To understand how hypospadias arises, one has to start from how the normal urethra is built. What is striking is that the account we now give as a matter of course was, until fairly recently, understood quite differently.
Rather than describe the process again, this piece follows which studies changed the prevailing view. The process itself is set out in Part 4 of the clinical series.
1954 — Glenister: the glanular urethra grows inward from the skin
Glenister TW. The origin and fate of the urethral plate in man. J Anat. 1954;88(3):413-425.
From serial sections of human fetuses, Glenister proposed that the urethra of the penile shaft forms from the urethral plate, while the distal urethra within the glans is formed by an ingrowth of surface ectoderm — the ectodermal ingrowth theory.
What this paper changed
It was a model grounded in human fetal anatomy that separated the origin of the shaft urethra from that of the glanular urethra. Generations of textbooks carried the distinction forward, and the statement that the glanular urethra is ectodermal still turns up in relatively recent writing.
What it could not show
Morphology alone, in the 1950s, could not establish cell lineage. The step from the distal urethral epithelium looks like skin to it came from skin was an inference, and there was then no tool with which to test it.
1999 — Kurzrock: squamous does not mean ectodermal
Kurzrock EA, Baskin LS, Cunha GR. Ontogeny of the male urethra: theory of endodermal differentiation. Differentiation. 1999;64(2):115-122.
This is where the direction changes. Cytokeratin immunohistochemistry was applied to 36 human fetal phallic specimens of 5–22 weeks' gestation. The urethral plate, an extension of the urogenital sinus, ran to the tip of the phallus and remained patent and continuous throughout. No evidence of a solid ectodermal ingrowth was seen.
The authors proposed instead that endodermal urethral epithelium differentiates into a stratified squamous phenotype during development — and did not leave the proposal at the level of observation. Recombining mouse fetal bladder epithelium with rat fetal genital tubercle mesenchyme under the renal capsule of athymic mice, the urothelium differentiated into stratified squamous epithelium. Direct experimental evidence that mesenchymal signalling can change epithelial phenotype.
What this paper changed
It broke the inferential link from "the glanular urethra is squamous" to "therefore it came from ectoderm". Epithelial appearance may be a product of differentiation rather than a marker of origin.
What it left standing
The paper still held that the entire urethra, glans included, is formed by extension and disintegration of the urethral plate together with ventral fusion of the urethral folds all the way to the tip. The origin had changed to endoderm; the mechanism was still fusion throughout. That part would be revised twenty years later, by the same laboratory.
2014 — Hadidi: look again at human fetuses and neither model suffices
Hadidi AT, Roessler J, Coerdt W. Development of the human male urethra: a histochemical study on human embryos. J Pediatr Surg. 2014;49(7):1146-1152.
Serial histological sections of 15 human embryos of 6–14 weeks. The spongy urethra was described as passing through three stages — solid epithelial plate, deep urethral groove, fused urethra — and the glanular urethra through four: solid epithelial plate, blind central canal, deep glanular groove, and a floor derived from the preputial lamella. Again, no ectodermal ingrowth was observed.
One incidental observation is worth keeping: at 14 weeks the urethral opening had migrated only to the middle of the shaft. The distal part is completed last, and late — which matches the clinical impression.
How to read it carefully
Hadidi's model of the glans is not the same as the direct-canalization model that follows from the UCSF group. He held that the glanular urethra first forms a blind canal, whose floor then opens into a groove, and that this floor is completed by the preputial lamella. Even in 2014 there was no settled account of how the glanular urethra is completed. The value of the paper lies less in establishing a new theory than in showing again that neither the simple fusion model nor the ingrowth model accounts for human urethral development.
2015 — Li: an opening zipper and a closing zipper
Li Y, Sinclair A, Cao M, Shen J, Choudhry S, Botta S, Cunha G, Baskin L. Canalization of the urethral plate precedes fusion of the urethral folds during male penile urethral development: the double zipper hypothesis. J Urol. 2015;193(4):1353-1359.
Human fetal specimens of 6.5–16.5 weeks were examined by optical projection tomography and serial sections. The solid urethral plate first canalizes distally into an open groove with two epithelial edges (the opening zipper), and those two epithelial surfaces then fuse behind it to form the tube (the closing zipper). The whole sequence took about ten weeks.
Proliferation and apoptosis staining carried the argument. Ki67 was localized to the dorsal aspect of the urethral plate, and caspase 3 was not observed. The groove opens by proliferation and rearrangement rather than by cell loss — a departure from earlier accounts that assumed dissolution of the urethral plate.
What this paper changed
The urethra does not simply arise from two walls meeting: it opens first, and closes behind. This is the point at which "fusion failure" stops being an adequate one-word account of hypospadias. The two zippers are different processes, and they can be disturbed separately, at different moments.
2018 — the same thing does not happen in the glans
Liu X, Liu G, Shen J, et al. Human glans and preputial development. Differentiation. 2018;103:86-99. · Liu G, Liu X, Shen J, Sinclair A, Baskin L, Cunha GR. Contrasting mechanisms of penile urethral formation in mouse and human. Differentiation. 2018;101:46-64.
A follow-up study of 48 normal human fetal penile specimens by scanning electron microscopy, optical projection tomography and serial sections. As the closing zipper completes the tube within the shaft (around 13 weeks), canalization of the urethral plate continues proximal-to-distal into the glans. But in the glans no open groove is formed. A lumen arises directly within the solid plate and advances toward the tip — direct canalization.
Through this, the urethral plate is initially attached to the ventral epidermis by an epithelial seam; the seam is remodelled and eliminated, establishing mesenchymal confluence ventral to the glanular urethra. The most ventral urethral plate cells are pinched off from the urethra, reabsorbed into the epidermis, and lose their markers. Cytokeratin 7, FoxA1 and uroplakin — all markers of endodermal epithelium — are expressed strategically during that remodelling, which the authors took to be androgen-regulated.
The prepuce first forms on the dorsal aspect of the glans at about 12 weeks, and only fuses in the ventral midline after the ventral urethral plate has been remodelled from proximal to distal. Here is the embryological basis for the clinical observation that ventral preputial closure comes last.
The companion paper from the same year points out that mouse and human are opposites. In the human, the glans forms by direct canalization and the shaft by fusion; in the mouse, the proximal urethra forms by direct canalization and the distal urethra by fusion. The authors note that the highest incidence of hypospadias in humans occurs at the junction of these two mechanisms.
It is also a reason for caution in reading mouse distal urethral morphogenesis straight across to the human.
What changed over sixty years
| Study | The prevailing idea at the time | What changed afterwards |
|---|---|---|
| Glenister, 1954 | Shaft from the urethral plate; distal glans by ectodermal ingrowth | The ectodermal origin of the distal urethra came into question |
| Kurzrock, 1999 | A continuous endodermal urethral plate running to the tip | Origin revised to endoderm — but the mechanism was still fusion throughout |
| Hadidi, 2014 | Human fetal morphology re-examined: three stages in the shaft, four in the glans | Limits of the simple fusion and ingrowth models; the glans still unsettled |
| Li, 2015 | Opening zipper plus closing zipper | Shaft morphogenesis separated into two sequential processes |
| Liu · Baskin · Cunha, 2018 | Glans by direct canalization; prepuce dorsal first, ventral last | Shaft and glans mechanisms separated; mouse and human patterns are reversed |
What I take from these papers
The important thing, to my mind, is not the name "double zipper". It is that the urethra is not built by a single mechanism.
In the shaft, a solid plate opens into a groove and fusion follows behind it. In the glans, a lumen forms directly within the plate, with no open groove at all. Ventral closure of the prepuce is interlocked in time with both, and comes last.
So the displaced meatus, the ventral skin deficiency, the dorsal hood and some of the curvature seen in hypospadias are better read not as four independent problems but as several developmental processes running in the same window and affected to different degrees. The tendency for findings to travel together as the meatus moves proximally, and conversely the relatively intact prepuce seen in some glanular cases, both sit within that frame. Curvature is the partial exception: skin, dartos, the urethral plate and corporal disproportion can each contribute, so it does not reduce to this frame alone.
One more thing. These sixty years also show how long it takes to change one sentence in a textbook. Ectodermal ingrowth was effectively refuted in 1999, and remained in textbooks and reviews for a long time afterwards. Some of what we now state as a matter of course is likely to be untested inference as well.
Hypospadias, Part 4 — How the Urethra Is Built
If this piece is about why we came to think as we do, Part 4 is about what we now say.
Hypospadias, Part 5 — Is It Really Increasing, and Why Does It Happen?
Part 5 takes up how the limits of the mouse model bear on aetiological research.
References
- Glenister TW. The origin and fate of the urethral plate in man. J Anat. 1954;88(3):413-425.
- Kurzrock EA, Baskin LS, Cunha GR. Ontogeny of the male urethra: theory of endodermal differentiation. Differentiation. 1999;64(2):115-122. doi:10.1046/j.1432-0436.1999.6420115.x
- Hadidi AT, Roessler J, Coerdt W. Development of the human male urethra: a histochemical study on human embryos. J Pediatr Surg. 2014;49(7):1146-1152. doi:10.1016/j.jpedsurg.2014.01.009
- Li Y, Sinclair A, Cao M, Shen J, Choudhry S, Botta S, Cunha G, Baskin L. Canalization of the urethral plate precedes fusion of the urethral folds during male penile urethral development: the double zipper hypothesis. J Urol. 2015;193(4):1353-1359. doi:10.1016/j.juro.2014.09.108
- Liu X, Liu G, Shen J, Yue A, Isaacson D, Sinclair A, Cao M, Liaw A, Cunha GR, Baskin L. Human glans and preputial development. Differentiation. 2018;103:86-99. doi:10.1016/j.diff.2018.08.002
- Liu G, Liu X, Shen J, Sinclair A, Baskin L, Cunha GR. Contrasting mechanisms of penile urethral formation in mouse and human. Differentiation. 2018;101:46-64. doi:10.1016/j.diff.2018.05.001
- Shen J, Cunha GR, Sinclair A, Cao M, Isaacson D, Baskin L. Macroscopic whole-mounts of the developing human fetal urogenital-genital tract: indifferent stage to male and female differentiation. Differentiation. 2018;103:5-13. doi:10.1016/j.diff.2018.08.003