High-throughput dia-PASEF workflow improves proteome coverage and quantitative performance in repeated-measures proteomics of cervicovaginal fluid within a preconception cohort

This study demonstrates that a high-throughput dia-PASEF proteomics workflow significantly outperforms conventional DDA-FracOffline methods by increasing protein coverage by approximately 50% and enhancing quantitative reproducibility, thereby enabling more robust longitudinal analysis of cervicovaginal fluid dynamics across the menstrual cycle.

The Gist

Imagine the female reproductive system as a bustling, high-stakes city. Every day, this city changes its mood, its traffic patterns, and its construction projects based on a complex schedule (the menstrual cycle). To understand how this city works, or to spot when a storm is coming (disease), scientists need to take "snapshots" of the city's atmosphere.

The fluid that surrounds the cervix and vagina (Cervicovaginal Fluid, or CVF) is like the city's atmospheric air. It carries tiny messengers (proteins) that tell the story of what's happening inside. However, reading these messages is incredibly difficult because the air is thick, the messengers are tiny, and they change rapidly from day to day.

This paper is about a team of scientists who built a super-powered microscope to read these messages better than ever before.

The Problem: The Old Camera vs. The New Drone

For a long time, scientists used a method called DDA (Data-Dependent Acquisition) to study this fluid. Think of this like an old-fashioned photographer taking pictures of a busy street.

• How it worked: The photographer would look at the crowd, pick the loudest or most obvious people to photograph, and snap a picture.
• The flaw: If the crowd moved fast (which it does in the menstrual cycle), the photographer would miss the quiet people in the back, or get confused and take the same picture twice. This resulted in blurry photos with lots of missing details (missing data) and inconsistent results.

The scientists in this paper wanted to try a new method called dia-PASEF. Think of this as a high-tech surveillance drone flying over the city.

• How it works: Instead of picking who to photograph, the drone scans everyone in the frame simultaneously, capturing every single person, from the mayor to the quietest pedestrian, regardless of how loud they are.
• The advantage: It creates a complete, crystal-clear map of the entire city, capturing details the old camera missed.

The Experiment: A Week in the Life

To test their new drone, the scientists studied six women who were trying to get pregnant. They collected fluid samples from these women every single day for a week right around the time of ovulation (the moment an egg is released).

They ran every single sample through both the "Old Camera" (DDA) and the "New Drone" (dia-PASEF) to see which one gave a better picture.

The Results: Why the Drone Won

The results were a landslide victory for the new method:

1. More Details: The "Drone" (dia-PASEF) found 50% more proteins than the "Old Camera." It didn't just see the loud, obvious proteins; it found the quiet, rare ones that might be the most important clues.
2. Less Guesswork: The old method had a lot of "missing data" (like a photo with big white spots where the camera failed). The new method had very few missing spots, making the data much more reliable.
3. Consistency: If you took the same photo twice with the old camera, the results might look different. The new drone was incredibly consistent, giving the same clear picture every time.
4. Spotting the Changes: Because the new method was so sensitive, it could detect tiny changes in protein levels from one day to the next. It was like being able to hear a whisper in a noisy room, whereas the old method only heard shouting.

Why This Matters: Finding the "Smoking Gun"

Why does finding more proteins matter?

• Fertility: It helps scientists understand exactly what happens in the body right before and after an egg is released, which could help couples trying to conceive.
• Disease Detection: The study showed that this new method could spot proteins linked to serious conditions like ovarian cancer and endometriosis much better than before. It's like upgrading from a magnifying glass to a high-resolution satellite image; you can spot the early warning signs of a fire before the building burns down.

The Bottom Line

This paper proves that by switching from an old, picky way of looking at biological fluids to a new, comprehensive "drone scan" method, scientists can finally see the full picture of women's reproductive health. It's a massive step forward for finding early biomarkers (warning signs) for diseases and understanding the complex, daily rhythms of the female body.

In short: They traded a shaky, picky camera for a steady, all-seeing drone, and now they can finally read the secret language of the female reproductive system.

Technical Summary

Problem Statement

Proteomic studies of the female reproductive system, specifically using cervicovaginal fluid (CVF), hold significant potential for advancing reproductive health and fertility research. However, large-scale biomarker discovery in CVF faces substantial challenges due to the fluid's complexity and the limitations of current mass spectrometry (MS) workflows.

• Limitations of Current Methods: Traditional Data-Dependent Acquisition (DDA) workflows, particularly those coupled with offline peptide fractionation (DDA-FracOffline), suffer from poor reproducibility, high rates of missing values, and limited quantitative accuracy. These issues are exacerbated in longitudinal repeated-measures study designs (tracking changes over time), where consistency is critical.
• The Gap: While Data-Independent Acquisition (DIA) methods like SWATH-MS have been explored, they often rely on fixed isolation windows that can introduce spectral interferences. There is a lack of systematic evaluation of 4D proteomics (integrating ion mobility) using dia-PASEF (parallel accumulation–serial fragmentation combined with DIA) for CVF analysis across consecutive days of the menstrual cycle.

Methodology

The study aimed to validate a high-throughput dia-PASEF workflow against a conventional DDA-FracOffline strategy for longitudinal CVF analysis.

• Cohort and Sampling:
  • Samples were drawn from the EARLY-PREG preconception cohort (NCT07358026).
  • A homogeneous subgroup of 6 women (non-conception cycles) was selected based on strict criteria (BMI, age, cycle length, reproductive history) to minimize biological variability.
  • Timeframe: Samples were collected daily across a 7-day, ovulation-anchored window (Day 0 = ovulation day, defined by LH surge; Days 0–6). This resulted in 42 CVF samples (7 days × 6 participants).
• Sample Preparation:
  • Proteins were extracted from both soluble and insoluble phases of CVF.
  • Samples underwent reduction, alkylation, and tryptic digestion.
• Mass Spectrometry Workflows:
  • Instrument: timsTOF Pro (Bruker) coupled with a NanoElute system.
  • DDA-FracOffline (Control): 200 µg of peptides were fractionated offline using high-pH reversed-phase chromatography into 8 fractions. Analysis used Data-Dependent PASEF mode (1 MS1 + 10 MS/MS scans).
  • dia-PASEF (Experimental): No offline fractionation. Analysis used DIA mode with 64 MS2 windows covering 400–1200 m/z and ion mobility range (0.57–1.46 1/K0). Cycle time was 1.80 seconds.
• Data Processing:
  • DDA: Processed via FragPipe (MSFragger) to generate a spectral library.
  • dia-PASEF: Processed via DIA-NN using the DDA-generated library.
  • Statistics: Bayesian t-tests for pairwise day comparisons; missing values imputed via missForest; pathway enrichment via ReactomePA and TISSUES 2.0.

Key Contributions

1. First Systematic Longitudinal CVF Proteome: This is among the first studies to apply dia-PASEF to CVF collected across consecutive days of the menstrual cycle, establishing a robust framework for time-dependent molecular profiling.
2. Methodological Benchmarking: Provides a direct, head-to-head comparison of dia-PASEF versus DDA-FracOffline in a complex biofluid, demonstrating the superiority of the DIA approach for longitudinal studies.
3. Biomarker Validation: Validates the workflow's ability to detect clinically relevant biomarkers (including FDA-proposed panels and literature-curated candidates for reproductive pathologies) with higher sensitivity and dynamic range.

Results

• Proteome Coverage:
  • dia-PASEF identified 4,229 quantifiable proteins, representing a ~50% increase over DDA-FracOffline (2,817 proteins).
  • While DDA identified more total proteins (9,593) due to deep fractionation, dia-PASEF provided significantly higher quantifiable yield with fewer missing values.
• Reproducibility and Variability:
  • Coefficient of Variation (CV): dia-PASEF showed superior quantitative consistency with a mean CV of 5.27 ± 0.68%, compared to 9.37 ± 1.75% for DDA-FracOffline.
  • Missing Values: dia-PASEF had a substantially lower proportion of missing values across samples compared to DDA (which reached up to 75% missing at low detection points).
  • Correlation: Biological replicates showed higher correlation with dia-PASEF (0.90) vs. DDA (0.70).
• Differentially Expressed Proteins (DEPs):
  • dia-PASEF detected significantly more DEPs in pairwise day comparisons (e.g., 952–1,116 DEPs) compared to DDA (161–460 DEPs).
  • This allowed for the capture of subtle, time-dependent molecular shifts across the luteal phase.
• Biomarker Detection:
  • FDA Panel: dia-PASEF detected 75% (91/121) of FDA-proposed biomarkers, compared to 63% (76/121) for DDA. It uniquely identified 15 FDA biomarkers (e.g., NT5E, ACE, VWF) that DDA missed.
  • Clinical Relevance: The workflow successfully tracked longitudinal variations of key proteins including IGFBP-4 (pregnancy/growth), MUC1 (barrier function), CA-125 (MUC16), and HE4 (WFDC2) (ovarian/endometrial cancer markers).
• Functional Enrichment: Pathway analysis revealed broader biological coverage with dia-PASEF, highlighting immune responses, cell signaling (RHO GTPases), and tissue-specific enrichment (uterus, cervix) that were less consistent in the DDA dataset.

Significance

• Advancement in Reproductive Research: The study establishes dia-PASEF as a superior, scalable, and reproducible workflow for longitudinal proteomics in CVF. This is crucial for understanding the dynamic molecular changes associated with fertility, early pregnancy, and gynecological diseases.
• Clinical Translation: By improving the detection of low-abundance proteins and reducing missing data, this method enhances the potential for discovering reliable biomarkers for conditions like endometriosis, ovarian cancer, and pregnancy complications.
• Efficiency: The dia-PASEF workflow offers a significant time advantage (minutes per sample vs. hours for DDA fractionation) without sacrificing depth, making it ideal for high-throughput clinical studies.
• Future Outlook: The authors propose this framework as a standard for future large-scale cohort studies to investigate time-dependent molecular signatures in the female reproductive system, paving the way for precision medicine in women's health.