关键词: 蛇毒, 物种鉴定, 特征肽段, 纳升级液相色谱-串联质谱, 毛细管电泳-串联质谱, 平行反应监控, 法医学

Abstract: Snakebite envenoming is a serious public health issue worldwide. An estimated 1.2 to 5.5 million cases occur annually, resulting in over 421000 envenomations and approximately 20000 deaths. In forensic practice, suspected snake venom poisoning cases are increasingly encountered. The critical tasks are to confirm whether envenomation has occurred and to identify the snake species. Current identification relies mainly on bite marks and clinical symptoms. Yet ambiguity arises in dry bites, intentional injection, or absent bite marks. Immunoassays, though widely used, suffer from cross-reactivity and require species-specific antibodies that limit their universal application. PCR-based method offers high accuracy but is constrained by DNA degradation and long amplicon requirements. Shotgun proteomics provides high-throughput peptide information. Yet conserved repeats among homologous venom proteins cause one-peptide-to-multiple-proteins problem, complicating species assignment.
To address these challenges, we developed a novel strategy based on characteristic peptides using nano-liquid chromatography-tandem mass spectrometry (nanoLC-MS/MS), coupled with parallel reaction monitoring (PRM) for targeted detection in complex matrices. The systematic selection and validation of multiple characteristic peptides per species greatly enhanced specificity and reliability. Four important venomous snakes in China, Trimeresurus stejnegeri, Ophiophagus hannah, Bungarus multicinctus, and Daboia russellii, were investigated. Venom proteins were digested with trypsin, and the resulting peptide mixtures were analyzed by nanoLC-MS/MS. The identified peptides were then subjected to BLAST searching against the Uniprot database to filter out those uniquely present in each species. For each snake, we selected 5 characteristic peptides that are strictly species-specific, which are absent from all other organisms in the database. Any single peptide is sufficient for positive identification, while the concurrent detection of several peptides further strengthens the confidence.
The feasibility of the approach was first demonstrated by extracted ion chromatograms of the characteristic peptides. Clear signals were observed only in the corresponding venom samples, with no interference from other snake venoms or from healthy human serum, confirming excellent selectivity. To improve sensitivity and quantitative capability, we applied PRM mode. PRM is a targeted MS/MS acquisition mode, which selectively monitors the fragmentation of predefined precursor ions. Compared with conventional data dependent acquisition, PRM increased the peak areas of characteristic peptides by approximately 3 to 132 folds. Furthermore, the venom samples were serially diluted (10, 100, and 1000 folds) in human serum matrix and analyzed by PRM. The characteristic peptides remained clearly detectable with good signal to noise ratios. The signal intensities correlated positively with venom concentrations, demonstrating robust matrix tolerance and the potential for semi-quantitative analysis. These results indicate that the PRM-based strategy is not only highly specific but also sensitive. This strategy is particularly valuable for forensic casework where sample amounts are often limited.
Moreover, we assessed the universality of the characteristic peptide concept by comparing LC-MS/MS with capillary electrophoresis-tandem MS (CE-MS/MS). Using the same data processing parameters, CE-MS/MS identified a greater number of peptides overall. CE-MS/MS identified a broader distribution of peptide lengths, especially more low molecular weight species. This difference is mainly owing to the absence of stationary phase interactions of CE. Importantly, two characteristic peptides of Trimeresurus stejnegeri were commonly identified by both techniques. This result confirmed that the characteristic peptide strategy is compatible with diverse separation platforms. This versatility enhances its practical applicability in different laboratories.
In summary, this work established a sensitive, reliable, and universally applicable proteomic method for snake venom species identification via characteristic peptides. The rigorous selection of multiple characteristic peptides overcame cross-reactivity and the peptide to protein ambiguity. The integration of PRM significantly boosted sensitivity and anti-interference ability in complex matrices while offering semi-quantitative potential. This method demonstrated the transferability across LC and CE based separation systems. Furthermore, recent advances in rapid enzymatic digestion hold promise to shorten the total analysis time to less than one hour. With the less analysis time, this method will be feasible for clinical emergency settings as well. This strategy provides a powerful new tool for forensic investigation of snakebite cases, aiding in source tracing and evidentiary support. It can be readily extended to other venomous species or toxin related proteins by updating the characteristic peptide library.

Key words: snake venom, species identification, characteristic peptide, nano-liquid chromatography-tandem mass spectrometry, capillary electrophoresis-mass spectrometry, parallel reaction monitoring, forensic science