Slides from journal club on de Koning & al (2020) "An eQTL in the cystathionine beta synthase gene is linked to osteoporosis in laying hens"

2. de Koning & al (2020) An eQTL
in the cystathionine beta
synthase gene is linked to
osteoporosis in laying hens
Virtual Animal Breeding Journal Club
Martin Johnsson

3. TODO
1. Housekeeping
2. Brief summary of paper
3. Discussion!

4. Housekeeping
• Google group animal-breeding-journal-club@googlegroups.com.
• Who wants to go next?
• Slides will be shared—somehow
• We’ll try to keep to ~60 minutes.

5. Code of conduct
• Presenters
– Engage objectively with the material
– Treat authors with respect
– Critiques aren’t attacks
– Offer solutions as well criticism
– Our goal is to learn
• Participants
– Treat presenters with respect
– Ask questions in good faith
– If a mistake is made, help fix it
– Assume good faith

6. Martin
• Postdoc at Swedish University of Agricultural Sciences; visiting at Roslin up
until recently.
• Interests:
– Population genomics
– Genetic mapping of molecular traits
– Deleterious variation
– Domestication

7. The paper
• Causative gene identification
• Bone biology
• The saga of a QTL

8. Causative gene identification
• QTL—eQTL paradigm
• cis- vs trans-acting eQTL
• Protein-coding vs regulatory

9. Bone biology
• Conserved bone pathways
• Enzyme activity
• Potential for biology-inspired interventions

10. Chicken bone in 1 min
• Hard cortical bone
Soft trabecular bone
Soft medullary bone (female birds only)
• Start laying—switch from making cortical
to medullary bone
• Keep laying—osteoporosis
Whitehead (2004)

11. Studies
1. Fine-mapping
2. eQTL and allelic imbalance
3. Biochemistry

12. The bone strength locus on chicken chromosome 1
• First detected in Dunn & al (2007)
• Refined from ~ 300 cM to 50 cM
confidence interval (tibiotarsal
breaking strength) with 27 new
markers
• Follow-up with targeted
genotyping in later generations

13. Gene expression
• Local expression QTL study (N = 8)
• 5 genes differentially expressed; four
local; CBS most significant and
appealing
• Allelic imbalance (RT-PCR, restriction,
gel quantification)

14. Cystathione beta synthase
• Enzyme that makes cystathione from homocysteine
and serine
• Mutations cause homocystinuria in humans
• High homocysteine may cause low bone strength

15. Homocysteine and bone composition

16. Enzyme function
• 489 Lys>Glut
• Enzyme kinetics assay in embryonic liver not significantly different

17. Plasma homocysteine levels different
• Higher homocysteine in low bone strength genotype (✅)
• Also higher expression (❓)
• One might predict lower expression -> higher homocysteine -> worse bones,
especially given no difference in enzyme function

18. Cis or trans?
“Expression data from heterozygotes for the low and high bone strength alleles
showed allelic imbalance in the expression of CBS. This clearly indicated that
the difference in expression was due to a cis-acting effect and not a trans-
acting effect. In other words, it was unlikely that there was any involvement of
a transcription factor transcribed from the region, acting on expression, as this
would affect both alleles.”

19. Protein-coding or regulatory?
“However, applying Occam’s razor, the simplest possibility for the observed differences in bone strength
between the genotypes might be the difference in the predicted amino-acid sequence of the CBS gene at
position 498 from a lysine (K) to a glutamine (Q). Differences in expression could be the result of
differences in feedback mechanisms if the enzyme activity was more or less active than the ‘wild type’
gene and protein. However, the results from the allelic imbalance study does not support the hypothesis
that the observed effects were due to differences in feedback because of a faulty copy of the CBS
enzyme. If a faulty copy of CBS was present, we would expect both copies to be equally affected by any
feedback in a heterozygous individual and this was not the case.”

20. Protein-coding or regulatory?
“Finally and more directly, we did not observe a difference in the activity of the two allozymes when they
were tested. The chicken and the turkey genomes predict a glutamine (Q) at position 498 in the CBS
protein, and glutamine has a non-charged polar side chain. Lysine (K), which has a charged side chain
seems universal in predictions from other bird genomes. Lysine is also present in the chicken at this
position when the glutamine codon is not present, as we have seen in this study. Reptiles feature
glutamine (Q) or aspartic acid (D) at this position in the CBS protein; these amino-acids possess a
charged side chain with a negative charge. In mammals, it seems that threonine (T), another non-
charged polar side chain amino-acid, is almost universal. Therefore, there is no indication that the charge
at the position is conserved or that the difference in charge at this position might lead to a large effect on
the enzyme’s activity. The results of the assays of enzyme activity in our study confirmed this with neither
Vmax nor Km differing between the forms.“

21. Summary
• Cis-acting eQTL on CBS seems like best candidate
• Direction of the difference in gene expression remains to be explained

22. Points for discussion
• Where to next? (Biology, application)
– “Validated markers that predict bone strength have been defined for selective breeding”
– “a gene was identified that may suggest alternative ways to improve bone health”
– “loci with a larger effect than the CBS loci have been observed, which suggests we can
make progress in finding the underlying mechanism for these loci if large enough samples
can be assembled”
• Systematic and interpretive QTG identification?

23. Thank you! Next?
MARTIN
Martin.Johnsson@slu.se
martinjohnsson.se
onunicornsandgenes.blog
THE JOURNAL CLUB
animal-breeding-journal-club@googlegroups.com

24. References
• Whitehead (2004), review of layer bone biology
https://www.ncbi.nlm.nih.gov/pubmed/14979569
• Zhou & al (2019) Paper about potential causative variant for the CBS eQTL
https://www.sciencedirect.com/science/article/abs/pii/S0378111919305311