Intercalated cells have been traditionally viewed as being important for the control of systemic acid-base homeostasis by excreting acid or base equivalents 
. However, in more recent years several links to the renal control of salt and electrolyte handling have emerged. Secretion of potassium by flow-activated maxiK channels expressed in intercalated cells has been observed 
. The identification of pendrin as major apical anion transporter in non-type A intercalated cells has linked this cell type to the reabsorption of chloride and its impact on blood pressure regulation 
. More recent work by Eladari and colleagues identified a Na+
-driven chloride/bicarbonate transporter, Slc4a8, that may in conjunction with pendrin mediate electroneutral NaCl reabsorption. Interestingly, the activity of Slc4a8 is also thiazide sensitive 
Here we demonstrate that expression of AE1 and pendrin is sensitive to dietary electrolyte intake and the aldosterone analogue DOCA. Several observations are important. First, regulation on mRNA and protein level do not always go in parallel, sometimes even in opposite directions calling for caution in drawing conclusions about protein expression from mRNA data. Of note, we did not determine transport activities in the present study, thus, our data on protein and cell abundances do not necessarily translate into changes in activity. However, the data on blood and urine composition indicate parallel changes suggesting that protein abundance and function may be closely linked. Second, AE1 mRNA and protein expression in cortex was unchanged in all groups whereas protein expression in kidney medulla was strongly regulated by NH4
/DOCA, and NaCl. Third, regulation of AE1 in the medulla was only observed if NaHCO3
was combined with DOCA, but not in the groups receiving DOCA, NaHCO3
alone. These data therefore suggest, that alkali treatment, sodium, potassium or DOCA alone do not affect AE1 protein expression. DOCA, or its natural analogue aldosterone, may therefore have a permissive effect on AE1 expression when combined with alkalosis. This effect may be of clinical relevance in patients with stimulated aldosterone levels and alkalosis as seen in volume contracted patients and would consequently contribute to maintenance of alkalosis or even enhance it. The increase in AE1 mRNA in medulla in NaHCO3
treated animals did not translate into more protein. NH4
Cl increased AE1 protein but not mRNA abundance in medulla. Consistently, we did not detect significant AE1 mRNA changes in a genome-wide transcriptome analysis of mouse kidney during acidosis in earlier experiments 
. Though, in rat kidney, AE1 is regulated by NH4
Cl both on mRNA and protein level 
. The reason for this species difference is presently unknown.
Our study further showed that AE1 is not regulated by dietary chloride intake, since KCl failed to change AE1 abundance. On the other hand, NaCl tended to increase AE1 mRNA and stimulated protein expression in kidney medulla. Since NaHCO3
alone had no effect on AE1 mRNA or protein expression, we reason that sodium delivery alone does not regulate AE1. NaCl availability has been shown to enhance the kidneys ability to excrete acid 
. Increased expression of AE1 under these conditions could contribute to removal of acid and generation of new bicarbonate.
Pendrin regulation by aldosterone and various electrolytes has been studied in some detail 
. Our results here confirm regulation of pendrin by chloride availability. Dietary supplementation with NH4
Cl, NaCl or KCl increased urinary chloride excretion and reduced pendrin protein abundance. Interestingly, these treatments had different effects on pendrin mRNA expression. KCl reduced mRNA levels, whereas NaCl increased it, and NH4
Cl had no effect. Thus, chloride may not directly affect pendrin mRNA but additional factors must be operative that regulate pendrin transcription or mRNA stability.
Mice were treated with bicarbonate in combination with either sodium or potassium as counter cations and in the absence or presence of DOCA, respectively. Bicarbonate alone did not regulate pendrin mRNA or protein over a period of 7 days since NaHCO3
alone did affect neither mRNA nor protein, whereas KHCO3
in combination with DOCA increased mRNA abundance. Only the combination of NaHCO3
/DOCA increased also pendrin protein expression. DOCA alone was without effect on mRNA and protein. Thus, the combination of aldosterone and bicarbonate is a potent stimulus to enhance pendrin expression. In the case of KHCO3
, the potassium load may increase serum aldosterone levels but to a lesser extent than in animals with DOCA administration. Our results are not in agreement with two previous reports. Verlander et al.
reported that DOCP alone enhanced pendrin expression but mice had received additional NaCl supplements and only one application of DOCP seven days before organ collection 
. We speculate that the additional NaCl load and the different application regimen may alter the response to the aldosterone analogue. Alternatively, the effect of DOCA or its analogues may be time-dependent and transient and only detectable after short treatments but not after 7 days 
. A second study addressed regulation of human pendrin transcription by aldosterone using HEK293 cells as cellular model. These cells are derived from human embryonic kidney and are not representative for collecting duct intercalated cells. In these cells, aldosterone suppressed pendrin transcription 
, while our results suggest that DOCA alone has no effect on pendrin mRNA abundance and that a second stimulus, i.e. bicarbonate, is needed to enhance pendrin mRNA. Thus, transcriptional regulation of pendrin differs between cell models and in vivo 
, which may be the result of the interaction of several factors affecting the collecting duct, and not all transcriptional changes translate into altered protein abundance demonstrating an additional level of control.
Remodeling of the collecting duct occurs in response to changes in electrolyte and/or acid-base status 
. Induction of acidosis with NH4
Cl has been shown to increase the relative number of type A intercalated cells in various rat and mouse models. AE1 protein expression increased during NH4
Cl only in the medulla but not in cortex. In contrast, we found a higher relative number of AE1 expressing cells in both cortex and medulla during NH4
Cl-loading suggesting that in the cortex the abundance of AE1 per cell may be stimulated. Pendrin abundance in the cortex was reduced during NH4
Cl loading and the relative number of pendrin positive cells decreased in parallel. However in KCl-loaded animals, pendrin protein abundance decreased without a significant chance in the relative number of pendrin expressing cells consistent with a possible reduction of pendrin expression per cell. In both cases, the cellular mechanisms responsible for sensing changes in acid-base status or chloride concentrations are unknown. Also the mechanisms triggering the reduction in protein expression or cell number have not been identified to date. Several mechanisms have been suggested to be involved in the remodeling of the collecting duct in response to NH4
Cl-induced acidosis including proliferation of type A intercalated cells and hensin-mediated interconversion of type A and non-A intercalated cells 
In summary, the abundance of the two major anion exchangers AE1 and pendrin expressed in the different subtypes of intercalated cells is heavily regulated by electrolyte intake, acid-base status, and the aldosterone analogue. In some instances the presence of two combined stimuli is required for regulation. Apparently, regulation involves changes in protein expression per cell as well as changes in the relative abundance of the two major intercalated cell subtypes. The adaptive regulation of the type A intercalated cell anion exchanger AE1 mainly follows changes in acid-base status whereas regulation of pendrin expression paralleled mostly chloride intake and acid-base status. Our data demonstrate differential regulation of these transporters and the two intercalated subtypes. The altered expression may support the renal response to changes in acid-base status and electrolyte intake but may also contribute in some instances to the maintenance of electrolyte and acid-base disturbances.