Cytomics Limited, Hong Kong Science and Technology Park, Hong Kong, China.
Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong, China.
Genes (Basel). 2021 Jun 25;12(7):971. doi: 10.3390/genes12070971.
Peripheral blood transcriptome is a highly promising area for biomarker development. However, transcript abundances (TA) in these cell mixture samples are confounded by proportions of the component leukocyte subpopulations. This poses a challenge to clinical applications, as the cell of origin of any change in TA is not known without prior cell separation procedure. We developed a framework to develop a cell-type informative TA biomarkers which enable determination of TA of a single cell-type (B lymphocytes) directly in cell mixture samples of peripheral blood (e.g., peripheral blood mononuclear cells, PBMC) without the need for subpopulation separation. It is applicable to a panel of genes called B cell informative genes. Then a ratio of two B cell informative genes (a target gene and a stably expressed reference gene) obtained in PBMC was used as a new biomarker to represent the target gene expression in purified B lymphocytes. This approach, which eliminates the tedious procedure of cell separation and directly determines TA of a leukocyte subpopulation in peripheral blood samples, is called the Direct LS-TA method. This method is applied to gene expression datasets collected in influenza vaccination trials as early predictive biomarkers of seroconversion. By using TNFRSF17 or TXNDC5 as the target genes and TNFRSF13C or FCRLA as the reference genes, the Direct LS-TA B cell biomarkers were determined directly in the PBMC transcriptome data and were highly correlated with TA of the corresponding target genes in purified B lymphocytes. Vaccination responders had almost a 2-fold higher Direct LS-TA biomarker level of TNFRSF17 (log 2 SMD = 0.84, 95% CI = 0.47-1.21) on day 7 after vaccination. The sensitivity of these Direct LS-TA biomarkers in the prediction of seroconversion was greater than 0.7 and area-under curves (AUC) were over 0.8 in many datasets. In this paper, we report a straightforward approach to directly estimate B lymphocyte gene expression in PBMC, which could be used in a routine clinical setting. Moreover, the method enables the practice of precision medicine in the prediction of vaccination response. More importantly, seroconversion could now be predicted as early as day 7. As the acquired immunology pathway is common to vaccination against influenza and COVID-19, these biomarkers could also be useful to predict seroconversion for the new COVID-19 vaccines.
外周血转录组是开发生物标志物的一个极有前途的领域。然而，这些细胞混合物样本中的转录丰度（TA）受到组成白细胞亚群比例的影响。这给临床应用带来了挑战，因为如果没有预先进行细胞分离程序，就无法知道 TA 变化的细胞起源。我们开发了一种框架来开发细胞类型信息性 TA 生物标志物，该生物标志物能够直接在外周血细胞混合物样本（例如外周血单核细胞，PBMC）中确定单个细胞类型（B 淋巴细胞）的 TA，而无需进行亚群分离。它适用于一组称为 B 细胞信息基因的基因。然后，在 PBMC 中获得的两个 B 细胞信息基因（靶基因和稳定表达的参考基因）的比值用作新的生物标志物，以代表纯化 B 淋巴细胞中的靶基因表达。这种方法消除了细胞分离的繁琐过程，直接在外周血样本中确定白细胞亚群的 TA，称为直接 LS-TA 方法。该方法应用于在流感疫苗接种试验中收集的基因表达数据集，作为血清转化率的早期预测生物标志物。使用 TNFRSF17 或 TXNDC5 作为靶基因，TNFRSF13C 或 FCRLA 作为参考基因，直接在 PBMC 转录组数据中确定直接 LS-TA B 细胞生物标志物，并且与纯化 B 淋巴细胞中相应靶基因的 TA 高度相关。接种疫苗的应答者在接种后 7 天的 TNFRSF17 直接 LS-TA 生物标志物水平几乎高出 2 倍（log 2 SMD = 0.84，95%CI = 0.47-1.21）。在许多数据集，这些直接 LS-TA 生物标志物在预测血清转化率方面的灵敏度大于 0.7，曲线下面积（AUC）大于 0.8。在本文中，我们报告了一种直接估计 PBMC 中 B 淋巴细胞基因表达的简单方法，该方法可在常规临床环境中使用。此外，该方法使预测疫苗接种反应的精准医学实践成为可能。更重要的是，现在可以在第 7 天预测血清转化率。由于获得性免疫途径对流感和 COVID-19 的疫苗接种都很常见，因此这些生物标志物也可用于预测新的 COVID-19 疫苗的血清转化率。
