Evidence for two distinct major protein components, PAR 1 and PAR 2, in the paraflagellar rod of Trypanosoma cruzi. Complete nucleotide sequence of PAR.

The previously identified major protein components of the paraflagellar rod in Trypanosoma cruzi, PAR 1 and PAR 2, were analyzed to determine if they are distinct proteins or different conformations of a single polypeptide as has been suggested for other trypanosomatids. Amino acid sequence analysis showed PAR 1 and PAR 2 to be two distinct polypeptides. Antibodies specific against either PAR 1 or PAR 2 were shown to each react with a distinct band in Western blots of paraflagellar isolates of T. cruzi and other trypanosomatids if rigorous protease inhibition was used. The PAR 2 message was isolated and characterized by Northern blot and nucleic acid sequence analysis. Preliminary analysis of the PAR 2 gene indicates that PAR 2 is a member of a multigene family with all members residing on a single chromosome.

lar rod (PFR)' is a single polypeptide of 600 amino acids that gives two bands in PAGE due to different conformations. They also characterized the gene coding for the PFR protein and determined that there were two identical copies in the genome.
Previously, we reported the isolation of two major immunologically distinct protein components of the paraflagellar rod in T. cruzi, the proteins PAR 1 and PAR 2 (4). These proteins show no immunological cross-reactivity with actin, tubulin, intermediate filament proteins, or other proteins present in mammalian cells. Here, we confirm that PAR 1 and PAR 2 are distinct proteins by amino acid sequence analysis. We also report the isolation and characterization of the PAR 2 message and that the PAR 2 gene is a member of a multigene family of a t least 30 members.

MATERIALS AND METHODS
Parasites-T. cruzi Esmeraldo clone 3 strain was obtained from James Dvorak, National Institutes of Health, Bethesda, MD. Growth and maintenance of epimastigotes and tissue culture derived trypomastigotes are as described elsewhere (5). Leishmania brasiliensis promastigotes (6) and T. brucei procyclics (7) were obtained as described.
Purification of T. cruzi Paraflagellar Proteins-Crude flagellar pellets from 2' . cruzi epimastigotes were prepared as described previously (4), except that all solutions employed contained either leupeptin (100 pg/ml), antipain (50 pg/ml), E-64 M), or a mixture with these three inhibitors. In brief, about 4 X 10" epimastigotes were harvested by centrifugation, washed twice with 0.02 M sodium phosphate, pH 7.4,0.9% sodium chloride (phosphate-buffered saline), and lysed in 20 ml of 1% Nonidet P-40 in 0.1 M Tricine, pH 8.5. Insoluble material was collected by centrifugation at 2,000 x g and was extracted once more with the same buffer. The pellet was then suspended in 10 ml of 1.0 M NaC1, 0.1% Triton X-100, in 0.1 M Tricine, pH 8.5. The DNA released at this step was sheared by stirring the suspension a t maximum speed for 30 s with a tissue grinder (Tissue-Tearon, Biospec Products, Bartlesville, OK). The suspension was centrifuged at 12,000 X g, and the resulting pellet corresponds to the crude flagellar fraction.
Crude flagellar fractions were successively extracted with 2.0 and 6.0 M urea in 10 mM Tricine, pH 8.5. As reported previously about 80% of the tubulin and 20% of the paraflagellar proteins were solubilized in 2.0 M urea, and the remaining paraflagellar proteins were solubilized in 6.0 M urea (4). The latter material was applied to a Mono-Q column equilibrated with the same buffer, and bound protein was eluted with a 0-500 mM NaCl gradient.
Polyacrylamide Gel Electrophoresis (PAGE) and Western Blot Analysis-For analysis of whole cell lysates, parasites were harvested from culture media by centrifugation, washed twice with phosphatebuffered saline, and solubilized by direct addition to boiling 2% SDS The abbreviations used are: PFR, paraflagellar rod; PAGE, polyacrylamide gel electrophoresis; Tricine, N-[2-hydroxy-l,lbis(hydroxymethyl)ethyl]glycine; HPLC, high performance liquid chromatography; PFGE, pulsed field gel electrophoresis; kb, kilobase. 21656 with hoiling continued for 5 min. These and all other samples were adjusted to the composition of the electrophoresis sample buffer (62.5 mM Tris, pH 6.8, 10% glycerol, 5% [j-mercaptoethanol, 20; sodium dodecyl sulfate, 0.001% bromphenol blue) prior to analysis. Onedimensional PAGE in 0.75-mm slab gels was done according to Laemmli (8). Prestained and "C-labeled molecular weight markers (Amersham Corp.) were included in the gels. Gels were either stained with Coomassie Brilliant Blue R (9) or processed for Western blot analysis (5) as previously described. Western blots were probed with a polyclonal antibody to PAR 1 (pcAbPAR 1) (4) or with a monoclonal antibody to PAR 2 (mAbPAR 2) (4).
Amino Acid Sequencing of Paraflagellar Proteins-Selected fractions from the Mono-Q column were concentrated in Centricon tubes t o a concentration of approximately 1.0 mg of protein/ml. 10 pl of 8mercaptoethanol were added to 190 pl of the protein solution and the mixture was incubated under Freon for 5 h a t 50 "C. The protein solution was then made 7.5% with 4-vinylpyridine and incubation continued a t room temperature for 1 h in the dark, followed by overnight dialysis against water and lyophilization.
The S-alkylated protein was dissolved in 100 pI of 75% formic acid and 100 pl of a freshly prepared CNRr solution (20% w/v in 75% formic acid) was added. The mixture was incubated in the dark for 2 h a t room temperature, diluted with 1.0 ml of water, and concentrated hy evaporation with a stream of nitrogen.
Cyanogen bromide fragments of the alkylated proteins were subjected to HPLC (LKR, Rromma, Sweden) on a reversed-phase column (Nucleosil CIR 300-10, column size 4 X 250 mm. Macherey-Nagel, Duren, Federal Republic of Germany) (10). The solvent system consisted of 0.1% (by volume) trifluoroacetic acid in water (A) and in acetonitrile (R). The proportion of R was increased from 0 to 50% within 110 min. The flow rate was 1.0 ml/min. The peptide fragments were monitored a t 206 nm, and the peaks were collected manually.
For amino acid sequencing of cyanogen hromide fragments, the Edman degradation method was carried out in a pulsed-liquid-phase sequencer (Applied Riosystems model 477A, Foster City, CAI. The phenylthiohydantoin derivatives were identified by the on-line HPLC system in which also the derivative of S-pyridylethyl cysteine is separated (11).
Nuclric Acid Isolation, Radiolabeling, Southern and Northern Transfrr, and Restriction Enzymes-Parasites were harvested and DNA, RNA, and poly(A)+ mRNA were isolated as described previously (12). Plasmid DNA was isolated by alkaline lysis miniprep as described (13). X-Phage DNA was prepared as described (14). DNA restriction fragments were radiolabeled with [n-'"PIdNTP using the HRL Nick Translation Kit as recommended by the manufacturer (GIRCO RRL). Agarose gel electrophoresis of DNA, Southern transfer, prehyhridization, hybridization, and filter washing were performed as described (15) except the gels were 1% agarose and the wash temperature was 68 "C. RNA was electrophoresed in a formaldehyde gel, blotted to nitrocellulose, baked, prehybridized, hybridized, and washed as described (16). All restriction enzymes were purchased from Roehringer Mannheim and used as recommended.
cDNA Library Construction and Screening-cDNA libraries were constructed in phage XgtlO or Xgtll using epimastigote poly(A)' rnRNA as described (17) or by using a cDNA synthesis system according to the manufacturer's instructions (Pharmacia LKR Riotechnology Inc.). The X-cDNA libraries were plated and transferred t o nitrocellulose as described (13). Filter hybridizations were carried out as descrihed (15) except that the wash temperature was 68 'C.
T h e Xgtl0 cDNA library was screened using radiolabeled TccPar2a. This fragment was isolated from a cDNA expression lihrary in X g t l l probed with a monoclonal antibody against the PAR 2 polypeptide (4, 17). The inserts present in phages showing positive hybridization were excised, subcloned into Rluescript KS+ (Stratagene Inc., La ,Jolla, CA), and characterized hy restriction enzyme mapping and direct nucleotide sequence analysis.
DNA Scqucncint-DNA sequence information was obtained by use of the dideoxy chain-termination method (18). Fragments to be sequenced were either subcloned into Rluescript KS+ or sequenced from the original Xgtl0 or Xgtll phage. Oligonucleotide sequencing primers were synthesized in the Gene Assemhler Plus (Pharmacia) according to the manufacturer's instructions.
Pulsed Field Gel I:lrctrophoresis (PFGE)-Epimastigotes were prepared and lysed in agarose blocks a t a concentration of 2 X 10" cells/ r n l as described (19). PFGE was carried out in a CHEF-DR I1 system supplied by Rio-Rad. A single agarose block containing epimastigotes as descrihed above was loaded in each well of a 1% agarose gel submerged in 0.5 X THE (90 mM Tris base, 90 mM boric acid, 2.5 mM EDTA, pH 8.0). Chromosome separation was performed over a period of 24-48 h a t 14 "C using 200 volts with a switch time of 60 and 90 s. Sacrharom.ycrs cereoisiae chromosomes were used as molecular weight markers. Gels were stained with ethidium bromide for 15 min and destained for 20 min hefore transfer of the DNA to nitrocellulose as described ahove.

RESULTS
Purification of Paraflagellar Proteins-A protein preparation enriched in paraflagellar proteins was obtained in the insoluble residue after successive extractions of T. cruti epimastigotes with buffer solutions containing 1% Nonidet P-40, 1.0 M NaCI, and 2.0 M urea, as described under "Materials and Methods." Extraction of that insoluble residue with 6 M urea produced a soluble fraction in which approximately SO?; of the total amount of protein corresponded to tubulin, and the other 50% to equimolar amounts of PAR 1 and PAR 2 ( Fig. 1, inset, lane c).
Previously, we reported that PAR 1 and PAR 2 have slightly different isoelectric points and molecular weights and that no immunological cross-reactivity between these two polypeptides could be demonstrated (4). These observations suggested that PAR 1 and PAR 2 correspond to two different paraflagellar rod components. In an attempt to obtain additional support for this contention through separation and further characterization of these two polypeptides, the protein fraction soluble in 6 M urea was fractionated by ion-exchange chromatography on a Mono-Q column. Fig. 1 shows a typical absorbance profile and the electrophoretic pattern of some of the fractions obtained from that column. These results indicate that the paraflagellar proteins are readily separated from tubulin and that fractions containing mixtures of approximately 80% PAR 2 and 20% PAR 1 (Fig. 1, inset, fraction 21 ) and 60% PAR 1 and 40% PAR 2 ( Fig. 1, inset,  (fraction 21) or PAR 1 (fractions 25 and 26) were subjected to mercaptolysis, alkylation, and cleavage with CNBr as described under "Material and Methods." The resulting peptides were fractionated by HPLC on a reversed-phase column.
As shown in Fig. 2, panel A , the HPLC profile from the preparation enriched for PAR 2 (fraction 21 from the Mono-Q column) contains 10-12 major peaks. In contrast, a more complex HPLC profile, with 18-20 major peaks, is obtained from the protein preparation containing 60% PAR 1 and 40% Par 2 (panel B ) . The positions of some of the peaks in panel B coincide with the positions of peaks in panel A , but several distinct peaks present in panel B are absent in panel A . These observations are consistent with the view that 1) PAR 1 and PAR 2 are two distinct polypeptides, 2) the major peaks observed in panel A represent peptide fragments derived from PAR 2, and 3) the peptides present in panel B but absent in panel A are derived from PAR 1.
The amino acid sequence of several CNBr peptides presumably derived from PAR 2 (Fig. 2, panel A ) was determined by direct amino acid sequence analysis. Unambiguous sequences were obtained from the peaks labeled with numbers 4, 7, 9, 11,13, 14, and 16 in Fig. 2, panel A . These sequences, which correspond to eight different peptides and a total of 217 amino acids, are shown in Fig. 3 under PAR 2. As will be shown in a different section of this paper, all of these sequences could be accounted for in the complete amino acid sequence of PAR 2 deduced from the nucleotide sequence of the PAR 2 gene, thus confirming the supposition that the major peptides in  Fig. 1) was recovered and subjected to mercaptolysis, alkylation, and CNBr cleavage as described under "Materials and Methods." The CNBr peptides were separated by reversed-phase on a Nucleosil C,, 300-10 (4 X 250mm) column. The solvent system consisted of 0.1% trifluoroacetic acid in water ( A ) or in acetonitrile ( B ) . The proportion of solvent B was increased from 0 to 50% within 110 min. The peaks were collected manually. Panel A corresponds to the CNBr peptides from PAR 2, and panel B to the peptides from the mixture of PAR 1 and PAR 2. Material from the numbered peaks was used for amino acid sequence analysis. The shaded areas in panel B are peptides identified by amino acid sequence as corresponding to PAR 2 peptides shown in panel A. to PAR 2. Thus, the amino acid sequence of peptide 7 in panel A was found in peak 13 in panel B, while the sequences of peptides 13 and 14 in panel A were found in peaks 23 and 24, respectively, in panel B. The positions of these PAR 2 peptides are identified as shaded areas in Fig. 2, panel B. Unambiguous amino acid sequences were obtained from the peaks labeled with numbers 4, 7,[11][12][13][14]20 and 24 in Fig. 2, panel B. These sequences, which correspond to eight different peptides and a total of 128 amino acids, are shown in Fig, 3 under PAR 1 & PAR 2. None of the latter sequences correspond to PAR 2 sequences, whether determined directly or deduced from the nucleotide sequence of the PAR 2 gene. Since PAR 1 is the only observable protein other that PAR 2 in fractions 25 and 26, it is very likely that these sequences represent portions of PAR 1.

Western Blot Analysis of T. brucei, T. cruzi, and L. brasi-
liensis Extracts-The above results, which directly show that PAR 1 and PAR 2 are two distinct polypeptides, agree with our previous observations that indicate lack of immunological cross-reactivity between these two major components of the paraflagellar rod of T. cruzi (4). Others have shown that monoclonal antibodies to the paraflagellar rod of T. brucei (20) or L. brasiliensis (21) react with the two major paraflagellar polypeptides detected in each of those parasites. In T. brucei these results have been interpreted either as indicative of common epitopes in two different polypeptides (20) or as due to the existence of a single polypeptide with two conformational variants that exhibit slightly different electrophoretic mobilities (3). As previously reported, antibodies to T. cruzi PAR 1 or PAR 2 each react with a single polypeptide in extracts of that parasite (4). These latter results, however, critically depend on efficient prevention of proteolytic activity during extract preparation. If proteolysis is not completely inhibited, antibodies to PAR 1 or PAR 2 can each react with several proteolytic fragments of the corresponding polypeptides (4). In an attempt to determine whether antibodies to T. cruzi PAR 1 and PAR 2 each react with a single polypeptide in lysates of T. brucei or L. brasiliensis, extracts of these organisms were prepared by direct solubilization of pelleted parasites in boiling 2% SDS solution, in our hands the most efficient procedure for immediate and efficient inactivation of proteolytic activity. These extracts, together with a T. cruzi extract included as a control, were fractionated by one-dimensional PAGE and processed for Western blot analysis. The results shown in Fig. 4 indicate that pcAbPAR 1 reacts with a single polypeptide of molecular mass about 70 kDa, while mAbPAR 2 reacts with a single polypeptide of molecular mass about 68 kDa in the extracts of T. brucei (lanes b and b'), T.  cruzi (lanes c and c'), and L. bradiemis (lanes d and d'). shown in Fig. 5. The portions of the deduced amino acid sequence that have been verified by direct amino acid sequence analysis of PAR 2 are also indicated. The presence of nucleotide sequence coding for all PAR 2 peptides determined by direct amino acid sequence analysis indicates that these cDNAs do indeed code for PAR 2.
Attempts to determine the putative NH,-terminal sequence of PAR 2 were unsuccessful, possibly because the protein is naturally blocked at the NH, terminus or was blocked during purification (e.g carbamylation by urea). Although direct amino acid sequence analysis of the NHr terminus of PAR 2 could not be obtained, we believe that the coding region begins The location of the initially isolated r D S A clone. Tccl'nr?a, is indicated, as is the single base difference hetween TccPar'Lh and Tccl'arZc. Amino acids that have heen verified hv direct amino arid sequence analysis are indicated hy an ocvrlinc. Signififant restriction enzyme sites are also indicated. at the indicated ATG for several reasons. 1) Translation starting at the indicated start site would produce a protein of 69,439 Da, consistent with the observed M , of 68,000. Translation starting at the next in-frame ATG would produce a protein of only 58,450 Da. 2) The sequence environment around the next most plausible ATG (AGCTG ATG C), where direct amino acid sequence data are available, does not conform to the consensus identified for ribosome-binding sites (23) (i.e. a purine, usually A at the highly conserved -3 position and a purine, usually G at the +4 position). In contrast, the environment around the first ATG contains a purine in both of these positions (AACCA ATG A). 3) The nucleotide sequence was examined using Fickett's Testcode for coding region determination (24). This analysis is based on the frequency of nucleotides in each of the three potential codon positions in all three frames. The Testcode results indicate that the entire region between the first in-frame ATG and the termination codon is probably a protein coding region (data not shown).
TccParZa, which was initially identified by its polypeptide product in an expression vector, was determined to be the 3' half of the gene, including a poly(A) tail. TccPar2b and TccPar2c do not contain poly(A) tails and neither included 5' mini-exon sequence (25). The only polymorphism found was in a poly(A) region 5' of the coding region. Clone Tcc-Par2b contained 10 As while TccPar2c contained 11.
Stage-specific Expression of the PAR 2 Gene-To determine the developmental expression pattern of the PAR 2 gene, a Northern blot containing trypomastigote and epimastigote poly(A)+ mRNA was hybridized with the '"P-labeled 738-base pair PstI fragment from TccPar2c (Fig. 6). to a Southern blot containing trypomastigote nuclear DNA digested with KspI (Fig. 7). Included on the Southern blot was TccPar2c subcloned into Rluescript and restricted with KspI in amounts equivalent to 2, . i , 10, 20. 30, 40. and 50 copies per haploid genome. Strong hybridization of the probe was observed to a genomic fragment of 1.7 kb. [{'hen the intensity of the hybridization signal in the genomic DNA is compared to that of the various equivalents in the cloned DNA, the 1.7-kb fragment is seen to occur approximatelv 30 times per haploid genome.  b l~~~E d~s~~~W k~~~~~~   70  80  90  100  110  120   SEPQKQKLWUHEKAQELLSSVEGGTKWNLTEAYDIKKIX~VCGLQLSVRELYKPEDKPN   SETQKQKLQAAHEKAHELLASVEGGTWSLTEAYDIKK~VCGLEMSVFSLYKPEDKPQ   I I I I I I I I I I I I I I I l l I I I I I I I I : I l I l I I I I I I I I I I l I : I I I I I I I I I I l I   130  140  150  160  170 I I I : I I I I I I : I I I l I l I I I I l I I l I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I   190    To determine if the multiple copies are dispersed in the genome, chromosome size DNA molecules from Esmeraldo epimastigotes were separated by PFGE and blotted to nitrocellulose. Hybridization of the chromosome blots with ["PI TccPar2a revealed a single band corresponding to a chromosome of 0.78 megabase size (Fig. 8), suggesting that all of the PAR 2 genes are located on a single chromosome.

HPAEEVESGNlIDRRSKllVEYRAHLAKQEEVKIAAEREE~RS~~S~YRGRTHPQITQ
Comparison of Paraflagellar Rod Proteins from Two Trypa-nosomes-The coding region nucleic acid and amino acid sequences of the PAR 2 protein from T. cruzi were compared to those of the PFR protein from T. brucei (3). At the nucleic acid level there was 83.3% identity between the two genes (data not shown). The results of the amino acid comparison are shown in Fig. 9. Identity was observed for 541 of 600 amino acids (90.2%) with an additional 37 conservative amino acid changes (total 96.3%).

DISCUSSION
The paraflagellar rod is a structure closely associated with the axoneme in the flagella of Trypanosomatids and Euglenoids. Although this structure has been thoroughly studied at the ultrastructural level (26), information about its molecular composition, organization, and function(s) is scant.
Two polypeptides, with molecular masses about 70 kDa, have been tentatively identified as components of the paraflagellar body of a variety of Trypanosomatids and Euglenoids (20). Whether these two putative paraflagellar polypeptides correspond to two different gene products, or to conformational variants of a single one, as suggested by Schlaeppi et al. (3) for T. brucei, has remained an open question. In a previous work (4), we directly identified the polypeptide PAR 2 as one of the major components of the T. cruzi paraflagellar rod and suggested that a second paraflagellar polypeptide, named PAR 1, was distinct from PAR 2. Three lines of evidence presented in this paper indicate that indeed PAR 1 and PAR 2 are two distinct paraflagellar rod components. 1) These two polypeptides can be partially separated by ionexchange chromatography. As shown in Fig. 1, protein preparations containing about 80% PAR 2 and 20% PAR 1 (inset, lane 21) and 60% PAR 1 and 40% PAR 2 (inset, lanes 25 and 26) were reproducibly obtained. 2) The HPLC profiles of CNBr cleavage products from protein preparations enriched in PAR 2 (Fig. 2, panel A ) or containing mixtures of PAR 1 and PAR 2 (Fig. 2, panel B ) are clearly different. These results are only compatible with PAR 1 and PAR 2 being two different polypeptides, since the HPLC profiles of conformational variants of the same gene product would be identical.
3) The results of amino acid sequencing analysis provide definitive evidence for PAR 1 and PAR 2 being two distinct polypeptides. First, all the amino acid sequences of peptides derived from protein preparations enriched in PAR 2 (Fig. 3) can be accounted for in the complete protein sequence deduced from the nucleotide sequence of the corresponding gene ( Figs. 5 and 9). The sequences determined directly correspond to the following amino acids deduced from the gene sequence: 552, and 557-600 (Fig. 3). Second, the amino acid sequences of peptides from mixtures of PAR 1 and PAR 2 (Fig. 3) include PAR 2 sequences (PAR 2 amino acids 100-119, 122-150 and 316-345 were identified), as well as sequences of at least eight major peptides unrelated to PAR 2. These latter peptides likely correspond to CNBr cleavage products from PAR 1, the only other major polypeptide detected in preparations of purified paraflagellar proteins.
These results show that PAR 1 and PAR 2, the major paraflagellar polypeptides of T. cruzi, are chemically distinct, in agreement with our previous results (4) which indicated these two proteins to be immunological distinct. In apparent contradiction with these data, other workers reported that monoclonal antibodies to the paraflagellar rod of T. brucei (20) or Leishmania (21) reacted with the two major paraflagellar polypeptides of each of those parasites, results that in one case were interpreted as indicative of common epitopes in those polypeptides (20). Two types of data, however, would   100-119,122-150, 174-187,189-226,228-251, 316-345, 524-argue against the latter interpretation. 1) We have previously shown that mAbPAR 2, a monoclonal antibody to PAR 2, reacts in Western blots with a single polypeptide when T. cruzi extracts are obtained under stringent proteolysis-free conditions. In contrast, mAbPAR 2 reacts with several peptides if proteolysis is not totally inhibited ( 4 , results that could be erroneously interpreted as indicative of the existence of different paraflagellar peptides with common epitopes.

2)
In Western blots of extracts of either T. brucei or L. brasiliensis prepared under proteolysis-free conditions, each of our antibodies (4) specific for PAR 1 (pcAbPAR 1) or PAR 2 (mAbPAR 2) reacted with a single polypeptide (Fig. 4). Although the combined results of these immunological and amino acid sequencing studies provide solid evidence for two major paraflagellar proteins in T. cruzi, the question of possible homologies between PAR 1 and PAR 2 will only be resolved when the complete amino acid sequence of PAR 1 becomes available. Also, we cannot currently exclude the possible existence of minor polypeptides with epitopes in common with the major paraflagellar proteins. In fact, our amino acid sequencing studies of PAR 2 point to the existence of sequence microheterogeneity (to be reported elsewhere). This observation opens the possibility that more than one of the genes in the PAR 2 tandem array may be expressed.
Previous immunological studies also demonstrated crossreactivity between paraflagellar components of different members of the Trypanosomatid family, and even with members of the more distant Euglenoid family (20). These observations predicted some degree of conservation in the primary structure of paraflagellar polypeptides, although the degree of conservation and the components involved were not identified. In this paper we present the complete amino acid sequence of T. cruzi PAR 2, one of the two major protein  (Table I). We anticipate that evolutionary conservation of primary structure may not be confined to PAR 2 but may extend to other paraflagellar components since, as mentioned before, pcAbPAR 1, a polyclonal antibody to T. cruzi PAR 1, reacts with a single polypeptide on Western blots of both T. brucei and L. brasiliensis (Fig. 4). This high degree of conservation, so far only proven for PAR 2, indicates that stringent structural requirements, relevant for either macromolecular organization or a highly specialized function(s) have been imposed upon the cross-reactivity of the paraflagellar components of Trypanosomatids and Euglenoids is intriguing, as the paraflagellar rod of the first group of parasites is organized as a compact filamentous structure, while in the latter group the paraflagellar rod is a hollow cylindrical body (27).