Dermatophagoides pteronyssinus is a predominant allergenic house dust mite used as a raw material to produce allergen extracts for diagnosis and allergen immunotherapy (AIT). Industrial mite cultures require close monitoring and quality control to maximize yield, avoid microbial contamination, and determine optimal harvest timing. Traditional visual monitoring is operator-dependent; therefore, the study tested genetic and microbiomics approaches to provide objective monitoring of mite growth and to characterise associated microbial communities during industrial culture production.
Five sequential, independent D. pteronyssinus cultures were established at LETI Pharma facilities. Each culture weighed 350 g and combined 175 g of a LETI Pharma inoculum with 175 g of culture medium (CM). The five cultures used culture media from five distinct CM batches supplied by the same provider. Media had been frozen at −20°C for 48 hours prior to use. Cultures were maintained for 27–34 days with continuous monitoring of temperature and relative humidity. Samples for counting, DNA extraction, allergen assays and microbiome sequencing were taken at defined time points (day 1, day 15, and final day [27–34]). Cultures were inactivated by freezing at temperatures below −20°C for 48 hours before final processing.
Manual quantification: At each sampling point, 5–10 g of culture material was processed for mite counting. Two trained operators independently counted mites in each sample; counts were averaged and expressed as mites per gram of culture.
qRT-PCR quantification: DNA was extracted from culture or CM samples using a NucleoSpin Plant II kit with minor modifications; two biological replicates per sample were prepared. Extracted DNA was diluted 1:200 and amplified using a TaqMan qRT-PCR assay on a Bio-Rad CFX384 system. Primers and a probe targeted a mite DNA sequence (forward: 5′-CATCCAACCAGAGTGGTATTTCC-3′; reverse: 5′-GCTATTGCGCATACTCCACCTA-3′; probe: 5′-TATGCAATCCTTCGGGCTATCCCATCA-3′). Thermocycling was 95°C for 20 s, then 40 cycles of 95°C for 3 s and 60°C for 30 s. Technical triplicates were run and mean Cq values were reported. qRT-PCR amplifications were performed independently at two research centers.
Der p 1 and Der p 2 levels were quantified by ELISA. For each culture sample, 5 g were extracted in PBS/NaCl (1:10) at 2–8°C for 4 hours with stirring, centrifuged, and supernatants assayed using commercial ELISA kits following the manufacturer’s instructions.
Microbiome analysis included internal (600 adult mites washed), external (20 mg of culture material), and CM samples, yielding 35 samples across the five culture batches and time points. Bacterial communities were profiled by sequencing the V3–V4 region of the 16S rRNA gene, and fungal communities by sequencing the ITS region. Sequencing was performed on an Illumina MiSeq platform. Read quality control included FastQC, primer removal, chimera filtering, and processing with PANDAseq and QIIME2. Chloroplast and mitochondrial 16S reads were removed (maximum contamination reported 0.6% mitochondria, 3.5% chloroplast). ASVs were defined (equivalent to 100% OTUs); analyses produced 2,195 ASVs for 16S and 917 ASVs for ITS. Raw sequence data are deposited in the European Nucleotide Archive (accession PRJEB80324). Bioinformatic and downstream analyses used MicrobiomeAnalyst and STAMP.
Mite counts and Cq values were evaluated in GraphPad Prism 9. Pearson correlations between mean Cq and mite density were computed per culture and across cultures; significance threshold was p < 0.05. Two-way ANOVA assessed effects of time and culture type on mite counts and Cq. Microbiome beta diversity analyses used Bray–Curtis distance matrices, PCoA and hierarchical clustering at the ASV level. Group differences in community composition were tested by PERMANOVA with 999 permutations. Taxonomic-level comparisons used STAMP with G-tests (and Fisher’s exact test when appropriate) and Bonferroni correction; adjusted p < 0.05 defined significance.
Visually quantified mite growth followed expected phases: initial latency, exponential increase, and plateau. Final mite densities varied among cultures: Culture 1 reached 181,783 mites/g, Culture 2 reached 146,226 mites/g, Culture 5 showed rapid early growth then slowed, and Cultures 3 and 4 reached lower final densities (99,434 and 106,865 mites/g, respectively).
Two-way ANOVA showed a significant effect of time on mite population (F(2,8) = 11.66, p = 0.0043), accounting for 57.5% of variance; culture type did not show a significant effect on mite counts (F(4,8) = 2.308, p = 0.1460).
qRT-PCR Cq values decreased in cultures with increasing mite density. Culture 1 had the lowest Cq reduction (−1.45 cycles), while Culture 5 showed the largest Cq change (−3.07 cycles); cultures 3 and 4 had modest Cq reductions (0.90 and 0.95 cycles). A global Pearson correlation between Cq and mite density was significant (r = −0.69; p = 0.0443), indicating inverse association between Cq and mite numbers. Two-way ANOVA for Cq showed significant effects of both time (F(2,8) = 5.716, p = 0.0287) and culture type (F(4,8) = 5.485, p = 0.0200), indicating that Cq values reflect temporal changes and culture-specific factors.
Der p 1 and Der p 2 were measured across cultures and time points. At the final time point, Cultures 1 and 4 had the highest Der p 1 levels. Der p 2 levels tended to be more similar across cultures at the end of the culture period. Detailed numeric values and time courses are reported in the supplementary figures of the original work.
Bacterial community analysis recovered three phyla across samples: Firmicutes, Actinobacteria, and Proteobacteria. Within Firmicutes, Staphylococcaceae and Bacillaceae were prominent; Actinobacteria signals derived from Micrococcales and Propionibacteriaceae; Proteobacteria reads included Burkholderiales and Enterobacteriales. At the genus level, the most abundant bacterial taxa were Staphylococcus, Virgibacillus, Kocuria, Streptomyces and Cutibacterium; additional reads were labelled as “Other” or “Not assigned.”
Fungal communities were dominated by Ascomycota and Basidiomycota. Ascomycota reads were associated with Saccharomycetes and Eurotiomycetes; Basidiomycota included genera such as Malassezia and Cryptococcus. The most common fungal genera included Aspergillus, Cryptococcus, Malassezia, and Cladosporium.
Sequence processing recovered between ~72,000 and ~326,000 non-chimeric reads per target region across samples; ASV-level abundance and taxonomy tables are provided in supplementary materials.
Unsupervised hierarchical clustering and PCoA based on bacterial and fungal taxonomic profiles revealed patterns by culture batch and sample origin. Bacterial clustering separated cultures 1–2 from cultures 3–5, with culture medium (CM) samples forming a distinct branch. Microbiome composition among the five cultures showed greater variability at the beginning of culture, trending toward more homogeneous composition as mite growth progressed.
PERMANOVA and further taxon-level comparisons were used to identify statistically different taxa between groups; detailed statistical outputs and adjusted p-values are available in the supplementary files.
Across cultures and over time, Staphylococcus, Virgibacillus, and Malassezia emerged as taxa most associated with culture progression. The study reports the absence of Gram-negative bacteria associated with endotoxin production in these cultures and notes the presence of non-pathogenic bacteria. The analyses distinguish internal (mite-associated) and external (culture surface/CM) microbial profiles, with trends toward convergence of microbiome composition among cultures as mite populations increased.
The authors developed and validated a qRT-PCR application for objective quantification of D. pteronyssinus in industrial cultures; mean Cq values inversely correlated with mite density and were affected by time and culture-specific conditions. Microbiome sequencing of internal, external, and CM samples described bacterial and fungal community composition during mite culture, identifying Staphylococcus, Virgibacillus, and Malassezia as key taxa linked to culture progression. The combined genetic and microbiomics approach can be used to support standardized harvesting decisions and quality control in the manufacture of mite-derived allergen extracts. The raw sequencing data and supplementary materials contain full methodological and numerical details.