PP001. Variability in MRNA expression of Jmjd6 and FLT-1 variants in normal and preeclamptic human placenta.

INTRODUCTION: Biomolecules such as soluble fms-like tyrosine kinase 1 (sFLT-1) have been implicated in the pathogenesis of preeclampsia with many studies reporting on their expression in human placenta.

OBJECTIVES: This study aimed to determine whether variation exists in the expression of different genes in human placenta based on collection site. Expression of different FLT-1 variants including the primate-specific sFLT-1e15a and a novel gene, Jumonji domain-containing protein 6 (Jmjd6) that may prove to have a role in the pathogenesis of preeclampsia, was selected as targets.

METHODS: Placental tissue was collected from one normotensive and one preeclamptic woman following caesarean section at 38 weeks. Twelve 1.5cm diameter×2mm thick samples were excised from various sites around the decidual surface. Quantitative PCR was used to determine the relative expression of the FLT-1 and Jmjd6 transcripts in the separate samples. Within a placenta, the first sample collected served as the reference and transcript expression in the remaining 11 samples was expressed relative to this sample. Between placentas, a pooled normal sample was used as a reference to determine the relative expression in preeclamptic compared to normal placental samples. One sample t -tests and coefficients of variation (CV) were used to explore the variation and Pearson's correlation coefficient was used to examine relationships.

RESULTS: Within the normal placenta, significant variation was seen in the 12 collection sites for sFLT-1 e15a (CV=45.1% p=0.008) and Jmjd6 (CV=30.4% p=0.019). The CVs for sFLT-1 i13 and mFLT-1 were 25.6% and 23.7% respectively. Within the preeclamptic placenta, significant variation was seen in the expression of all FLT-1 variants; mFLT-1 (CV=66.9% p=0.023), sFLT-1 i13 (CV=64.8% p=0.033) and sFLT-1 e15a (CV=61.1% p=0.001) across different collection sites. Significant variation was also seen between preeclamptic placenta sites and a normotensive pool; mFLT-1 (CV=66.9% p=0.012), sFLT-1 e15a (CV=61.1% p=0.005) and Jmjd6(CV=65.2% p=0.029). Using cumulative moving means, the minimum number of samples required to obtain a zero difference in means for all transcripts in a data subset was 8 for the normal placenta and 6 for the preeclamptic placenta. Overall, the expression of Jmjd6 and all FLT-1 variants was increased in the samples from the preeclamptic placenta compared to normal. Expression of mFLT-1 was highly correlated with sFLT-1 i13 and sFLT-1 e15ain preeclamptic (r=0.808 p=0.001; r=0.841p=0.001) but not normal placenta, and Jmjd6 was not correlated with any transcript in either placenta.

CONCLUSION: This study demonstrates significant variation in expression levels of several new and commonly investigated genes across sites in both normal and preeclamptic human placenta. These data show samples should be obtained from no less than 8 separate sites when pooling samples for expression analysis. Further, given that many studies examine relationships between different colocalised molecules, it may also be prudent to examine expression levels in each site separately to ensure that no relationships are missed.