General enquiries on this form should be made to:Defra, Procurements and Contracts Division (Science R&D Team)Telephone No. 0207 238 5734E-mail: research.competitions@defra.gsi.gov.uk

SID 5 Research Project Final Report

SID 5 (Rev. 05/09) Page 1 of 18

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code AC0220

2. Project title

Factors associated with oestrous expression in dairy cows

3. Contractororganisation(s)

The University of NottinghamSchool of BiosciencesSutton Bonington CampusLoughboroughLE12 5RD     

54. Total Defra project costs £ 75,355(agreed fixed price)

5. Project: start date................ 01 October 2009

end date................. 31 May 2010

SID 5 (Rev. 05/09) Page 2 of 18

6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They

should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.Poor reproductive performance is the major cause of wastage in the dairy industry and increases its environmental impact. Pregnancy depends not only on conception rate, but also on heat (oestrus) detection. If oestrus cannot be detected accurately, then a high proportion of cows will either not be inseminated at all, or will be mated at a time when they are unable to conceive. Heat detection rate has been declining for many years in the UK, and is currently about 50%. Coupled with an estimated pregnancy rate of 40%, this means that only 20% of ovulations actually result in the birth of a calf: 80% of opportunities are missed. There is a clear practical need to improve oestrus expression and detection.

In previous projects we identified a number of genetic markers (SNPs) associated with fertility traits and a preliminary study suggested that some variation in strength of oestrus expression may be explained by these SNPs. The overall aim of the current project was to increase the evidence base for associations between genetic markers and strength of oestrous expression.

Objective 1 was to extend our database of genotyped cows to find further animals with the oestrogen receptor alpha SNP. To meet this objective, DNA was obtained from blood samples of 202 high-yielding dairy cows and analysed for 41 SNPs. Variations in SNP values were investigated for associations with strength of oestrous expression, as indicated by pedometer activity. Activity at oestrous was associated with 11 SNPs in 5 genes. These genes were Signal Transducer & Activator of Transcription 5A, and receptors for Activin, Gonadotrophin releasing hormone, Prolactin and Oestrogen. Each of these has a role in regulation of the oestrous cycle, but we were particularly interested in mutation of oestrogen receptor alpha. In the current study, 8.5% of cows had a mutation in the promoter region of oestrogen receptor alpha, which increased activity at oestrus. A change in the oestrogen receptor is consistent with changes in activity at Objective 2 was to investigate the possible significance of the oestrogen receptor alpha SNP 173 by determining whether the relevant DNA sequence is transcribed and appears in messenger RNA.To meet this objective, samples of bovine reproductive tissues were obtained from an abattoir and subjected to mRNA analysis. Transcription was confirmed in foetal ovaries and testes, and in adult ovarian and uterine tissues, but not in adult kidney tissue used as a negative control.

During the course of the research we discovered that the DNA sequence of interest contains a secondary structure known as a G-quadruplex, which occurs when DNA folds into a cuboid shape. This is of significance, because G-quadruplexes are absent from all other mammalian oestrogen receptor genes investigated at this position. We had previously shown that a G-quadruplex in the human oestrogen receptor gene, but at a different position, has significant effects on translation of the oestrogen receptor

SID 5 (Rev. 05/09) Page 3 of 18

protein. Using synthetic mRNA with and without the G-quadruplex, we found that G-quadruplex removal caused an approximate doubling in rate of translation. Mutation of the G-quadruplex, so that the secondary structure could not be formed, led to a 10-fold increase. When the effects of the SNP mutations were investigated there were minor differences between the sequences, but these were evident only in the absence of the G-quadruplex forming sequence.

Objective 3 was to carry out preliminary experiments to develop and test methods for pheromone collection from dairy cows.To meet this objective, vaginal mucus samples were collected from 16 cows in oestrus, and from the same cows two weeks later when not in oestrus. Unfortunately, the swaps used to obtain non-oestrus samples contained a biocidal chemical that interfered with detection of other volatile compounds. For cows in oestrus, a volatile compound was detected in mucus and identified as hexadecanal. This is known to act as a pheromone in insects, but such a role not been reported in mammals. None of the compounds previously suggested as pheromones in cattle (acetic acid, propionic acid and iododecane) could be detected in mucus samples from this study. Further work is required to investigate volatiles in fresh mucus samples, because it is possible that some might have undergone oxidation during storage.

Conclusions and future workThis study has confirmed that differences in oestrous expression among cows have genetic components. We have provided further evidence that genetic markers associated with fertility traits are associated also with activity at oestrus, thus offering the possibility of marker-assisted selection. Examination of pedigree data showed that oestrous activity is heritable, although heritability is low. Nevertheless, identification of a significant effect of oestrous activity on likelihood of pregnancy following insemination confirms the value of selecting for this trait.

Study of oestrogen receptor alpha confirmed that the genetic mutation at position 173 is transcribed into mRNA. Discovery of the G-quadruplex in this DNA sequence, and its significant effect on translation of the oestrogen receptor protein, suggests that presence of this structure could have important implications for gene expression. Further work is required to see if this structure is ubiquitous, or if cows without it might have improved oestrous expression and reproductive performance.

Preliminary work on pheromones identified one compound that might have pheromonal activity, but this requires confirmation in future studies. It is likely that other pheromone compounds are present in vaginal mucus and other secretions, but this preliminary study has indicated that future work needs to start from fresh samples. If a range of pheromones can be identified, their association with strength of oestrous expression can be tested.

Overall, this short project has provided a considerable amount of novel information about factors associated with oestrous expression in dairy cows. As well as providing tools that could be used immediately in selection programmes to improve oestrous detection, this information increases our understanding of the underlying physiology and suggests new avenues to explore in future research.

Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with

details of the outputs of the research project for internal purposes; to meet the terms of the contract; and

SID 5 (Rev. 05/09) Page 4 of 18

to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability; the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).

IntroductionGenetic selection for production efficiency over the past 30 years has led to dramatic increases in milk yield of dairy cows. This allowed national milk production to remain relatively constant with a greatly reduced number of cows. The number of cows in the national herd is the main driver for the environmental impact of dairy farming, in terms of methane, ammonia, nitrogen and phosphorus (Garnsworthy, 2004). It is predicted that the genetic trend for higher milk yield will continue for the foreseeable future, so cow numbers and environmental impact would be expected to continue falling (Garnsworthy and Thomas, 2005; Maas et al., 2008). However, there is a negative correlation between milk yield and fertility, so increased milk yield has been accompanied by a decline in reproductive efficiency, leading to premature culling and reduced lifetime performance. Reduced performance offsets efficiency gains from genetic merit for milk yield and could halt the reduction in cow numbers. In addition, outputs of methane, nitrogen and phosphorus during the rearing phase are spread over fewer units of lifetime milk production. Thus, it is extremely important that the fertility issue is addressed urgently.

Poor reproductive performance is the major cause of wastage in the dairy industry and increases its environmental impact: • 33% of dairy cows are culled each year, after an average of only 3 lactations• over 50% of cows culled are perfectly healthy; they simply fail to conceive• there is a national shortage of heifer replacements and imports from abroad pose risks for biosecurity• the UK is no longer self-sufficient in milk and fewer crossbred animals enter the beef industry• poor fertility adds 20% to methane and ammonia emissions from dairy herds• projected reductions in methane emissions by 2050 will be halved if fertility is not addressed.

Pregnancy failure or delay reduces the lifetime milk yield of cows and thereby increases the environmental footprint of dairy systems. Pregnancy depends not only on conception rate, but also on heat (oestrus) detection. If oestrus cannot be detected accurately, then a high proportion of cows will either not be inseminated at all, or will be mated at a time when they are unable to conceive. Heat detection rate has been declining for many years in the UK, as in other countries, and is currently about 50%. Coupled with an estimated pregnancy rate of 40%, this means that only 20% of ovulations actually result in the birth of a calf: 80% of opportunities are missed. Given the importance of accurate and successful insemination for both continued milk production and breeding of replacements, and their influence on environmental efficiency, there is a clear and overwhelming practical need to improve oestrus expression and detection.

Genetic improvement relies on identifying superior animals to select for breeding the next generation. For any trait, rate of genetic progress can be accelerated by identification of genetic markers associated with better performance in that trait. This is particularly useful for female traits, such as milk yield, ovulation rate and pregnancy, because genetic markers are carried by both sexes and can be identified at any time after birth. A genetic marker involves a change in just one nucleotide (A, T, C or G) in an individual’s DNA sequence, called a single nucleotide polymorphism (SNP). This change may affect the amino acid sequence of the protein produced by that DNA sequence, which can affect the performance or disease resistance of an animal. In previous Defra-funded projects, we identified a number of SNPs associated with fertility traits (LK0639) and a preliminary study (AC0205) suggested that some of the variation in strength of oestrus expression may be explained by these SNPs. The overall aim of the current project was to increase the evidence base for associations between genetic markers and strength of oestrous expression and to investigate changes in pheromones associated with oestrus.

ObjectivesThe objectives set out in the contract were:

SID 5 (Rev. 05/09) Page 5 of 18

Objective 1 – to extend our database of genotyped cows to find further animals with the oestrogen receptor alpha SNP. Between 50 and 150 cows will be sampled for whole blood for genotyping at 17 previously identified loci.Objective 2 – to investigate the possible significance of the oestrogen receptor alpha SNP 173 by determining whether the relevant DNA sequence is transcribed, and appears in messenger RNA.Objective 3 – to carry out preliminary experiments to develop and test methods for pheromone collection from dairy cows.

Objective 1 has been exceeded. Over 200 cows were genotyped and 41 SNPs were investigated.Objective 2 has been met in full. Transcription of the DNA sequence for SNP 173 was confirmed in mRNA from a number of tissues. Additional work has discovered the existence of g-quadruplex formation in bovine DNA.Objective 3 has been met, although we had hoped to make greater progress. A preliminary experiment was conducted involving sampling of 16 cows during oestrus and dioestrus, but only one putative pheromone could be identified in an initial analysis of samples. By agreement with the Defra Project Officer, further analytical work was postponed due to staffing complications. When the analysis was revisited, however, further volatiles could not be detected, possibly due to oxidation during storage. We plan to perform further analysis on fresh samples at a later date and results will be presented in a supplementary report to Defra.

REPORT ON OBJECTIVE 1Objective 1 was to extend our database of genotyped cows to find further animals with the oestrogen receptor alpha SNP.

MethodsBlood samples were collected from 202 high-yielding (10,000 litres/cow/year) dairy cows at the University of Nottingham Dairy Centre. Samples were sent for DNA extraction and genotyping by primer extension, which was performed commercially by KBiosciences Ltd. Purified genomic DNA was subsequently analysed for 41 SNPs. Candidate genes were selected on the basis of a) their involvement in hypothalamic/ovarian/uterine function (GnRH receptor, LH receptor, FSH receptor, oestrogen receptors α and β, activin), b) known roles in energy metabolism and milk synthesis (leptin, GH, prolactin, fatty acid synthetase, PPARγ, STAT1, STAT5) and c) roles in central nervous pathways directly controlling oestrous behaviour (neuropeptide Y and its receptor). In some cases the same gene product is expected to be involved in more than one of these processes through functions in different organs (e.g. the oestrogen receptors in the ovary, uterus and hypothalamus).

DNA samples collected under this objective will be stored for use in a future collaborative programme with our partners in the Ruminant Genetic Improvement Network. The aim will be to perform genome-wide scanning of a large population of animals to identify genetic markers for methane emissions, as well as important production and reproduction traits.

Variations in SNP values were investigated for associations with strength of oestrous expression, as indicated by activity monitors. Cows were milked between 1 and 3 (mean 2.8) times daily through a Lely Astronaut automatic (robotic) milking system. Each cow was fitted with a neck collar that provided identification and also activity data every 2 hours, which were downloaded at milking. Activity values are reported as an index that quantifies all animal movements (walking, running, laying, standing up, head movements etc). The primary measure for strength of oestrus expression was maximum activity on the day of oestrus. Maximum activity on the day of oestrus as a percentage increase over rolling average activity for the 4 days preceding oestrus was examined also, but percentage increase in activity was more variable within cows (Coefficient of Variation 26% versus 16%) and showed few significant relationships with SNP and other data. Therefore, only results for maximum activity on the day of oestrus are reported.

Activity data at oestrus were not available for 6 cows. These cows were pregnant when purchased shortly before blood sampling, so they did not express oestrus during the study period. Consequently, the final analysis was performed using 196 cows. Data were recorded from March 2008 (when the robotic milking system was installed) until April 2010, which included a total of 1016 oestruses with an average of 5.2 oestruses from 1.4 lactations per cow. A total of 740 inseminations were recorded, resulting in 242 pregnancies.

Activity data were analysed using the generalised linear mixed model (GLMM) procedure within the Genstat statistical software package. A Poisson error distribution and a logarithm link function were chosen to account for the non-normal distribution of activity data. Cow identity was declared as a random effect in all models to account for repeated measures at successive oestruses and lactations within individuals. The fixed effects examined (in addition to SNPs) were daily milk yield around oestrus (average from 5 days before to 5 days after oestrus), average daily milk yield from calving to day of oestrus, season of calving and oestrus (1 = Jan-Mar, 2 = Apr-Jun, 3 = Jul-Sep, 4 = Oct-Dec), lactation number, and parity (1 = lactation 1, 2 = lactation 2, 3 = lactation >=3). When fitted individually in the model, each of these factors except season of calving was significantly associated with oestrus activity. When fitted together in the model, however, partial confounding meant that all factors could be dropped through non-significance except parity and season of oestrus. In other words, parity accounted for most

SID 5 (Rev. 05/09) Page 6 of 18

of the effects of lactation number and milk yield on activity; season of oestrus accounted for some of the effect of milk yield on activity. The final model used to test the effects of each SNP on activity included SNP, Parity and Season of oestrus as fixed effects, and Cow as a random effect.To test the effects of SNPs on outcome of inseminations (pregnant or not), the same model was used, but the error distribution was changed to binomial and the link function to logit.

To study genetic effects on activity at oestrus the pedigree of cows was traced back 4 generations (using the Holstein UK database) in order to construct a relationship matrix for genetic evaluation; the pedigree file included 893 animals. Variance components and fixed effect solutions were calculated using ASREML software (Gilmour et al., 2009). Heritability, repeatability, and genetic and phenotypic correlations among activity traits (Baseline activity (BASE), activity at oestrus (STEPS), and increase in activity (STEPS%)) were estimated in a series of univariate and bivariate analyses. A single-trait mixed model was used to evaluate effects of covariance components for individual animals, solving for both permanent environmental effects and genetic effects on STEPS and STEPS%. The model included fixed effects of oestrus number, daily milk yield around oestrus, average daily milk yield from calving to day of oestrus, parity and season of oestrus as follows:

where y is the trait of interest which was modelled as dependent on: fixed effects b, random animal effect a, permanent environmental effects pe, and random residual effect e. X, Z, and W are incidence matrices relating oestrus records to fixed, animal and permanent environmental effects, respectively. Bivariate analyses were undertaken to estimate heritability and genetic and phenotypic correlations between the three traits BASE, STEPS and STEPS%, comparing two traits in each analysis. The model used was:

Where yi is the vector of oestrous records of the ith trait, bi is the vector of the fixed effects for the ith trait, ai is the vector of random animal effects for the ith trait, and ei is the vector of the random residual effects. The same model was used to compare all traits by fitting the three traits simultaneously.

Results and discussionEffects of parity, season of oestrus and milk yield on maximum activity at oestrus are shown in Table 1.1. Increasing parity was negatively associated with strength of oestrus and activity of cows beyond their second lactation averaged 10 units lower during oestrus than cows in their first lactation. There are two possible explanations for this observation. Firstly, the effect of parity could be confounded partially by the negative effect of milk yield; secondly, older cows might be more dominant and therefore offer some resistance to being chased around the pen during oestrus.

Table 1.1. Effects of parity, season and milk yield on activity at oestrusParameter Value Effects¹ s.e. P-value Activity²Parity 1 4.26 0.022 <0.001 70.9

2 -0.055 67.13 -0.162 60.5

Season of oestrus³ 1 4.16 0.022 0.004 63.82 0.061 67.83 0.077 68.94 0.031 65.8

Milk yield at oestrus (kg/d)

constant 4.20 0.016 0.002effect -0.0036 0.00115 -0.24 Activity units per kg

Milk yield from calving to oestrus (kg/d)

constant 4.20 0.016 0.003effect -0.0035 0.00117 -0.23 Activity units per kg

1. Effects are on the natural logarithm scale and are relative to Parity 1, Season 1 or Milk Yield 0.2. Activity values are back-transformed means from fitted models.3. Seasons: 1 = Jan-Mar; 2 = Apr-Jun; 3 = Jul-Sep; 4 = Oct-Dec.

Oestrus expression was stronger during summer (April to September) than during winter (October to March) by an average of three activity units. This is counter-intuitive because conception rate is usually lower in summer. In commercial herds, oestrous expression is often higher following turnout to grass in the spring. In the current study, however, cows were housed throughout the year, so we can provide no obvious explanation for this result.

SID 5 (Rev. 05/09) Page 7 of 18

Average milk yield, both at oestrus and from calving to oestrus, had a negative effect on strength of oestrus expression. This observation is consistent with the long-term trend for a decline in oestrus expression as milk yield has increased. There was no association between oestrus activity and stage of lactation or oestrus number.

Effects of SNPsAssociations between oestrus activity and SNPs where P<0.1 are presented in Table 1.2.

Table 1.2. Effects of SNPs on activity at oestrus

Gene/SNP ID Genotype¹ # Cows P-value Effect² s.e. Prediction³Activin Receptor Type 2ACT_IIB_45 0 T:T 60 0.064 Constant 4.225 0.026 68.4

1 C:T 97 Effect -0.040 0.0241 65.72 C:C 30 63.1

ACT_IIB_46 0 G:G 65 0.078 Constant 4.225 0.026 68.41 T:G 93 Effect -0.035 0.0235 66.02 T:T 30 63.8

ACT_IIB_86_END 0 A:A 65 0.067 Constant 4.220 0.026 68.01 G:A 92 Effect -0.036 0.0231 65.62 G:G 30 63.3

ACT_IIB_95 0 A:A 60 0.043 Constant 4.223 0.03 68.21 G:A 96 Effect -0.0425 0.024 65.42 G:G 31 62.7

Signal Transducer & Activator of Transcription 5A STAT5a_12195 0 C:C 61 0.019 Constant 4.214 0.026 67.6

1 G:C 99 Effect 0.0358 0.0252 70.12 G:G 27 72.6

Gonadotrophin releasing hormone receptorbGNRHR_ex1_SNP_340 0 C:C 66 0.052 Constant 4.228 0.026 68.6

1 C:T 89 Effect 0.0309 0.0232 70.72 T:T 31 72.9

bGNRHR_ex1_SNP_286 0 G:G 154 0.043 Constant 4.219 0.026 68.01 G:A 32 Effect 0.0594 0.041 72.12 A:A 1 76.5

bGNRHR_ex1_SNP_490 0 C:C 66 0.065 Constant 4.221 0.026 68.11 C:T 88 Effect 0.0277 0.0227 70.02 T:T 32 72.0

bGHRHR_prom_SNP_1189 0 T:T 66 0.070 Constant 4.225 0.026 68.41 C:T 89 Effect 0.0276 0.0231 70.32 C:C 32 72.3

Prolactin receptorPRLR_Ser18Asn 0 A:A 142 0.052 Constant 4.226 0.027 68.4

1 A:G 0 Effect -0.0320 0.02032 G:G 39 64.2

Oestrogen receptor-αbERA_prom_SNP173 0 G:G 172 0.046 Constant 4.224 0.026 68.3

1 G:A 16 Effect 0.0743 0.0579 73.62 A:A 0

1. Genotype values: 0 = wild type; 1 = heterozygote; 3 = double mutation. 2. Effects are on the natural logarithm scale and are relative to wild type.3. Predicted activity values are back-transformed means from fitted models.It should be noted that some SNPs in the activin and GnRH genes are inherited together (where the number of cows in each category is approximately equal), so it is not possible to separate effects of individual SNPs.

SID 5 (Rev. 05/09) Page 8 of 18

Four SNPs within the activin receptor type 2 gene were associated with reductions in activity at oestrus (one P<0.05; three P<0.1). Activin plays a complementary role to inhibin in regulation of follicle growth and steroidogenesis. It is not surprising, therefore, that polymorphisms in the activin receptor type 2 gene could influence oestradiol synthesis and hence activity at oestrus.

Mutation in the Signal Transducer & Activator of Transcription 5A gene significantly (P=0.019) increased activity at oestrus. STAT proteins are involved in cytokine signalling pathways and regulate gene transcription in response to cytokines and growth factors (signalling molecules). In cattle, SNP mutation of STAT5A 12195 from CC to GG was associated with reduced embryo survival rate (Khatib et al. 2008).

Four SNPs in the gonadotrophin releasing hormone receptor gene were associated with increased activity at oestrus (one P<0.05; three P<0.1). GnRH is the main driver of the oestrous cycle, so changes in its receptor would be predicted to alter FSH and LH release, thereby leading to changes in follicular dynamics, oestradiol production and strength of oestrous expression. Only one of these SNPs (286) was found to affect oestrous activity in our previous study, but there the G>A mutation significantly reduced activity at oestrus. It should be noted, however, that there was only one cow with the AA mutation in the current study and no cow in the previous study.

A SNP in the prolactin receptor was negatively associated with activity at oestrus. Prolactin reduces steroid synthesis during lactation, so changes in its receptor could reduce circulating levels of oestradiol and hence expression of oestrus.

The main focus of Objective 1, and also Objective 2, was to provide further information on the oestrogen receptor alpha SNP (bERA_prom_SNP173). In the current study, 16/188 cows were found to have a single mutation (G>A) at position 173 of the promoter region of oestrogen receptor alpha. This frequency (8.5%) is higher than that observed in our previous study (2.4%), allowing more reliable conclusions to be drawn. In neither study was a cow identified with a double mutation (AA), so we speculate that this might be a lethal condition. The mutation G>A led to an average increase of 5 activity units at oestrus (P=0.046). In our previous study, the same mutation was not significantly related to activity at oestrus. The current finding, being based on a larger population and a higher frequency, is considered to be more reliable. A change in the oestrogen receptor is entirely consistent with changes in activity at oestrus because elevated plasma oestrogen concentrations provide the main signal for onset of oestrous behaviour.

Insemination outcomesStrength of oestrus expression influences the ability to detect oestrus and therefore whether or not a cow is inseminated. We were also interested to see whether strength of expression, as measured by activity at oestrus, influenced the outcome of inseminations. Data were restricted to 740 oestruses that were accompanied by insemination, and a binomial GLMM model was used to examine effects of fixed factors on probability of pregnancy resulting from an insemination.

Probability of pregnancy was not affected significantly by milk yield, days after calving, parity or season of oestrus when these factors were fitted individually. When adjusted for season, activity at oestrus increased (P = 0.041) probability of pregnancy by 0.002 per unit increase in activity over the range observed in the current study (Table 1.3). This suggests that increased activity at oestrus not only increases the likelihood of detecting oestrus, but is likely to also enhance the chance of conception through either better timing of insemination or a better developed ovarian follicle at ovulation.

Significant effects of SNPs in the FSH, GH, LH and prolactin receptors on probability of pregnancy were observed when adjusted for activity at oestrus (Table 1.3). A SNP in the FSH receptor reduced the probability of pregnancy by 9 percentage units. SNPs in the GH, LH and prolactin receptors increased the probability of pregnancy by up to 19 percentage units. It is interesting to note that none of these SNPs had direct significant effects on strength of oestrous expression, although some were found on the same genes as SNPs that did alter activity. These results suggest that, as found in our previous study, SNPs can have varying effects on both activity at oestrus and other parameters of fertility.

Table 1.3. Effects of activity at oestrus and SNPs on insemination outcomeActivity P-value Effect² s.e. Pregnancy

SID 5 (Rev. 05/09) Page 9 of 18

probability³Activity at oestrus (units) 0.041 Constant -0.3546 0.2507

Effect 0.0085 0.0042 +0.002/unit

Gene/SNP ID Genotype¹ # Cows P-value Effect² s.e.Pregnancy probability³

FSH Receptor FSHR_S596S 0 C:C 164 0.039 Constant -0.6664 0.0811 0.36

1 C:T 36 Effect -0.4344 0.2094 0.272 T:T 0

Growth hormone receptorGHRA257G_ex10 0 A:A 172 0.005 Constant -0.6615 0.0812 0.34

1 A:G 29 Effect 0.7807 0.2719 0.532 G:G 0

LH receptorLHR_L490L 0 C:C 142 0.031 Constant -0.6584 0.0812 0.34

1 C:T 54 Effect 0.3843 0.1773 0.432 T:T 2 0.53

Prolactin receptorPRL_89398_g_a_R 0 G:G 143 0.039 Constant -0.6566 0.0814 0.34

1 A:G 50 Effect 0.3335 0.1616 0.422 A:A 5 0.50

1. Genotype values: 0 = wild type; 1 = heterozygote; 3 = double mutation. 2. Effects are on the logit scale and are relative to wild type.3. Predicted probability of pregnancy are back-transformed means from fitted models.

Genetic effectsResults for single-trait analysis of oestrous activity are shown in Table 1.4. Variance due to animal, permanent environment (repeated measures within cows) and unexplained effects accounted for similar proportions of total variance for activity at oestrus (STEPS) and increase in activity above baseline (STEPS%). Heritability estimates are low-moderate for both traits; similar to heritability of other fertility traits. Repeatability (correlation between repeated measures on the same cow) was much higher for STEPS (0.41) than for STEPS% (0.14), probably because STEPS% is influenced by variation in both basal activity and STEPS.

Table 1.4.Variance components, heritability and repeatability of activity at oestrus (STEPS) and increase in activity above baseline (STEPS%)

STEPS STEPS%Variance component: Animal 1.003 1.004 Permanent environmental 1.01 1.002 Residual 1.02 1.004Heritability 0.07 (SE=0.101) 0.096 (SE = 0.078)Repeatability 0.41 (SE=0.04) 0.14 (SE = 0.04)

Bivariate analysis of BASE, STEPS and STEPS% is shown in Table 1.5. Each analysis fits a pair of traits to the pedigree and phenotypic data to estimate heritability of each trait whilst adjusting for variation in the other trait. The analysis also indicates genetic and phenotypic correlations between traits. Activity during oestrus (STEPS) had high genetic and phenotypic correlations with basal activity (BASE). Increase in activity above baseline (STEPS%) was only moderately correlated with STEPS and showed no correlation with BASE. These results indicate that cows which are generally more active show greater activity during oestrus, as might be expected. They also indicate that increase in activity is related more to activity during oestrus than to basal activity.Heritability estimates were moderate for all three traits when fitted to bivariate models. They are higher than those estimated from single-trait analysis due to the correlations between traits.

Table 1.5. Heritability (diagonals, bold), genetic correlations (below diagonals) and phenotypic (above diagonals) correlations between pairs of activity traits

SID 5 (Rev. 05/09) Page 10 of 18

 a BASE STEPSBASE 0.25 ± 0.02 0.82 ± 0.05STEPS 0.7 ± 0.01 0.19 ± 0.02

 b BASE STEPS%BASE 0.25 ± 0.02 0.05 ± 0.02STEPS% 0.05 ± 0.08 0.16 ± 0.02

 c STEPS STEPS%STEPS 0.24 ± 0.02 0.49 ± 0.017STEPS% 0.35 ± 0.07 0.16 ± 0.02BASE = baseline activity; STEPS = activity at oestrus; STEPS% = increase in activity above baseline.Data were analysed using bivariate models which adjust for covariance between pairs of traits.

Multivariate analysis of BASE, STEPS and STEPS% is shown in Table 1.6. This analysis fits all traits at once to the pedigree and phenotypic data to estimate heritability of each trait whilst adjusting for variation in the other traits. Genetic and phenotypic correlations among traits were similar to those seen with paired traits analysed with bivariate models. Heritability estimates were lower for all three traits, compared with bivariate models due to the lack of correlation between BASE and STEPS%, suggesting that there is no need to include both traits in genetic evaluations. Nevertheless, the heritability estimates do indicate that genetic progress in oestrous expression could be made by selecting animals for activity traits.

Table 1.6. Estimates of heritability (diagonal, bold), genetic correlations (below diagonal) and phenotypic correlations (above diagonal) for baseline activity, activity at oestrus, and increase in activity above baseline

BASE = baseline activity; STEPS = activity at oestrus; STEPS% = increase in activity above baseline.Data were analysed using a multivariate model which adjusted for covariance between traits

REPORT ON OBJECTIVE 2Objective 2 was to investigate the possible significance of the oestrogen receptor alpha SNP 173 by determining whether the relevant DNA sequence is transcribed, and appears in messenger RNA.

The oestrogen receptor plays a central role in many reproductive processes in all mammals. In relation to bovine fertility, this receptor is involved in ovarian and uterine function at all the important stages of the reproductive cycle, including the onset of ovarian cyclicity post partum, mammary development, preparation of the uterus for implantation of the blastocyst, maternal recognition of pregnancy and oestrous behaviour.

Given the key role of this molecule in such a wide range of vital reproductive functions, it is important that we should understand the control of its expression and function. However these are complicated processes, due to the complexity of the oestrogen receptor gene, which consists of a number of exons that may or may not be transcribed, and which have different regulatory 5’ sequences controlling their expression.

We have concentrated on one of these exons (exon C). At the outset of the project this sequence was known to be of particular interest because it contains a single nucleotide polymorphism (SNP), which may have an effect on the expression or function of the receptor protein. However during the course of the research we discovered that the sequence of interest also contains a secondary structure known as a G-quadruplex. This is of additional significance, because G-quadruplexes are absent from all other mammalian oestrogen receptor genes investigated at this position. There is a similar G-quadruplex in the human exon C, but at a different position, which we have previously shown to have significant effects on translation of the oestrogen receptor protein.

We therefore asked the following questions:a) Is the SNP-containing sequence in exon C transcribed into RNA? (This was milestone 2 in the application).b) Does the G-quadruplex in the bovine gene affect transcription or translation of the gene?

SID 5 (Rev. 05/09) Page 11 of 18

  BASE STEPS STEPS%BASE 0.11 ± 0.02 0.76 ± 0.04 0.06 ± 0.02STEPS 0.65 ± 0.37 0.06 ± 0.01 0.62 ± 0.03STEPS% -0.004 ± 0.16 0.37 ± 0.25 0.09 ± 0.03

Figure 2.1 shows the bovine oestrogen receptor exon C sequence compared with other species. Despite the high level of homology in this region between the sequences in a range of species, the bovine sequence is characterized by the insertion of 22 bases with a high guanine content. The additional sequence shown in panel B) in the bovine is the PQS number 1 in panel C), and this corresponds to the 1 in panel A). This sequence, in combination with the nucleotides immediately adjacent to it, results in 5 runs of guanine residues, which are capable of forming a G-quadruplex secondary structure. Close to the 3’ end of this run of guanines is the SNP, which is a G>A replacement mutation, and immediately following the SNP is the first of 2 E-boxes. These are short DNA sequences recognized by a large and ubiquitous family of transcription factors involved in basic regulation of transcription, as well as very specific tasks like embryo development and muscle formation.

To determine whether the sequence containing the SNP is transcribed (milestone 1) we carried out multiplex RT-PCR using a number of bovine tissues. Oligonucleotide sequences were derived from the sequence located in the promoter of exon C in the Bos taurus ESR1 gene. Bovine foetal and adult uterine tissues obtained from local slaughterhouses were frozen in liquid nitrogen. RNA was extracted using a Macheray and Nagel Nucleic Acid Purification kit. DNA contamination of RNA was eliminated by digestion with DNAse included in the kit. Additionally, 100 ng RNA before reaction was subjected to DNA digestion (2units, 10 min at 37°C) with Turbo DNnase (Ambion).

Figure 2.1. (a) The bovine ESR1 gene. The gene spans 141,328 nucleotides encompassing 9 coding exons (1-9) and the 3 known non-coding 5' upstream region exons (A, B and C). The positions of the 5 putative quadruplex forming sequences are shown by the arrows; (b) A section of the ESR1 promoter sequence showing alignment across a number of different species; the G-rich tracts (IV) are marked on the bovine sequence. The E-box motif (underlined) is conserved across these species, as is the G-rich section immediately prior to it (bold). The bovine gene includes a G-rich insert of 22 nucleotides not present in the other species listed, which has the potential to adopt an intramolecular quadruplex motif. The BLAST sequence alignment was performed using the ESR1 gene information for Bos taurus (Cow), Homo sapiens (Human), Canis familiaris (Dog), Pan troglodytes (Chimpanzee), Oryctolagus cuniculus (Rabbit) and Equus caballus (Horse); (c) Positions and sequences of the 5 putative quadruplex-forming sequences (PQS) in the bovine gene.

SID 5 (Rev. 05/09) Page 12 of 18

The sensitive, multiplex RT-PCR (GeXP, A21019, A25295, Beckman) utilised a conversion from a reaction initiated with specific primers to PCR based on universal primers. The universal sequence flanked the 5’ end of each of the primers and added an additional 37 nucleotides to the length of each amplicon. The sequence of specific primers for all genes and size of amplicons given are in Table 2.1. To exclude the possibility that amplicons arise due to DNA contamination of RNA, additional duplex reactions were performed on RNA used in the study. As DNA contamination controls, pairs of primers were used for the bovine gonadotrophin releasing hormone receptor (GnRHR) promoter (167 nucleotides), as well as 2 pairs for exon C/intron boundaries (235 and 237 nucleotides). Each of these primer pairs was used in a separate reaction spiked with mRNA coding for the kanamycin resistance gene (a non-mammalian internal marker gene). The sequences of exon/intron boundary primers are given in Table 2.1.The primers included in the multiplex PCR were designed to cover: the region of exon C, ESR1 and IGF1 as RNA condition controls and the GnRHR promoter as a DNA contamination control. Kanamycin resistance gene mRNA added to the reaction acted as an internal positive control for the system. DNase treated RNA (60 ng) was used in every reaction according to the manufacturer’s protocols. Products were run on capillary electrophoresis and analyzed using Beckman SEQ and Profiler software. The amplicons of the ESR1 gene and exon C were re-amplified in PCR and sequenced.

Table 2.1. PCR primer sequences

Primer NameProduct Size LeftSequence RightSequence

quadruplex 137 cggcaacagtgtgtctgttc aagccttctgggatccacttH2A 155 tccagtgttggtgattccag gcagaaatttggttggttggGNRHR_prom_DNA cont 167 cagtcgtgtccggacttctt atgcatgcttacggagaaggexC 182 cagacagcaagcctctcctt atgtgtttgcatgtgggatgESR1 225 cccaactcctcctcatcctc ctggctctgattcacgtcctIGF1 265 tcttctatctggccctgtgc acatctccagcctcctcagaKan(r) 325 atcatcaggattgcattcgattcctgtttg attccgactcgtccaacatc

Exon/itron boundary2 235 tccagcagggtaggactgtt ttgcttagaattcgcgaggtExon/itron boundary3 237 caagcccatggaacatttct ctgctttcaggaccgtaagg

Bovine foetal ovaries (week 21, 26, 31) and testes (week 17, 19 and 35), as well as adult ovarian follicles and uterine tissues (endometrum and myometrum) exhibited the presence of exon C transcripts. Furthermore a fragment containing the quadruplex-forming sequence, which is present in the bovine genome, is also present in these transcripts (Fig. 2.2a). Bovine kidney, used as a negative control tissue, did not express either of the fragments (Figure 2.2b,c), despite expressing the oestrogen receptor at a low level. Quadruplex/exon C expression was also tested against the exon/intron boundary and GnRHR promoter, neither of which resulted in amplification.

G-quadruplex formation results in secondary structures in DNA which, if they are copied into RNA (i.e. transcribed) interfere with the process of translation of the RNA sequence into protein. This occurs in the human oestrogen receptor, as well as in other transcripts. The data on expression (Figure 2.2) show that the G-quadruplex is present in transcripts containing exon C. To determine whether the G-quadruplex affects translation rate, we prepared synthetic mRNA coding for luciferase, with or without the G-quadruplex, and examined the rate of translation in vitro.

SID 5 (Rev. 05/09) Page 13 of 18

A

0 10000 20000 30000 40000 50000 60000 70000 80000 90000

100000 110000 120000 130000 140000 150000 160000 170000 180000 190000

120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 Size (nt)

H2A

ESR1

Kan Quadruplex ExC IGF1

B

H2A ESR1 Kan

C

Kan

Figure 2.2. Expression of exon C in bovine tissues. A multiplex RT-PCR reaction investigating transcription of the quadruplex forming region of the bovine ESR1 promoter: A) An overlay graph representing expression of exon C and its 5’ upstream region in several bovine foetal and adult tissues. B) The same multiplex reaction performed with bovine kidney RNA. The 144nt peak is a non-specific product. C) The duplex reaction for DNA contamination. Expected products are GnRHR promoter, 167 nucleotides and exon C/intron boundaries at 235 or 237 nucleotides. Kan = kanamycin resistance gene mRNA used as an internal control.

SID 5 (Rev. 05/09) Page 14 of 18

Plasmid DNA with T7/with or without the quadruplex or Mutation/SNP/E-box was linearised (at the SalI site) and 1μg of each construct was used for in vitro transcription using the mMessage mMachime T7 ultra kit (Ambion) according to the manufacturer's instructions. The RNA produced using the T7 promoter was capped and poly(A) tailed and used in the in vitro translation procedure after confirming RNA integrity and size on 1% agarose gels as suggested in Ambion’s instructions. One μg of the transcripts with incorporated regulatory sequences containing ±Quadruplex or Mutation/SNP/E-box upstream of the luciferase gene was denatured and used as a template in the in vitro translation reaction with a rabbit reticulocyte lysate extract (Promega), according to the manufacturer’s protocol. The outcome of the translation reaction (2.5 μl of cell lysate) containing luciferase reporter protein was measured using a MicroLumenotPlus LB 96V (Berthold Technologies) with a Biotium kit.

+QxG +QxA +QxA/G -QxG -QxA -QxA/G MutQ0

3

5

8

10

13

15

18

20

23

25

28

RNA sequence

Luci

fera

se a

ctiv

ity

Figure 2.3. Translational efficiency assayed in vitro investigating the role of the SNP and/or quadruplex formation on levels of RNA translation using a luciferase reporter gene containing bov-ESR1 inserts. To prepare RNA the insert required was cloned downstream of a T7 promoter sequence, and expressed using RNA polymerase. The quadruplex forming sequence was 12 nucleotides from the T7 promoter site. +Qx and –Qx indicates inclusion or absence of the 25 nucleotide quadruplex sequence insert respectively, and /A or /G are with either the A or G polymorphism (or G+A, representing a ‘heterozygous’ mixture of both alleles). Mut Q contains a GGG to AAA mutation, which was shown to block quadruplex formation.

As shown in Fig. 2.3, removal of the G-quadruplex forming sequence caused an approximate doubling in rate of translation. Mutation of the quadruplex, so that the secondary structure could not be formed, caused a greater increase, to about 10-fold over the baseline rate with the G-quadruplex. When the effects of the SNP mutations were investigated there were minor differences between the sequences, but these were evident only in the absence of the G-quadruplex forming sequence.

We conclude therefore that the exon C sequence is expressed in a variety of tissues, and that the G-quadruplex forming sequence, but not the SNP, affects translation of transcripts containing exon C.

The practical significance of this work is as follows: (a) the results show that the SNP at 173 is unlikely to result in any alteration in oestrogen receptor function; (b) oestrogen receptor function is affected by translational control as well as transcription modulation;(c) the opportunity should be taken to examine the G quadruplex sequence as a possible target for transgenesis, as its removal would be expected to modify oestrogen receptor function in a specific and controllable manner.

It is clear therefore that this preliminary work has resulted in a far better understanding than hitherto not only of oestrogen receptor functionality, but of gene expression generally.

SID 5 (Rev. 05/09) Page 15 of 18

REPORT ON OBJECTIVE 3Objective 3 was to carry out preliminary experiments to develop and test methods for pheromone collection from dairy cows.

During oestrus dairy cows, like many species, produce pheromones. These volatile organic compounds cause behavioural reactions in other cattle and have been identified in saliva, urine and vaginal mucus from cows during oestrus. There is little information in the literature on pheromones produced by cattle. In the longer term we plan to investigate levels of these compounds in cows exhibiting different strengths of oestrous activity. In this initial study we proposed to investigate methods to collect pheromone samples from individual cows around the time of oestrus, for gas analysis.

An experiment was conducted using vaginal mucus as a source of pheromones. Mucus was collected during routine artificial insemination from 16 cows known to be in oestrus. In each case, oestrus had been observed previously by farm staff, was accompanied by an increase in pedometer activity, and was confirmed from visual and physical diagnosis by the experienced inseminator. Mucus is secreted in significant quantities only during oestrus; for comparison, therefore, vaginal swabs were taken from the same cows 10-14 days later. Mucus samples and swabs were stored at -80 °C until ready for analysis.

Mass spectrometry analysis involved using an Atmospheric Pressure Chemical Ionization Mass Spectrometer (APCI-MS). This sucks in headspace volatiles and gives a signal in real time. In Figure 3.1 the MS is sampling continuously and samples were placed on the MS one after the other. The first sample was an oestrus one (at 1.5 min) and this was followed by a non-oestrus one (at 2.25 min). We were expecting a bigger signal for the oestrus samples, but, this was not the case and we got a much bigger signal for the non-oestrus samples. This was repeated for the second pair of test samples.

0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00 5.50 6.00 6.50Time0

100

%

rob 230210 Scan ES+ 40_240 1.00Da

2.07e5

Figure 3.1. APCI-MS of oestrus (1.5 and 3.25 min) and non oestrus (2.25 and 4.25 min) mucus samples.

In addition to looking at changes in the summed signal we can get a mass spectral profile formed by the volatiles entering the MS (note: they are not separated and typically gain a proton, +1 in mass fragmentation is minimal) this is shown in Figure 3.2. The non-oestrus samples showed large peaks at m/z 139 and 121 and swamped any signal for the oestrus samples. No signal stood out as bigger for the oestrus samples (scanning up to mass 240).

Samples were then run through a gas chromatogram mass spectrometer (GC-MS) to separate the volatile compounds. Again, a large peak was observed for the non-oestrus samples, which was not present in samples collected at oestrus. A search of the spectral library produced a match that suggested the big peak in the non-oestrus samples might be phenoxy ethanol. This is a bactericidal compound which must have been present on the swabs used to collect samples during the non-oestrus period. Clearly this was an artefact of sampling protocols and not a difference in pheromones.

Further examination of the GC-MS spectra revealed only one peak that appeared to be present during oestrus, but not at non-oestrus. This compound was identified as hexadecanal, which is listed as a pheromone in the insect world (particularly moths). We could find no report of hexadecanal acting as a pheromone in mammals, although it is involved in Sphingosine 1-phosphate cell signalling and there is some suggestion of reproductive links (Hla, 2003). Substances that have been suggested to act as pheromones in cattle include acetic acid, propionic acid and iododecane (Sankar and Archunan, 2008). When looking for them using key ions in their spectra, there was no evidence for these substances in samples taken at oestrus.

SID 5 (Rev. 05/09) Page 16 of 18

40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210m/z0

100

%

0

100

%

rob 230210 80 (1.492) Scan ES+ 4.58e4

59

rob 230210 121 (2.244) Scan ES+ 4.58e4139

121

4745

9559 89 109

non-oestrus

Oestrus

Figure 3.2. Spectral profiles from APCI-MS analysis of oestrus (top) and non-oestrus (lower) mucus samples.

It is unlikely that hexadecanal was the only odiferous volatile substance present in the oestrus samples. Attempts to analyse these samples at a later date for further volatile compounds were unsuccessful, however, possibly due to oxidation of compounds during 6-months of storage, which takes place even at -80 °C. Time constraints precluded collection of further samples within this project, but we intend to pursue the concept of pheromone analysis and will provide a supplementary report for Defra in due course.

ConclusionsIn this preliminary study, we have learnt some key methodological lessons: 1. If swabs are used in future studies, they need to be screened in advance for substances that might interfere with identification of volatile compounds of interest.2. It is unlikely that sufficient vaginal mucus can be obtained from cows not in oestrus, so future studies should concentrate on differences in mucus composition among cows in oestrus only, or else investigate presence of pheromones in other fluids, such as saliva or milk.3. Due to the ephemeral nature of putative pheromone compounds, future studies need to work with fresh samples or samples stored for a short period of time.

One putative pheromone, hexadecanal, was identified in vaginal mucus collected during oestrus, but further work is required to confirm if this substance does act as a pheromone in dairy cows, if similar compounds are present, and if they are related to strength of oestrous expression.

ReferencesGarnsworthy, P.C. (2004) The environmental impact of fertility in dairy cows: a modelling approach to predict methane and ammonia emissions. Animal Feed Science and Technology, 112, 211-223.

Garnsworthy, P.C. and Thomas, P.C. (2005) Yield trends in UK dairy and beef cattle. In Yields of Farmed Species (Eds J. Wiseman and R. Sylvester-Bradley), Nottingham University Press, Nottingham.

Gilmour, A.R., Gogel, B.J., Cullis, B.R., and Thompson, R. (2009) ASReml User Guide Release 3.0 VSN International Ltd, Hemel Hempstead, HP1 1ES, UK www.vsni.co.uk

Hla, T. (2003) Signaling and biological actions of sphingosine 1-phosphate. Pharmacological Research, 47, 401-407.

Khatib, H., Monson, R.L., Schutzkus, V., Kohl, D.M., Rosa, G.J. and Rutledge, J.J. (2008) Mutations in the STAT5A gene are associated with embryonic survival and milk composition in cattle. Journal of Dairy Science, 91, 784–793.

Maas, J.A., Garnsworthy, P.C. and Flint, A.P.F. (2008) Modelling responses to nutritional, endocrine and genetic strategies to increase fertility in the UK dairy herd. The Veterinary Journal, 180, 356-362.

Sankar, R. and Archunan, G. (2008) Identification of putative pheromones in bovine (Bos taurus) faeces in relation to estrus detection. Animal Reproduction Science, 103, 149–153.

SID 5 (Rev. 05/09) Page 17 of 18

References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.

The following article, arising from work under Objective 2, has been published online:

Kamila Derecka, Graham D. Balkwill, Thomas P. Garner, Charlie Hodgman, Anthony P. F. Flint and Mark S. Searle. (2010). Occurrence of a quadruplex motif in a unique insert within exon C of the bovine estrogen receptor α gene (ESR1). Biochemistry, 49 (35), 7625–7633. DOI: 10.1021/bi100804f

SID 5 (Rev. 05/09) Page 18 of 18