Intro to PCOS

Question 1

How does PCOS affect the NHS?

Answer 1

• Most common endocrine disorder in women of reproductive age, but it has ramifications beyond that.
• Arguably the most prevalent medical condition in women (presents in many different ways)
• Common in SE asian population.
• Patients with PCOS have systemic metabolic manifestations with multiple symptomatology: endocrine, gynaecological, diabetic, dermatological, eating disorder, psychiatry
• Presents in many different ways; can come in through different clinics
• Complex symptomology; makes diagnosis difficult, leads to issues in treatment too
• A lot of these women have insulin resistance (IR), which is likely to have a life-long impact on patient. The IR is the main metabolic concern as it has long-term problems rather than fertility issues. Even though PCOS can be seen as affecting women in their reproductive years, the insulin resistance and metabolic manifestations and associations of that have lifelong impact on them.
• Annual economic cost of diagnosis and treatment of PCOS in USA recently calculated to be $4.36 billion and 40% of this was IR/T2D-related = VERY expensive. In the US a significant problem is also because of obesity, though the UK is catching up. It costs the NHS a lot of money.
• IR/T2D = insulin resistance and type 2 diabetes.

Question 2

What are polycystic ovaries?

Answer 2

• The ovary contains increased numbers (>12) of small antral follicles (2-9mm) visible on ultrasound
• Ultrasound shows a necklace of antral follicles laid around the periphery, a thickened stroma and enlarged ovary overall.
• Polycystic ovaries are characterised by this increased number of small antral follicles visible on ultrasound. Would not expect to be seen in somebody with normal ovaries in that part of the menstrual cycle.
• There is a disorder of follicle growth at all stages
  1) It has been suggested that maybe women with PCOS start off with an increased proportion of primordial follicles compared to normal ovaries, however it has been confirmed that there are an increased number of activated (primary) follicles.
  2) Arrested antral follicle growth before they mature
  3) Lower rates of atresia » antral follicles persist (visible on u/s)
• There is increased activation of growth, so more primary follicles, but then they tend to slow down and stop in the antral phase. There is an accumulation of antral follicles, they arrest, but they don’t die off. Normally, once dominant follicle selection occurs, the remaining antral follicles die off by atresia. Women who have PCOS have lower rates of atresia.
• In some cases there is a failure of dominant follicle selection and therefore anovulation.
• In a good proportion of women, they also do not ovulate = anovulatory PCOS. The classification for anovulatory PCO can include one or two ovulations a year. There is a spectrum of presentation = some women will have polycystic ovaries and ovulate; others are anovulatory.
• Lower rates of atresia = arrested antral follicles remain

Question 3

How was PCOS initially described?

Answer 3

• First described in 1935 by Stein and Leventhal. PCOS is a much older condition than 1935. There are some references to it, even in literature and readings from Roman ages (they wouldn’t have been able to look inside, but just the external presentation, i.e. description of symptoms of women, which we would now classify as PCOS). They thought it was cysts present, but they are now known to be antral follicles that persist. Ovarian cysts are completely different (PCO is an unfortunate name).
• Syndrome described in 1935 as obesity, hirsutism (excess hair) and anovulation in the presence of bilaterally enlarged sclerocystic (hardened) ovaries. In the initial paper, there were only 10 women (all overweight).
• Initially diagnosis by laparotomy (open surgery), visual examination, post morterm = no ultrasound or laparascopes. Now diagnosed routinely by U/S (since 1980s).
• As ultrasound came into prevalence, they found that many women appeared to have polycystic-like ovaries. They look like polycystic ovaries, but they wouldn’t necessarily have all the symptoms. This caused massive confusion as to how it was diagnosed. Lead to discrepancies in diagnosis (diagnosed differently in Europe and the USA); makes it harder to use/compare older studies.
• Spectrum of presentation has led to lack of consensus regarding the definition
• Have to exclude disorders that mimic PCOS before diagnosing it.
• Diagnosis of exclusion i.e. disorders that mimic PCOS:
  1) Non-classical adrenal hyperplasia (most common is deficiency of 21-hydroxylase → ↑17-hydroxyprogesterone production & androgens)
  2) Hyperprolactinemia, thyroid disease, Cushing’s syndrome
  3) Ovarian hyperthecosis (very rare) - nests of luteinized theca cells (tumour or overgrowth of theca cells). Excess androgen production due to nests of luteinized theca cells scattered throughout ovarian stroma. Ovary slightly enlarged but devoid of antral follicles – hence distinct from PCOS. Serum testo levels>women with PCOS but

Question 4

What is the standard diagnostic criteria used for PCOS?

Answer 4

• Rotterdam Criteria
• For a woman to be considered as having PCOS under the Rotterdam criteria, two out of the following three criteria are needed = polycystic ovary (some women may still be ovulatory so have a DF present), hyperandrogenism and ovulatory dysfunction.
  1) Polycystic Ovaries = either 12 or more follicles measuring 2-9mm diameter and/or increased ovarian volume >10ml in either ovary & no DF >10mm. Technique (operator) and equipment dependent. Transvaginal imaging not always appropriate, e.g. under 18’s (when PCOS often starts to present), people who can’t consent
  2) Hyper-androgenism = clinical/biochemical evidence. Biochemical = measured, e.g. serum or blood androgens (currently researching a standardised assay; can’t have normative data/normal ranges without a standardised assay), Clinical = how symptoms present. Clinical hyperandrogenism includes hirsutism, acne, or male pattern alopecia or biochemical signs of hyperandrogenism i.e. elevated levels of total or free testosterone. Assays not standardized across labs; normative data not clearly defined; clinical hyperandrogenism difficult to quantify; ethnicity. Need normative scoring for each population to account for ethnic differences.
  3) Ovulatory Dysfunction = Oligomenorrhea/anovulation. Frequent bleeding <21d or infrequent bleeding >35d. To confirm ovulation serum progesterone level at mid-luteal phase (d21-22) of cycle (values ≥7ng/ml needed for regular luteal function). Anovulation is either frequent bleeding at intervals <21d or infrequent bleeding at intervals of >35d. Normal ovulation is hard to define – mid-luteal Progesterone <3-4ng/ml indicates oligo-anovulatory. To confirm whether someone has ovulated, the gold standard is to measure serum progesterone which comes from the CL. This is not feasible to do regularly; need tests that can be used in other health settings, e.g. primary care GP clinics, rather that serum progesterone to measure ovulatory dysfunction.
• Need a second “hit” to cause PCOS – Usually insulin/insulin resistance that tips a woman over and makes her become anovulatory. Rare for a woman to become completely anovulatory with PCOS. They will be late a couple of times a year, but not very frequently and it may get worse with age.
  Some women can have ovaries that look polycystic on ultrasound and may tip over, but don’t have any other signs (present with normal androgen levels, no hirsutism/acne and regular cycles). This is why the Rotterdam criteria helps; requires PCO plus one of the other two. Similar to diagnosis of exclusion, e.g. woman may seem to present with PCO, but it can be explained by high prolactin from blood tests.
  Adolescent girls undergoing puberty can go through a phase of ovaries looking polycystic so have to be careful when scanning them.

Question 5

Define PCO

Answer 5

• By Ultrasound:
  1) Normal ovary has no more than 5 antral follicles in an ovary with a small amount of stroma in a woman with regular cycles
  2) PCO = In at least one ovary ≥ 12 follicles of 2-9mm diameter arranged peripherally around an enlarged core of dense stroma - ovarian volume >10mls, without a dominant follicle
  3) PCOS = PCO on scan plus one or more symptoms. Women can appear to have PCO but two or more of the criteria need to be met for PCOS.
• The terms cyst and follicles used interchangeably, but this is incorrect – they are NOT cysts in PCO, but instead follicles – has lead to confusion.
• Scanning is not a science but an art ie. very operator dependent. It is easy to miss follicles and it also depends on where you are in the cycle. Also age – adolescents can have multi-follicular morphology until puberty is complete. Ideal would be to scan sequentially.

Question 6

Compare the morphology of the PCO of a normal ovary, an anovulatory PCO and an ovulatory PCO in each stage of the cycle.

Answer 6

• In the early follicular phase, in a normal menstrual cycle, scans would show a small, round ovary with about 5 or so antral follicles seen. By the mid follicular phase, a DF starts to emerge among the group of follicles. Coming close to ovulation, the other antral follicles should be dying off (reduce in number), while the big DF is clearly present and gets ready to ovulate.
• The anovulatory polycystic ovary more or less does not change throughout the menstrual cycle, i.e. enlarged ovary with white, echogenic stroma with a necklace of follicles. The dominant follicle is not seen, there are lots of antral follicles (2 to 9mm) around the edge, enlarged core and stroma, remains persistent.
• An ovulatory PCO starts off very much like the anovulatory PCO. Then, by the mid follicular phase, a dominant follicle starts to be seen which can ovulate. The other small antral follicles will remain and persist; this is what makes the difference between ovulatory PCO and someone without PCO who has normal cycles.

Question 7

What is the prevalence of PCO?

Answer 7

• Very common condition
• PCO present in
  1) 32% of patients with amenorrhoea (no ovulation/periods)
  2) 87% with oligomenorrhoea (irregular cycles)
  3) 87% with hirsutism and regular cycles
  4) 75% of bulimics?
  5) 22% of ‘normal’ population (those who consider themselves normal)
• These studies have been repeated a lot and averages from many studies calculated
• It is the most common cause of anovulatory infertility, one of the most common reasons why women would present at an IVF clinic or for any type of assisted reproduction.

Question 8

What studies have looked at PCO in the normal population?

Answer 8

• ‘normal’ population’ = those who consider themselves normal
  1) Polson et al. 1988 (only used ultrasound)found that 22% of 257 women who considered themselves normal (had normal cycles) had PCO on scan. However, 75% had irregular cycles on closer observation, 19% had hirsutism.
  2) Hull et al. 1987 biochem (measured androgens) and cycle history. 20% of women had biochemical marker for PCO.
  3) Clayton et al 1993 used ultrasound and biochemistry. 23% of “normal” women
  4) Rodin 1995 (SGUL) found that 52% of women in UK from Indian sub-continent
  5) Numerous studies repeated since on prevalence = PCO approx. 20% and PCOS 5-10% depending on definition.

Question 9

Describe the aetiology of PCOS.

Answer 9

• Familial aggregation
  1) Sisters more likely to be affected
  2) first-degree relatives have higher rates of metabolic abnormalities (including insulin resistance, decreased beta-cell function etc)
  3) Male relatives of women with PCOS increased prevalence of metabolic syndrome (collection of related conditions, e.g. abnormal lipids, cholesterol etc.) & obesity compared to general US male population
• Monozygotic twins twice as likely to both have PCOS than dizygotic.
• common finding of raised androgen led to belief that PCOS is caused by an inherited disorder -most likely a gene in the steroid biosynthetic pathway
• Many candidate genes were investigated: all ‘obvious’ ones ruled out
• Now know it is a complex polygenic disease – involves subtle interaction with environmental factors (intra- & extra-uterine)
• Difficult to use family history as the Rotterdam criteria was developed relatively recently (2003/2004); can only carry out prospective studies with relatively younger cohorts of women and track them forward. These types of studies have become more popular and easier with the consensus published.
• As raised androgens is common, a lot of genetic studies went into looking at all the genes involved in the androgen-producing pathway. The steroids pathway has different enzymes that are encoded for by different genes.
• There is a definite impact and interaction with the environment. In utero, if the mother has PCOS, the babies are more prone to having these effects/traits and extra uterine as seen in the adult Rhesus monkey experiment. This classic experiment showed that by altering the monkeys’ diet to such an extent, it also altered the ovarian phenotype. Having the predisposing genes increases the likelihood of getting PCOS and it is going to present in puberty and in post puberty.

Question 10

What candidate genes were highlighted in the Chinese studies?

Answer 10

• 1st Genome-wide association study (GWAS) identified causative genes in Han Chinese women (2011) = 744 women with PCOS & 895 controls
• 3 loci linked and candidate genes within these loci are:
  1) LHCGR
  2) FSHR
  3) THADA….linked to T2D
  4) DENND1A …linked with obesity
  Chen et al (2011) Nature Genetics 43:55-59
  Zhao H & Chen ZJ (2013) Hum. Reprod. 19:644-654
• (October 2014) GWAS confirmed variants in DENND1A, THADA, FSHR & INSR were associated with PCOS in Europeans (Brower et al (2014) JCEM)
• Confirming the biological relevance of PCOS-associated variants by molecular analysis (e.g. expression analysis, targeted genetic disruption in cell culture or organism) is critical to confirming findings from GWAS – rarely done.
• Past GWAS studies very poor – due to bad design, lack of consensus of PCOS. 1st Chinese one used Rotterdam criteria fulfilling all 3 criteria
• There have been other genome wide studies, but this one was good because they used a large sample with the Rotterdam criteria (a lot of consistency; all diagnosed by the same criteria, led to further studies that could be used for comparison)
• Candidate studies = comparing women with PCOS to women without to identify candidate genes.
• Genome wide association study is comparing women who have PCOS to women who don’t. Candidate genes associated with the population of PCOS are identified (shared variants that are not necessarily present in the normal population). These genes were very consistently associated with the loci (the areas on the chromosomes that had come up). This was also confirmed in a similar study done on PCOS in the European population. The candidate genes were linked; two genes in particular have been highlighted in the aetiology of PCOS.
• Part of the problem with these GWAS studies is that they are just associative studies. They are good, but they don’t give any actual idea of the relevance of those genes or whether they actually have a biological effect or the mechanism. Need to take these candidate genes and do more studies in the laboratory, e.g. on cells, cell lines, tissues, to prove those genes actually cause the disease rather than are associated.

Question 11

How was the association between PCOS and DENND1A.V2 proven?

Answer 11

• Theca cells were transfected with DENND1A isoform and treated with/without forskolin (to stimulate cAMP). Usually difficult to obtain a theca cell line. This cell line has now stopped working.
• They forced the expression of a variant associated with PCOS in a normal theca cell line. It resulted in increased androgen and progesterone production.
• cAMP is the secondary messenger for both the LH and FSH pathways; will induce all of the downstream signalling events no matter how it is activated. Stimulates the whole steroid pathway. Able to measure steroid production.
• Measured production of various androgens and progestegins
• DENND1A.V2 overexpression recapitulated hyperandrogenic theca cell function. Proved the theca cells could produce excess androgens.
• The different graphs measure different steroids (various androgens and progesterones).
• Proved that this variant made women produce more androgens and progesterones. It is intrinsic to the theca in women with PCOS; this variant increases androgen and progesterone production
• Can sometimes have an association that does not have a true effect; can be linked to another gene close by.
• Forced expression of DENND1A.V2 in normal theca cells results in augmented androgen and progestin production.

Question 12

How is the LH:FSH ratio altered in PCOS?

Answer 12

• Consistent feature of PCOS is disordered gonadotrophin secretion leading to downstream ovarian consequences
• Elevated/upper-normal mean LH and Low/low-normal FSH = altered ratio LH:FSH.
• Rapid GnRH frequency → favouring rapid LH pulse secretion
• Elevated LH pulse frequency and increased LH:FSH ratio impairs downstream ovarian folliculogenesis and alters steroid hormone production. - In PCOS the “FSH threshold” that is required for follicle maturation is not reached (due to this dysregulated gonadotrophins and also increased AMH, which alters FSH sensitivity) causing follicles to arrest in antral stages.
• What may be more important in PCOS is the LH:FSH ratio. Favours LH being higher.
• Consistent features of PCOS are high androgens and disordered follicle growth. Often linked to disordered gonadotrophin secretion.
• They have an altered LH to FSH ratio which favours LH being higher (FSH may be higher, LH may be higher, sometimes both occurs). However, this was not in every single case (not consistent), so it never became part of the diagnostic criteria. It is still present in the majority of women.
• This can be seen in experiments. Measuring LH from control lean and obese women and then corresponding PCOS groups (important to match groups by BMI). Regular blood samples measured LH (faithfully reproduced as GnRH). Women with PCOS have much higher and dysregulated LH pulses, whether they are lean or obese, compared to the relevant controls. This means that body weight is not really having an impact; it is an intrinsic effect which is altering the LH pulses (intrinsic effect on GnRH).

Question 13

Why does dysregulated gonadotrophin secretion occur?

Answer 13

• Impaired negative regulation of GnRH pulse generator
• High Testosterone impairs negative feedback by Progesterone in presence of oestradiol. Intrinsic high testosterone in the polycystic ovaries will impair the negative feedback by progesterone in the presence of oestradiol. Normally, progesterone in the luteal phase negatively feeds back onto the GnRH and high testosterone impairs that. Progesterone normally acts as a brake, but it is prevented from acting so it is dysregulated.
• Proof: block AR with flutamide → progesterone then able to↓ LH & FSH. Interestingly if use flutamide (which blocks the AR) can reverse this insensitivity to progesterone – implicates testo as a causative factor in altered steroid hormone feedback.
• Also see that in late puberty girls (nearing the end) without hyperandrogonaemia respond to progesterone with ↓LH pulse frequency overnight, which did not occur in HA girls at same pubertal stage. Thinking about puberty, there are often changes which occur at night and changes in pulse frequency. This change in frequency can be blocked with progesterone. This will not occur in girls who have hyperandrogenaemia.
• High testosterone is intrinsic in the PCO.

Question 14

Why does dysregulated gonadotrophin secretion occur?

Answer 14

• Impaired negative regulation of GnRH pulse generator
• There is rapid GnRH frequency and high LH
• High Testosterone impairs negative feedback by Progesterone in presence of oestradiol. Intrinsic high testosterone in the polycystic ovaries will impair the negative feedback by progesterone in the presence of oestradiol. Normally, progesterone in the luteal phase negatively feeds back onto the GnRH and high testosterone impairs that. Progesterone normally acts as a brake, but it is prevented from acting so it is dysregulated.
• Proof: block AR with flutamide → progesterone then able to↓ LH & FSH. Interestingly if use flutamide (which blocks the AR) can reverse this insensitivity to progesterone – implicates testo as a causative factor in altered steroid hormone feedback.
• Also see that in late puberty girls (nearing the end) without hyperandrogonaemia respond to progesterone with ↓LH pulse frequency overnight, which did not occur in HA girls at same pubertal stage. Thinking about puberty, there are often changes which occur at night and changes in pulse frequency. This change in frequency can be blocked with progesterone. This will not occur in girls who have hyperandrogenaemia.
• High testosterone is intrinsic in the PCO.

Question 15

What is the effect of high LH levels in PCOS?

Answer 15

• LH levels in PCOS show normal biological variation
• The higher LH will drive thecal cell hyperplasia and the hyper-androgenemia, but HA is also intrinsic and can be independent of LH.
• The OC pill works by feedback to reduce LH. Hence, with PCO+OC pill, some women respond, but some don’t – hence adding further weight to evidence for disrupted feedback.
• In women who have polycystic ovaries, the median is higher but the spread is larger (the spread overlaps with normal ovaries, but some also have very high LH). LH is suppressed in women with normal ovaries who have been put on the oral contraceptive pill. In women with PCO, the oral contraceptive pill does not do very much. LH binds to the LH receptor on theca and drives androgen production, as well as theca growth = higher LH drives theca cell hyperplasia (and hyperandrogenaemia). Therefore, high LH makes the androgen situation worse with PCOS, but these women often have high androgens anyway as a result of the intrinsic defect in the theca itself.

Question 16

What is the effect of PCOS on androgen levels?

Answer 16

• Increased androgen
  1) Most consistent biochemical abnormality in women with PCOS is hypersecretion of androgens
  2) ideal to measure free (T) i.e. SHBG and total testosterone to work out free T. SHBG = sex hormone binding globulin. In the circulation, testosterone is bound to SHBG. SHBG binds androgens (testosterone) and prevents it from acting. Therefore, only free testosterone will exert its effect and is ideal to measure.
  3) anov> ov>normal. Women who have anovulatory PCOS have higher free androgen levels than women with ovulatory PCOS, which is more than normal.
  4) increased androgen production by the ovary, even in ovPCO
  5) Increased LH leads to increased androgen production
• May lead to hirsutism and acne = consistently reported as most distressing symptoms in young women. Hirsutism and acne are the most distressing symptoms for a lot of young women. This is because of the way androgens are converted in the skin.

Question 17

What symptoms are caused by high testosterone levels?

Answer 17

• Androgens play a critical role in impaired follicle growth as we will come to see.
• Androgen-producing tumour can arise from adrenal or ovary = dramatic progression in symptoms of severe hirsutism and virilization
• In women with normal ovaries, testosterone reaches levels of around 2 nmol/l
• PCO with regular cycles = slightly elevated androgens
• Hirsute but ovulate = elevated androgens and large error bars
• Anovulatory women have higher androgens
• Anovulatory and hirsute shows a big spread in androgens
• Increasing androgens = increasing severity of syndrome and presentation.
• Will never get male levels of androgens in women with PCOS.

Question 18

How are androgens linked to hirsutism?

Answer 18

• Testosterone converted to DHT at hair follicle!!!
• For treatment, sometimes give 5alpha-inhibitors
• DHT more potent androgen
• 5a-reductase may be higher in PCOS
• Not just absolute levels of testosterone per se but the sensitivity to AR – see this with acne. is AR in some women more effectively/ excessively coupled to its downstream signalling pathways, which will make the endpoint of that pathway more excessive?

Question 19

Where is all of this excess androgen coming from?

How would we find out?

Answer 19

• Excess androgens coming from either the ovary and/or cells of adrenal cortex.
• Would be the theca cells of the ovaries
• Has to be carried out as a diagnosis of exclusion.
• Can use GnRH agonist or antagonist to switch off the ovary. This would not be clinically relevant, only potentially used in research
• Suppress ovary with GnRH analogues and measure adrenal androgens. Then stimulate adrenals with synacthen and measure cortical and 17-OHP. If >33 then has CAH.
• Dexamethasone test to rule out adrenal function = if the androgen levels are suppressed, it is coming from the adrenals. If they are unaffected, it is coming from the ovaries.

Question 20

Where do androgens come from in women?

Answer 20

• In normal women, the adrenal glands and ovaries secrete androgens in response to ACTH and LH respectively. Approximately half the androgen production stems from direct secretion and half from peripheral conversion by enzymes in skin, liver and adipose tissue. In women with PCOS the ovary is main source of androgen, though adrenals do contribute in about 30-50% of women
• Can access androgens from the adrenals, but PCOS refers to excess androgens from OVARY (coming from the adrenals is a different problem; diagnosis by exclusion).
• Excess androgens from the adrenals still act on peripheral tissue, hair follicles and the skin. This is why a diagnosis of exclusion has to be carried out.

Question 21

How is steroid production by theca affected?

Answer 21

• Theca grown in vitro and steroids measured. Comparing the androgen output/1000 theca cells from normal and PCOS.
• Androstenedione is being measured. Androstenedione is the main androgen from ovary and not testo.
• It’s not just androgens that are higher but also progesterone. Progesterone is being measured in relation to androgens due to steroid conversion.
• Theca was cultured from follicles in vitro for some length of time. The media can be removed, and the steroids produced from the cells into the media can be measured. When compared to women with normal ovaries, androstenedione production was significantly higher in women with PCOS. This was also shown for progesterone. This is because they are part of the same pathway, meaning the whole pathway is dysregulated and progesterone is also affected. This shows an intrinsic dysfunction in theca cells; these cells were not treated with anything (theca was dissected out from normal follicles and follicles from PCO, grown in culture for 48h and then measuring the steroids in the medium).
• Intrinsic dysfunction in theca cells

Question 22

Due to ↑ in number of arrested follicles will see a slight ↑E2. Why is there no increase in DF levels in spite of ↑T?

Answer 22

• Pathway shows that progesterone and androstenedione are all in the same pathway; some are precursors, some can be secreted straight into blood. Both enzymes, 17α-hydroxylase and C17-20-lyase are encoded by the same gene, CYP17. This gene makes two different enzymes that are similar in characteristics.
• Common misconception is that high testosterone and androstenedione etc. means there is high oestradiol. There is a slight increase in oestradiol, but this is a reflection more of the increased number of antral follicle seen in PCOS. Each follicle produces oestrogen so there is a slight increase, but it is NOT because of more androgens. Dominant follicle levels of oestrogen are NOT seen. It is though it could be because maybe the levels of androgen receptor may not have increased in granulosa cells, so the excess testosterone floating around is not able to bind to saturate the existing androgen receptors. There is also no increase in aromatase either.
• Some papers say there is an increase in oestradiol and that is what is causing the feedback effect and the change. Generally, it is not; it is about the androgens and high androgens does not necessarily mean excess oestrogen (can be normal/a little high).

Question 23

How is CYP17 activity affected by PCOS?

Answer 23

• The enzymes involved in androgen production are higher with increased promoter activity and decreased degradation of mRNA (i.e. increased mRNA stability).
• Stable phenotype of PCOS theca
• CYP17 promoter is more active
• CYP17 mRNA degrades more slowly
• all of the steroids in that pathway are raised in PCO.
• Found that CYP17 is more active, producing more of its product (the enzymes), but the mRNA degrades more slowly. Normally, the gene is transcribed into mRNA, leaves the nucleus to produce protein and then is degraded away. In theca from PCOS, that mRNA does not degrade very quickly. If it is present, it is able to make more protein.
• Have to think about all of these various factors that come into play in an exam question. There are many reasons and one thing leads to another.

Question 24

How does insulin act as a co-gonadotrophin with LH?

Answer 24

• In a lot of instances, women have insulin resistance. Therefore, they have high insulin.
• That high insulin acts as a co-gonadotrophin with LH. Insulin doesn’t bind to the LH receptor, it binds to its own receptor but stimulates similar downstream pathways and merges the signalling pathways with LH. They each bind to their own receptor, activate their pathways which merge and increase androgen production.
• hyperinsulinemia will augment hyper-androgenemia
• One thing leads to another; these women have a perfect storm of events which makes things worse for them.

Question 25

PCO Summary

Answer 25

• Polycystic ovaries are genetically acquired and may have a foetal origin
• There is a defect in the way follicles grow
• There is a basic defect in steroid metabolism
• The gene/s causing this are becoming known
• Endocrine disturbances result in miscarriage, anovulation, infertility and hyper-androgenism

Question 26

At what stage do the follicles increase in number? Are the number of follicles at earlier stages increased?

Answer 26

• At the antral follicle stage, follicles can be counted and assessed using ultrasound. If the arrest is occurring here, we can know about it (and this is what has happened)
• Everything feeds in starting from the early stages. To find out if the number of follicles are increased at earlier stages, the tissue is soaked in solution to fix it. Enters stasis (stops at whatever stage it is in). It goes through a process of being put through a graded series of alcohols, embedded in a wax block and cut into very thin sections. Then it can be stained and counted – this was carried out by Hughesden (1982).
  1) Hughesden 1982 counted follicle numbers in sections from 17 PCOs and 17 normal ovaries and found two times the number of follicles in all of the growing stages in PCO compared to normal ovaries. There were errors in his counting methods, but still borne out by other studies that an increase even in preantral follicles.
  2) Webber et al (2003)..Lancet counted follicles in biopsies and found six times more primary follicles, no significant increase in primordials
  3) Maciel GA et al, (2004)…JCEM Counted follicles in sections = ‘stockpiling’ of primary follicles.
• Increased follicle activation and recruitment but growth of follicles arrested before they mature
• Later studies found an increase in primary follicles but no significant increase in primordial follicles. It would seem that in women with PCOS, there is increased activation. From the resting pool, an increased number of follicles are activated to start growing, but then they arrest before they mature. They arrest and slow down at the primary follicle stage but even in the early antral stage.

Question 27

What might cause increased numbers of follicles ?

What are the candidates?

Answer 27

• Not FSH, because no increase in FSH levels.
• Androgens seem likely candidate for increasing follicle numbers early in folliculogenesis. Androgens involved in stimulating primordial follicle initiation and increasing number of small antral follicles. LH hypersecretion amplifies androgen production by theca. AR expression found in GC at all follicle stages. Experiments in monkeys demonstrate that exogenous androgens increase the number of classic primary follicles. The mechanism is unclear but appears to be mediated by androgen receptors expressed in the granulosa cells. In this context, experiments in monkeys suggest that the androgen-induced increase in the pool of primary follicles might involve the increased expression of IGF-I and IGF-I receptor mRNAs in the oocytes. In both normal and PCOS ovaries, androgen receptors are expressed in granulosa cells of preantral follicles. Furthermore, treating women with exogenous androgens leads to the development of bilateral polycystic ovaries. Thus, a model for describing changes in the rate of folliculogenesis in PCOS, which takes into account the local effect of increased ovarian androgens on the accumulation of classic primary follicles, could be proposed. Androgen seemed to be important in the early stage (in early activation from primordial to primary to the pre antral stage). It is known that androgens are involved in that process = LH hypersecretion will amplify androgen production and androgen receptors are also found at all stages of the preantral follicles in granulosa cells.
• Increased numbers at the primary stage persist to antral stages? Low/normal FSH:LH ratio reduce normal maturation. It is thought that the abnormal ratio, where FSH (needed for follicle growth) is always a bit lower and LH is a bit higher, it reduces normal maturation. There is also a lower rate of atresia.
• Intra-ovarian factors involving follicular
  recruitment & growth also contribute. AMH & others TGF-β superfamily members. AMH production high from granulosa cells of PCO . Still looking at all the factors that affect whether the increased number of primary follicles just persist to the antral stage. There are also other intraovarian factors, particularly AMH. It is known that AMH levels are high in PCO, not just because there are more follicles, but also because each granulosa cell in the polycystic ovary makes more AMH. The increased amount of AMH will contribute to the follicle arrest seen in PCO.
• It is well recognized that the rate of LH release is increased in women with PCOS (54, 55) and that increased plasma LH contributes to increased androgen production by the theca interstitial cells (9). That this LH alteration might be a part of the mechanism of primary follicle build-up in PCOS ovaries comes from a rather startling study in mice (56). It was found that treating neonatal mice with pregnant mare serum gonadotropin (PMSG) markedly increases the number of growing primary follicles, independent of changes in number of the primordial follicles. Thus, high levels of gonadotropin (PMSG) are capable of slowing the rate of primary follicle growth in the neonatal mouse.

Question 28

What is the Hypothesis of PCOS origin relating to AMH?

Answer 28

• Excess AMH in utero may affect development of female foetus
• This AMH arises from mother and not the foetus
• In women with normal fertility AMH levels would drop during pregnancy
• In pregnant women with PCOS, AMH levels are elevated
• Treated pregnant mice with AMH → altered neuroendocrine phenotype of female offspring and induced PCOS-like phenotype
• “Elevated prenatal anti-Müllerian hormone reprograms the fetus and induces polycystic ovary syndrome in adulthood”
  Tata B et al (2018) Nature Medicine 24(6):834-846
• This paper said that the reprogramming is actually due to AMH. They had pregnant women with PCO and pregnant women without. They measured AMH levels when both groups were between 16 to 19 weeks pregnant. They found that women with PCOS had higher levels of AMH compared to those who did not. They claimed that this AMH could be the reason why the neuroendocrine pathway was being altered in foetus. They carried out some definitive experiments by treating pregnant mice (can’t look at cause and effect directly in humans) with AMH and showed that they had this altered neuroendocrine phenotype, altered GnRH pulsatility, LH etc. and these mice developed PCOS-like phenotypes.

Question 29

What is the Hypothesis of PCOS origin relating to androgens?

Answer 29

• Exposure of developing hypothalamus to excess androgen before final programming of steroid feedback and other regulatory mechanisms alters GnRH pulsatility and feedback
• It is believed that androgens definitely have a role to play
• There’s still a bit of a question mark over the foetal exposure to antigens, but it could be there is reprogramming things. While there may not be enough crossing over to masculinise the foetus, it may cause reprogramming events (not only of the hypothalamus and pituitary, but also reprogrammes the pancreas, fat etc. so they are predisposed to insulin resistance or metabolic syndrome etc.). In adulthood, they will then present with that PCOS phenotype. It could be that it is just affecting the GnRH pulsatility; the jury is still out, but we are closer than we were a few years ago in thinking about this.

Question 30

What evidence suggests androgens are a cause of PCOS?

Answer 30

• Excess foetal T (exposure of female animals to elevated androgens in utero) induces PCOS-like traits that manifest in offspring during adulthood:
  1) Sheep models had increased LH pulsatility and impaired E2/P feedback
  2) Offspring of T-exposed monkey mothers after puberty → LH hypersecretion, ovulatory dysfunction, hyper-androgenism and IR. 50% have enlarged ovaries and increased follicles counts
  3) Adolescent girls with HA have similar pattern of rapid LH pulse secretion before menarche
  4) Obesity in pubertal girls also alters LH pulses
  5) Pregnant women with PCOS? Maternal T is raised but is fetus exposed? High levels of SHBG & aromatase activity in placenta, prevent maternal T crossing over. Maybe excess secretion coming from foetus itself?
• The other effect that makes it a little more complicated. The study in the macaque monkeys used adult monkeys. There are numerous animal models whereby exposing the maternal animal, in this case the mother sheep, to high testosterone levels (so the foetal lamb was exposed to high androgens in utero), then the female lamb displayed PCOS-like traits after reaching adulthood.
• Therefore, exposure in the womb to high androgen levels can induce PCOS in these animals. In these sheep, it was not just about the phenotype of PCOS, they also showed that increased LH pulsatility and impaired feedback
• Similarly, with monkeys, the pregnant monkeys were exposed to testosterone and the offspring were tracked from birth past puberty. They have all the characteristic features of PCOS.
• Measuring androgen levels of girls in puberty, those with hyperandrogenaemia also have this rapid LH pulse secretion before menarche. Obesity alters this. Girls in going through puberty can go through a vicious cycle which makes things worse for them.
• The key issue is whether the foetus can be exposed to testosterone. If a woman has PCOS, gets pregnant, has raised androgens, will that androgen cross over the placenta into the foetus? If it crossed, it would masculinise the female foetus. The body has to ensure this does not happen. There is lots of aromatase and SHBG in the placenta to prevent this from happening. The question remains whether what happens in animal models happens in humans (clearly it can be induced). Is the mechanism that testosterone is crossing the placenta and then somehow affecting the foetus yet not masculinising the female foetus?
• It is now thought that maybe it comes from the foetus itself. Comes back to a genetics; there are some genes that are present in the foetus of a woman who has PCOS which are going to predispose a female foetus to developing PCOS and there is something in the womb that triggers it.