PRELIMINARY STUDIES ON THE USE OF SYNTHETIC OLIGONUCLEOTIDE PROBES BASED ON CONSERVED PROTEIN SEQUENCES FOR IDENTIFYING GENES IN MOSQUITOES AND MALARIA PARASITES.

The isolation and sequencing of particular genes is a necessary step in determining the amino acid sequence of the corresponding proteins and the regulation of their synthesis. In one approach to isolating genes, limited amino acid sequence is used to deduce the corresponding coding sequence. Oligonucleotides of this sequence are then chemically synthesized, radioactively labelled and used as hybridization probes to detect target coding sequences in a clone bank. The redundancy of the genetic code poses a limitation to this approach and probe sequences are usually based on regions of protein sequence that are rich in amino acids with a restricted number of codons eg. methionine and tryptophan. To cover all codon possibilities, it is sometimes necessary to synthesize probes containing a mixture of different individual nucleotides. One way to overcome this problem is to use a base analogue that can pair with any one of the four natural bases at ambiguous positions with or without forming hydrogen bonds. Inosine, which is found in the 5 or "wobble" position of tRNA anticodons, forms base pairs with A, C and U in mRNA. 2'-deoxyinosine is therefore an useful insert at ambiguous codon positions in synthetic DNA probes.1


INTRODUCTION
The isolation and sequencing of particular genes is a necessary step in determining the amino acid sequence of the corresponding proteins and the regulation of their synthesis. In one approach to isolating genes, limited amino acid sequence is used to deduce the corresponding coding sequence. Oligonucleotides of this sequence are then chemically synthesized, radioactively labelled and used as hybridization probes to detect target coding sequences in a clone bank. The redundancy of the genetic code poses a limitation to this approach and probe sequences are usually based on regions of protein sequence that are rich in amino acids with a restricted number of codons eg. methionine and tryptophan. To cover all codon possibilities, it is sometimes necessary to synthesize probes containing a mixture of different individual nucleotides. One way to overcome this problem is to use a base analogue that can pair with any one of the four natural bases at ambiguous positions with or without forming hydrogen bonds. Inosine, which is found in the 5 or "wobble" position of tRNA anticodons, forms base pairs with A, C and U in mRNA. 2'-deoxyinosine is therefore an useful insert at ambiguous codon positions in synthetic DNA probes.1 ~rolcases in malaria parasites are needed for degrading ol, haemoglobin, protein processing,2 invasion of host erythrocytes3 and immune evasion. Parasite proteases may therefore be useful targets for the rational design of inhibitory drugs.
Analogues of insect hormones may modulate the physiology of insects and are . therefore potentially useful as agents for controlling insect vectors of disease. Many insect hormones are peptides, and in some cases are structurally related to vertebrate peptide hormones, indicating a common ancestral gene. Examples are some of the prothoracicotropic hormones (PlTH) of Bombyx mon (silkworm) that stimulate ecdysone release from the prothoracic glands. The 4400 m.wt PTTH-I1 is structurally homologous to ihe insulin A chain and insulinlike growth factor 14. Such conserved sequences may be used to produce oligonucleotides to identify homologous neuropeptide coding genes in insects such as mosquitoes that are of medical importance.
We report here on experiments to test the feasibility of using synthetic oligonucleotides to determine the presence of genes coding for a 26 kDa P. falcipamm protein, serine protease, cathepsin D like aspartyl protease, cysteine protease and a P1TH-11-like peptide in the human malaria parasite Plusmodillm falcipamm Welch and a laboratory vector of human malaria, Anopheles farauti Laveran.

Protein sequence and oligonucleotide synthesis
Conserved amino acid sequences were selected for probe synthesis by examining published sequences of proteins for suitable sequences with minimal codon multiplicity. The temperature at which 50% of hybrids dissociate, Td, was calculated using the formula : (a) Insulin like insect neuropeptide4 amino acid sequence Synthesis of oligonucleotides was performed on an automated synthesizer (Applied Biosystems, CA), and cleavage and deprotection in concentrated NHIOH at 6 0 '~ for 16h. The oligonucleotides were purified by gel filtration on Sephadex G-50 columns using. lOmM triethylammoniurn bicarbonate buffer pH 7. Fractions absorbing strongly at 260 nm were pooled and freeze dried.

preparation of DNA
Parasite DNA was obtained from irt vitro cultures of P. falcipamm FC27 isolate. 0.5 ml schizonts at > 50% parasitaemia prepared by gelatine sedimentation8: were used'for DNA preparation. Three ml of An. farauti pupae obtained fr0m.a laboratory colon? were used to prepare mosquito DNA.
Parasites and pupae were resuspended in 1 -2 ml TE (10 mM Tris.Cl, 1 mM EDTA pH 7.4) to which was added 20 -50 ml of a solution containing 0.5 M EDTA pH 8, 100 jig ml" proteinase K and 0.5% sarkosyl (Sigma, MO) and incubated for 3h at 5 0 '~ to digest protein. The preparations were extracted with phenol (containing 0.1% 8-hydroxyquinoline as a free radical scavenger) by standard procedures10 three times and the aqueous layer with DNA dialysed against several changes of a buffer containing 50 mM Tris.Cl, 10 mM EDTA, 10 mM NaCl, pH 8. The DNA was then treated for 3h at 37% with 100 pg ml-' of bovine pancreatic ribonuclease A (Boehringer, CA ) that had been pre-heated to 100'~ to destroy contaminating DNAase.
The DNA was then re-extracted with phenol twice and once with water saturated ether to remove residual phenol. The DNA was precipitated by adding 2 vols of ice cold ethanol and incubated for 16h at -70'~. The precipitates collected by centrifugation were resuspended in TE and dialysed against TE at 4'~. The concentrations of the resulting mosquito and parasite DNA were 0.175 mglml and 0.075 mglml respectively.
Restriction enzyme digestion and electrophoresis 10 pg of mosquito and parasite DNA were completely digested with 10 units of the following restriction enzymes Bam HI, Dra 1, Eco R1, Hind 111, Pst 1 and Rsa 1 (BRL, Gaithersberg, MD) by standard procedures'0. Digested DNA was separated by electrophoresis on 0.7% agarose gels using Hind I11 digest as a size marker and the separated DNA, denatured and transferred to nylon membrane by Southern blotting1'.

Dot blots
10, 1 and 0.1 pg aliquots of the DNA, diluted in TE, were heated at 9 5 '~ for 5min to denature the DNA and then cooled on ice for 5min. An 8 x 11 cm piece of Hybond N nylon membrane (Amersham, UK) was pre-wetted in 0.1 M Tris pH 7.4, and samples of mosquito and parasite DNA were blotted onto the membrane using a dot-blot apparatus (Biorad, CA). The DNA in the blots were again denatured in 1.5M NaCI, 0.5M NaOH for 1 min followed by neutralization in 1.5M NaCI, 1M Tris pH 8 for 1 min. Dried membranes were UV irradiated for 3 min to cross-link the DNA to nylon. The membranes were treated with pre-hybridizing ~olution'~ containing 5 x Denhard's solution," 0.1% sodium dodecyl sulphate, 6mM EDTA, 0.9M NaCl, U)mM Tris pH 8 and 10% dextran sulphate at 2 5 '~ for 4h. 100 pl of solution was used per sq. cm of membrane. For hybridization, the pre-hybridization solution was changed once and the labelled oligonucleotides added directly to the pre-hybridization solution. Hybridization was performed at 2 5 '~ for 16h. The membrane; were then washed 3x at 2 5 '~ for 20 min each in 6x concentrated saline sodium citrate1° (6SSC) containing 0.1% SDS followed by three washes at 2 5 '~ of lOmin each in inSC, 0.1% SDS. The membranes were dried, wrapped in thin plastic and autoradiographed with X-ray film overnight.

Southern blots
Southern blots were hybridized with 3 2~ labelled oligonucleotides in an identical manner to dot blots.

RESULTS
The results of dot blots with the different labelled oligonucleotides are presented in Figure 1. The probes directed against the 26 kDa P. falcipanim antigen, the insulinlike insect neuropeptide, cysteine protease and serine protease hybridised to sequences present in both parasite and mosquito DNA. The latter three oligonucleotides appeared to react more strongly with mosquito DNA rather than parasite DNA. The aspartyl protease showed a weak reaction with mosquito but not parasite DNA.
In subsequent Southern blot analysis of P.falc@amnt DNA, the serine protease specific probe hybridized to multiple fragments in many restriction enzyme digests ( Figure 2).

DISCUSSION
The results demonstrate the potential use of 3 2~labelled oligonucleotides for identifying specific genes. The dot blot patterns indicate the probable presence of genes for cysteine and serine proteases and for insulin like peptides or proteins in P. falcipanml and An. farauti. The hybridization of the serine protease probe to several but discrete DNA fragments after restriction enzyme digestion may be due to the presence of different genes with the corresponding conserved sequences. This is similar to what is observed in higher organisms where genes for several different serine proteases and serine proteaserelated proteins are present in the genome.
However, hybridization of the probes to irrelevant coding or non-coding sequences in mosquito or parasite DNA, due to probe redundancy, cannot be excluded.
False-positive hybridization can be minimized by increasing the stringency of the hybridization and washing conditions eg. by using higher temperatures. Since the codon usage frequencies for P.falcipmm have been determined recently1', it is now also possible to synthesize less-redundant oligonocleotide probes that can further reduce false-positive hybridization.
The function of the cytoplasmic 26 kDa antigen in ~larmodiu&* is not known. The presence of homologous sequences in mosquitoes suggests that this may be a gene conserved across different organisms or that the probe sequence is very degenerate?
Polymerase chain reaction based amplification of the genes using conserved oligonucleotide sequences may be used to isolate the corresponding genes as the next step. An alternative procedure is to use the DNA probes to isolate clones from a DNA library prepared by digesting DNA with appropriate restriction enzymes. From the data presented, Dra 1, Hind I11 and Rsa 1 libraries of P. falcipamm genomic DNA may be suitable for isolating genes for serine proteases. With both procedures, a number of irrelevant coding or non-coding sequences homologous to the probes might be isolated together with the relevant genes.