Hamilton, Lorna Sheradyn

Genome-wide association study of Prolactin in cerebrospinal fluid and plasma

Faculty Mentor: Dr. John Kauwe, Department of Biology: Bioinformatics

Prolactin is the major hormone involved in milk lactation for pregnant women and has
been discovered to play a role in a variety of biological functions. Several other
functions include the immune system, reproductive system, maternal behavior, insulin
production, and stimulating neurogenesis.1 2 Using prolactin levels from both
cerebrospinal fluid and blood plasma from nearly 500 individuals, we will be able to
conduct a genome-wide association study to identify single-nucleotide polymorphisms
(SNPs), which will lead to a better understanding of what genes are playing a significant
role in prolactin’s production and regulation.

For my study we used data gathered from two different sources, Alzheimer’s Disease
Neuroimaging Initiative (ADNI) and Knight-Alzheimer’s Disease Research Center at
Washington University School of Medicine (WU). The data was from approximately 500
participants, with ages ranging from 50 to 90, with an average age of 75.

We used the bioinformatics tool PLINK to clean the data and conduct a linear
regression. The linear regression finds correlation between levels of prolactin in CSF
and plasma and produces a list of SNPs. The list of SNPs produced by PLINK gives me
the SNPs that most significantly impact levels of prolactin in individuals. Then we
conducted meta-analyses to find which SNPs are similarly significant in both ADNI and
WU. We next looked for SNPs significant in both the plasma and CSF. After we made
this final list of the most significant SNPs, I used annotation tools such as wANNOVAR
and Regulome DB to collect all relevant information about each SNP.

The results for this project were really interesting, in plasma we discovered 37 SNPs
and in cerebrospinal fluid we discovered 666 SNPs. We also discovered 12 significant
SNPs in both the cerebrospinal fluid and blood plasma. These 12 SNPs were the most
interesting because they influenced prolactin levels in two separate body fluids. 6 of
these SNPs were located on the same chromosome as the prolactin gene (PRL), but
were found about 6 million base pairs away. Another annotation tool called
PathwayCommons, showed us that two of the genes our SNPs are in, SULF1 and
TRIB2, interact with PRL through using similar transcription factors.

The SNPs located on the same chromosome as PRL (chromosome 6) are in or around
the genes ZSCAN8 and ZSCAN9. Most interestingly, 3 of the 6 SNPs are located in
psuedogenes and still have an effect on prolactin levels. The other 6 chromosomes are
spread out across genome; they are located on chromosomes 2, 7, 8, and 17. An
interesting discovery was that several transcription factors regulate SULF1, TRIB2, and
PRL. These transcription factors include PBX1, XBP1, TCF3, LEF1, VSX1, PITX2, and
LHX3.

Using tools to effectively analyze our data, we were able to generate lists of SNPs that
were most significant in affecting levels of prolactin in plasma, cerebrospinal fluid, and in
both. After annotating our list of SNPs, we were able to get a much better picture of
how our body regulates prolactin through the use of other genes. The list of SNPs we
gathered and other important information can then lead to future research and a more
conclusive idea of what is working with prolactin in our bodies. This foundational
research will be important for everyone because of the influence the hormone plays in
several biological functions, and especially for women because of the relevance it has
on pregnancy and milk lactation.

¹ Freeman, Marc E., et al. “Prolactin: Structure, Function, and Regulation of Secretion.”
Physiological Reviews 80.4 (2000): 1523-631. Web
² Larsen, C. M., and D. R. Grattan. “Prolactin, Neurogenesis, and Maternal Behaviors.”
Brain, behavior, and immunity 26.2 (2012): 201-9.