Arginine 121 is a crucial residue for the specific cytotoxic activity of the ribotoxin a-sarcin Manuel Masip, Javier Lacadena*, Jose ´ Miguel Manchen ˜ o, Mercedes On ˜ aderra, Antonio Martı ´ nez-Ruiz†, A ´ lvaro Martı ´ nez del Pozo and Jose ´ G. Gavilanes Departamento de Bioquı ´ mica y Biologı ´ a Molecular, Facultad de Quı ´ mica, Universidad Complutense, Madrid, Spain. a-Sarcin, a cyclizing ribonuclease secreted by the mould Aspergillus giganteus, is one of the best characterized members of a family of fungal ribotoxins. This protein induces apoptosis in tumour cells due to its highly specific activity on ribosomes. Fungal ribotoxins display a three- dimensional protein fold similar to those of a larger group of microbial noncytotoxic RNases, represented by RNases T1 and U2. This similarity involves the three catalytic residues and also the Arg121 residue, whose counterpart in RNase T1, Arg77, is located in the vicinity of the substrate phos- phate moiety although its potential functional role is not known. In this work, Arg121 of a-sarcin has been replaced by Gln or Lys. These two mutations do not modify the conformation of the protein but abolish the ribosome- inactivating activity of a-sarcin. In addition, the loss of the positive charge at that position produces dramatic changes on the interaction of a-sarcin with phospholipid membranes. It is concluded that Arg121 is a crucial residue for the characteristic cytotoxicity of a-sarcin and presumably of the other fungal ribotoxins. Keywords: Aspergillus protein; cytotoxin; protein –lipid interaction; ribotoxin; a-sarcin. a-Sarcin is the best characterized member of a family of fungal cytotoxic ribosome-inactivating proteins (ribotoxins) that cleave one single phosphodiester bond at an evolu- tionarily conserved sequence of the larger rRNA [1], impair- ing the binding of elongation factors EF-1 and EF-2 [2]. This well known exquisite ribonucleolytic action in cell-free systems has been also recently observed in intact cells, namely human rhabdomyosarcoma cells, where the toxic effect produces cell death via apoptosis [3]. a-Sarcin is internalized by the target cells via endocytosis involving acidic endosomes and the Golgi [3,4]. This traffic might be related to the known ability of a-sarcin to interact with model membranes. In fact, besides its specific ribonuclease activity, this highly polar protein binds to acidic lipid vesicles promoting their aggregation and fusion [5,6] and resulting in translocation of the protein across the model membranes [7,8]. Ribotoxins of the a-sarcin family display a high degree of sequence similarity although they are secreted by a variety of different moulds [9–11], i.e. restrictocin and mitogillin, two other members of this family, display 85% sequence identity to a-sarcin, which is only one residue longer [12–14]. These ribotoxins belong to a larger group of micro- bial extracellular RNases, represented by the noncytotoxic RNases T1 and U2 [10,11,15], based on their sequence and three-dimensional structure similarities. Thus, a-sarcin [15], restrictocin [16] and RNase T1 [17] display an identical architecture and connectivity of the secondary structure elements in the three proteins. The catalytic mechanisms of a-sarcin and RNase T1 are similar [18–20] and related to that of bovine pancreatic RNase A [21]. They behave as cyclizing RNases, the overall reaction being composed of two steps, cyclization followed by cleavage of the 2 0 ,3 0 - cyclic product formed [18,21–23]. In RNase T1, His92 acts as the general acid and Glu58 as the general base during the first step of the reaction. The hydrolysis of the cyclic derivative is performed by the same groups, but their roles are reversed [23]. In a-sarcin the same roles are fulfilled by His137 and Glu96, although differences between the two enzymes regarding the individual pK a values of these catalytic residues and optimum pH have been observed [19,20]. Analysis of the three dimensional structure of RNase T1 complexed with different inhibitors or substrate analogues has revealed the presence of some other side chains in the vicinity of the substrate phosphate moiety [22,24]. This is the case for His40 and Arg77, their counter- parts in a-sarcin being His50 and Arg121 [10,11] occupying geometrically conserved positions [15,25] (Fig. 1). Site- directed mutagenesis experiments have revealed that His40 Correspondence A. Martı ´ nez del Pozo or J. G. Gavilanes, Departamento de Bioquı ´ mica y Biologı ´ a Molecular I, Facultad de Quı ´ mica, Universidad Complutense, 28040 Madrid, Spain. Fax: 1 34 91 3944159, Tel.: 1 34 91 3944158, E-mail: alvaro@bbm1.ucm.es or ppgf@bbm1.ucm.es Enzymes: mitogillin and restrictocin (RNMG_ASPRE, P04389, EC 3.1.27 ); ribonuclease A (RNP_BOVIN, P00656, EC 3.1.27.5); ribonuclease T1 (RNT1_ASPOR, P00651, EC 3.1.27.3); ribonuclease U2 (RNU2_ASTSP, P00654, EC 3.1.27.4); a-sarcin (RNAS_ASPGI, P00655, EC 3.1.27.10). All accession codes are for the SWISS-PROT data base. *Present address: Facultad de Biologı ´ a, Universidad SEK, 40003 Segovia, Spain. †Present address: Centro de Investigaciones Biolo ´ gicas, CSIC, Vela ´ zquez 144, 28006 Madrid, Spain. Note: a web page is available at http://www.bbm1.ucm.es (Received 2 July 2001, revised 4 September 2001, accepted 1 October 2001) Abbreviations: ApA, adenylyl(3 0 !5 0 )adenosine; Myr 2 GroPGro, dimyristoylglycerophosphoglycerol; Nbd-Myr 2 GroP Etn, [N-(7-nitro- 2-1,3-benzoxadiazol-4-yl)-dimyristoylglycerophosphoethanolamine; rhodamine-GroP Etn, N-(lissamine rhodamine B sulfonyl)- diacylglycerophosphoethanolamine; RNase, ribonuclease. Eur. J. Biochem. 268, 6190–6196 (2001) q FEBS 2001 of RNase T1 [23,26,27] and His50 of a-sarcin [20] are involved in catalysis rather than in substrate binding. The role of Arg77 has not yet been established as all attempts to isolate any RNase T1 Arg77 mutant have been unsuccessful [23]. Hydroxylamine treatment has been used to produce point mutations in recombinant mitogillin produced by yeast [28]. Only a few transformants survived upon induction, suggesting that the affected residues, one of them Arg120 (the counterpart of Arg121 in a-sarcin), has a crucial role in the ribonucleolytic catalysis [28]. The work herein presented deals with the production, purification and characterization of two a-sarcin mutants where Arg121 has been substituted by Lys (R121K) or Gln (R121Q). The results obtained show that this residue is involved in the specific ribonucleolytic activity of ribotoxins against ribosomes and also in the interaction with membranes. EXPERIMENTAL PROCEDURES DNA manipulations All materials and reagents were of molecular biology grade. Cloning procedures and bacteria manipulations were carried out according to standard methods [29] as described pre- viously [20,30,31]. Oligonucleotide site-directed mutagen- esis was used to replace Arg121 with Lys (R121K) or Gln (R121Q) as described previously [20,30–32]. The muta- genic primers used were 5 0 -CCTGGCCCGGCGAAGGTCA TCTACACC-3 0 for R121K and 5 0 -CCTGGCCCGGCG CAGGTCATCTACACC-3 0 for R121Q (the site of mutation is underlined). The E. coli strains used were BW313 {(HfrKL16 pol45 [LysA (61– 62)] dut1 ung1 thi1 relA1] to obtain the uridine-rich ssDNA, DH5aF 0 ({[F 0 ] endA1 hsdR17 (r – K m – K ) supE44 thi-1 recA1 gyrA (NaI R ) relA1 D(lac- ZYA-argF ) U169 deoR [B80 dLac D(lacZ ) M15]}) to obtain the expression constructs, and BL21(DE3)(F 0 ompT[lon] hsd B (r – B m – B )) to produce the proteins. The thioredoxin producing plasmid (pT-Trx) [33] was a generous gift of S. Ishii, from the Riken Tsukuba Life Science Center (Riken, Japan). Proteins production and purification BL21(DE3) cells cotransformed with pT-Trx and the corre- sponding a-sarcin mutant plasmid were used to produce the mutant proteins, which were purified as described for the wild-type protein [30,34]. Fungal wild-type a-sarcin was produced and purified as previously reported [30,35,36]. Polyacrylamide electrophoresis, tryptic digestions, acid hydrolysis of proteins, peptide maps, and amino-acid analysis were also performed according to standard pro- cedures described previously [20,30,35,37]. Spectroscopic characterization Proteins were dissolved in either 50 m M sodium phosphate, pH 7.0, or 50 m M sodium acetate, pH 5.0, both containing 0.1 M NaCl, as required and centrifuged at 14 000 g for 5 min. Absorbance measurements were carried out on a Uvikon 930 spectrophotometer at 100 nm : min 21 scanning speed, at room temperature, and in 1-cm optical-path cells. CD spectra were obtained on a Jasco 715 spectropolarimeter Fig. 1. Diagrams corresponding to the three-dimensional structure of a-sarcin constructed from the atomic coordinates deposited in the PDB (accession code [1DE3], http://www.rcsb.org/pdb). Images were generated by the MOLMOL program [52]. (A), Ribbon diagram of the whole protein molecule where selected residues (His50, Glu96, His137, Arg121, in black; Tyr48, Trp51, Tyr106, in grey) are shown. (B), Detail of the residues not labelled in part A. (C), Active site residues of a-sarcin (black)/RNase T1 (grey), after fitting of the backbone atoms of the residues involved in the five-strands central b sheet of both proteins. The protein view is maintained through the three diagrams. q FEBS 2001 Role of Arg121 on cytotoxicity of a-sarcin (Eur. J. Biochem. 268) 6191 at 0.2 nm : s 21 scanning speed; 0.1- and 1.0-cm optical-path cells were used in the far- and near-UV wavelength range, respectively. Mean residue weight ellipticities were expressed in units of degrees : cm 2 : dmol 21 . Extinction coef- ficients E 0.1% (280 nm, 1-cm optical path) were calculated from the corresponding absorbance spectra and amino-acid analyses. Thermal denaturation profiles were obtained by measuring the temperature dependence of the ellipticity at 220 nm in the range of 25–85 8C; the temperature was continuously changed at a rate of 0.5 8C : min 21 . The tem- perature values at the midpoint of the thermal transition (T m ) were calculated assuming a two-state unfolding mechanism [38–40]. Fluorescence emission spectra were obtained on a SLM Aminco 8000 spectrofluorimeter at 25 8C in 0.2-cm optical-path cells, as described previously [20]. Ribonucleolytic activity The specific ribonucleolytic activity of a-sarcin against intact ribosomes was followed by detecting the release of the 400 nucleotide a-fragment [1,41,42] from a cell-free reticulo- cyte lysate (Promega) as described previously [20,30,31]. The production of the 400 nucleotide a-fragment was visualized by ethidium bromide staining after electrophor- esis on 2.4% agarose. The activity of the purified proteins against polyadenylic acid was assayed after subjecting the samples to an electrophoretic procedure in 15% polyacryl- amide gels containing 0.1% SDS and 0.3 mg : mL 21 poly(A). This method, designated as a zymogram, was based on one previously described [31,43]. Proteins exhibiting ribonuclease activity appear as colourless bands after proper destaining treatment. Hydrolysis of adenylyl(3 0 !5 0 )adenosine (ApA) by a-sarcin and the mutant variants was performed as described elsewhere [18] at pH 5.0. The reaction products were resolved by HPLC with a phosphate-methanol gradient as described [18,44]. All of these assays were performed with the corresponding controls to test potential nonspecific degradation of the substrates, which did not occur under the conditions used. Protein–lipid interactions Dimyristoylglycerophosphoglycerol (Myr 2 GroPGro) was purchased from Avanti Polar Lipids Inc. Vesicles were formed by hydrating a dry lipid film with Tris buffer (15 m M Tris, pH 7.5, containing 0.1 M NaCl and 1 mM EDTA) for 60 min at 37 8C. The lipid suspension was subjected to five cycles of extrusion through two stacked 0.1 mm (pore diameter) polycarbonate membranes [45]. The average diameter of the vesicle population was 100 nm (85% of the vesicles in the range 75–125 nm), as determined by electron microscopy studies [45]. Phospholipid concentration was determined as described [46]. Aggregation of phospholipid vesicles was monitored by measuring the increase in the absorbance at 360 nm of a suspension of Myr 2 GroPGro vesicles in Tris buffer (30 m M final lipid concentration) after addition of a small aliquot of a freshly prepared solution of protein [5]. Intermixing of membrane lipids was measured by fluorescence energy transfer assays [47] as described [45,48]. A decrease in the donor-to-acceptor, [N-(7-nitro-2-1,3-benzoxadiazol-4-yl)-dimyristoylglycero- phosphoethanolamine (Nbd-Myr 2 GroP Etn) and N-(lissamine rhodamine B sulfonyl)-diacylglycerophosphoethanolamine (rhodamine-GroP Etn), respectively, fluorescence energy transfer indicates lipid-mixing between membranes. Bind- ing of the proteins to the lipid vesicles was analysed by measuring the free protein concentration in the supernatant obtained by centrifugation (160 000 g for 20 min, Beckman Airfuge) of protein –vesicle mixtures at different protein/ lipid ratios. In this particular case, the buffer used was 15 m M Mops, pH 7.0, containing 0.1 M NaCl and 1 mM EDTA. The protein concentration was calculated by densitometering the Coomasie-stained SDS-gels resulting from subjecting supernatant aliquots to SDS/PAGE. A calibration plot, volumogram obtained (density or quantity of a spot calculated from its volume made of the sum of all pixel intensities composing the spot obtained with a photo documentation system UVI-Tec using the software facility UVISOFT UVI BAND) vs. protein concentration (determined by amino-acid analysis of acid-hydrolysed protein samples) was used for these calculations. Other experimental details were as reported previously [45,48,49]. Control assays without protein were always performed. RESULTS Purification and structural characterization Both mutant variants of a-sarcin, R121K and R121Q, were purified to homogeneity according to their behaviour on SDS/PAGE, and displayed a single immunoreactive band when stained with anti-(a-sarcin) Ig in Western-blots ana- lysis. Yields were about 5.0 mg per litre of E. coli culture for the R121K variant but much lower for R121Q (less than 0.3 mg per L). Amino-acid analysis as well as tryptic map of the purified mutants were consistent with the residue substitutions planned. The calculated E 0.1% (280 nm, 1-cm optical path) were 1.30 and 1.12 for R121K and R121Q, respectively, in reasonable agreement with the value of 1.34 reported for the wild-type protein [37]. The far-UV CD spectra of both mutant variants were indistinguishable from that already reported for the natural fungal a-sarcin [20,30,36,37] (Fig. 2A). Small differences were observed in the near-UV wavelength range (Fig. 2B). Only minor differences were also observed in terms of protein Fig. 2. Circular dichroism spectra of fungal wild-type a-sarcin (X) and its R121K (W) and R121Q (K) mutant variants. (A) Far- and (B) near-UV. Mean residue weight ellipticities (u MRW ) are expressed in units of degrees : cm 2 : dmol 21 . 6192 M. Masip et al. (Eur. J. Biochem. 268) q FEBS 2001 fluorescence emission. Thus, the fluorescence emission of R121K coincided with that of the wild-type protein while R121Q showed identical Trp emission but an increased (1.5-fold) tyrosine contribution (Fig. 3). R121K and R121Q displayed a decreased stability in comparison to the wild-type protein. Denaturation of these proteins was studied by analysing the thermal variation of their far-UV circular dichroism properties (Fig. 4). The T m value for the wild-type protein was 62 8C at pH 5.0, the optimum pH for both enzyme activity and stability [19,20,50]. The T m of the R121K and R121Q mutant variants was 56 8C and 53 8C, respectively, which corre- sponded to a decreased stability of 8.36 and 12.54 kJ : mol 21 , respectively, according to the calculated DDG values. Enzymatic characterization Neither of the two mutants produced the a-fragment result- ing from the specific ribonucleolytic activity of these fungal ribotoxins against ribosomes. Forty nanograms of the wild- type protein was enough to completely release the a-fragment from the larger rRNA under the standard assay conditions [20,30,31], but no cleavage was observed even when 100 ng of the mutant variants were assayed. In addition, although wild-type a-sarcin degrades polyadenylic acid on a zymo- gram assay, R121K and R121Q did not cleave the homo- polynucleotide even when 500 ng of protein were assayed, the lowest detection limit of this assay being 50 ng of wild- type protein [20]. A quantitative analysis of the ribonuclease activity of a-sarcin has been developed by using the dinucleotide ApA as substrate [18]. Both mutant variants hydrolysed ApA. They displayed a K m value similar to that of the wild-type protein although they exhibited a lower catalytic efficiency (Table 1), thus suggesting that the mutated residue is involved in catalysis rather than in substrate binding. Interaction with phospholipid vesicles The interaction of a-sarcinwithmodelphospholipid vesicles through electrostatic and hydrophobic interactions is well documented. The ribotoxin promotes aggregation of vesicles and mixing of phospholipids from different bilayers [5,6]. The mutant variant R121K promoted the same effects than the wild-type protein and these became saturated at the same protein/phospholipid molar ratio, although R121K displayed a slightly higher affinity for the target vesicles (1.3-fold that of the wild-type protein) (Fig. 5). However, Fig. 3. Fluorescence emission spectra of wild-type a-sarcin (WT) and its R121K and R121Q mutant variants. (1), Fluorescence emission spectra for excitation at 275 nm; (2), fluorescence emission spectra for excitation at 295 nm (tryptophan contribution) normalized at wavelengths above 380 nm where the tyrosine emission is negligible; (3) calculated difference spectra of (1) minus (2) (tyrosine contribution). Fluorescence emission is expressed in arbitrary units considering the intensity at the wavelength of the emission maximum of the wild-type protein for excitation at 275 nm as 1.0. Fig. 4. Thermal denaturation profiles of fungal wild-type a-sarcin (X) and its R121K (W)andR121Q(K)mutantvariants. Measurements were performed by continuously recording the mean residue weight ellipticity at 220 nm, expressed in units of degrees : cm 2 : dmol 21 . Table 1. Activity of a-sarcin and its mutant variants against ApA at pH 5.0. Kinetic parameters (^ SD) determined from the transesterifi- cation of ApA by linear regression analysis of double reciprocal plots from three different determinations [18]. Protein K m (mM) k cat (s 21 ) k cat /K m (M 21 : s 21 ) Wild-type 40 ^ 4 (27.0 ^ 1.0) Â 10 25 6.7 ^ 0.7 R121K 36 ^ 4 (1.2 ^ 0.2) Â 10 25 0.3 ^ 0.1 R121Q 27 ^ 4 (7.9 ^ 0.3) Â 10 25 2.3 ^ 0.4 q FEBS 2001 Role of Arg121 on cytotoxicity of a-sarcin (Eur. J. Biochem. 268) 6193 R121Q exhibited a low affinity, 0.2-fold that of the wild- type a-sarcin, which corresponded to a very low affinity to aggregate vesicles and promote lipid-mixing between bilayers (Fig. 5). DISCUSSION Both purified mutant forms of a-sarcin display the same global fold as the wild-type protein according to the spec- troscopic and structural characterization performed. Thus, the modified activities of these mutants to cleave RNA substrates or to interact with phospholipid vesicles cannot be considered as a consequence of an altered native confor- mation. In fact, no secondary structure variations have been detected and only slight differences were observed in terms of extinction coefficient and environment of aromatic amino acids (near-UV circular dichroism and fluorescence emission spectra). Tyr48, Trp51 and Tyr106 are located in the vicinity of the mutated 121st position (their a-carbons are about 11 A ˚ distant in the same side of the central b sheet of the protein) (Fig. 1) [15]. Thus, such differences might be attributed to changes in the local environment of these residues. In fact, the side-chain of Trp51 has been proven to be responsible for most of the ellipticity signal of a-sarcin observed in the near-UV wavelength region but it does not show fluorescence emission [51]. Therefore, the near-UV ellipticity variations observed for both mutant variants arise from local changes affecting Trp51, while the change on the fluorescence of R121Q, 1.5-fold increased Tyr emission, is related to Tyr48 and/or Tyr106. Both mutant variants were isolated in a very different yield (< 5mg : L 21 and < 0.3 mg : L 21 for R121K and R121Q, respectively), but this cannot be related to a differ- ent protein stability. Although mutation results in a less stable protein form, the stability change DDG is similar in both cases, 28.36 and 212.54 kJ : mol 21 for R121K and R121Q, respectively. The cytotoxic action of a-sarcin can be dissected into two different steps. First, the protein must enter the cells, crossing the phospholipid bilayer barrier. This inter- nalization occurs in the target cells by endocytosis [3], which would require a membrane interaction. Further, the toxin cleaves ribosomes, impairing protein biosynthesis and producing cellular death via apoptosis [3]. Conse- quently, both abilities have been studied for R121Q and R121K. Substitution of Arg121 by Lys or Gln abolishes the activity of a-sarcin to release the a-fragment from ribosomes. This lack of ribonuclease activity against the specific substrate is also observed against the homopoly- nucleotide poly(A). These results indicate that this residue plays an important role in the rRNA catalytic cleavage by a-sarcin, which would not be simply related to the bearing positive charge. Both mutant variants of a-sarcin cleave ApA, with slight variation in the K m value but k cat being reduced by about one order of magnitude (Table 1). A very similar result has been obtained before for the H50Q variant, while mutation of either Glu96 or His137, residues acting as the general acid and base on the catalysis by a-sarcin [19,20], rendered proteins with no detectable activity against ApA [20]. Cleavage of ApA by a-sarcin is a low-specificity reaction requiring high amounts of protein and long incu- bation times [18]. Thus, using this dinucleotide substrate only allows us to obtain information about the minimal requirements needed for a-sarcin to cleave a phosphodiester bond but it appears to be a more sensitive assay than those using rRNA or poly(A). From this point of view, this Arg residue although not essential for catalysis would contribute to position the substrate in the optimum conformation for cleavage. In fact, in the crystal structure of restrictocin the active site residues His49, Glu95, Arg120 and His136 (equivalent to His50, Glu96, Arg121 and His137 in a-sarcin) cluster together and point towards a tetrahedral-shaped electron density which was interpreted as an inorganic phosphate group derived from the crystallization buffer [16]. A very similar situation occurs with Arg77 in the active site of RNase T1, where the phosphate group can accept hydrogen bonds from Arg77(N1H/NhH) [22– 24], although its role in catalysis has not been elucidated because no mutant forms at this position have been obtained [23]. Thus, Fig. 5. Effect of fungal wild-type a-sarcin (X) and its R121K (W) and R121Q (K) mutant variants on phosphatidylglycerol vesicles. (A), Protein binding to the vesicles. Protein bound is expressed as percentage of the total protein present in the assay vs phospholipid/ protein molar ratio. (B), Aggregation of lipid vesicles. Aggregated vesicles scatter more light than the un-aggregated vesicles and the process can be measured from the resulting apparent DA 360 promoted by the proteins in a vesicle suspension. Results are expressed as relative DA 360 values, considering the maximum increase produced by the wild- type protein as unit, vs. protein/phospholipid molar ratio. (C), Lipid- mixing from different bilayers. This assay is performed with a mixture of fluorescence-labelled and unlabelled vesicles (1 : 9 ratio, labelled to unlabelled, respectively). The lipid mixing results in dilution of the two fluorescence probes and consequently in decrease of the relative fluorescence energy transfer (RET). Results are expressed as relative RET values, considering the maximum decrease produced by the wild- type protein as unit, vs. protein/phospholipid molar ratio. In all cases, the results shown are the average of three independent experiments. 6194 M. Masip et al. (Eur. J. Biochem. 268) q FEBS 2001 in the absence of the Arg guanidinium group, the protein would not be able to accommodate polymeric RNA, such as rRNA or poly(A), in the proper conformation within the active site. ApA, a much smaller substrate, would fit more easily into the cleavage pocket although not under optimal structural constraints. This would explain the observed reduction in k cat . In fact, it has been already proposed that the mechanism of action of a-sarcin against ribo- somes might differ in some details from that against dinuc- leotides [18,19]. The behaviour of a-sarcin against ApA as a function of pH, altogether with the characterization of the individual pK a values of the active site residues, were consistent with the existence of two mechanisms for the hydrolysis of ApA with different optimum pH and different orientations for the dinucleotide within the active site [19]. The ability of R121K and R121Q variants to interact with phospholipid model membranes was studied by analysing their ability in aggregating vesicles and mixing phospho- lipids from different vesicles. The results obtained indicate that the loss of a positive charge in the position corre- sponding to Arg121 side-chain has a dramatic effect on the a-sarcin–membrane interaction. Regarding to this, the region around Trp51, located at around 11 A ˚ from position 121, is involved in the interaction with the lipid bilayer, as this Trp residue is located in the hydrophobic core of the bilayer following protein– vesicle interaction [51]. The membrane affinity of R121Q is largely diminished, while that of R121K is not modified (Fig. 5A). The decreased binding affinity of the R121Q mutant is correlated with a reduced ability to aggregate lipid vesicles that requires the formation of vesicle–protein–vesicle complexes [45]. Conse- quently, the lipid-mixing from different bilayers within the vesicle aggregates also occurs in a very reduced extent (Fig. 5C). From the results obtained, it can be concluded that Arg121 of a-sarcin, a conserved residue in its fungal ribo- toxins family, plays a crucial role in the cytoxicity shown by this protein. Firstly, because this residue is required for the specific ribonuclease activity of the protein against ribo- somes. and secondly, because this residue is essential for the binding of the protein to the membranes which would be a required step in crossing the cell membrane barrier via endocytosis. This essential character of Arg121 for mem- brane binding may be a consequence of its essential role in the ribonucleolytic mechanism. The dramatic membrane affinity decrease due to the loss of only one positive charge in a protein with 24 basic residues and 17 acid residues [12] and its location at the ribonucleolytic active site may suggest some kind of relationship. It would not be unreasonable to assume that proteins evolved to interact with nucleic acids, such as RNases, would have developed structural deter- minants to recognize polyphosphate lattices, i.e. RNA, and this ability, in some cases, may allow a recognition of phospholipid surfaces, considered as two-dimensional phosphate networks. ACKNOWLEDGEMENTS This work has been supported by grant BMC2000-0551 from Direccio ´ n General de Investigacio ´ n (Ministerio de Ciencia y Tecnologı ´ a), Spain. M. M. is recipient of a fellowship from Ministerio de Educacio ´ n, Cultura y Deporte, Spain. REFERENCES 1. Schindler, D.G. & Davies, J.E. (1977) Specific cleavage of ribo- somal RNA caused by a-sarcin. Nucleic Acids Res. 4, 1097–1100. 2. Brigotti, M., Rambelli, F., Zamboni, M., Montanaro, L. & Sperti, S. (1989) Effect of a-sarcin and ribosome inactivating proteins on the interaction of elongation factors with ribosomes. Biochem. J. 257, 723–727. 3. Olmo, N., Turnay, J., de Buitrago, G.G., Lo ´ pez de Silanes, I., Gavilanes, J.G. & Lizarbe, M.A. (2001) Cytotoxic mechanism of the ribotoxin a-sarcin. Induction of cell death via apoptosis. Eur. J. Biochem. 268, 2113– 2123. 4. Turnay, J., Olmo, N., Jime ´ nez, J., Lizarbe, M.A. & Gavilanes, J.G. (1993) Kinetic study of the cytotoxic effect of a-sarcin, a ribosome inactivating protein from A. giganteus, on tumor cell lines: protein biosynthesis inhibition and cell binding. Mol. Cell. Biochem. 122, 39–47. 5. Gasset, M., Martı ´ nez del Pozo, A., On ˜ aderra, M. & Gavilanes, J.G. (1989) Study of the interaction between the antitumour protein a-sarcin and phospholipid vesicles. Biochem. J. 258, 569–575. 6. Gasset, M., On ˜ aderra, M., Thomas, P.G. & Gavilanes, J.G. (1990) Fusion of phospholipid vesicles produced by the anti-tumour protein a-sarcin. Biochem. J. 265, 815– 822. 7. On ˜ aderra, M., Manchen ˜ o, J.M., Gasset, M., Lacadena, J., Schiavo, G., Martı ´ nez del Pozo, A. & Gavilanes, J.G. (1993) Translocation of a-sarcin across the lipid bilayer of asolectin vesicles. Biochem. J. 295, 221–225. 8. On ˜ aderra, M., Manchen ˜ o, J.M., Lacadena, J., De los Rı ´ os, V., Martı ´ nez del Pozo, A. & Gavilanes, J.G. (1998) Oligomerization of the cytotoxin a-sarcin associated to phospholipid membranes. Mol. Membr. Biol. 15, 141–144. 9. Wirth, J., Martı ´ nez del Pozo, A., Manchen ˜ o, J.M., Martı ´ nez-Ruiz, A., Lacadena, J., On ˜ aderra, M. & Gavilanes, J.G. (1997) Sequence determination and molecular characterization of gigantin, a cyto- toxic protein produced by the mould Aspergillus giganteus IFO 5818. Arch. Biochem. Biophys. 343, 188–193. 10. Martı ´ nez-Ruiz, A., Kao, R., Davies, J. & Martı ´ nez del Pozo, A. (1999) Ribotoxins are a more widespread group of proteins within the filamentous fungi than previously believed. Toxicon 37, 1549–1563. 11. Martı ´ nez-Ruiz, A., Martı ´ nez del Pozo, A., Lacadena, J., On ˜ aderra, M. & Gavilanes, J.G. (1999) Hirsutelin A displays significant homology to microbial extracellular ribonucleases. J. Invertebr. Pathol. 74, 96–97. 12. Sacco, G., Drickamer, K. & Wool, I.G. (1983) The primary structure of the cytotoxin a-sarcin. J. Biol. Chem. 258, 5811–5818. 13. Lo ´ pez Otı ´ n, C., Barber, D., Ferna ´ ndez Luna, J.L., Soriano, F. & Me ´ ndez, E. (1985) The primary structure of the cytotoxin restrictocin. Eur. J. Biochem. 149, 621–634. 14. Ferna ´ ndez-Luna, J.L., Lo ´ pez Otı ´ n, C., Soriano, F. & Me ´ ndez, E. (1985) Complete amino acid sequence of the Aspergillus cytotoxin mitogillin. Biochemistry 24, 861 –867. 15. Pe ´ rez-Can ˜ adillas, J.M., Santoro, J., Campos-Olivas, R., Lacadena, J., Martı ´ nez del Pozo, A., Gavilanes, J.G., Rico, M. & Bruix, M. (2000) The highly refined solution structure of the cytotoxic ribonuclease a-sarcin reveals the structural requirements for substrate recognition and ribonucleolytic activity. J. Mol. Biol. 299, 1061–1073. 16. Yang, X.J. & Moffat, K. (1996) Insights into specificity of cleavage and mechanism of cell entry from the crystal structure of the highly specific Aspergillus ribotoxin, restrictocin. Structure 4, 837–852. 17. Pace, C.N., Heinemann, U., Hahn, U. & Saenger, W. (1991) Ribonuclease T1: structure, function and stability. Angew. Chem. Int. 30, 343 –360. 18. Lacadena, J., Martı ´ nez del Pozo, A., Lacadena, V., Martı ´ nez-Ruiz, A., Manchen ˜ o, J.M., On ˜ aderra, M. & Gavilanes, J.G. (1998) The q FEBS 2001 Role of Arg121 on cytotoxicity of a-sarcin (Eur. J. Biochem. 268) 6195 cytotoxin a-sarcin behaves as a cyclizing ribonuclease. FEBS Lett. 424, 46–48. 19. Pe ´ rez-Can ˜ adillas, J.M., Campos-Olivas, R., Lacadena, J., Martı ´ nez del Pozo, A., Gavilanes, J.G., Santoro, J., Rico, M. & Bruix, M. (1998) Characterization of pK a values and titration shifts in the cytotoxic ribonuclease a-sarcin by NMR. Relationship between electrostatic interactions, structure and catalytic function. Bio- chemistry 37, 15865–15876. 20. Lacadena, J., Martı ´ nez del Pozo, A., Martı ´ nez-Ruiz, A., Pe ´ rez- Can ˜ adillas, J.M., Bruix, M., Manchen ˜ o, J.M., On ˜ aderra, M. & Gavilanes, J.G. (1999) Role of histidine-50, glutamic acid-96 and histidine-137 in the ribonucleolytic mechanism of the ribotoxin a-sarcin. Proteins 37, 474–484. 21. del Cardayre ´ , S.B., Ribo ´ , M., Yokel, E.M., Quirk, D.J., Rutter, W.J. & Raines, R.T. (1995) Engineering ribonuclease A: production, purification and characterization of wild-type enzyme and mutants at Gln11. Protein Eng. 8, 261–273. 22. Hahn, U. & Heinemann, U. (1994) Structure determination, modeling and site-directed mutagenesis studies. In Concepts in protein engineering and design (Wrede, P. & Schneider, G., eds), pp. 109–168. Walter de Gruyter GmbH & Co, Berlin, Germany. 23. Steyaert, J. (1997) A decade of protein engineering on ribonuclease T1. Atomic dissection of the enzyme–substrate interactions. Eur. J. Biochem. 247, 1– 11. 24. Zegers, I., Loris, R., Dehollander, G., Haikal, A.F., Poortmans, F., Steyaert, J. & Wyns, L. (1998) Hydrolysis of a slow cyclic thiophosphate substrate of RNase T1 analyzed by time-resolved crystallography. Nat. Struct. Biol. 5, 280–283. 25. Campos-Olivas, R., Bruix, M., Santoro, J., Martı ´ nez del Pozo, A., Lacadena, J., Gavilanes, J.G. & Rico, M. (1996) Structural basis for the catalytic mechanism and substrate specificity of the ribonuclease a-sarcin. FEBS Lett. 399, 163 –165. 26. Doumen, J., Gonciarz, M., Zegers, I., Loris, R., Wyns, L. & Steyaert, J. (1996) A catalytic function for the structurally conserved residue Phe 100 of ribonuclease T1. Protein Sci. 5, 1523–1530. 27. Loverix, S., Laus, G., Martins, J.C. & Steyaert, J. (1998) Recon- sidering the energetics of ribonuclease catalysed RNA hydrolysis. Eur. J. Biochem. 257, 286 – 290. 28. Kao, R., Shea, J.E., Davies, J. & Holden, D.W. (1998) Probing the active site of mitogillin, a fungal ribotoxin. Mol. Microbiol. 29, 1019–1027. 29. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 30. Lacadena, J., Martı ´ nez del Pozo, A., Barbero, J.L., Manchen ˜ o, J.M., Gasset, M., On ˜ aderra, M., Lo ´ pez-Otı ´ n, C., Ortega, S., Garcı ´ a, J.L. & Gavilanes, J.G. (1994) Overproduction and purification of biologically active native fungal a-sarcin in Escherichia coli. Gene 142, 147–151. 31. Lacadena, J., Manchen ˜ o, J.M., Martı ´ nez-Ruiz, A., Martı ´ nez del Pozo, A., Gasset, M., On ˜ aderra, M. & Gavilanes, J.G. (1995) Substitution of histidine-137 by glutamine abolishes the catalytic activity of the ribosome-inactivating protein a-sarcin. Biochem. J. 309, 581–586. 32. Kunkel, T.A., Roberts, J.D. & Zakour, R.A. (1987) Rapid and efficient site-specific mutagenesis without phenotypic selection. Methods Enzymol. 154, 367–382. 33. Yasukawa, T., Kanei-Ishii, C., Mackaura, T., Fujimoto, J., Yamamoto, T. & Ishii, S. (1995) Increase of solubility of foreign proteins in Escherichia coli by coproduction of the bacterial thioredoxin. J. Biol. Chem. 270, 25328– 25331. 34. Garcı ´ a-Ortega, L., Lacadena, J., Lacadena, V., Masip, M., De Antonio, C., Martı ´ nez-Ruiz, A. & Martı ´ nez del Pozo, A. (2000) The solubility of the ribotoxin a-sarcin, produced as a recombinant protein in Escherichia coli, is significantly increased in the presence of thioredoxin. Lett. Appl. Microbiol. 30, 298–302. 35. Olson, B.H. & Goerner, G.L. (1965) Alpha-sarcin, a new antitumor agent. I. Isolation, purification, chemical composition, and the identity of a new amino acid. Appl. Microbiol. 13, 314–321. 36. Martı ´ nez del Pozo, A., Gasset, M., On ˜ aderra, M. & Gavilanes, J.G. (1988) Conformational study of the antitumor protein a-sarcin. Biochim. Biophys. Acta 953, 280– 288. 37. Gavilanes, J.G., Va ´ zquez, D., Soriano, F. & Me ´ ndez, E. (1983) Chemical and spectroscopic evidence on the homology of three antitumor proteins: a-sarcin, mitogillin and restrictocin. J. Protein. Chem. 2, 251 –261. 38. Becktell, W.J. & Schellman, J.A. (1987) Protein stability curves. Biopolymers 26, 1859–1877. 39. Pace, C.N. & Laurents, D.V. (1989) A new method for determining the heat capacity change for protein folding. Biochemistry 28, 2520–2525. 40. Pace, C.N. (1990) Conformational stability of globular proteins. Trends Biochem. Sci. 15, 14–17. 41. Endo, Y. & Wool, I.G. (1982) The site of action of a-sarcin on eukaryotic ribosomes: the sequence at the a-sarcin cleavage- site in 28S ribosomal ribonucleic acid. J. Biol. Chem. 257, 9054–9060. 42. Endo, Y., Hubert, P.W. & Wool, I.G. (1983) The ribonuclease activity of the cytotoxin a-sarcin: the characteristics of the enzymatic activity of a-sarcin with ribosomes and ribonucleic acids as substrates. J. Biol. Chem. 258, 2662– 2667. 43. Blank, A. & Sugiyama, R.H. & Dekker, C.A. (1982) Activity staining of nucleolytic enzymes after sodium dodecyl sulfate- polyacrylamide gel electrophoresis: use of aqueous isopropanol to remove detergent from gels. Anal. Biochem. 120, 267–275. 44. Shapiro, R., Fett, J.W., Strydom, D.J. & Vallee, B.L. (1986) Isolation and characterization of a human colon carcinoma- secreted enzyme with pancreatic ribonuclease-like activity. Biochemistry 25, 7255–7264. 45. Manchen ˜ o, J.M., Gasset, M., Lacadena, J., Ramo ´ n, F., Martı ´ nez del Pozo, A., On ˜ aderra, M. & Gavilanes, J.G. (1994) Kinetic study of the aggregation and lipid-mixing produced by a-sarcin on phosphatidylglycerol and phosphatidylserine vesicles: stopped- flow light-scattering and fluorescence energy transfer measure- ments. Biophys. J. 67, 1117 –1125. 46. Barlett, G.R. (1959) Colorimetric assay methods for free and phosphorylated glyceric acids. J. Biol. Chem. 234, 466 –468. 47. Struck, D.K., Hoekstra, D. & Pagano, R.E. (1981) Use of resonance energy transfer to monitor membrane fusion. Biochemistry 20, 4093–4099. 48. Manchen ˜ o, J.M., Gasset, M., Albar, J.P., Lacadena, J., Martı ´ nez del Pozo, A., On ˜ aderra, M. & Gavilanes, J.G. (1995) Membrane interaction of a b-structure-forming synthetic peptide comprising the 116–139th sequence of the cytotoxic protein a-sarcin. Biophys. J. 68, 2387–2395. 49. Manchen ˜ o, J.M., Martı ´ nez del Pozo, A., Albar, J.P., On ˜ aderra, M. & Gavilanes, J.G. (1998) A peptide of nine amino acid residues from a-sarcin cytotoxin is a membrane-perturbing structure. J. Peptide Res. 51, 142 –148. 50. Garcı ´ a-Ortega, L., Lacadena, J., Manchen ˜ o, J.M., On ˜ aderra, M., Kao, R., Davies, J., Olmo, N., Martı ´ nez del Pozo, A. & Gavilanes, J.G. (2001) Involvement of the NH 2 -terminal b-hairpin of the Aspergillus ribotoxins on the interaction with membranes and non- specific ribonuclease activity. Protein Sci. 10, 1658–1668. 51. De Antonio, C., Martı ´ nez del Pozo, A., Manchen ˜ o, J.M., On ˜ aderra, M., Lacadena, J., Martı ´ nez-Ruiz, A., Pe ´ rez-Can ˜ adillas, J.M., Bruix, M. & Gavilanes, J.G. (2000) Assignment of the contribution of the tryptophan residues to the spectroscopic and functional properties of the ribotoxin a-sarcin. Proteins 41, 350–361. 52. Koradi, R., Billeter, M. & Wu ¨ trich, K. (1996) MOLMOL: a program for display and analysis of macromolecular structures. J. Mol. Graph. 14, 51–55. 6196 M. Masip et al. (Eur. J. Biochem. 268) q FEBS 2001 . ribonuclease activity of a- sarcin has been developed by using the dinucleotide ApA as substrate [18]. Both mutant variants hydrolysed ApA. They displayed a K m value similar to that of the wild-type. and base on the catalysis by a- sarcin [19,20], rendered proteins with no detectable activity against ApA [20]. Cleavage of ApA by a- sarcin is a low-specificity reaction requiring high amounts of. Arginine 121 is a crucial residue for the specific cytotoxic activity of the ribotoxin a- sarcin Manuel Masip, Javier Lacadena*, Jose ´ Miguel Manchen ˜ o, Mercedes On ˜ aderra, Antonio Martı ´ nez-Ruiz†, A ´ lvaro