Báo cáo khoa học: Functional characterization of front-end desaturases from trypanosomatids depicts the first polyunsaturated fatty acid biosynthetic pathway from a parasitic protozoan ppt

Functional characterization of front-end desaturases from trypanosomatids depicts the first polyunsaturated fatty acid biosynthetic pathway from a parasitic protozoan Karina E. J. Tripodi, Laura V. Buttigliero, Silvia G. Altabe and Antonio D. Uttaro Instituto de Biologı ´ a Molecular y Celular de Rosario (IBR), CONICET, Departamento de Microbiologı ´ a, Facultad de Ciencias Bioquı ´ micas y Farmace ´ uticas, Universidad Nacional de Rosario, Santa Fe, Argentina Trypanosoma brucei, T. cruzi and Leishmania spp. are parasitic protozoa belonging to the order Kinetoplast- ida. They are causative agents of several highly disab- ling and often fatal diseases, including African sleeping sickness, Chagas disease and leishmaniasis. The drugs used in the treatment of trypanosomiasis and leish- maniasis are toxic, and in some cases have low effect- iveness, which makes imperative the development of new chemotherapeutic compounds [1,2]. For a number of years it was believed that trypanosomatids are dependent on the host for their lipid content. However, recently it was reported that they are able to synthesize de novo their own fatty acids (FA) through a prokaryotic or type II FA synthase [3]. This pathway is inhibited in T. brucei by the antibiotic thiolactomycin, which kills the parasite, making it a promising chemotherapeutic target [3]. Keywords front-end desaturases; Leishmania; lipids; Trypanosoma; trypanosomatids Correspondence A. D. Uttaro, IBR-CONICET, Dpto Microbiologı ´ a, Facultad de Ciencias Bioquı ´ micas y Farmace ´ uticas, Universidad Nacional de Rosario, Suipacha 531 (2000) Rosario, Santa Fe, Argentina Fax: +54 341 4390465 Tel: +54 341 4350661 E-mail: toniuttaro@yahoo.com.ar (Received 30 September 2005, revised 28 October 2005, accepted 3 November 2005) doi:10.1111/j.1742-4658.2005.05049.x A survey of the three kinetoplastid genome projects revealed the presence of three putative front-end desaturase genes in Leishmania major, one in Trypanosoma brucei and two highly identical ones (98%) in T. cruzi. The encoded gene products were tentatively annotated as D8, D5 and D6 desaturases for L. major, and D6 desaturase for both trypanosomes. After phylogenetic and structural analysis of the deduced proteins, we predicted that the putative D6 desaturases could have D4 desaturase activity, based mainly on the conserved HX 3 HH motif for the second histidine box, when compared with D4 desaturases from Thraustochy- trium, Euglena gracilis and the microalga, Pavlova lutheri, which are more than 30% identical to the trypanosomatid enzymes. After cloning and expression in Saccharomyces cerevisiae, it was possible to functionally characterize each of the front-end desaturases present in L. major and T. brucei. Our prediction about the presence of D4 desaturase activity in the three kinetoplastids was corroborated. In the same way, D5 desatu- rase activity was confirmed to be present in L. major. Interestingly, the putative D8 desaturase turned out to be a functional D6 desaturase, being 35% and 31% identical to Rhizopus oryzae and Pythium irregulare D6 desaturases, respectively. Our results indicate that no conclusive pre- dictions can be made about the function of this class of enzymes merely on the basis of sequence homology. Moreover, they indicate that a com- plete pathway for very-long-chain polyunsaturated fatty acid biosynthesis is functional in L. major using D6, D5 and D4 desaturases. In trypano- somes, only D4 desaturases are present. The putative algal origin of the pathway in kinetoplastids is discussed. Abbreviations BHT, 2,6-di-tert-butyl-p-cresol; FA, fatty acid; PUFA, polyunsaturated fatty acid. FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS 271 Trypanosomatids contain the usual range of lipids found in their eukaryote host (i.e. triacylglicerols, phospholipids, plasmalogens, sterols) but a higher pro- portion of polyunsaturated fatty acids (PUFAs) [4,5]. This suggests a high membrane fluidity that may be essential for the parasites in order to adapt themselves to the dramatic changes in temperature and chemical parameters experienced during their complex life cycles. Bloodstream T. brucei has an elevated amount of linoleic acid (18:2) and 22:5 and 22:6 PUFAs, but lower levels of oleate (18:1) and C16 FAs than the human host. Procyclic T. brucei showed lower levels of 16:0 and 18:1, but higher ratios of 18:0 and 18:2 than those present in the growth medium. Remarkably, they possess 22:5, a PUFA absent from culture media [4,5]. These and other [6,7] data indicate that trypano- somatids are able to elongate and desaturate de novo synthesized and exogenously acquired FAs. Leishmania has a high proportion of 18:3 [4], although at present there is no conclusive evidence for the isomeric nature of these molecules. Meyer & Holz [8] found 18:3 n-6 as the major FA in L. tarentolae grown on defined media, whose synthesis would require a D6 desaturase, indicating the presence of the ‘animal pathway’ for PUFA biosynthesis. Korn and coworkers [9,10] found 18:3 n-3 as the main PUFA, whereas the n-6 form was detected only in trace amounts. This latter finding could indicate the pres- ence of a D15 or x3 desaturase, compatible with the ‘plant type’ pathway. Currently, it is assumed that both pathways could coexist in trypanosomatids, as observed for Euglena, the organism most related to kinetoplastids [4]. In both cases, the presence of the key enzyme D12 or oleate desaturase is imperative for the de novo synthesis of these PUFAs. Early work did not show conclusive evidence for the presence of enzymes involved in PUFA biosynthesis in trypano- somatids, with the exception of oleate desaturase. Linoleate (18:2 n-6) synthesis from radioactive satur- ated or monounsaturated FAs has been described for T. cruzi [6] and Leishmania spp. [8,10]. We have recently confirmed the presence of this activity, by the cloning and functional characterization of an oleate desaturase from T. brucei [11]. Orthologous genes are present in T. cruzi and L. major. Oleate desaturase is absent in mammals, so the biosynthesis of PUFAs goes through the desaturation and elongation of essential FAs taken from the diet, linoleic and a-linolenic (18:3 n-3) acids, which allow the production of n-6 and n-3 series of PUFAs, respectively. Essential FAs are first desaturated at position 6 by a D6 desaturase, then elongated to the corresponding 20:3 n-6 and 20:4 n-3 PUFAs. A D5 desaturase produces 20:4 and 20:5 that are finally converted to 22:5 n-6 and 22:6 n-3 PUFAs by the Sprecher pathway [12], which involves two elon- gation steps up to C24 FAs, a desaturation by the same D6 desaturase and one b-oxidation cycle in peroxisomes. At present, the Sprecher pathway has only been observed in mammals. An alternative route was detec- ted in some protozoa, including Euglena gracilis and marine microalgae [13]. Here, C20 FAs are elongated to C22 and then desaturated by a D4 desaturase to the same final 22:5 n-6 and 22:6 n-3 PUFAs. Euglena exhibits another variation in the first steps of the pathway. C18 FAs are elongated to C20, and then a D8 desaturase produces the same kind of C20 FAs that will be the substrate of D5 desaturase. Although D8 desaturase was cloned and functionally characterized only from E. gracilis [14], the so-called ‘Euglena pathway’ was suggested to be present also in the ciliate Tetrahymena [15], the marine microalga Isochrysis galbana [16] and the oyster protozoan para- site Perkinsus marinus [17]. The enzymes involved in PUFA biosynthesis (i.e. D6, D8, D5 and D4 desaturases) are named ‘front-end’ desaturases, because they introduce a double bond between a pre-existent olefinic bond and the carboxyl end of the FA molecule. On the other hand, methyl- end desaturases, such as D12 (x6) and x3 desaturases, introduce a double bond towards the methyl end of the aliphatic chain. Front-end desaturases share struc- tural features that make them easy to recognize. They contain a fused cytochrome b5 as an N-terminal domain and three histidine boxes in the desaturase (C-terminal) domain. The third histidine box has the QX 2 HH instead of the HX 2 HH motif present in other types of desaturases [18]. The genome projects of trypanosomatids revealed the presence of front-end desaturase genes, tentatively annotated as D8, D5 and D6 desaturases, on the basis of sequence similarity. However, functional predictions are never conclusive for desaturases, meaning that bio- chemical characterization is essential for the correct assignment of enzyme regioselectivities. Here, we describe the cloning and functional charac- terization of all the front-end desaturases detected in the T. brucei and L. major genomes. This work allowed us to assign the correct enzymatic activity to each desaturase and to depict, for the first time, the com- plete pathway for PUFA biosynthesis in a protozoon. Moreover, it would help to resolve an old controversy related to the isomeric nature of the PUFAs synthes- ized by these organisms. Front-end desaturases of trypanosomatids K. E. J. Tripodi et al. 272 FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS Results Identification of putative front-end desaturase genes in trypanosomatids, and a sequence-based structural analysis of the encoded proteins We performed, periodically, blast searches for desatu- rase genes in the databases of the three trypanosoma- tid genome projects, using, as queries, the sequences of previously characterized desaturases present in the public domain. Three open reading frames encoding amino acid sequences with high similarity to front-end desaturases were detected in the L. major genome. They have the GeneDB accession codes LmjF 36.6950, LmjF 07.1090 and LmjF 14.1340, and were tentatively annotated as D8, D5 and D6 desaturases, respectively. Entries for T. brucei (Tb 10.6k15.3610) and T. cruzi (Tc00.1047053510181.20 and Tc00.1047053507609.40, which are 98% identical) share 65–53% identity at the protein sequence level with LmjF 14.1340 and were also annotated as D6 desaturases. The protein encoded by LmjF 36.6950 showed 35.6% identity with Rhizopus oryzae D6 desaturase (AAS93682) and 31% with E. gracilis D8 desaturase (AAQ19605). The LmjF 07.1090 protein is 43.1% identical to Thraustochytrium sp. D5 desaturase (AAM09687), 40.2% to Ostreococcus tauri D6 desatu- rase (AAW70159), 31.3% to Homo sapiens D6 desatu- rase (AAH04901) and 30.2% to Danio rerio D5 ⁄ D6 bifunctional desaturase (Q9DEX7). The protein enco- ded by LmjF 14.1340 and its trypanosome orthologue are 31.5–40.1% identical to Pavlova lutheri D4 desatu- rase (AAQ98793), 36.3–37.6% identical to Phaeodacty- lum tricornutum D5 desaturase (AAL92562) and 34– 34.2% identical to E. gracilis D4 desaturase (AAQ19605). All structural features of front-end desaturases are present in each of these trypanosomatid proteins, including the three highly conserved histidine boxes and the cytochrome b5 N-terminal domain (Fig. 1). Interestingly, the second histidine box of LmjF 14.1340, Tb 10.6k15.3610, Tc00.1047053510181.20 and Tc00.1047053507609.40 has an additional amino acid, with the motif HX 3 HH, as found in D4 desaturases from E. gracilis, Pav. lutheri and Thraustochytrium sp. (AAN75710). In contrast, one other enzyme known to have D4 desaturase activity, from I. galbana (AAV33631), possesses an HX 2 HH motif in its second histidine box, characteristic of all front-end desaturases, with the aforementioned exceptions. In a phylogenetic analysis [16], I. galbana desaturase was located in a cluster distinct from all other D4 desaturases, indica- ting that it has evolved in an independent way, but reached the same regioselectivity by a convergent evo- lutionary process. Considering I. galbana an exception, the HX 3 HH motif could be used to a priori character- ize LmjF 14.1340 and the Trypanosoma proteins as front-end D4 desaturases. Functional characterization of front-end desaturases Functional characterization was carried out by deter- mining the FA profiles of Saccharomyces cerevisiae transformed either with vector p426GPD alone or with the vector containing an insert harbouring the putative L. major or T. brucei front-end desaturases. Yeast cul- tures were supplemented with the predicted substrates for each desaturase. The absence of any endogenous PUFA makes yeast a suitable expression host for char- acterizing enzymes involved in PUFA biosynthesis. The FA composition of the yeast transformed with p426GPD showed the four main FAs normally found in S. cerevisiae, namely 16:0, 16:1D 9 , 18:0 and 18:1D 9 plus the supplemented FA that was incorporated from the culture medium. Yeasts expressing the LmjF 14.1340 and Tb 10.6k15.3610 genes, and supplemented with the 22:4 n-6 substrate for D4 desaturase, cis- 7,10,13,16 docosatetraenoic acid (22:4 n-6), showed an additional peak when compared with the FA profile of the control harbouring the empty vector (Fig. 2). This Fig. 1. Sequence alignment of front-end desaturases by CLUSTALW [32]. Conserved motifs are highlighted in boxes. I, II and III are the first, second and third histidine boxes, respectively; cytb5 is the heme-binding motif. LmC36, Leishmania major LmjF36.6950 (CAJ09677); D8Eg, Euglena gracilis AAD45877; D6Hs, Homo sapiens AAD31282; D6Pi, Pythium irregulare AAL13310; LmC7, L. major LmjF07.1090 (CAJ07076); D5Ts, Thraustochytrium sp. AAM09687; LmC14, L. major LmjF14.1340 (CAJ03208); TbC10, Trypanosoma brucei Tb10.6k15.3610 (EAN78117); Tc20, T. cruzi Tc00.1047053510181.20 (EAN90580); D4PL, Pavlova lutheri AAQ98793; D4Ts, Thraustochytrium sp. AAN75710. K. E. J. Tripodi et al. Front-end desaturases of trypanosomatids FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS 273 peak was identified as 22:5D 4,7,10,13,16 , and this was confirmed by unequivocal determination of the double bond positions through analysis of the mass spectrum of the dimethyloxazoline derivative (Fig. 2, inset). This result was a confirmation of our predictions based on primary structural analysis. When the n-3 substrate, cis-7,10,13,16,19 docosatetraenoic acid, was added to the culture, it was converted to the corresponding 22:6 D4 product to a similar extent as the n-6 substrate (i.e. 4.5–6%) (Table 1). No other FA was shown to be used as a substrate, indicating that L. major and T. brucei D4 desaturases are highly specific for D7 PUFAs. Using the same approach, the expressed desaturase from LmjF 07.1090 was, as predicted, able to use only FAs that represent the substrates characteristic for D5 desaturases, namely cis-8,11,14 eicosatrienoic acid (20:3 n-6) and cis-8,11,14,17 eicosatetraenoic acid (20:4 n-3), converting them to 20:4D 5,8,11,14 (Fig. 3) and 20:5D 5,8,11,14,17 , respectively (Table 1). Fig. 2. Gas chromatography analysis of fatty acid methyl esters from yeasts expressing Leishmania major Lm14.1340 or Trypano- soma brucei Tb10.6k15.3610 with exogenous substrate 22:4 n-6. Cultures of yeast strain HH3 transformed with either the empty vector (HH3p426) or the vector harbouring the desaturases (HH3p426-Lm14, HH3p426-Tb10) were supplemented with 22:4D 7,10,13,16 . The product of D4 desaturase activity (i.e. 22:5 n-6) is highlighted in a box, and the spectrum of the correspond- ing dimethyloxazoline derivative is shown. The double bond in position 4 is defined by the fingerprint ion at m ⁄ z ¼ 152* [38]. Each feeding experiment was repeated at least twice, and the results of a representative experiment are shown. Analysis of fatty acid methyl esters was performed by using a SE-30 col- umn, as described in the Experimental procedures. Table 1. Activity of desaturases on n-3 and n-6 substrates after expression in Saccharomyces cerevisiae strain HH3. The percentage of con- version was calculated taking into account the area resulting from the integration of substrate and product peaks in the respective chromato- gram after running fatty acid methyl ester samples in a PE-WAX column, as described in the Experimental procedures. Data represent the mean and standard deviation of three independent determinations. Substrate fatty acid Activity (% of conversion) HH3p426-Tb10 HH3p426-Lm14 HH3p426-Lm07 HH3p426-Lm36 HH3p426 16:1 (D 9 )– – – + a– 18:1 (D 9 )– – – + a– 18:2 (D 9,12 ) – – – 8.55 ± 0.87 – 18:3 (D 9,12,15 ) – – – 6.30 ± 0.72 – 20:2 (D 11,14 )––––– 20:3 (D 8,11,14 ) – – 2.93 ± 0.48 – – 20:3 (D 11,14,17 )––––– 20:4 (D 8,11,14,17 ) – – 3.74 ± 0.41 – – 22:4 (D 7,10,13,16 ) 4.45 ± 0.65 5.97 ± 0.66 – – – 22:5 (D 7,10,13,16,19 ) 5.83 ± 0.32 4.50 ± 0.52 – – – a Conversion percentage less than 0.5%. –, No activity detected. Fig. 3. Gas chromatography analysis of fatty acid methyl esters from yeast expressing Leishmania major Lm07.1090 with exogen- ous substrate 20:3 n-6. Cultures of yeast strain HH3 transformed with either the empty vector (HH3p426) or the vector harbouring the desaturase (HH3p426-Lm07) were supplemented with 20:3D 8,11,14 . The product of D5 desaturase activity (20:4, n-6), is in a box and the spectrum of its corresponding dimethyloxazoline derivative is shown. Analysis was performed as described in Fig. 2. The double bond in position 5 is confirmed by the fingerprint ion at m ⁄ z ¼ 153* [38], while the remainder are located by the gaps of 12 atomic mass units, as indicated. Front-end desaturases of trypanosomatids K. E. J. Tripodi et al. 274 FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS Finally, the protein encoded on LmjF 36.6950, and tentatively annotated as D8 desaturase, was assayed using the substrates typical for this class of regioselec- tivity: 20:2D 11,14 and 20:3D 11,14,17 [14]. No activity at all was evident with these FAs, but a small peak of 16:2D 6,9 was detected on the chromatogram, suggestive for the action of a D6 desaturase on the endogenous 16:1D 9 FA (Fig. 4). Curiously, although 18:2D 9 is a rel- atively abundant endogenous substrate, we were able to detect its D 6 desaturation product only in very small amounts (less than 0.1%). The addition of linoleic (18:2D 9,12 )ora-linolenic (18:3D 9,12,15 ) acids resulted in the formation of the corresponding 18:3D 6,9,12 and 18:4D 6,9,12,15 FAs, indicating that the real regioselectiv- ity of the protein encoded by LmjF 36.6950 corres- ponds to that of a D6 desaturase (Fig. 4 and Table 1). Evolutionary relationships of front-end desaturases from trypanosomatids In order to gain information about the evolution of front-end desaturases in trypanosomatids, we com- pared them with desaturases from a variety of organ- isms (i.e. algae, fungi, marine protists, nematodes, vertebrates, plants and mosses). A similar detailed ana- lysis has been performed by Sperling et al. [18], but we included a number of recently described sequences, plus the trypanosomatid sequences, and obtained the phylogenetic tree depicted in Fig. 5. D4 desaturases from L. major, T. brucei and T. cruzi group together with D4 and D5 desaturases from fungi, algae and mosses, but form a subgroup with algae desaturases (i.e. D4 from Pav. lutheri and D5 desaturases from Phaeod. tricornutum and Thalassiosira pseudonana). Interestingly, E. gracilis D4 desaturase is located in the other subgroup, comprising also the highly related D4 desaturases from the alga Thal. pseudonana and Thrau- stochytrium sp., and the D5 desaturases from fungi. As previously reported, the D6 desaturase from cyanobac- teria is also present in this group [18]. In contrast, L. major D5 desaturase is separated from this cluster; it is found in a heterogeneous group that includes D5 desaturase from Thraustochytrium sp., D4 desaturase from I. galbana and the O. tauri D6 acyl-CoA desatu- rase [19]. Lastly, D6 desaturase from L. major was found in another major branch, mainly containing D6 desatu- rases from lower and higher eukaryotes and the D8 desaturase from E. gracilis. It is reasonably clear from the analysis of this branch that D5 desaturases in nem- atodes and vertebrates may have diverged from an ancestral D6 desaturase in each of these lineages after gene duplication. L. major D6 desaturase is located in a subgroup containing D6 desaturases from lower euk- aryotes and shows most relationship with E. gracilis D8 desaturase and the nematode desaturases. The remaining desaturases in this branch form two other groups; one containing vertebrate desaturases and the other formed by desaturases from higher plants and from the alga Thal. pseudonana. Discussion We have functionally characterized all front-end desaturases detected in the genome databases of the three trypanosomatids. The completion of these genome projects [20] allowed us to unravel the biosynthetic pathway for PUFAs in these parasitic protozoa. In L. major, the pathway involves the consecutive action of D6, D5 and D4 desaturases on C18, C20 and C22 FAs, respectively. The n-6 and n-3 isomers were used equally well by the enzymes. Recent evidence from our laborat- ory showed that D12 and x3 desaturases are present in L. major (A. Alloati and A. D. Uttaro, unpublished results), indicating that the biosynthesis of 18:2 n-6 and 18:3 n-3 FAs is functional in Leishmania (Fig. 6A). It was shown previously that Leishmania sp. mainly accumulates 18:3 FAs, but their isomeric nature remained controversial [4]. Our results confirm that Fig. 4. Gas chromatography analysis of fatty acid methyl esters from yeast expressing Leishmania major Lm36.6950. Cultures of yeast strain HH3 transformed with the desaturase insert (HH3p426-Lm36) were fed (+) or not (–) with 18:3D 9,12,15 (A) and 18:2D 9,12 (B). The products of D6 desaturase activity are boxed: 18:4 n-3 (A); 18:3 n-6 (B); and 16:2 (A and B). In each case, dimethyl- oxazoline derivatives were obtained and the double bond positions in products were ascertained (data not shown). Analysis was per- formed as described in Fig. 2. K. E. J. Tripodi et al. Front-end desaturases of trypanosomatids FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS 275 both 18:3D 6,9,12 (n-6) and 18:3D 9,12,15 (n-3) can be syn- thesized by L. major, but we cannot anticipate the proportion of each isomer from our heterologous expression experiments. This ratio will depend on the relative expression of the D6 and x3 desaturases in different Leishmania species and on the growth condi- tions. Interestingly, the groups of Holz and Korn, although both working with the same organism, L. tarentolae, reached opposing conclusions and sugges- ted the presence of the ‘animal’ or ‘plant’ pathway, respectively [8,9]. Now, these conclusions have to be considered as an oversimplification, as new evidence showed that alternative pathways are present in lower eukaryotes [13,14,16,18,21]. Algae synthesize PUFAs by two different ways, using the combination of D6, D5 and D4 desaturases plus D 6 and D5 elongases, or by D8, D5 and D4 desaturases plus D9 and D5 elongases. The only complete set of desaturases characterized to date from an alga is that for Thal. pseudonana [22], which has a pathway similar to the one described here for L. major. A complete characterization of all the desaturases involved in PUFA biosynthesis in other algae or proto- zoa is still lacking. The pathway involving D8 desaturas- es was proposed to be present in some algae, such as Isochrysis sp., and in some marine protists [16,17], but only conclusively proved in the protozoan E. gracilis [14], which is the organism most related to Kinetoplast- ids, both grouped in the Euglenozoa. Interestingly, L. major D6 desaturase showed high similarity to E. gracilis D8 desaturase (Fig. 5). On the other hand, the phylogenetic analysis locates L. major D4 desaturase in a subgroup with the same enzymes from trypano- somes and the microalga Pav. lutheri, and separated from E. gracilis D4 desaturase. Unfortunately, D4 desaturase is the only known Pavlova desaturase [23], Fig. 5. Phylogenetic analysis of front-end desaturases. The phylo- genetic tree was created using the neighbour-joining method, with 10 000 replicates, in MEGA-3 [33]. For the analysis we used sequences of desaturases from the trypanosomatids characterized here, D 4 Leishmania major (D4Lm14: CAJ03208), D5 L. major (D5Lm07: CAJ07076); D6 L. major (D6Lm36: CAJ09677), D4 Try- panosoma brucei (EAN78117) along with the following sequences: D4 desaturases from T. cruzi (EAN90580), Euglena gracilis (AAQ19605), Isochrysis galbana (AAV33631), Pavlova lutheri (AAQ98793), Thraustochytrium sp. (AAN75710), Thalassiosira pseu- donana (AAX14506); D5 desaturases from Caenorhabditis elegans (AAC95143), Mortierella alpina (AAC72755), Phaeodactylum tri- cornutum (AAL92562), Pythium irregulare (AAL13311), Thraustochy- trium sp. (AAM09687), Phytophthora megasperma (CAD53323), Mus musculus (AAH26848), Thal. pseudonana (AAX14502), Physc- omitrella patens (CAH05235), Homo sapiens (AAF29378); D8de- saturases from E. gracilis (AAD45877); D6 desaturases from C. elegans (AAC15586), Rhyzopus oryzae (AAS93682), Thal. pseu- donana (AAX14504), P. tricornutum (AAL92563), P. irregulare (AAL13310), P. patens (CAA11033), Ceratodon purpureus (CAB94993), H. sapiens (AAD31282), M. musculus (AAD20017), Ostreococcus tauri (AAW70159), Borago officinalis (AAD01410), Primula vialii (AAP23036), Synechocystis sp. (Q08871) and D5 ⁄ D6 bifunctional desaturase from Danio rerio (AAG25710). Numbers represent bootstrap values. The bar represents the percentage of substitutions. Fig. 6. Hypothetical routes followed by Leishmania (A) and trypano- somes (B) for polyunsaturated fatty acid (PUFA) biosynthesis. FAD, fatty acid desaturase; Elo, elongase; Elo5, D5 elongase; Elo6, D6 elongase; AA, arachidonic acid; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid. Light grey ovals are front-end desaturases identified in the present work. Grey ovals represent desaturases already characterized by our group. White ovals correspond to activ- ities that await characterization. Front-end desaturases of trypanosomatids K. E. J. Tripodi et al. 276 FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS making it impossible to establish the relationship between L. major D6 and D5 desaturases and related enzymes from this alga. L. major D5 desaturase is also phylogenetically related to algal enzymes such as the O. tauri D6 [19] desaturase and the D5 desaturase from the marine protist Thraustochytrium sp. [24]. Hannaert et al. have recently proposed that the high number of plant-like genes detected in the trypano- somatid genomes could be explained by an early event of endosymbiosis into the Euglenozoa, where an alga acted as the secondary endosymbiont [25]. After the divergence, euglenids have retained a chloroplast as the only remnant of the alga, whereas kinetoplastid ancestors have lost the chloroplast, and retained only nuclear and chloroplast genes that were transferred to the nucleus. We have shown here the phylogenetic relationship of the trypanosomatid front-end desatu- rases with algal and Euglena enzymes. However, it is expected that all trypanosomatid front-end desaturases would be related more to the Euglena counterpart than to any other organism in order to support that hypo- thesis, which is not the case for D4 desaturases. It may indicate that more than one event of secondary endo- symbiont acquisition could have occurred, as suggested previously [26]. The only front-end desaturase detected in T. brucei, characterized as a D4 desaturase, is able to use n-3 and n-6 FA isomers. Two highly identical protein sequences that share 66% of identity with the T. brucei enzyme were detected in the T. cruzi database, most probably representing the products from orthologous genes. In addition, we have previously characterized an oleate (D12) desaturase from T. brucei [11], and also detected the T. cruzi counterpart. No other methyl-end desaturases could be identified in the trypanosome genomes. This indicates that these parasites are only able to synthesize 18:2D 9,12 (n-6), which is the main FA in both species. Other PUFAs detected in total lipid extracts from T. brucei are 22:5 and 22:6. They are present in trace amounts, but in a higher propor- tion with respect to the levels found in the host or in culture media [4,5]. This last observation is in agree- ment with the existence of a D4 desaturase acting on exogenous n-6 and n-3 substrates. The pathways shown in Fig. 6 involve the presence of putative D6 and D5 elongases in L. major, whereas in trypanosomes, poss- ibly a D5 elongase or no elongase activity would be present. As a consequence, the nature of the exogenous substrates for trypanosome D4 desaturases can only be speculated. A probable candidate is the relatively abundant PUFA in the mammalian host, arachidonic acid (20:4; n-6), which would imply the presence of a D5 elongase (Fig. 6B). We have shown here that the prediction of front-end desaturase regioselectivity cannot be based merely on sequence similarity or phylogenetic analysis (Fig. 5). However, we can gain some structural insights, as new desaturases are functionally characterized. It is the case for D4 desaturases, with four new enzymes conclusively characterized, including that recently reported from Thal. pseudonana [22] and the three trypanosomatid enzymes described here. As discussed previously, Isochrysis sp. D4 desaturase appears to have evolved independently, probably from a D5 desaturase, conser- ving the consensus motifs for the three histidine boxes characteristic of D6 and D5 desaturases, which includes the motif HX 2 HH in the second box. The other seven D4 desaturases present a variation in this box with the motif HX 3 HH. This structural characteristic may thus be taken as a diagnostic feature to identify D4 desatu- rases, although the presence of the canonical HX 2 HH cannot rule out this kind of regioselectivity. Unsaturated FAs have a key role in maintaining the correct membrane fluidity in poikilothermic organisms, which is necessary for the mobility and function of embedded proteins and in giving shape to membrane curvatures, which in turn are required for the forma- tion of organelles, the vesicular system and the nuclear envelope. They represent more than 70% of total FAs in trypanosomatids [27]. This fact could indicate that the unsaturated FA biosynthetic pathway is essential in these parasitic protozoa. This is not unexpected con- sidering the complexity of their life cycles, where they have to adapt to dramatic changes of temperature and morphology. It may be very interesting to evaluate the possibility of using this pathway as a putative chemo- therapeutic target. The high proportion of C18 mono- and polyunsaturated FAs can be ascribed to the physiological processes described above, but the role of C20 and C22 PUFAs is more difficult to explain. In mammals, PUFAs are converted to eicosa- noids, like prostaglandins and leukotrienes. They are potent mediators involved in numerous homeostatic biological functions and inflammation. Two recent reports highlighted the significance of prostaglandins in trypanosomatid life cycles [28,29]. Prostaglandins are produced from arachidonic acid, both in T. brucei bloodstream forms and in L. major promastigotes, and a role of prostaglandins in the pathogenesis of the dis- eases was suggested. An additional function for C22 FAs could be assigned as trypanosomes have con- served a D4 desaturase. This function cannot be antici- pated, although it is known that some PUFAs act as second messengers in other organisms. Recent experi- ments on T. brucei and L. major showed that free arachidonic acid might induce the mobilization of K. E. J. Tripodi et al. Front-end desaturases of trypanosomatids FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS 277 Ca 2+ and other nonspecific effects. This role could also be shared by linoleic acid and linolenic acid [30]. Experimental procedures a-Linolenic (18:3, D 9,12,15 ); linoleic (18:2, D 9,12 ); cis-11, 14-eicosadienoic; cis-11,14,17-eicosatrienoic; cis-8,11,14-eico- satrienoic and cis-7,10,13,16-docosatetraenoic acids (all more than 99% pure); Tergitol (type Nonidet P-40); sodium methoxide; ampicillin, yeast nitrogen base; glucose; amino acids; 2-amino-2-methyl-1-propanol (95%) and 2,6-di-tert-butyl-p-cresol (BHT) were obtained from Sigma (Sigma-Aldrich, St. Louis, MI, USA). Cis-7,10,13,16,19-do- cosapentaenoic acid and cis-8,11,14,17-eicosatetraenoic acid were from Cayman Chemical Company (Ann Arbor, MA, USA). All organic solvents were purchased from Merck (Whitehouse Station, NJ, USA). Cloning, sequencing and sequence analysis Front-end desaturase gene sequences were retrieved from the databases of the trypanosomatid genome projects and analysed using tools available online (http://www.genedb.org and http://www.ncbi.nlm.nih.gov/BLAST). Promastigote L. major (Friedlin strain) and procyclic T. brucei (strain 427) cells were grown in SDM-79 medium supplemented with 10% fetal bovine serum and hemin [31]. Genomic DNA from both sources was prepared by stand- ard methods. Four sets of primers were designed to amplify three types of front-end desaturases: Lm14 forward, 5¢-CG GGATCCATGAACCAGTGTTGCCAC-3¢ and Lm14 reverse, 5¢-CG GAATTCTAGGCGCTCTTCCGCTTC-3¢ for Lm14.1340; Lm07 forward, 5¢-CG GGATCCATGGCC CTCGACAATGTCC-3¢ and Lm07 reverse; 5¢-CC AAGCT TAGTTCCCAGCAACGATGAA-3¢ for Lm07.1090; Lm36 forward, 5¢-CG GGATCCATGGTCTTCGAGCTCACTC- 3¢ and Lm36 reverse, 5¢-CC AAGCTTCTACTTCCCGCTC TTGGCCTC3¢ for Lm36.6950; and Tb10 forward, 5¢-CG GGATCCATGAGTTCGGTAAAGAGTAAAG-3¢ and Tb10 reverse, 5¢-CC AAGCTTCACAACCGTTTGTCTTC TAT-3¢ for Tb 10.6k15.3610. The underlined sequences rep- resent BamHI sites for forward primers and HindIII sites for reverse primers, with the exception of Lm14 reverse, where EcoRI was introduced; oligonucleotides also include the natural initiation and stop codons. Amplifications were carried out in a 50 lL volume under the following condi- tions: initial incubation at 94 °C for 4 min, followed by 30 cycles of denaturation at 94 °C for 1 min, annealing at 60 °C (for Lm14.1340, Lm07.1090 and Tb10.6k15.3610 desaturases) or 58 °C (for Lm36.6950 desaturase) for 30 s, and extension at 72 °C for 2 min. Amplified fragments were cloned using pGEM-T Easy vector (Promega, Madison, WI, USA), according to the manufacturer’s procedure, and transformed into competent Escherichia coli (XL1-Blue). Plasmids purified from positive clones were sequenced completely. Phylogenetic analyses Available front-end desaturase protein sequences were aligned using clustalw [32]. Positions with gaps were removed. Phylogenetic analyses were carried out by the neighbour-joining method using the program mega3, ver- sion 3.0 [33] with 10 000 bootstrap samplings or by mini- mum evolution with 5000 bootstrap replicates. Both methods gave very similar tree topologies. Expression of front-end desaturase genes Cloned sequences were ligated into the BamHI and HindIII (or EcoRI) sites of p426GPD, a yeast expression vector containing the glyceraldehyde-3-phosphate dehydrogenase constitutive promoter [34]. A selectable marker gene in this vector confers uracil prototrophy to the host. Either the vector alone, or the vector harbouring different desaturases, was used to transform S. cerevisiae strain HH3 (MATa, trp1-1, ura3-52, ade2-101, his3-200, lys2-801 , leu2-1 [35]) by electroporation. Transformed clones were selected on min- imal agar plates lacking uracil [36]. In order to determine enzyme activities, transformed yeasts were cultured for 2 days at 30 °C in 0.67% (w ⁄ v) yeast nitrogen base, 2% (w ⁄ v) glucose and leucine, and tryptophan, lysine, adenine and histidine (all at 20 mgÆL )1 ), and inoculated. Cultures were diluted to an absorbance at 600 nm of 0.2, and grown for 72 h at 20 °C with constant agitation. FAs were prepared in eth- anol containing BHT at a stock concentration of 2% (w ⁄ v) and added to a final concentration of 0.002% (w ⁄ v) into 20 mL cultures containing 0.2% (v ⁄ v) Tergitol (type Nonidet P-40). Fatty acid analysis Twenty millilitre cultures were centrifuged at 5000 g for 5 min, and pelleted cells were washed twice with an equal volume of distilled water. Afterward, lipids were extracted as described by Bligh & Dyer [37]. The organic phase was dried under N 2 , and fatty acid methyl esters were obtained by incubation with 1 mL of 0.5 m sodium methoxide in methanol for 20 min at room temperature. Following neut- ralization with 6 m HCl and extraction with 2 mL of hex- ane, the solvent was evaporated to dryness under a N 2 atmosphere. Alternatively, dimethyloxazoline derivatives were prepared by adding 0.25 g of 2-amino-2-methyl-1-pro- panol to up to 2 mg of lipid sample, as described by Chris- tie [38]. In both cases, after evaporation of the solvent, the product was dissolved in isohexane containing BHT (50 p.p.m.) for GC-MS analysis. Front-end desaturases of trypanosomatids K. E. J. Tripodi et al. 278 FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS The composition of fatty acid methyl esters was analysed by running samples through a polyethylene glycol column (PE-WAX; 30 m · 0.25 mm inside diameter; Perkin Elmer, Norwalk, CT, USA) isothermically at 180 °C, or through an SE-30 column (25 m · 0.22 mm inside diameter; Scientific Glass Engineering, Ringwood, Australia) for 4 min isother- mically at 164 ° C, followed by a temperature increase, at a rate of 4 °CÆmin )1 , to 215 °C. Gas chromatographic analysis was performed with a Perkin Elmer AutoSystem XL gas chromatograph, and GC-MS was carried out in a Perkin El- mer mass detector (model TurboMass) at a ionization volt- age of 70 eV with a scan range of 20–500 Da. Retention times and mass spectra of peaks obtained were compared with those for standards (Sigma) or with those available on NBS75K (National Bureau of Standards database, Perkin Elmer). Dimethyloxazoline derivatives were submitted to GC through an SE-30 column. After holding the tempera- ture at 150 °C for 3 min, the column was temperature-pro- grammed to increase to 320 °C at a rate of 4 °CÆmin )1 . Helium was the carrier gas, at a constant flow rate of 1 mLÆ min )1 or 1.2 mLÆmin )1 for the SE-30 and the PE-WAX col- umn, respectively. MS was used in the electron impact mode at 70 eV with a scan range of 40–500 Da. Acknowledgements We wish to thank Mo ´ nica Hourcade and Daniel Elı ´ as for technical assistance, and Paul A. M. Michels for comments and suggestions on the manuscript. We also acknowledge the Trypanosoma brucei, T. cruzi and Leishmania major genome projects, The Institute of Genomic Research (TIGR) and The Sanger Institute, for the availability of sequence data. S.G.A. and A.D.U. are members of Carrera del Investigador Cien- tı ´ fico, CONICET, Argentina. K.E.J.T. has a postdoc- toral fellowship from ANTORCHAS and CONICET, Argentina. This work was supported by Fondo Nac- ional de Ciencia y Tecnologı ´ a, SECyT, Argentina, by grants PICT 99 No.1-7160 and PICT 03 No.1- 13842. References 1 Garcia LS (2001) Diagnostic Medical Parasitology, 4th edn. ASM Press, Washington, DC. 2 World Health Organization (2001) African trypanosom- iasis. Fact Sheet Number 259. WHO Publications, Geneva. 3 Morita YS, Paul KS & Englund PT. (2000) Specialized fatty acid synthesis in African trypanosomes: myristate for GPI anchors. Science 288, 140–143. 4 Haughan PA & Goad LJ (1991) Lipid biochemistry of trypanosomatids. In Biochemical Protozoology (Coombs GH & North MJ, eds), pp. 312–328. Taylor and Frances, London, Washington DC. 5 Mellors A & Samad SA (1989) The aquisition of lipids by african trypanosomes. Parasitol Today 5, 239–244. 6 de Lema MG & Aeberhard EE (1986) Desaturation of fatty acids in Trypanosoma cruzi. Lipids 21 , 718–720. 7 Florin-Christensen M, Florin-Christensen J, de Isola ED, Lammel E, Meinardi E, Brenner RR & Rasmussen L (1997) Temperature acclimation of Trypanosoma cruzi epimastigote and metacyclic trypomastigote lipids. Mol Biochem Parasitol 88, 25–33. 8 Meyer H & Holz GG Jr (1966) Biosynthesis of lipids by kinetoplastid flagellates. J Biol Chem 241, 5000–5007. 9 Korn ED, Greenblatt CL & Lees AM (1965) Synthesis of unsaturated fatty acids in the slime mold Physarum polycephalum and the zooflagellates Leishmania tarento- lae, Trypanosoma lewisi and Crithidia sp. A comparative study. J Lipid Res 79, 43–50. 10 Korn ED & Greenblatt CL (1963) Synthesis of alpha- linolenic acid by Leishmania enriettii. Science 142, 1301– 1303. 11 Petrini GA, Altabe SG & Uttaro AD (2004) Trypano- soma brucei oleate desaturase may use a cytochrome b5- like domain in another desaturase as an electron donor. Eur J Biochem 271, 1079–1086. 12 Sprecher H & Chen Q (1999) Polyunsaturated fatty acid biosynthesis: a microsomal-peroxisomal process. Prosta- glandins Leukot Essent Fatty Acids 60, 317–321. 13 Pereira SL, Leonard AE & Mukerji P (2003) Recent advances in the study of fatty acid desaturases from ani- mals and lower eukaryotes. Prostaglandins Leukot Essent Fatty Acids 68, 97–106. 14 Wallis JG & Browse J (1999) The Delta8-desaturase of Euglena gracilis: an alternate pathway for synthesis of 20-carbon polyunsaturated fatty acids. Arch Biochem Biophys 365, 307–316. 15 Lees AM & Korn ED (1966) Metabolism of unsaturated fatty acids in protozoa. Biochemistry 5, 1475–1481. 16 Pereira SL, Leonard AE, Huang YS, Chuang LT & Mukerji P (2004) Identification of two novel microalgal enzymes involved in the conversion of the omega3-fatty acid, eicosapentaenoic acid, into docosahexaenoic acid. Biochem J 384, 357–366. 17 Chu FL, Lund ED, Harvey E & Adlof R (2004) Arachi- donic acid synthetic pathways of the oyster protozoan parasite, Perkinsus marinus: evidence for usage of a delta-8 pathway. Mol Biochem Parasitol 133, 45–51. 18 Sperling P, Ternes P, Zank TK & Heinz E (2003) The evolution of desaturases. Prostaglandins Leukot Essent Fatty Acids 68, 73–95. 19 Domergue F, Abbadi A, Zahringer U, Moreau H & Heinz E (2005) In vivo-characterisation of the first acyl-CoA Delta6-desaturase from the plant kingdom. Biochem J 389, 483–490. K. E. J. Tripodi et al. Front-end desaturases of trypanosomatids FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS 279 20 Ivens AC, Peacock CS, Worthey EA, Murphy L, Aggarwal G, Berriman M, Sisk E, Rajandream MA, Adlem E, Aert R, et al. The genome of the kinetoplastid parasite, Leishmania major. Science 309, 436–442. 21 Qiu X (2003) Biosynthesis of docosahexaenoic acid (DHA, 22: 6-4, 7,10,13,16,19): two distinct pathways. Prostaglandins Leukot Essent Fatty Acids 68, 181–186. 22 Tonon T, Sayanova O, Michaelson LV, Qing R, Harvey D, Larson TR, Li Y, Napier JA & Graham IA (2005) Fatty acid desaturases from the microalga Thalassiosira pseudonana. FEBS J 272, 3401–3412. 23 Tonon T, Harvey D, Larson TR & Graham IA (2003) Identification of a very long chain polyunsaturated fatty acid Delta4-desaturase from the microalga Pavlova lutheri. FEBS Lett 553, 440–444. 24 Qiu X, Hong H & MacKenzie SL (2001) Identification of a Delta 4 fatty acid desaturase from Thraustochy- trium sp. involved in the biosynthesis of docosahexanoic acid by heterologous expression in Saccharomyces cere- visiae and Brassica juncea. J Biol Chem 276, 31561– 31566. 25 Hannaert V, Saavedra E, Duffieux F, Szikora JP, Rigden DJ, Michels PA & Opperdoes FR (2003) Plant- like traits associated with metabolism of Trypanosoma parasites. Proc Natl Acad Sci USA 100, 1067–1071. 26 Waller RF, McConville MJ & McFadden GI (2004) More plastids in human parasites? Trends Parasitol 20, 54–57. 27 Villasuso AL, Aveldan ˜ o M, Vicario A, Machado- Domenech EE & Garcia de Lema M (2005) Culture age and carbamoylcholine increase the incorporation of endogenously synthesized linoleic acid in lipids of Try- panosoma cruzi epimastigotes. Biochim Biophys Acta 1735, 185–191. 28 Kabututu Z, Martin SK, Nozaki T, Kawazu S, Okada T, Munday CJ, Duszenko M, Lazarus M, Thuita LW, Urade Y et al. (2003) Prostaglandin production from arachidonic acid and evidence for a 9,11-endoperoxide prostaglandin H2 reductase in Leishmania. Int J Parasi- tol 33, 221–228. 29 Kubata BK, Duszenko M, Kabututu Z, Rawer M, Szal- lies A, Fujimori K, Inui T, Nozaki T, Yamashita K, Horii T et al. (2000) Identification of a novel prosta- glandin f (2alpha) synthase in Trypanosoma brucei. J Exp Med 192, 1327–1338. 30 Catisti R, Uyemura SA, Docampo R & Vercesi AE (2000) Calcium mobilization by arachidonic acid in try- panosomatids. Mol Biochem Parasitol 105, 261–271. 31 Brun R & Schoeneberger M (1979) Cultivation and in vitro cloning of procyclic culture forms of Trypanosoma brucei in a semi-defined medium. Acta Trop 36, 289– 292. 32 Thompson JD, Higgins DG & Gibbson TJ (1994) clus- talw: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, posi- tion specific gap penalties and weight matrix choice. Nucleic Acids Res 22, 4673–4680. 33 Kumar S, Tamura K & Nei M (2004) mega3: Inte- grated software for Molecular Evolutionary Genetics Analysis and sequence alignment. Brief Bioinform 5, 150–163. 34 Mumberg D, Muller R & Funk M (1995) Yeast vectors for the controlled expression of heterologous proteins in different genetic backgrounds. Gene 156, 119–122. 35 Castro O, Chen LY, Parodi AJ & Abeijon C (1999) Uridine diphosphate-glucose transport into the endo- plasmic reticulum of Saccharomyces cerevisiae: in vivo and in vitro evidence. Mol Biol Cell 10, 1019–1030. 36 Ausubel FM & Frederick M (1991) Current Protocols in Molecular Biology. Wiley, New York. 37 Bligh EG & Dyer WJ (1959) A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37, 911–917. 38 Christie WW (1997) Structural analysis of fatty acids. In Advances in Lipid Methodology – Four (Christie WW, ed.), pp. 119–169. Oily Press, Dundee. Front-end desaturases of trypanosomatids K. E. J. Tripodi et al. 280 FEBS Journal 273 (2006) 271–280 ª 2005 The Authors Journal compilation ª 2005 FEBS . Functional characterization of front-end desaturases from trypanosomatids depicts the first polyunsaturated fatty acid biosynthetic pathway from a parasitic. Sayanova O, Michaelson LV, Qing R, Harvey D, Larson TR, Li Y, Napier JA & Graham IA (2005) Fatty acid desaturases from the microalga Thalassiosira pseudonana.