Insulin protects acinar cells during pancreatitis by preserving glycolytic ATP supply to calcium pumps

Acute pancreatitis (AP) is serious inflammatory disease of the pancreas. Accumulating evidence links diabetes with severity of AP, suggesting that endogenous insulin may be protective. We investigated this putative protective effect of insulin during cellular and in vivo models of AP in diabetic mice (Ins2Akita) and Pancreatic Acinar cell-specific Conditional Insulin Receptor Knock Out mice (PACIRKO). Caerulein and palmitoleic acid (POA)/ethanol-induced pancreatitis was more severe in both Ins2Akita and PACIRKO vs control mice, suggesting that endogenous insulin directly protects acinar cells in vivo. In isolated pancreatic acinar cells, insulin induced Akt-mediated phosphorylation of 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 2 (PFKFB2) which upregulated glycolysis thereby preventing POA-induced ATP depletion, inhibition of the ATP-dependent plasma membrane Ca2+ ATPase (PMCA) and cytotoxic Ca2+ overload. These data provide the first mechanistic link between diabetes and severity of AP and suggest that phosphorylation of PFKFB2 may represent a potential therapeutic strategy for treatment of AP.

3) figure 6g POA concentration should be 0 microM in the first PACIRKO column, if the legend is correct Reviewed by Professor Robert Sutton Reviewer #3: Remarks to the Author: Acute pancreatitis (AP) is serious inflammatory disease of the pancreas. Accumulating evidence links diabetes with severity of AP, suggesting that endogenous insulin may be protective. In the present study, the authors investigated this putative protective effect of insulin during cellular and in vivo models of AP in diabetic mice (Ins2Akita) and Pancreatric Acinar cell-specific Conditional Insulin Receptor Knock Out mice (PACIRKO). Caerulein and palmitoleic acid (POA)/ethanol-induced pancreatitis was more severe in both Ins2Akita and PACIRKO versus control mice, suggesting that endogenous insulin directly protects acinar cells in vivo. In isolated pancreatic acinar cells, insulin induced Akt-mediated phosphorylation of 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 2 (PFKFB2) which upregulated glycolysis thereby preventing POA-induced ATP depletion, inhibition of the ATP-dependent plasma membrane Ca2+ ATPase (PMCA) and cytotoxic Ca2+ overload. The authors concluded that these data provided the first mechanistic link between diabetes and severity of AP and suggested that phosphorylation of PFKFB2 may represent a novel therapeutic strategy for treatment of AP.

Major Criticisms
1. There was a high basal Akt and PFKFB2 phosphorylation in untreated acinar cells from PACIRKO mice compared to acinar cells from IRlox/lox mice. There was no further increase in Akt or PFKFB2 phosphorylation following insulin treatment of acinar cells from PACIRKO mice. The authors speculate that the loss of insulin receptors was responsible for this insulin insensitivity. This reviewer requests the authors to examine the expression and phosphorylation levels of insulin receptor substrates, such as IRS-1 and IRS-2, because there are another up-stream signals for these molecules. For example, incretin, such as GLP-1, increases cAMP, thereby stimulating phosphorylation of CREB and up-regulating IRS-2. Information on these molecules may be needed to access the high basal Akt level in untreated acinar cells from PACIRKO mice. 2. In isolated pancreatic acinar cells, insulin induced Akt-mediated phosphorylation of 6phosphofructo-2-kinase/fructose-2,6-biphosphatase 2 (PFKFB2) which upregulated glycolysis thereby preventing POA-induced ATP depletion, inhibition of the ATP-dependent plasma membrane Ca2+ ATPase (PMCA) and cytotoxic Ca2+ overload. The authors speculate that phosphorylation of PFKFB2 acts as a "volume control" for glycolytic flux and thus ATP production and is thus likely to be the major molecular mechanism for the protective effects of insulin during pancreatitis. Then, the authors may provide with the results on glycolytic flux in acinar cells under normal and AP conditions. 3. In the present study, the authors characterize mouse models whose insulin signaling is impaired in acinar cells. On the other hand, there are many models whose insulin signaling may be upregulated due to hyperinsulinemia to compensate for systemic insulin resistance. Given the role of insulin signaling in acinar cells, such models may be protected against caerulein and POA/ethanolinduced pancreatitis. Is this true? 4. The authors propose that insulin infusion with the aim of reducing pancreatic injury may prove effective for the treatment of severe AP, as long as there is tight moment-to-moment glycaemic control (hyperinsulinaemic euglycaemic clamp). This reviewer asks the authors to examine the amount of infused insulin to sufficiently reduce pancreatic injury. Is it possible to protect against pancreatic injury and simultaneously avoid hypoglycemia?

Reviewer #1 (Remarks to the Author):
The study demonstrates the importance of insulin and insulin receptors on pancreatic acinar cells in reducing the severity of acute pancreatitis. This is an important finding. The primary phenomenon was clearly demonstrated, the authors provided convincing evidence of the cellular mechanism responsible for the insulin protection (increased efficiency of PMCA and consequent reduction in Ca2+ toxicity) and made a reasonable and systematic attempt to define the molecular mechanism.
Authors Response: We would like to thank reviewer #1 for these comments and are pleased that they found merit in our study

Minor points:
The graphical abstract is somewhat overloaded and can be simplified.

Authors Response:
This has now been modified by removing the additional data and simplifying the cartoon

The described effect of insulin on NAD(P)H fluorescence in cells treated by CCCP and IAA (in presence / absence of insulin receptors) is an interesting observation. It would be useful to discuss how this relates to expected ATP changes.
Authors Response: ATP was actually measured directly using the firefly luciferase assay, which is summarised in Figure 7a. These data show that insulin causes a rightward shift in the palmitoleic acid (POA) concentration-ATP response curve. This reaffirms the notion that insulin preserves ATP production and thus prevents POA-induced ATP depletion. However, in the context of the NADH autofluorescence experiments we used CCCP-induced NADH depletion as a surrogate for mitochondrial bioenergetics and IAA-induced NADH depletion as a surrogate for glycolytic bioenergetics. This is because CCCP is a protonophore and mitochondrial uncoupler that dissipates the mitochondrial proton gradient and thus mitochondrial membrane potential which in turn reduces the driving force for ATP depletion by the ATP synthase. Hence, CCCP is described as a mitochondrial uncoupler, as it uncouples ATP synthesis from the electron transport chain/oxygen consumption. This means that CCCP causes the electron chain and oxygen consumption (respiration) to "work in overdrive" at a maximum rate in an attempt to maintain the proton gradient against this futile cycle of maintaining ATP synthesis. This in turn causes mitochondrial NADH consumption (derived from Krebs cycle intermediates) to "fuel" the electron transport chain and leads to eventual mitochondrial NADH depletion. Therefore, CCCP-induced NADH depletion could be described as an indirect measure of mitochondrial bioenergetics and thus ATP production. IAA is an inhibitor of GAPDH, which is the major source of glycolytic NADH. Therefore, IAA-induced NADH depletion could be described as a measure of glycolytic bioenergetics and thus ATP production. These data show that by reducing CCCP-induced NADH depletion and increasing IAA-induced NADH depletion, insulin is effectively inducing a shift from mitochondrial bioenergetics towards glycolytic bioenergetics, which would further explain the rightward shift in the POA concentration ATP response curve to the right ( Figure  7a). We have now added a further explanation of this in the revised manuscript (line 193-203, page 8).
However, we accept that NADH autofluorescence, whilst a convenient surrogate for mitochondrial vs glycolytic bioenergetics, is not a direct measure of ATP and indeed may be affected by numerous other factors that affect NADH and is confounded by NAD(P)H autofluorescence. Therefore, we further investigated whether this insulin-induced switch in mitochondrial vs glycolytic bioenergetics could be translated to ATP. This was achieved using the same experimental paradigm but instead cells were loaded with magnesium green (MgGreen), to assess ATP depletion. Since most cellular ATP exists as MgATP, any ATP depletion will cause a corresponding increase in free Mg 2+ , which is detected by MgGreen.
Although this is still not a direct measure of ATP concentration per se, it is a much closer representation of ATP depletion that can be correlated with NADH autofluorescence. These experiments essentially showed the same phenomena; insulin caused a decrease in CCCPinduced increase in MgGreen fluorescence (ATP depletion) and increase in IAA-induced increase in MgGreen fluorescence (ATP depletion). This further supports the notion that insulin switched metabolism from mitochondria towards glycolytic ATP production that was sufficient to maintain cellular ATP even in the face of POA-induced impaired mitochondrial metabolism. These new data have now been included in Supplementary Information (Methods, line 164-175, page 6-7; Results, line 269-284; Figure S6) and described in the Results (line 213-227, page 8-9). Its also worth noting in the context of these experiments that our previous studies in rat pancreatic acinar cells (ref 5, Mankad et al 2012& ref 6, Samad, et al 2014 show that; 1-Insulin prevented POA-induced MgGreen fluorescence in isolated pancreatic acinar cells 2-The insulin switch from mitochondrial metabolism towards glycolysis also translated to PMCA activity (in situ Ca 2+ clearance assay). Insulin markedly reduced the CCCP-induced inhibition of PMCA activity but potentiated the bromopyruvate (glycolytic inhibitor)-induced inhibition of PMCA activity. Again this is consistent with a switch in metabolism which not only translates to ATP production but also PMCA activity.

Authors Response:
We have now investigated this further and provide new data showing that the high basal phosphorylation of PFKFB2 in PACIRKO mouse acinar cells is most likely due to an increased expression and phosphorylation of IRS1 and IRS2 ( Figure S9), upstream of activation of PI3K/Akt. This will inevitably lead to a higher basal phosphorylation of Akt and consequently PFKFB2. The mechanism for this upregulation of IRS1/IRS2 remains unclear but may reflect some kind of compensatory mechanism in response to deletion of the insulin receptor (IR). However, it is important to note that insulin treatment of PACIRKO mouse acinar cells failed to increase phosphorylation of IRS1/IRS2 ( Figure S9 This study provides sufficient evidence for the significant changes in the severity of acute pancreatitis but in the further studies (not this paper) it would be perhaps useful to characterise the effect on the early stages of pancreatic damage (e.g. pancreatic trypsin activity).

Authors Response:
We agree that additional measures of pancreatic injury in addition to distant organ injury would further strengthen these data. However, due to the logistical issues of the COVID-19 pandemic and in the interest of reducing animal use, it was decided that these experiments could not justify such an incremental increase in knowledge.

Reviewer #2 (Remarks to the Author):
There has been speculation as to the role of islet function and especially insulin secretion in acute and chronic pancreatitis, not least because of the close anatomical relationship between pancreatic endocrine and exocrine tissue, with arterial blood feeding first into islets and a portal circulation within the pancreas that then supplies acinar tissue. This study specifically addresses the potential role of insulin in pancreatic acinar cell injury driving acute pancreatitis. The authors provide substantial evidence obtained in a number of different and corroborative ways in multiple murine models for a direct effect of insulin via insulin receptors in driving anaerobic glycolysis that contributes to pancreatic acinar cellular ATP supply, known to be critically impared by mitochondrial injury in acute pancreatitis. Authors Response: We would like to thank reviewer #2 for their positive and pragmatic comments. On reflection, we also agree that additional measures of both pancreatic injury in addition to distant organ injury would further strengthen our data. However, due to the logistical issues of the COVID-19 pandemic and in the interest of reducing animal use, we decided that these experiments could not justify such a small incremental increase in knowledge.

There are several textual issues: 1) line 58 would read better as 'reduced insulin sensitivity (type-2 diabetes)' not 'reduced insulin effectivenes (spelling) (type-1 diabetes)' 2) line 240 remove 'similar' as this appears later in the sentence 3) line 280 there are a number of other mechanisms likely to account for the association of a raised mortality in chronic pancreatitis with diabetes mellitus 4) line 282 and 283 switch to past tense inappropriate 5) line 316 'collatoral' (spelling) 6) line 336 'priveleged' (spelling) 7) line 338 'globl' (spelling) 8) line 361 suggest (and likely insulin receptors within acinar cells)
The 'conclusion' is repetitive and first section unnecessary so it is suggested that lines 387-391 are removed. Concerning the figures: 1) figure 3biv the ordinate label should be Relative 18S rRNA, if the legend is correct 2) figure 4biv the ordinate label should be Relative 18S rRNA, if the legend is correct 3) figure 6g POA concentration should be 0 microM in the first PACIRKO column, if the legend is correct Authors Response: We apologise for these minor errors, which have now been corrected Reviewer #3 (Remarks to the Author): Acute pancreatitis (AP) is serious inflammatory disease of the pancreas. Accumulating evidence links diabetes with severity of AP, suggesting that endogenous insulin may be protective. In the present study, the authors investigated this putative protective effect of insulin during cellular and in vivo models of AP in diabetic mice (Ins2Akita) and Pancreatric Acinar cell-specific Conditional Insulin Receptor Knock Out mice (PACIRKO). Caerulein and palmitoleic acid (POA)/ethanol-induced pancreatitis was more severe in both Ins2Akita and PACIRKO versus control mice, suggesting that endogenous insulin directly protects acinar cells in vivo. In isolated pancreatic acinar cells, insulin induced Akt-mediated phosphorylation of 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 2 (PFKFB2) which upregulated glycolysis thereby preventing POA-induced ATP depletion, inhibition of the ATPdependent plasma membrane Ca2+ ATPase (PMCA) and cytotoxic Ca2+ overload. The authors concluded that these data provided the first mechanistic link between diabetes and