Training immunophenotyping deep learning models with the same-section ground truth cell label derivation method improves virtual staining accuracy.

Introduction: Deep learning (DL) models predicting biomarker expression in images of hematoxylin and eosin (H&E)-stained tissues can improve access to multi-marker immunophenotyping, crucial for therapeutic monitoring, biomarker discovery, and personalized treatment development. Conventionally, these models are trained on ground truth cell labels derived from IHC-stained tissue sections adjacent to H&E-stained ones, which might be less accurate than labels from the same section. Although many such DL models have been developed, the impact of ground truth cell label derivation methods on their performance has not been studied. Methodology: In this study, we assess the impact of cell label derivation on H&E model performance, with CD3+ T-cells in lung cancer tissues as a proof-of-concept. We compare two Pix2Pix generative adversarial network (P2P-GAN)-based virtual staining models: one trained with cell labels obtained from the same tissue section as the H&E-stained section (the 'same-section' model) and one trained on cell labels from an adjacent tissue section (the 'serial-section' model). Results: We show that the same-section model exhibited significantly improved prediction performance compared to the 'serial-section' model. Furthermore, the same-section model outperformed the serial-section model in stratifying lung cancer patients within a public lung cancer cohort based on survival outcomes, demonstrating its potential clinical utility. Discussion: Collectively, our findings suggest that employing ground truth cell labels obtained through the same-section approach boosts immunophenotyping DL solutions.

The epigenetic reader PHF21B modulates murine social memory and synaptic plasticity-related genes.

Synaptic dysfunction is a manifestation of several neurobehavioral and neurological disorders. A major therapeutic challenge lies in uncovering the upstream regulatory factors controlling synaptic processes. Plant homeodomain (PHD) finger proteins are epigenetic readers whose dysfunctions are implicated in neurological disorders. However, the molecular mechanisms linking PHD protein deficits to disease remain unclear. Here, we generated a PHD finger protein 21B-depleted (Phf21b-depleted) mutant CRISPR mouse model (hereafter called Phf21bΔ4/Δ4) to examine Phf21b's roles in the brain. Phf21bΔ4/Δ4 animals exhibited impaired social memory. In addition, reduced expression of synaptic proteins and impaired long-term potentiation were observed in the Phf21bΔ4/Δ4 hippocampi. Transcriptome profiling revealed differential expression of genes involved in synaptic plasticity processes. Furthermore, we characterized a potentially novel interaction of PHF21B with histone H3 trimethylated lysine 36 (H3K36me3), a histone modification associated with transcriptional activation, and the transcriptional factor CREB. These results establish PHF21B as an important upstream regulator of synaptic plasticity-related genes and a candidate therapeutic target for neurobehavioral dysfunction in mice, with potential applications in human neurological and psychiatric disorders.

Developmental IL-6 Exposure Favors Production of PDGF-Responsive Multipotential Progenitors at the Expense of Neural Stem Cells and Other Progenitors.

Interleukin-6 (IL-6) is increased in maternal serum and amniotic fluid of children subsequently diagnosed with autism spectrum disorders. However, it is not clear how increased IL-6 alters brain development. Here, we show that IL-6 increases the prevalence of a specific platelet-derived growth factor (PDGF)-responsive multipotent progenitor, with opposite effects on neural stem cells and on subsets of bipotential glial progenitors. Acutely, increasing circulating IL-6 levels 2-fold above baseline in neonatal mice specifically stimulated the proliferation of a PDGF-responsive multipotential progenitor accompanied by increased phosphorylated STAT3, increased Fbxo15 expression, and decreased Dnmt1 and Tlx expression. Fate mapping studies using a Nestin-CreERT2 driver revealed decreased astrogliogenesis in the frontal cortex. IL-6-treated mice were hyposmic; however, olfactory bulb neuronogenesis was unaffected. Altogether, these studies provide important insights into how inflammation alters neural stem cells and progenitors and provide new insights into the molecular and cellular underpinnings of neurodevelopmental disorders associated with maternal infections.