This review aims to report current understanding of paracellular calcium and magnesium reabsorption from the proximal tubule and thick ascending limb. This is an interesting review and the paper is well written.
Nevertheless, a number of concerns were identified that should be addressed.

Comments and concerns are listed below:
Page 3,
-Line 17 : Bone magnesium does not play a major role in the defense of blood magnesium concentration.
-Line 18: What are the arguments to say that blood calcium concentration is subject to tighter control than magnesium concentration ? Extracellular magnesium concentration is also tightly controlled (even if the hormone(s) regulating extracellular magnesium concentration remains unknown).
- Figure 1: The % of filtered calcium and magnesium reabsorbed in each tubular segment are different in the text and in the figure.
- Line 47: Remove “both”
- Line 49: Replace regulated by controlled?
- As it is written for Ca and Mg page 3, it could be understood that only their reabsorption in the DCT/CNT is controlled. However, paracellular reabsorption in the TAL is also controlled.

Page 4
-Line 27-40: As written, it is not understood that the transepithelial voltage depends on the diffusion potential of the apical and the basolateral membranes.

Page 6
- Line 22: Could you provide more details concerning the statement of a “significantly reduced permeability to Mg as compared to Ca” ? Did the authors directly compare PMg and PCa to assess that PMg is significantly lower than PCa? If so, what were the values and the ratio?
- Line 29: “we are unaware of epithelial cell models examining PMg”: Cldn2 was recently overexpressed in MDCK C7 cells and Mg relative to chloride permeability ratio was studied. (Breiderhoff, T. et al. 2022. Claudin-10a Deficiency Shifts Proximal Tubular Cl(-) Permeability to Cation Selectivity via Claudin-2 Redistribution. J Am Soc Nephrol.)
- Line 32-34: Could you provide references corresponding to the statement “moreover Cldn2 and Cldn12 single KO mouse models do not have hypermagnesesuria, nor altered fecal Mg content... Further, Cldn2 and Cldn12 double knockout mice do not have altered urinary or fecal magnesium excretion on a standard rodent diet"
According to the cited references :
- urinary Mg excretion nor fecal Mg excretion were not studied in the cldn12 KO model
- fecal mg content was not measured in the cldn2 KO model
- fecal mg content was not measured in the cldn2-cldn12 double KO model.
- Line 38: “the absence of hypomagnesuria in the cldn2 and cldn12 KO strains is consistent with cldn2 and cldn12 being the primary cldn responsible for mg reabsorption from this segment”. Do you mean the absence of hypermagnesuria….”?
- Line 48-52: According to reference 34, cldn13 is not detectable in neonatal or adult proximal convoluted tubules. According to that, information and discussion should be corrected.
- Factors known to affect divalent cations reabsorption in the proximal tubule should be discussed.

page 7, line 51-52: “the higher lumen positive Vte in CTAL are best observed when tubules are microperfused at low flow rates….rather than alterations of active transport” : could you provide references? Was Vte compared in the same study with different flow rates? with the same tubule length ? Was NaCl concentration similar in the bath and in the lumen or was it lower in the lumen than in the bath ?

Page 8
- Line 13: Calciuria is variable in Bartter syndrome resulting from CLCNKB pathogenic variants (Konrad M. et al, Kidney international 2021). Maybe you wanted to say …. “And hypercalciuria and nephrocalcinosis are only infrequent observed”.
Low urinary calcium excretion has also been reported in patients with Bartter Syndrome 3, and these patients mimic the Gitelman syndrome phenotype. It should be precised that ClC-Kb is also expressed in the DCT.
In addition the classification with 5 types of Bartter syndrome should be used (Konrad M. et al, Kidney international 2021 PMID: 33509356 , DOI: 10.1016/j.kint.2020.10.035).
- Line 28-40: Could you provide references that have directly compared measurements of TER and/or of paracellular ion permeability in ISOM and CTAL to assess reported differences ? ref 58 is a review. Milatz and al compared 3 ISOM TAL to CTAL/OSOM TAL (Milatz, S.,2017. Proc. Natl. Acad. Sci. U.S.A. 114: E219–E227) and Plain et al did not compare statistically the results obtained in ISOM TAL and CTAL (Plain, A., et al. 2016. Pflugers Arch. 468: 293–303).
-Line 19- 46: Luminal ion concentrations in the early DCT, just downstream the late cortical TAL have been measured during micropuncture experiments. Luminal ion concentrations in the early medullary TAL can be estimated based on a calcium and magnesium reabsorption equaling 20–25% and 60–70% of filtered load, respectively and the lack of substantial water reabsorption in the TAL. Could you discuss this point and compare results obtained with this estimate to results already reported in the review ?
- Line 47-50: Considering that the bulk of calcium and magnesium transport occurs in the cortical TAL, a discussion regarding the concentration of Ca and Mg in the cortical interstitial fluid should be added.

Page 10
- Line 12: Nephrocalcinosis has not been shown to be the cause of the renal failure.
- Line 44-46 : “this finding are in line with cldn16 and 19 contributing permeability to divalent cations”. However, in the study conducted by Milatz et al, the association concerns the expression of cldn16 and PMg/PNa. PMg is not correlated with cldn10b or cldn16 expression and PNa decreases with the reduction of cldn10b. These results should be discussed.
- Other authors have reported that Cldn16/Cldn19 would be a non-specific cation channel, having an indirect effect on divalent cations reabsorption by decreasing PNa/PCl, thereby reducing the diffusion potential/the transepithelial voltage. This part of literature should be discussed.

Page 11
- Line 9: The type of effect on PCa and PMg should be described.
- Line 10- 13: Individual paracellular ion permeabilities were not measured. There were only ratios that need to be discussed.
- Line 13: A decrease in PNa should not increase Vte when the concentration of NaCl is lower in the lumen than in the interstitium at the end of CTAL. What about the role of active transcellular transport of NaCl on Vte ?
- Line 16: HELIX patients don’t have nephrocalcinosis, this should be discussed as the phenotype of HELIX patients.
- Line 18: Individual paracellular permeabilities to ions were not measured. There were only ratios that should be discussed. PMg/PNa is higher in Cldn10 CLdn16 double KO TAL compared to WT TAL.
- Line 22-24: literature regarding the effect of PTH should be more detailed.
- Line 42-44: What is the demonstration that the effect of extracellular calcium on calcium and magnesium reabsorption involves CaSR ?
- Line 44-45: The effect of calcium on the activity of the apical potassium channel should also be discussed.

Page 13
- Line 12-14: Could you provide data from studies that have measured calcium and magnesium concentrations in glomerular ultrafiltrate ?
- Line 15-16: “the driving forces for divalent cations reabsorption from the proximal tubule would otherwise be the same “: Do you mean that the chemical gradient is the same for Ca and Mg in the early proximal tubule (because Vte is the same for both ion)? The factor ([X]l - [X]int exp(-ζ)) involved in ion fluxes can be calculated and discussed (where ζ = VteZF/RT ; [X]l = [ion] in the lumen; [X]int = interstitial ion concentration )

Other:
Figure 2 and 3 are not mentioned in the text
Why paracellular Na fluxes are not shown in figure 2 and 3 ? (but only transcellular ?)
Page 6 line 48: Clnd should be replaced by Cldn
Page 8 line 8: KCNJ1 should be replaced by KCNJ1
Replace mutation by pathogenic variant
Page 8 line 38: stipe should be replaced by stripe
Page 9 line 20: delete one “the” : “reaches the the TAL”… “
Page 9 : line 22 add a space: “this segment”
Page 9 line 34 : add a space: “has been”
Page 13 line 24: add a space: “assess magnesium”
Page 14 line 21: remove .,

The authors have responded to the majority of queries. The following revisions should be considered:

In the abstract, line 11: Free ionized calcium and magnesium is filtered at the glomerulus.
Calcium and magnesium complexed to anions are also ultrafilterable.

Page 4: The appearance of a lumen positive potential occurs as a consequence of the preferential reabsorption of bicarbonate along the early proximal tubule by processes involving the sodium hydrogen exchanger isoform 3 (NHE3), the apical and cytosolic carbonic anhydrases, and sodium bicarbonate cotransporter 1 (NBCe1), which leads to the concentration of chloride in the tubular lumen, since HCO3 - is the primary counterion moving with sodium in the early portions of the proximal tubule. Chloride is then reabsorbed via a paracellular pathway.
Relative permeabilities to anions and cations should be discussed to explain the transepithelial voltage.

Page 6 line 44: In fact, the absence of hypomagnesuria in the Cldn2 knockout mice is consistent with CLDN2 being a primary claudin responsible for magnesium reabsorption from this segment, as it would be relatively easy given the small amount of magnesium reabsorbed from the proximal tubule for more distal segments to compensate.
If Cldn 2 increases PMg, one would expect hypermagnesuria (not hypomagnesuria) when Cldn2 is lacking.

Page 7 line 7: The relative expression of Cldn2 and Cldn12 between neonatal and adult renal tissue has not been reported, but it is tempting to speculate that increased expression of these cation permeable claudins may contribute greater magnesium permeability to the proximal tubule early in life.
Aubuazza et al found no difference in claudin 2 and 12 mRNA expression in 1-day-old neonatal and adult proximal convoluted tubules (Abuazza, G. et al. 2006. American journal of physiology. Renal physiology. 291: F1132-1141).

Page 11: However, these animals also suffer from hyperaldosteronism, which differs from both Cldn16 deficient mice (where Cldn16 is removed by targeted deletion) and FHHNC patients.
Aldosterone concentration was not measured in Cldn16 KO mice.

Page 12: SiRNA knockdown of Cldn19 in mice leads to renal wasting of calcium and magnesium. However, hyperaldosteronism is also observed, as with the Cldn16 SiRNA knockdown model and in contrast to Cldn16 deficient animals and FHHNC patients.
Aldosterone concentration was not measured in Cldn16 KO mice.

page 12 line 19: double deletion of Cldn16 and Cldn10 reduces PCa/PNa in the TAL to a level similar to the control animals, while PMg/PNa is attenuated in comparison with Cldn10 deficient animals.
PCa/PNa is also lower in dKO CTAL than in C10 KO CTAL. PMg/PNa is higher in dKO CTAL compared to control animals.
To clarify, comparaisons to control animals and single KO TAL could be discussed for both PCa/PNa and PMg/PNa.

Other:

Page 4: CLDN2 and -CLDN12 confer cation permeability to this segment, while claudin-10a confers anion permeability.
-CLDN12 should be replaced by CLDN12

Page 6 line 56 Clnd9 should be replaced by Cldn

Page 7 line 32
Interestingly, the bulk of paracellular calcium and magnesium transport occurs in the more cortical portions of the TAL, with limited divalent cation reabsorption from especially the inner stipe of outer medulla (ISOM) TAL epithelium
inner stipe should be replaced by inner stripe.

Page 8 line 23: The antenatal forms of Bartter’s syndrome occur due to mutations in NKCC2 and ROMK encoded by the SLC12A1 and KCNJ1 genes and are characterized by marked calcium wasting and nephrocalcinosis
KCNJ1 should be replaced by KCNJ1 (in italics)

page 8: resulting from mutations in the CLCNKB gene, often develop hypomagnesemia whereas hypercalciuria and nephrocalcinosis is only infrequently observed.
“is” should be replaced by are.