Role of Calbindin-D28k in Diabetes-Associated Advanced Glycation End-Products-Induced Renal Proximal Tubule Cell Injury

Round 1

Reviewer 1 Report

Review ofmanuscript No. cells-517752 entitled „Role of calbindin-D28k in diabetes-associated advanced glycation end-products-induced renal proximal tubule cell injury”

„Abstract”

This study was conducted using animal model, so conclusion that ” Calbindin-D28k may serve as a marker for proximal tubule injury/fibrosis in diabetic nephropathy.” Is somewhat a bit too significant.

In section „Introduction” the statemnat that „… (AGEs) are generally considered as the poisons for diabetic patients.” is too general, and the term “poison” is inadequate in these aspect; should be changed.

Lines 75 and further– this section i san introduction to the examined scientific problem ,so statement that „In this study, we found that the protein expressions of AGEs and calbindin-D28k were significantly and markedly  induced in the kidneys of db/db diabetic mice.” Should be done in section „Discussion” or „Sumary”, not in section „Introduction”

Later lines till 84 – is it adescription  results of own study or date derived from literatrur? Appropraiate references should be added.

„Materials and Methods”

More detailed information about C57BL/6J db/db mice should be adde, and confirmation that are diabetic mice should be added.

Line 98 – what concentration or proportion (stok) of antibodies for AGEs and calbindin-D28k were  used?

Why such solutions  and proportion of „Bovine serum albumin (BSA, 1 mg/ml) and D-glucose (1 mg/ml)” was applied. What is efficiency of of glycation proces, and obtained AGEs?

Line 135 – there is a lack of information about the sused solutions of secondary antibodies

Line 144 – technical mistake

What conditios to siRNA examination was used?

 „Results”

Lines 169-170 and 230 and futrher lines, and 254  ….– in this section results of earler studies did not showed and comment.

In the text of the manuscript should be given general information what the subsequent Figures presents.

The organization of Figure 2 should be slightly improved for greater readability.

Figure 3 – additionall expalantion should be added to beter identification and interpretation of obtained results.

Generally the section “Results” is rich, and rather well-presented and succinctly described.

“Discussion”

Information about role of AGEs in pathogenesis and development of diabetic nephropathy and its pathways of action is rather well known and described. For example – lines 317 – 318 – given information are not new.

Generally section “Discussion” is well constructed, but the significance of own obtained results by authors should be more emphasized and highlighted.

“Reference”

Reference number 1 should be more current. Reference 3 should be change. Reference number 6 as well as especially 18  should be more current.

This manuscript after minor changes is suitable to publication.

Author Response

Reviewer#1

This manuscript after minor changes is suitable to publication.

1. „Abstract”: This study was conducted using animal model, so conclusion that ” Calbindin-D28k may serve as a marker for proximal tubule injury/fibrosis in diabetic nephropathy.” Is somewhat a bit too significant.  

Response: We appreciate the reviewer's comment. We have deleted this sentence in Abstract of this revised manuscript according to the suggestion of reviewer.

2. “Introduction”: (1). the statemnat that „… (AGEs) are generally considered as the poisons for diabetic patients.” is too general, and the term “poison” is inadequate in these aspect; should be changed. (2). Lines 75 and further– this section i san introduction to the examined scientific problem ,so statement that „In this study, we found that the protein expressions of AGEs and calbindin-D28k were significantly and markedly  induced in the kidneys of db/db diabetic mice.” Should be done in section „Discussion” or „Sumary”, not in section „Introduction”. (3). Later lines till 84 – is it a description  result of own study or date derived from literature? Appropraiate references should be added.

Response: We appreciate the reviewer's comment. We have revised the Introduction section in this revised manuscript according to the suggestion of reviewer: (1). We have revised the word “poisons” to “risk factors”. (2). We have moved this sentence to Discussion section. (3). This description is the results of this study. We have deleted this description and added a description for our aim for this study.

3. „Materials and Methods”: (1). More detailed information about C57BL/6J db/db mice should be adde, and confirmation that are diabetic mice should be added. (2). Line 98 – what concentration or proportion (stok) of antibodies for AGEs and calbindin-D28k were used? (3). Why such solutions and proportion of „Bovine serum albumin (BSA, 1 mg/ml) and D-glucose (1 mg/ml)” was applied. What is efficiency of of glycation proces, and obtained AGEs? (4). Line 135 – there is a lack of information about the sused solutions of secondary antibodies. (5). Line 144 – technical mistake. What conditions to siRNA examination was used?

Response: We appreciate the reviewer's comment. We have revised the Materials and Methods section in this revised manuscript according to the suggestion of reviewer: (1). We have added more detailed information about C57BL/6J db/db mice. (2). We have added the concentration proportion of antibodies for AGEs, calbindin-D28k, and others using for Western blot or IHC. (3). We have corrected the description for AGEs preparation. Bovine serum albumin (BSA, 300 mg/ml) and D-glucose (0.5 M) were incubated in phosphate buffer (0.2 M, pH7.4) at 37^oC according to the previous studies. The AGEs were identified by an Ultraflex-III MALDI-TOF/TOF mass spectrometer and the concentration of AGEs was determined by a BCA protein assay kit. (4). We have added the information for secondary antibodies. (5). We have revised the description for siRNA transfection and added the information for siRNA source.

4.  „Results”: (1). Lines 169-170 and 230 and futrher lines, and 254  ….– in this section results of earler studies did not showed and comment. (2). In the text of the manuscript should be given general information what the subsequent Figures presents. (3). The organization of Figure 2 should be slightly improved for greater readability. (4). Figure 3 – additional explanation should be added to beter identification and interpretation of obtained results. (5). Generally the section “Results” is rich, and rather well-presented and succinctly described.

Response: We appreciate the reviewer's comment. We have revised the Results section in this revised manuscript according to the suggestion of reviewer: (1). We have added the description and comment for the results of earlier studies. (2). We have given information and describes what the subsequent figure presents. (3). We have re-organized the Figure 2. (4). We have added the additional explanation for Figure 3. (5). We carefully revised the Results section according to suggestion comments.

5. “Discussion”: (1). Information about role of AGEs in pathogenesis and development of diabetic nephropathy and its pathways of action is rather well known and described. For example – lines 317 – 318 – given information are not new. (2). Generally section “Discussion” is well constructed, but the significance of own obtained results by authors should be more emphasized and highlighted.

Response: We appreciate the reviewer's comment. We have revised the Discussion section in this revised manuscript according to the suggestion of reviewer. We have added the recent reports for diabetic nephropathy and AGEs (2018, 2019). We also revised the Discussion to emphasize and highlight the significance of our results.

6. “Reference”: Reference number 1 should be more current. Reference 3 should be change. Reference number 6 as well as especially 18 should be more current.

Response: We appreciate the reviewer's comment. We have revised the References section for Refs #1, 3, 6, and 18 in this revised manuscript according to the suggestion of reviewer.

Reviewer 2 Report

The current study showed that diabetes could induce calbindin as an adaptive mechanism to lower AGE-induced ER stress and fibrosis in diabetic kidney. 

1-Since calbindin is known to regulate calcium homeostasis in distal tubule and authors shows that AGE increased calbindin in distal tubular cell, Why did the authors focus all the study in proximal cells rather than distal cells?

2- if calbindin plays a role on calcium homeostasis , the authors did not show any data on renal calcium handling even no data for serum or urine calcium if any in diabetic mice? 

3- if upregulation of calbindin provides a protective role via decrease ER stress and fibrosis , have the authors looked at whether overexpressing calbindin in diabetic kidney  lowers renal fibrosis , ER stress,  inflammation and injury ?

4- The authors should look at marker of renal injury in diabetic mice perhaps albuminuria and creatinine clearance to show if silencing calbindin signal will lead to relevant renal injury.

5- ER stress is known to activate JNk/NFKB inflammatory signal . Have the authors looked at the effect of silencing calbindin on JNK/ NFKB signal? 

Author Response

Reviewer#2

The current study showed that diabetes could induce calbindin as an adaptive mechanism to lower AGE-induced ER stress and fibrosis in diabetic kidney. 

1- Since calbindin is known to regulate calcium homeostasis in distal tubule and authors shows that AGE increased calbindin in distal tubular cell, Why did the authors focus all the study in proximal cells rather than distal cells?

Response: We appreciate the reviewer's comment. In db/db diabetic mice, we have unexpectedly found that AGEs and calbindin-D28k protein expression could be induced in the proximal renal tubules. Calbindin-D28k is known to be high-concentration in the renal distal tubules, but not in the proximal tubules. We further found that AGEs markedly induced the calbindin-D28k expression in the proximal tubule cells. Therefore, we have interest in studying this phenomenon. We investigated the role of calbindin-D28k in AGEs-induced renal proximal tubule cell injury using models of human renal proximal tubule cells and db/db diabetic mice.

We added this explanation for our study motivation in the Discussion of this revised manuscript.

2-  if calbindin plays a role on calcium homeostasis , the authors did not show any data on renal calcium handling even no data for serum or urine calcium if any in diabetic mice? 

Response: We appreciate the reviewer's comment. In this study, we unexpectedly found that calbindin-D28k expression could be markedly induced in proximal renal tubule cells in diabetic kidneys or AGEs-treated proximal tubule cells. The role of calbindin-D28k in AGEs-induced renal proximal tubule cell injury is still unclear. Therefore, we focused on the mechanisms of proximal tubule cell injury. We also agree with the comment by reviewer that the role of calbindin-D28k in calcium homeostasis of renal proximal tubules in diabetic mice is important. This topic deserves further investigation in the future. We added a description that “further investigation also needs to clarify the role of calbindin-D28k in calcium homeostasis of renal tubules during diabetes.” in the Discussion of this revised manuscript.

3-  if upregulation of calbindin provides a protective role via decrease ER stress and fibrosis , have the authors looked at whether overexpressing calbindin in diabetic kidney  lowers renal fibrosis , ER stress,  inflammation and injury ?

Response: We appreciate the reviewer's comment. In this study, we found that AGEs by itself stimulated the calbindin-D28k overexpression in the renal proximal tubule cells. Therefore, we tried to use the siRNA-calbindin-D28k transfection, but not overexpressing calbindin-D28k, to clarify the role of calbindin-D28k in AGEs-treated proximal tubule cells.

4- The authors should look at marker of renal injury in diabetic mice perhaps albuminuria and creatinine clearance to show if silencing calbindin signal will lead to relevant renal injury.

Response: We appreciate the reviewer's comment. In this study, we found silencing calbindin-D28k with siRNA could enhance the fibrotic signals in AGEs-treated HK2 cells and there was marked fibrosis induction in the kidney of db/db diabetic mice; therefore, in the experiments for silencing calbindin-D28k in diabetic mice, we focused on the fibrosis signals and found that renal fibrosis in tubular area was markedly enhanced in db/db mice with calbindin-D28k siRNA delivery. We have a mistake for blood biochemical analysis. We have no time and no animals to redo these experiments at the present state.

5- ER stress is known to activate JNk/NFKB inflammatory signal. Have the authors looked at the effect of silencing calbindin on JNK/ NFKB signal? 

Response: We appreciate the reviewer's comment. We have added the NFκB (p65) data on Figure 5B in this revised manuscript according to the suggestion of reviewer.

Reviewer 3 Report

This manuscript explores that role of diabetic AGE induction of calbindin-D28k in renal pathology using both the diabetic db/db mouse model and an HK-2 cell line. Though the results in the cell line are robust, additional work is needed to support the conclusion that calbinding-D28k is expressed in the proximal tubule in this diabetic model.

Major Comments:

To enhance rigor and reproducibility please provide all details about animal housing (i.e. diet, light/dark hours, etc.)

Figure 1.  The tubules that stain positive for calbindin-D28k in these sections look like distal tubules rather than proximal tubules.  The tubules that are staining here clearly lack a brush border, and have an open lumen, characteristics of distal tubules. Could it simply be that the sections of kidney shown in this representative image simply show more distal tubules? Double labeling for both calbindin-D28k/AGE and a proximal tubule marker should be done.  Further, since the western blot was performed on the entire kidney, that data can only support an upregulation overall, but does not point to proximal tubules uniquely.

Was the Western blot here for RAGE, rather than AGEs?  AGE antibody was not listed in methods for Western blots.

Figure 2.  Was the BSA and other controls used here prepared the same way as the AGE components.

Figure 7.  Trichrome staining cannot provide a direct assessment of collagen.  This needs to be amended to appropriately reflect what could be stained.  Further, there is no distinction here between proximal and distal tubules.  In order to link these changes to proximal tubules it would be better to stain for collagen and a proximal specifically.

Line 416.  There is no substantial evidence that calbindin-D28k is expressed in proximal tubules at enhanced levels.   This is an overstatement and needs to be appropriately clarified in text throughout manuscript and in title unless evidence is presented that the tubules that exhibit enhanced staining are indeed proximal tubules (see comments above).

Though there does seem to be evidence of an effect in HK-2 cells, it does not seem to be relevant in the animal model.   It could be for many reasons that need to be explored, such as:

HK-2 cells used may no longer maintain the proximal tubule phenotype, or that there are some other regulatory mechanisms in play. 

Perhaps the concentrations used in the cell culture model were supraphysiological. Alternatively, AGE’s were not quantified in kidney from the mouse model.

AGE’s resulting from BSA proteins may be entirely different from that              of mice.

Author Response

Reviewer#3

This manuscript explores that role of diabetic AGE induction of calbindin-D28k in renal pathology using both the diabetic db/db mouse model and an HK-2 cell line. Though the results in the cell line are robust, additional work is needed to support the conclusion that calbinding-D28k is expressed in the proximal tubule in this diabetic model.

Major Comments:

1. To enhance rigor and reproducibility please provide all details about animal housing (i.e. diet, light/dark hours, etc.)

Response: We appreciate the reviewer's comment. We have added the information for animal housing in the Methods section of this revised manuscript according to the suggestion of reviewer.

2. Figure 1. The tubules that stain positive for calbindin-D28k in these sections look like distal tubules rather than proximal tubules. The tubules that are staining here clearly lack a brush border, and have an open lumen, characteristics of distal tubules. Could it simply be that the sections of kidney shown in this representative image simply show more distal tubules? Double labeling for both calbindin-D28k/AGE and a proximal tubule marker should be done. Further, since the western blot was performed on the entire kidney, that data can only support an upregulation overall, but does not point to proximal tubules uniquely.

Response: We appreciate the reviewer's comment. We have added the data for double labeling of a proximal tubule marker (AQP-1) and calbindin-28k in the kidney of db/db mice in this revised manuscript according to the suggestion of reviewer. As shown in Figure 1E, calbindin-D28k-positive staining co-localized with the AQP-1 (proximal tubular marker) staining in the renal proximal tubular regions of db/db diabetic mice, but not control mice [there was only calbindin-D28k-positive stain, but not AQP-1-positive stain, in control kidney, indicating that calbindin-D28k was only located in renal distal tubules of control mice; in diabetic kidney, the calbindin-D28k were located in both distal and proximal tubules]. These results confirmed that calbindin-D28k could be induced in the renal proximal tubules of diabetic mice. Moreover, the brush borders in proximal tubules of kidney sections may shed during diabetic nephropathy, which stained tubule lacks the brush borders. In Figure 7 of this revised manuscript, we tried to magnify the scale to observe the proximal tubules in kidney sections. We found that calbindin-D28k staining was clearly shown in proximal tubules in which the brush borders in some proximal tubule cells were shed (Fig. 7B). Similarly, the Masson’s trichrome staining was clearly shown in proximal tubules of kidney sections at higher magnification (Fig. 7C, lower panel).

3. Was the Western blot here for RAGE, rather than AGEs? AGE antibody was not listed in methods for Western blots.

Response: We appreciate the reviewer's comment. We have added the information of AGEs antibody for Western blot in the Methods of this revised manuscript according to the suggestion of reviewer.

4. Figure 2. Was the BSA and other controls used here prepared the same way as the AGE components.

Response: We appreciate the reviewer's comment. The BSA used here was prepared the same way as the AGE components.

5. Figure 7. Trichrome staining cannot provide a direct assessment of collagen. This needs to be amended to appropriately reflect what could be stained. Further, there is no distinction here between proximal and distal tubules. In order to link these changes to proximal tubules it would be better to stain for collagen and a proximal specifically.

Response: We appreciate the reviewer's comment. We agree with the comment of reviewer for collagen stain. However, Masson's trichrome staining is known to be widely used to stain collagen for fibrosis studies, for examples: Matsui et al. (Diabetes. 2017 Jun;66(6):1683-1695; diabetic nephropathy; reference#6 of this revised manuscript); Magalhães et al. (Sci Rep. 2017 Dec 5;7(1):16915; CKD patients’ kidney biopsies); Zhang et al. (Mol Med Rep. 2019 Apr;19(4):3305-3313; UUO nephropathy). Therefore, could you kindly allow us to cite the reference to support this method.

6. Line 416. There is no substantial evidence that calbindin-D28k is expressed in proximal tubules at enhanced levels. This is an overstatement and needs to be appropriately clarified in text throughout manuscript and in title unless evidence is presented that the tubules that exhibit enhanced staining are indeed proximal tubules (see comments above). Though there does seem to be evidence of an effect in HK-2 cells, it does not seem to be relevant in the animal model. It could be for many reasons that need to be explored, such as: HK-2 cells used may no longer maintain the proximal tubule phenotype, or that there are some other regulatory mechanisms in play. Perhaps the concentrations used in the cell culture model were supraphysiological. Alternatively, AGE’s were not quantified in kidney from the mouse model. AGE’s resulting from BSA proteins may be entirely different from that              of mice.

Response: We appreciate the reviewer's comment. We provided the evidence to support our conclusion in this revised manuscript according to the suggestions of reviewer.

We have added the data for double labeling of a proximal tubule marker (AQP-1) and calbindin-28k in the kidney of db/db mice in this revised manuscript according to the suggestion of reviewer. As shown in Figure 1E, calbindin-D28k-positive staining co-localized with the AQP-1 (proximal tubular marker) staining in the renal proximal tubular regions of db/db diabetic mice, but not control mice [there was only calbindin-D28k-positive stain, but not AQP-1-positive stain, in control kidney, indicating that calbindin-D28k was only located in renal distal tubules of control mice; in diabetic kidney, the calbindin-D28k were located in both distal and proximal tubules]. These results confirmed that calbindin-D28k could be induced in the renal proximal tubules of diabetic mice. Moreover, the brush borders in proximal tubules of kidney sections may shed during diabetic nephropathy, which stained tubule lacks the brush borders. In Figure 7 of this revised manuscript, we tried to magnify the scale to observe the proximal tubules in kidney sections. We found that calbindin-D28k staining was clearly shown in proximal tubules in which the brush borders in some proximal tubule cells were shed (Fig. 7B). Similarly, the Masson’s trichrome staining was clearly shown in proximal tubules of kidney sections at higher magnification (Fig. 7C, lower panel).

Moreover, the HK2 cells were obtained from American Type Culture Collection that the cell lines authentication may need to be identified.

We also discussed the concentrations of AGEs used in the cell cultures in this revised manuscript.

AGEs with various concentrations have been shown to affect the cell growth and function in various cell types. AGEs (AGE-BSA) at 100 μg/ml have been found to reduce the viable cell numbers and induced apoptotic cell death at 250 μg/ml in retinal pericytes [47]. It has been shown that AGEs (AGE-BSA, 25-200 μg/ml) can induce apoptosis in pancreatic islet endothelial cells [48]. AGEs [AGE-human serum albumin (HSA)] at 100 μg/ml have been found to induce apoptosis in some cancer cell lines [49]. Chiang et al. (2016) have found that 40-160 μg/ml AGEs (AGE-BSA) significantly reduced mesangial cell viability and induced mesangial cell apoptosis [12]. In this study, we found that AGEs (AGE-BSA) at the concentrations of 25-100 μg/ml induced the calbindin-D28k expression in HK2. AGEs induced the expression of fibrotic molecules, ECM proteins, epithelial-mesenchymal transition (EMT) markers, and endoplasmic reticulum (ER) stress-related molecules in HK2 cells. The concentrations of AGEs used in this study are reasonable.