Sand Fly Saliva Test Shows Protein Factory Choice Can Make or Break Exposure Assays

Sand flies are among the smallest and most inconspicuous carriers of human disease, yet the pathogens they transmit, including Leishmania parasites, impose a heavy burden on millions of people across the tropics and subtropics. One of the central challenges in controlling these diseases is simply knowing who has been bitten, and how often. Because infection risk rises with repeated exposure to infected vectors, researchers have long sought reliable biological markers that reveal a person’s or an animal’s history of sand fly bites. A new study published in the journal Parasites & Vectors by Barbora Dvořáková, Iva Kolářová and colleagues at Charles University in Prague, together with collaborators in Turkey, the Czech Academy of Sciences and the National Institutes of Health in the United States, has now examined a deceptively simple question with far-reaching consequences: does the cellular factory used to manufacture a diagnostic protein change how well that protein performs as a marker of exposure?

The marker in question is SP32, a salivary protein produced by the female sand fly Phlebotomus papatasi, the principal vector of Leishmania major in many parts of the Middle East, North Africa and Asia. When a sand fly takes a blood meal, it injects a cocktail of salivary proteins that modulate host hemostasis and immunity. Over time, hosts that are bitten repeatedly mount antibody responses against these proteins, and those antibodies can be detected in serum. Serological assays built on salivary antigens therefore offer a window into exposure that does not depend on catching flies in the act or on the presence of parasites. Recombinant versions of individual salivary proteins, produced in the laboratory rather than dissected from thousands of fly salivary glands, have become the preferred reagents for such assays, and SP32 produced in human embryonic kidney cells, known as HEK cells, has recently been validated for assessing human exposure to P. papatasi.

Recombinant proteins, however, are not all created equal. The same gene expressed in different host cells can yield molecules with different folding patterns, different post-translational modifications and, consequently, different antigenic properties. The research team set out to compare three expression systems side by side. The first was HEK cells, the mammalian platform used in previous validation studies. The second was Escherichia coli, the workhorse of molecular biology, prized for its speed and low cost but incapable of many eukaryotic protein modifications. The third was Sf9 cells, derived from the fall armyworm Spodoptera frugiperda and infected with baculovirus vectors, an insect-cell system that the authors reasoned should provide the most natural environment for producing a recombinant sand fly protein, since the protein’s native context is itself an insect.

To evaluate the three versions of SP32, the researchers assembled an unusually rich panel of sera spanning three host species. Human and dog serum samples came from individuals and animals naturally exposed to sand flies in endemic areas of Turkey, collected with ethical approval from Ege University and informed consent from participants and dog owners. Mouse sera, by contrast, came from laboratory animals experimentally exposed to a single insect species under controlled conditions approved by Czech authorities. This combination of natural and experimental exposure is critical: field sera capture the messy reality of real-world exposure, where hosts may be bitten by several sand fly species and other biting insects, while the murine samples allow the specificity of each antigen to be tested with precision, because the exact biting history of every animal is known.

The benchmark against which all three recombinant proteins were judged was salivary gland homogenate, the crude natural antigen prepared from dissected P. papatasi salivary glands. Using enzyme-linked immunosorbent assays, or ELISA, the team measured how strongly antibodies in each serum sample reacted with each version of SP32, and then analyzed the results with receiver operating characteristic curves, which quantify sensitivity and specificity across a range of diagnostic thresholds. The results pointed clearly in one direction. SP32 produced in Sf9 insect cells performed as well as, or better than, the HEK-derived and E. coli-derived versions in terms of sensitivity, specificity, correlation with the reference antigen and species specificity. In human sera, SP32-Sf9 emerged as the most effective marker of exposure among the three recombinant candidates.

Intriguingly, the assay parameters were even stronger with canine sera than with human sera, a finding with practical implications for veterinary and One Health applications. Dogs are reservoir hosts for several Leishmania species, including L. infantum transmitted in the Mediterranean basin by sand flies such as Phlebotomus perniciosus, and monitoring exposure in dog populations is a cornerstone of leishmaniasis surveillance. A serological marker that performs robustly in dogs could therefore support both epidemiological studies and the evaluation of vector control interventions, such as insecticide-treated collars and topical repellents, whose success is often measured by reductions in bite exposure rather than infection alone.

The mechanistic explanation for the differences between the three proteins lies in post-translational modifications, the chemical decorations that cells add to proteins after they are synthesized. The researchers found that SP32 produced in HEK cells carries O-glycosylation, the attachment of sugar chains to oxygen atoms on serine or threonine residues, a modification typical of mammalian cell machinery. Neither the E. coli product, which lacks all glycosylation because bacteria do not perform it, nor the Sf9 product carried this modification. These sugar chains, foreign to the immune system of a host that has only ever encountered the native insect protein, may alter which epitopes are presented to antibodies and may even introduce artificial epitopes that distort the assay. The absence of O-glycosylation in the Sf9 product, combined with an insect-like folding environment, may explain why this version most faithfully mimicked the natural antigen.

For the field of vector biology, the study delivers a lesson that extends well beyond a single protein or a single fly species. Salivary proteins from many disease-transmitting arthropods, including mosquitoes, ticks and triatomine bugs, are under investigation as exposure markers and, in some cases, as vaccine candidates aimed at blocking transmission. In each case, the choice of expression system involves a trade-off. Bacterial expression is cheap and fast but yields unmodified proteins that may lack critical native features. Mammalian expression offers sophisticated processing but can add modifications that the native protein never carries. Insect-cell expression, as this work demonstrates, can hit a sweet spot for proteins of insect origin, providing eukaryotic folding machinery without imposing mammalian-specific glycosylation. The authors conclude that the expression system and its characteristics should be carefully selected and tailored to the purpose of recombinant protein production, weighing advantages and disadvantages rather than defaulting to whichever platform is most convenient.

The work also carries practical weight for public health programs in endemic regions. Cost matters: dissecting salivary glands from laboratory-reared flies is laborious and expensive, and a validated recombinant antigen that can be produced at scale opens the door to large serological surveys. Because Sf9-based production matched or exceeded the performance of the more expensive mammalian platform, the findings suggest that high-quality exposure assays need not depend on costly mammalian cell culture. At the same time, the study was funded in part by the European Commission’s CLIMOS project, which examines the climate-driven spread of sand fly-borne diseases, a reminder that as changing environmental conditions push vectors into new territories, tools to map human and animal exposure will become only more valuable.

Ultimately, the study is a reminder that in molecular diagnostics, the identity of a protein is defined not only by its amino acid sequence but by the cell that makes it. Two recombinant proteins with identical sequences can behave differently in a serological assay because of subtle differences in glycosylation, folding or purity. By systematically comparing three production platforms against a natural reference antigen, across three host species and with both natural and controlled exposure histories, Dvořáková, Kolářová and their colleagues have provided a template for how diagnostic antigens should be evaluated before deployment. For communities living under the threat of sand fly-borne disease, that rigor translates into more trustworthy answers to a vital question: how much exposure is really occurring, and where should control efforts be aimed next.

Subject of Research: Effect of recombinant protein expression systems on the performance of Phlebotomus papatasi salivary protein SP32 as a serological marker of sand fly exposure

Article Title: The protein expression system impacts the efficacy of Phlebotomus papatasi SP32 as a marker of exposure in humans, dogs, and mice

Article References: Dvořáková, B., Pyrihová, E., Kolarčíková, T., Arserim, S. K., Pekagirbas, M., Yetismis, K., Oliveira, F., Valenzuela, J. G., Özbel, Y., Töz, S., Alblová, M., Volf, P., & Kolářová, I. (2026). The protein expression system impacts the efficacy of Phlebotomus papatasi SP32 as a marker of exposure in humans, dogs, and mice. Parasites & Vectors. https://doi.org/10.1186/s13071-026-07743-0

Image Credits: AI Generated

DOI: 10.1186/s13071-026-07743-0

Keywords: Phlebotomus papatasi, SP32, recombinant protein, Sf9 cells, HEK cells, Escherichia coli, serology, exposure marker, ELISA, sand flies, leishmaniasis, glycosylation