Waye Young:
The major mechanism that causes the development of an axial osmotic gradient in the outer medulla is the active transport of Na and Cl by the thick ascending limb of the loop of Henle. The transporter that mediates this is the NKCC2, present in the apical membrane of the cells of thick ascending limb. Inhibition of this transporter is the mechanism underlying the potent diuretic effects of furosemide (aka frusemide).
In the inner medulla, the accumulation of urea in the interstitium contributes substantially to the osmolality of the inner medullary interstitium. You are right - urea permeability of the medullary collecting duct is ADH dependent. Urea cycles from out of the inner medullary collecting duct into the medullary interstitium and accumulates there. The thin ascending limb of the loop of Henle is permeable to urea. Urea that passively enters the loop of Henle from the medullary interstitium remains in the tubular lumen until it gets to the inner medullary collecting duct. From studies of urea transporter gene knockout studies, there is
evidence that urea accumulation in the medullary interstitium
contributes to the antidiuresis that can be achieved, especially under
conditions of fluid deprivation. Also, in mice with urea transporters
knocked out and placed on a high protein diet, the increased delivery of
urea to the inner medullary collecting duct results in a urea mediated
osmotic diuresis.
The fluid entering the descending limb of the loop of Henle is iso-osmotic (i.e. with respect to plasma). 25% of filtered Na and Cl is reabsorbed in the loop of Henle, whereas only 15% of the filtered water load is reabsorbed from the loop of Henle. This is why fluid entering the distal tubule is always hypotonic, no matter what the final osmolality of urine is. The final osmolality of urine is determined in the collecting duct and depends on the level of activation of water reabsorption by vasopressin.
To specifically answer your question, furosemide inhibition of the NKCC2 in the thick ascending limb abolishes the hyperosmolality normally developed there. Even if ADH were present and acted upon the collecting ducts, a polyuria would nevertheless result because the hypertonicity that drives water reabsorption from the collecting ducts would be lacking. The polyuria is considerable because 25% of the filtered NaCl is normally reabsorbed in the thick ascending limb and if this is inhibited - downstream segments of the nephron cannot compensate for it. Furthermore, the macula densa also reabsorbs Na and Cl via the NKCC2 - the transporter that furosemide inhibits. Thus, furosemide also shuts tubuloglomerular feedback down, and adaptive decrements in GFR in response to polyuria are inhibited. This is why loop diuretics are 'high ceiling' diuretics.
ADH is reported to have a chronic stimulatory effect on the NKCC2 and other transport mechanisms in the thick ascending limb as well [Ch 9 in Brenner and Rector's The Kidney, 2008] but I haven't come across anything specific as to how much ADH deficiency or ADH receptor mutations affects this process per se; presumably the acute antidiuretic effects of ADH with regard to water balance are largely accounted for by its actions in the collecting ducts both in humans and experimental animals.
You suggest failure of the vasa recta as a possible mechanism for loss of the gradient. That is correct. Normally, the very low blood flow in the inner medulla and countercurrent exchange of solutes between the ascending and descending limbs of the vasa recta allows the hypertonicity in the inner medulla to be maintained.
I can't however see how inhibiting the NKCC2 in the TAL would
inhibit water and or solute absorption in the PCT. In a state that results in profound diuresis, typically reactive neurohumoral mechanisms operate to enhance solute and fluid recovery from the PCT and this may ameliorate the polyuria to some extent. Profound diuresis reflexly activates the renin-angiotensin system and renal sympathetic outflow by inducing hypovolemia. For example, angiotensin II enhances Na, Cl and H20 reabsorption from the PCT. This mechanism was postulated to explain the paradoxical antidiuretic effect of thiazide diuretics given to individuals with diabetes insipidus, but other mechanisms possibly operate [see
http://jasn.asnjournals.org/content/15/11/2948.full].
Prakash