The Peer review that wasn't

A brief overview

Sophia Siedlberg


Issue 17
Spring 2002

Something strikes me as very strange about an article that consistently appears in numerous journals about "MPH" (Male pseudohermaphroditism) or rather what that term is meant to imply in terms of the paper in question. The paper, that causes me considerable annoyance is:

"The Identification of 5 alpha Reductase-2 and 17β Hydroxysteroid Dehydrogenase-3 Gene defects in Male Pseudohermaphrodites from a Turkish kindred." (Selcuk Can, Yuan Shan Zhu, Li Qun, Qi Ling, Mellisa D Katz, Suat Akgun, Cedric H.L.Shackleton and Juilanne Imperato-McGinley.

The last individual on the credits list is someone I am already very critical of for having conducted studies of 5-alpha patients without considering the fact that we are talking a limited gene pool here, and yet McGinley is cited very often as having laid down the "definitive" clinical model of 5 Alpha patients. Implying that all those affected would, at puberty, self identify as male, thus justifying bad surgery that was conducted of 5 alpha patients as children.

I feel that it is now important to explain that McGinley has only described cases within limited gene pools and has consistently done so. In that sense none of her material can be seen as universally relevant when dealing with a broader spectrum of affected individuals. To begin with in a limited gene pool, and working with Mendelian methodology, excludes the finer points that would be observed were polygenic expression taken into account. Also the term "Male" seems to be applied universally. You find numerous clinical guidelines, of the "intersex =social emergency" variety citing McGinley all the time to justify a male identification. Even the more respectable guidelines tend to use McGinley as a key reference, though one team who authored such a document did say themselves that McGinley was working with the methodology I am myself critical about.

The limit of this model is demonstrated purely by virtue of the description given in the first paragraph, on the molecular level.

"Using single strand DNA conformational polymorphism analysis and DNA sequencing, a new mutation in exon 5 of SRD5A2 gene was detected in certain male pseudohermaphrodites from this kindred. This single base deletion (adenine) resulted in a frame shift at amino acid position 251 resulting in the addition of 23 amino acids at the carboxyl-terminal of this 254-amino acid isozyme."

For the purpose of clarification, what is being stated is that a "stop" codon is read as a codon that codes for another amino acid due to it being shifted back by one base. (A-T being deleted). So the next stop codon occurs at a further 69 bases along the nucleotide.

So why are we talking about the carboxyl terminal, if it is two amino acids back from the carboxyl terminal in the peptide to begin with? How then were the additional amino acids "added from" the carboxyl terminal onwards. It does not make much sense. OK a grammatical error perhaps. We don't all read the UIPAC nomenclature guides 24 hours a day.

This lack of clarity makes me wonder. If the mutation does actually occur at the portion of the nucleotide that codes for position 251 as opposed to 253. This means what? that the cleavage site is where the mutation occurs? OK then, what restriction digests were used in the first place would be my immediate question, (To eliminate the possibility of an artefact). The point being that in order to add 23 amino acids to the resulting peptide chain, you would need to have deleted a specific type of codon in the nucleotide. This is implied, (By virtue of a stop codon being further along, accounting for the extra 23 residues in the resulting peptide) but seems unclear.

What gets me is the proximity of all these affected loci to the end or beginning of any given sequence, nucleotide or peptide.

"Transfection expression of the mutant isozyme in CV1 cells showed a complete loss of enzymatic activity in the conversion of [14C]testosterone to dihydrotestosterone, without a change in the messenger ribonucleic acid level compared to that of the wild-type isozyme."

Which is perhaps why I am toying with not so obvious ideas about restriction digests and artefacts.

It offers a possible answer by means of comparison, but does not cover the specific point. It is evident that, when quoting another study. And a more recent one, I hasten to add, things obviously are not quite as simple.

"A transition at the second nucleotide of codon 85 in exon 1 (GGC --> GAC - Quoted as DNA not mRNA) substituting glycine for aspartic acid, can occur"(Vilchis F, Mendez JP, Canto P, Lieberman E, Chavez B. 2000)

As with any sensible approach one has to look at alleles in a broader context than just one mutation, in one familial group, when considering things on a single gene basis, yet alone with respect to polygenic expression. While the many variants of AIS for example are covered by the AR mutations Database, covering a single AR gene. (Lenore Beitel 2000), it seems evident that 5 alpha reductase deficient conditions are not covered in quite the same way. However the polygenic element is allegedly taken into account.....

"Analysis of the 17Beta HSD3 gene in other male pseudohermaphrodites from this kindred revealed a single point mutation (G" border=0 src="JCEM -- Can et al_ 83) at the boundary between intron 8 and exon 9, disrupting the splice acceptor site of exon 9"

Not a dissimilar case scenario, just involving a different gene. Curious when you consider that this time a frame shift is not mentioned. But a mutation magically happens on a specific site - again. Suggesting perhaps an artefact of a restriction digest. Note the term "Boundary". And this is just a superficial observation of the document. Again my question still stands here: why is it always on a specific locus and as such codon of a sequence? At present I don't have the means to demonstrate the possibility of an artefact, but does this codon/locus specific anomaly occur even in a familial group? Well the possibility is there, but rather tentative. There is an interesting point here but not one that would justify such generalisation about 5 aRD.

OK so what have we here, two things I suppose. Firstly there is another gene at work with respect to the clinical presentations - bringing into question the universal idea of McGinley's model. This of course is one problem with working on the limited basis of a single familial group. Secondly 17Beta HSD3 is described as a point mutation, as with the other example I gave, meaning that we have the combined effect of a frame shift in SRD5A2 and a point mutation in 17Beta HSD3, being offered up as the root cause of the "classical" McGinley Phenotype. (Which is thus polygenic by the way)

The "complete loss of enzymic activity" as stated by McGinley et al implies, to the casual observer that this is the "worst case scenario" (the most "feminised" - more on that later) The scenario is that the individual starts life as phenotypical female, reaches puberty, spontaneously turns male and then becomes a "heterosexual, healthy young man" But this does not add up. While the scenario described does apply to some people with 5 alpha deficiencies, it does not automatically add up in all cases. And the phenotype/history is due to a complex polygenic expression, (which is hinted at) not just a frameshift in SRD5A2.

This is where I find myself feeling a little cautious about McGinley's work. I cannot work out in my mind how even when she has to admit that polygenic expression is taking place, she still makes a generalisation that runs through her entire theoretical model of 5 aRD people.

There is too much specificity in this for anyone to even begin trying to draw up a universal conclusion, even within the localised group she is working with. Also does she give us the data concerning one individual who didn't end up identifying as male? It seems too arbitrary and too set an agenda, almost as if she is trying to prove a specific point.

One sensible set of guidelines (Crieghton. S 2001) suggests that the patient be given a choice as to how they identify during puberty. Which is fine considering it was a surgeon that wrote them who has the patient's interest at heart, the problem of course is that, this may not be the sole clinical presentation of 5 alpha. However I know that certain less than reputable clinics will offer some form of masculinising surgery on atypical cases who present with partial virilisation during infancy.

The Dissenter?

What I find odd in the study is how at the very onset, there is a sense that McGinley et al seem determined to make the case for a "masculine bias". One of the affected participants of another study declined a physical examination. The discussion concerning this is quite enlightening.

"A-IV-3 is a 45-yr-old subject, born with ambiguous genitalia and raised as a female, although known in the community as an affected child. The subject did not change to a male gender role at puberty and remains the only known affected person not to do so in the community. A moderate amount of facial hair was present. Blood and urine samples were provided, but a physical examination was declined."

While it is acknowledged this time that the individual was indeed born with ambiguous genitalia, they declined further examination by McGinley et al beyond blood and urine samples. My question is why? Well lets look at the wording shall we?

"The subject did not change to a male gender role at puberty and remains the only known affected person not to do so in the community."

I get the impression that there was at some point a sense that they were expected to continue on to a male identity.

But in her assessment of the situation we find, (When presenting her "papers").

"As an example of how androgens can continue to influence gender identity through puberty, despite contrary social forces, Dr. Imperato-McGinley pointed to her work with people who have an inherited deficiency in 5a-reductase. This enzyme is necessary to convert testosterone into dihydrotestosterone. These individuals are born with male genes and male internal organs but lack external male genitalia.

In areas of the world in which cases have been observed without medical intervention, these male pseudohermaphrodites are declared female at birth and are raised as girls.

But at the time of puberty, they experience a surge in testosterone and gender confusion, said Dr. Imperato-McGinley, an endocrinologist and professor of medicine at the New York Hospital-Cornell Medical Center in New York.

They begin to 'feel like men,' are attracted to women, note testes in the inguinal canal or the labioscrotal folds, and have increased muscle mass and other signs of masculinization.

Among 18 such adults who she and her associates interviewed, 17 had converted unequivocally to having a male gender role identity by early adulthood." (see footnote)

So, by her criteria, all five alpha cases are heterosexual males. And while she mentions the evidence to bring her claim into question, it seems as if there is either reticence to acknowledge that evidence in full, or to make it appear as if the individual who contradicts her paradigm is somehow wary of examination.

The difficulty I have here is the need on someone's part to make a case for infant masculinisation where there is an ambiguous phenotype. Notice how we are guided very carefully to the conclusion that all "should" be male. I know that this would not be so, at least on this premise because there are tentative whisperings that DHT is the hormone that regulates apoptosis (prevents it) in the BSTc. (Swaab 2000)

Now the problem I can foresee if that proves to be the case, is that the lack of DHT is perhaps more likely to result in a female identity. Obviously I would be making generalisations myself were I to try to justify that, but what it does illustrate is that no one can predict "gender identity" on a limited study of this kind. (Which has been a problem with Swaab's research, but the point is, while it has been made clear by Swaab, the same cannot be said for McGinley)

It is very interesting to note that, while technically the problem of limited study groups applies to both McGinley's and Swaab's research, and, while McGinley does appear a bit sloppy, her work is readily accepted as a text book worthy study and taken to be some standard, while Swaab's work is not (with respect to gender identity). Given that Swaab indirectly (as he works with TS people not IS people) points towards a feminine identity and McGinley a masculine identity, are we witnessing the ignoring of science and the rantings of gender politics again?

Curious that!

The Dominican Republic.

Again McGinley seems to be obsessed with the idea that all people with 5 aRD are male. The term "Guevedoces" (meaning "Penis at twelve") is applied, and the common assumption from this, in my experience, has been that again people with 5 aRD are somehow spontaneously going to "change sex" at the age of 10, but also again to bring into focus my question, why a limited gene pool. Surely if the study was intended to represent all people with 5 aRD they would have at least tried to look further afield than simply concentrating on small (and possibly carefully selected or vetted) communities. For example, how many people in the West have been diagnosed with this condition, and have either been born with ambiguous genitalia and sought surgical intervention later on in life, to either sex, were surgically altered as children or who simply continued as women after puberty? I am deeply sceptical of McGinley's motives.

In the case of the "Guevedoces" we find photographs that are made to look as if they aim to present the same person and continually using the term "Boy" when it is patently obvious that the childhood photographs clearly show a phenotypic female. Also the adult "male" does seem to have somewhat smaller than usual genitalia for a man. It also appears evident that some surgery had taken place, and it was not just a case of "spontaneous" masculinisation.

The reality of course, veiled in implied rather than direct statement is that these were all individuals from the same family. But most striking is the way the "serially documented" photographs provide the onlooker with a clean cut picture of development. I suspect that each individual photographed would vary considerably from the implied path of development as described by these images.

In conclusion I would be cautious about drawing any conclusions on the studies conducted by McGinley, given that there is also a bias towards maintaining a paradigm, not just about phenotypic expression but "psychosexual" orientation. In fact this part is the part I am most suspicious of. Having the condition myself, and having a pathological hatred of being even thought of as male does somewhat undermine McGinley's "vision" of how people with 5 aRD "should" be. Masculinisation by proxy is not the same as masculinisation by nature. And as a consequence of my stating this I will say that while I do find McGinley's studies personally offensive, and her assumptions somewhat repulsive, I can only say from an objective standpoint that I find her work lacking credibility from a simple set of observations I made of her studies that appear to show an endocrinologist too eager to "define" those people she has studied as male. And no account whatsoever was given to the possibility that while there are people with 5 aRD individuals who may identify and develop into men, they may also be those who otherwise may develop into a number of gender expressions and phenotypic forms.

Also there is some strange artefact-like quality to some of the lab results McGinley quotes or obtains for herself. The extended peptide as presented in one example shows clearly that a deletion, at a cleavage site or a boundary seems to be a common factor. Why I ask is this the case? Also the total non-functioning of an enzyme does not say much. It merely implies a "worst case scenario". The jury is still out on this because no one truly knows what different mutations at different points may produce. You could have any product from an altered enzyme acting on a given steroid. Rather than simply "no product" at all. Then there is the polygenic element. What expressions of other genes will affect the outcome? As for gender identity - well, despite this set of conclusions on McGinley's part being based on a bad set of statistical rules (limited gene pools, ignoring individuals who don't fit etc. etc.), the medical profession still hangs on to the "Male" bias with some degree of conviction. Why?

My real problem with this is two fold, and it has been sailing close to the wind. I have recently read statements about the clinical management of 5 alpha people which say "5 alpha people identify as male 100% of the time in the absence of childhood feminising surgery" followed very rapidly with "Phalloplasty is very good these days". Also if the McGinley phenotype develops a fully grown penis at the age of ten, what purpose would earlier attempts at phalloplasty serve? If anything it would cause more complications surely. Which can lead to two very serious problems:

1 That infant genital surgery to masculinise a 5 alpha child is more likely to be considered, which is ethically no different from surgically feminising a child. It is still risky, it is still fraught with clinical complications, it is still a cosmetic "fix" and the individual may grow up feeling harmed by it.

2 That the treatment of adults wishing to have feminising surgery may find their choices limited by this assumption that they want to be male. I actually fear the HBSOC being applied in cases like this. Where those identifying as male may well be given surgery upon the initial request. A very restrictive practice indeed.

The reality is that the individual who has this condition should have the right to exercise all choices that are available to them, and not have them restricted by false assumptions based on arbitrary research that in reality carries little weight. Which leaves me asking the difficult question "What was McGinley trying to do with her initial research?" The answer I suspect was to resolve the idea that as a "socio-medical" emergency an intersex child must be sexed. This study therefore was to identify gender identity rather than the actual health of people with 5 alpha. As such this study, which has found its way into most medical textbooks urgently needs to be reviewed and critically studied again - if not repeated. Because if it is not I can foresee the potential for some clinics to try to cosmetically "sex" a child who is 5 alpha without considering the consequences, basically putting everyone back to square one when it comes to trying to find an alternative to the current situation of sexing children - the only difference being the sex chosen, this achieves little or nothing.

This is a situation where orthodoxy is playing some very questionable games with the lives of affected individuals. This is really a good example of needing to remove gender politics from the clinical arena, simply because this illustrates very clearly that gender politics and medicine do not mix in situations like this. An intersexed individual is not a "pychosocial" or "sociomedical" case, but an individual human being who deserves the right to control their own body, not to have their body controlled by feminists and phallocrats.

As for the genetics, I am stuck in this rut of reading the McGinley's "locii effect" each time I look at the data I am presented with from the archives carrying her work. Why do most of her observations, from all these small gene pools end up presenting us all with a mutation at or near the 3' or 5' of the nucleotide sequences or at the Amino or Carboxy ends of the resulting peptides? Either she is looking at one gene pool that is spread around the world (improbable) is tracking a given set of alleles carrying the same mutations across these small gene pools (more probable) or is having a bit of a problem in the lab (also more probable).

But even were the second probability to be true, it does not follow that all cases of 5 alpha reductase deficiency are down to these alleles. I don't profess to know myself what the truth actually is, but even with my limited resources and understanding in some areas, I smell a rat with "Gender agenda" written on it's back.

Footnote: For those who are unfamiliar with McGinley's work, her study of children of the Dominican Republic, members of an extended family, is where she made her name and is quoted in every school psychology textbook. Nevertheless, the study has been severely criticised by several authors, notably Gilbert Herdt. There is a brief account at "Imperato-McGinley"

Other articles by Sophia Siedlberg:

Web page copyright GENDYS Network. Text copyright of the author. Last amended 16.01.02