24 h versus 18 h for the TS-3R. The proline-rich repeated extensions, as those analyzed here, are likely to be open extended structures exposing a maximum surface area per residue (50). This molecular framework is usually associated, among other functions, with the insertion of binding motifs (50, 51) or with the conformation of repetitive

epitopes involved in immune evasion, a common theme among pathogens (52–54). Here we present evidence indicating that peptidic motifs allowing permanence in blood of the associated proteins can also be embedded in this structure. To search for the requirements concerning the proline spacing, a construction where the prolines were located at 10 residues distance was designed (HRM-P). This repeat retained the ability to extend the half-life of TS in blood (Fig. 4B), showing that this distance between proline residues works properly for this function. It should be emphasized, however, that neither the presence of the proline-rich framework nor the overall negative charge of the repetitive sequence guarantees the stabilization phenomenon described here, since molecules combining both attributes such as the intracellular T. cruzi antigens Tc1, Tc30, and Tc36 (12) and the hydrophilic extension of T. carassii TS (Fig. 5) did not significantly extend the TS half-life. Taken together, our results point out that sequences enriched in hydrophilic and negatively charged residues and inserted in a proline-rich framework constitute likely candidates to create chimeric proteins with extended half-life in blood. In support of this hypothesis, the extensions from the sialidases of A. viscosus and S. pneumoniae that display these features were able to extend the TS permanence in blood although they slightly resemble the SAPA, Tc13, or HRM molecules (Fig. 5). By another hand, the assayed E. coli EspF-derived domain that partially address these requirements have an intermediate effect; meanwhile, the negative charged extension from the TS of T. carassii not having any homology did not works at all. In addition, these findings suggest that the acquisition of an extended domain to improve the pharmacokinetic performance of the virulence factors might constitute a convergent strategy displayed by several pathogens. The approach followed here can be proposed as a viable route to increase the half-life of proteins in blood through the use of recombinant fusion proteins that can be readily expressed in microbial systems with the potential for scaling up production. Besides, the high metabolic rate of blood proteins in mice (55)

Improving Pharmacokinetics of Proteins suggests an even better performance of these repeats in humans or cattle. Acknowledgments—We are indebted to Drs. R. Schauer (Biochemisches Institut, Universita¨ t Kiel, Kiel, Germany) and P. Roggentin (Federal Dairy Research Centre, Kiel, Germany) for generously providing the plasmid containing the gene encoding the A. viscosus sialidase; to Dr. G. Gutkind and Dr. C. Nowicki (Facultad de Farmacia y Bioquı´mica, Universidad de Buenos Aires) for providing S. pneumoniae and the plasmid encoding the TAT, respectively; to Dr. M. Rivas (Servicio Fisiopatogenia INEI-ANLIS “Dr. Carlos G. Malbra´ n,” Buenos Aires, Argentina) for the E. coli O145:NM enteropathogenic strain; and Drs. A. C. C. Frasch and A. Ja¨ ger (IIB-UNSAM) for providing a genomic sample from T. carassii. We also appreciate the collaboration of Dr. J. Caramelo (Instituto Luis F. Leloir, Buenos Aires, Argentina) in computational analysis of the proteins and image capture and F. Fraga in taking care of the animals. REFERENCES 1. Peters, T. (1996) All about Albumin: Biochemistry, Genetics, and Medical Applications, pp. 188 –284, Academic Press, Inc., San Diego 2. Monfardini, C., and Veronese, F. M. (1998) Bioconjugate Chem. 9, 418 – 450 3. Rifai, A., Fadden, K., Morrison, S. L., and Chintalacharuvu, K. R. (2000) J. Exp. Med. 191, 2171–2182 4. Ni, H., Blajchman, M. A., Ananthanarayanan, V. S., Smith, I. J., and Sheffield, W. P. (2000) Thromb. Res. 99, 407– 415 5. Chitlaru, T., Kronman, C., Zeevi, M., Kam, M., Harel, A., Ordentlich, A., Velan, B., and Shafferman, A. (1998) Biochem. J. 336, 647– 658 6. Chitlaru, T., Kronman, C., Velan, B., and Shafferman, A. (2001) Biochem. J. 354, 613– 625 7. Hopkins, C. R., and Trowbridge, I. S. (1983) J. Cell Biol. 97, 508 –521 8. French, A. R., Tadaki, D. K., Niyogi, S. K., and Lauffenburger, D. A. (1995) J. Biol. Chem. 270, 4334 – 4340 9. Ghetie, V., and Ward, E. S. (2000) Annu. Rev. Immunol. 18, 739 –766 10. Fallon, E. M., Liparoto, S. F., Lee, K. J., Ciardelli, T. L., and Lauffenburger, D. A. (2000) J. Biol. Chem. 275, 6790 – 6797 11. Kim, J. K., Tsen, M. F., Ghetie, V., and Ward, E. S. (1994) Eur. J. Immunol. 24, 542–548 12. Buscaglia, C. A., Alfonso, J., Campetella, O., and Frasch, A. C. (1999) Blood 93, 2025–2032 13. Veronese, F. M., and Harris, J. M. (2002) Adv. Drug Deliv. Rev. 54, 453– 456 14. Markwardt, F., Richter, M., Walsmann, P., Riesener, G., and Paintz, M. (1990) Biomed. Biochim. Acta 49, 1103–1108 15. Ghetie, V., Popov, S., Borvak, J., Radu, C., Matesoi, D., Medesan, C., Ober, R. J., and Ward, E. S. (1997) Nat. Biotechnol. 15, 637– 640 16. Chapman, A. P., Antoniw, P., Spitali, M., West, S., Stephens, S., and King, D. J. (1999) Nat. Biotechnol. 17, 780 –783 17. Cohen, O., Kronman, C., Chitlaru, T., Ordentlich, A., Velan, B., and Shafferman, A. (2001) Biochem. J. 357, 795– 802 18. Sarkar, C. A., Lowenhaupt, K., Horan, T., Boone, T. C., Tidor, B., and Lauffenburger, D. A. (2002) Nat. Biotechnol. 20, 908 –913 19. Ashkenazi, A., and Chamow, S. M. (1997) Curr. Opin. Immunol. 9, 195–200 20. Syed, S., Schuyler, P. D., Kulczycky, M., and Sheffield, W. P. (1997) Blood 89, 3243–3252 21. Osborn, B. L., Sekut, L., Corcoran, M., Poortman, C., Sturm, B., Chen, G., Mather, D., Lin, H. L., and Parry, T. J. (2002) Eur. J. Pharmacol. 456, 149 –158 22. Frick, I. M., Wikstrom, M., Forsen, S., Drakenberg, T., Gomi, H., Sjobring, U., and Bjorck, L. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 8532– 8536

3381

23. Talay, S. R., Grammel, M. P., and Chhatwal, G. S. (1996) Biochem. J. 315, 577–582 24. Dennis, M. S., Zhang, M., Meng, Y. G., Kadkhodayan, M., Kirchhofer, D., Combs, D., and Damico, L. A. (2002) J. Biol. Chem. 277, 35035–35043 25. Frasch, A. C. (2000) Parasitol. Today 16, 282–286 26. Campetella, O. E., Uttaro, A. D., Parodi, A. J., and Frasch, A. C. (1994) Mol. Biochem. Parasitol. 64, 337–340 27. Buscaglia, C. A., Campetella, O., Leguizamo´ n, M. S., and Frasch, A. C. (1998) J. Infect. Dis. 177, 431– 436 28. Leguizamo´ n, M. S., Mocetti, E., Garcia Rivello, H., Argibay, P., and Campetella, O. (1999) J. Infect. Dis. 180, 1398 –1402 29. Mucci, J., Hidalgo, A., Mocetti, E., Argibay, P. F., Leguizamo´ n, M. S., and Campetella, O. (2002) Proc. Natl. Acad. Sci. U. S. A. 99, 3896 –3901 30. Campetella, O., Sa´ nchez, D. O., Cazzulo, J. J., and Frasch, A. C. C. (1992) Parasitol. Today 8, 378 –381 31. Affranchino, J. L., Iba´ n˜ ez, C. F., Luquetti, A. O., Rassi, A., Reyes, M. B., Macina, R. A., Aslund, L., Pettersson, U., and Frasch, A. C. (1989) Mol. Biochem. Parasitol. 34, 221–228 32. Henningsen, M., Roggentin, P., and Schauer, R. (1991) Biol. Chem. HoppeSeyler 372, 1065–1072 33. Camara, M., Boulnois, G. J., Andrew, P. W., and Mitchell, T. J. (1994) Infect. Immun. 62, 3688 –3695 34. Agu¨ ero, F., Campo, V., Cremona, L., Jager, A., Di Noia, J. M., Overath, P., Sa´ nchez, D. O., and Frasch, A. C. (2002) Infect. Immun. 70, 7140 –7144 35. McNamara, B. P., and Donnenberg, M. S. (1998) FEMS Microbiol. Lett. 166, 71–78 36. Elliott, S. J., O’Connell, C. B., Koutsouris, A., Brinkley, C., Donnenberg, M. S., Hecht, G., and Kaper, J. B. (2002) Infect. Immun. 70, 2271–2277 37. Leguizamo´ n, M. S., Campetella, O. E., Gonza´ lez Cappa, S. M., and Frasch, A. C. (1994) Infect. Immun. 62, 3441–3446 38. Pollock, R. R., French, D. L., Metlay, J. P., Birshtein, B. K., and Scharff, M. D. (1990) Eur. J. Immunol. 20, 2021–2027 39. Alvarez, P., Leguizamo´ n, M. S., Buscaglia, C. A., Pitcovsky, T. A., and Campetella, O. (2001) Infect. Immun. 69, 7946 –7949 40. Frank, R., and Overwin, H. (1996) Methods Mol. Biol. 66, 149 –169 41. Leguizamo´ n, M. S., Russomando, G., Luquetti, A., Rassi, A., Almiron, M., Gonza´ lez-Cappa, S. M., Frasch, A. C., and Campetella, O. (1997) J. Infect. Dis. 175, 1272–1275 42. Yeung, M. K. (1993) Infect. Immun. 61, 109 –116 43. Yeh, P., Landais, D., Lemaitre, M., Maury, I., Crenne, J. Y., Becquart, J., Murry-Brelier, A., Boucher, F., Montay, G., Fleer, R., Hirel, P., Mayaux, J., and Klatzmann, D. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 1904 –1908 44. Adams, G. P., and Schier, R. (1999) J. Immunol. Methods 231, 249 –260 45. Kitamura, K., Takahashi, T., Yamaguchi, T., Noguchi, A., Takashina, K., Tsurumi, H., Inagake, M., Toyokuni, T., and Hakomori, S. (1991) Cancer Res. 51, 4310 – 4315 46. Pedley, R. B., Boden, J. A., Boden, R., Begent, R. H., Turner, A., Haines, A. M., and King, D. J. (1994) Br. J. Cancer 70, 1126 –1130 47. Clark, R., Olson, K., Fuh, G., Marian, M., Mortensen, D., Teshima, G., Chang, S., Chu, H., Mukku, V., Canova-Davis, E., Somers, T., Cronin, M., Winkler, M., and Wells, J. A. (1996) J. Biol. Chem. 271, 21969 –21977 48. Osborn, B. L., Olsen, H. S., Nardelli, B., Murray, J. H., Zhou, J. X., Garcia, A., Moody, G., Zaritskaya, L. S., and Sung, C. (2002) J. Pharmacol. Exp. Ther. 303, 540 –548 49. Gandhi, R. R., and Bell, D. R. (1992) Am. J. Physiol. 262, H999 –H1008 50. Williamson, M. P. (1994) Biochem. J. 297, 249 –260 51. Kay, B. K., Williamson, M. P., and Sudol, M. (2000) FASEB J. 14, 231–241 52. Verra, F., and Hughes, A. L. (1999) Mol. Biol. Evol. 16, 627– 633 53. McKean, P. G., Trenholme, K. R., Rangarajan, D., Keen, J. K., and Smith, D. F. (1997) Mol. Biochem. Parasitol. 86, 225–235 54. Hoft, D. F., Kim, K. S., Otsu, K., Moser, D. R., Yost, W. J., Blumin, J. H., Donelson, J. E., and Kirchhoff, L. V. (1989) Infect. Immun. 57, 1959 –1967 55. Mordenti, J., Chen, S. A., Moore, J. A., Ferraiolo, B. L., and Green, J. D. (1991) Pharm. Res. (N. Y.) 8, 1351–1359