The Effects Of Diabetic Nephropathy Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Diabetic nephropathy (DN) also referred to as Kimmelsteil-Wilson disease or Diabetic glomerulosclerosis, is a clinical syndrome often presenting as one of the main microvascular complications in patients with long-standing diabetes. It is characterised by a progressive rise in urine albumin secretion "microalbuminuria on at least 3 occasions separated by 3 to 6 months" (Clinical evidence, 200?). This is usually accompanied by hypertension and a decline in glomerular filtration, eventually progressing to end-stage renal failure (ESRF) (Marshall, 2004).


DN is the leading cause of glomerulosclerosis and ESRF worldwide (Dronavalli et al, 2008) and is also classed as an independent risk factor for cardiovascular disease (Marshall, 2004). It is the principal diagnosis in 25-50% of patients starting renal replacement therapy for ESRF (Tang, 2010). Approximately 20-30% of all diabetic go on to develop nephropathy (Soldatos et al, 2008) but a much smaller proportion of type 2 diabetes sufferers are later diagnosed with ESRD. However due to the significantly increased prevalence of this form of the disease, these patients form more than 50% of all those currently starting dialysis (American Diabetes Association, 2004). Ethnicity appears to play a key role in these outcomes, with Hispanics, Native-Americans and Africa-Americans being at greater risk of developing ESRD than non-Hispanic Caucasians with type 2 diabetes (American Diabetes Association, 2004).


There is substantial evidence suggesting diabetic nephropathy is the result of interactions between metabolic and haemodynamic pathways (Fig. 1.6) As diabetes is characterised by hyperglycaemia it is possible that glucose is a biochemical factor involved in the pathogenesis of DN. As shown in Fig 1.6 increased production of glucose-derived proteins or AGEs (advanced glycation end products) may lead to extracellular matrix deposition and renal tissue injury (Cooper et al, 1998). Soulis et al, 1996 found increased levels of AGEs in the kidney in diabetic patients or those with poor renal function. The use of aminoguanidine (AGE inhibitor) was shown to decrease renal AGE levels and the development of albuminuria and mesangial expansion (Cooper et al, 1998).

The role of glucose in the polyol pathway may also play a part in the development of DN (Goldfarb et al, 1991). However blocking this pathway using aldose reductase inhibitors has produced inconsistent findings (Cooper et al, 1998).

The enzyme protein kinase C (PKC) has been shown to display enhanced activity in diabetic tissues

such as the glomerulus (Cooper et al, 1998). The inhibition of the PKC pathway using LY333531

prevents albuminuria and hyperfiltration in diabetic rats (Ishii et al, 1996, as cited in Cooper et al,

1998. However as yet these findings are only restricted to the murine model and have not been

identified in humans thus limiting their potential therapeutic implications.

Fig 1.6: Diagram summarising possible interplay between metabolic and haemodynamic pathways in the pathophysiology of DN (Cooper et al, 1998).

Although in patients DN often often co-exists with hypertension, studies in animal models of diabetes indicate that even when systemic hypertension is not present, the pressure within the kidneys (intraglomerular pressure) is raised (Cooper, 1998). Thus increased intraglomerular pressure appears to be the major haemodynamic factor at play in DN. The raised pressure which is due to constriction of the efferent glomerular arteriole, is thought to trigger glomerular damage as a result of direct pressure effects and also indirectly by increasing proteinuria (Marshall, 2004). Experimental studies in which renal mesangial cells are exposed mechanical stretch to mimic the effects of glomerular hypertension, have shown that this initiates biochemical changes such as activation of p38 MAPK via a PKC-dependent mechanism which in turn induces TGF-B1 and fibronectin expression (Gruden et al, 2000). [EXPAND]


Angiotensin II


Renal structural abnormalities (fig and table Wolf et al) +fig Marshall

Cytokines (Caramori et al, Dronavalli et al)

Oxidative stress (ROS- table Wolf et al)

Growth factors e.g. TGFB, VEGF, CTGF (Caramori et al, Dronavalli et al, Soldatos)

Ang/Tie-2 lead on to Nogo-B (Gnudi)


Fig. 1.7: Progression of renal damage over time in type 1 diabetes (renal biopsy images from electron microscopy). Increase in GBM thickness, narrowing of capillary loops and mesangial ECM expansion seen from normal (left) to diseased kidney (right). (Wolf, 2004).

Angiotensin II

The role of podocytes and mesangial cells

Renal structure and function

The human kidney is composed of over a million structural units or nephrons. Each nephron consists of …. Tubules,… glomeruli…arterioles…GFB

The function of the kidney is to …

The effect of diabetes on the kidneys is predominantly on the GFB therefore proteinuria, inc BP, reduced GFR

Thus research into the mechanisms underlying DN, have largely focused on the GFB and its individual components. This barrier consists of a number of components the most important of which, with regard to DN, are podocytes, mesangial cells and the glomerular basement membrane (GBM).


Podocytes are….

Research has shown that… exposure of podocytes to high glucose causes…

Mechanical stretch

Dalla et al, 2001 suggested that mesangial expansion, measured as mesangial fractional volume [Vv(mes/glom)] is the major structural feature of type 1 and 2 diabetes that is most closely correlated with glomerular filtration rate (GFR) and the presence of proteinuria and hypertension. Fioretto et al, 2007 also propose a role for mesangial and interstitial expansion in the glomerular injury associated with nephropathy in type 1 diabetes.

GBM is….

GBM thickening feature of both T1 and 2DM ?

Podocytopenia or a reduction in the number of podocytes is a feature of both type 1 and 2 diabetes (Li et al, 2007). This change is the result of an imbalance between podocyte proliferation and loss. The former can be explained by impaired DNA synthesis and hypertrophy of podocytes, whereas the latter in the due to podocyte apoptosis and detachment from the GBM (Li et al, 2007). In a population of Pima Indians with type 2 diabetes analysis of renal biopsies revealed that podocytopenia was accompanied by an increase in the width of podocyte foot processes (Pagtalunan et al, 1997 as cited in Li et al, 2007).

The effects of DN on the renal glomerular cells can be studied in various ways. As arterial hypertension is one of the key features of DN, in an attempt to mimic the effects of this on the kidney, many researchers have studied the effects of mechanical stretch on glomerular cells.

Mesangial cells

One group found that the exposure of mesangial cells to continuous stretch/relaxation cycles was associated with an increase in gene and protein expression of extracellular matrix (ECM) components such as fibronectin, collagen I, III, IV and Laminin (Yasuda et al, 1997 as cited in Lopes de Faria et al, 2010).

In addition the ECM accumulation which follows MC exposure to stretch was significantly increased in a high glucose environment. Mechanical stretch is also thought to increase the expression of the cytokine TGF-B1 which is known to contribute to ECM accumulation in DN (Riser et al, 1999 as cited in Lopes de Faria et al, 2010).

It has also been shown that human mesangial cells exposed to mechanical stretch, display increased MCP activity as well as decreased CCR2 expression. Thus stretch exposure in MCs appears to exert pro-inflammatory effects which may contribute to the ECM production seen in DN (Giunti et al, 2006 as cited in Lopes de Faria et al, 2010). Moreover it was found that simultaneous exposure of MCs to stretch and high glucose enhanced these pro-inflammatory effects via an NF-κB (nuclear factor -kappa B) pathway (Gruden et al, 2005). This group also reported that mechanical stretch and the ... Angiotensin II (Ang II) caused an increase in the production of vascular endothelial growth factor (VEGF).

Gnudi et al, 2005 found that stretch exposure in HMCs caused an increase in the expression of the facilitative glucose transporter-1 (GLUT-1). It also enhanced glucose transport via a PKC-TGF-B dependent pathway which in turn led to excess ECM accumulation. This increase remained even when MCs were cultured in normal glucose medium. This finding is of particular interest as it suggests or rather confirms the idea of interplay between glucose-mediated metabolic pathways and the haemodynamic effects of mechanical stretch in the development of DN.


The effects of mechanical stretch have also been studied on podocytes; the other major cell type that constitutes the glomerular filtration barrier. This stimulus is reported to exert a number of effects on podocytes including increased glucose uptake, hypertrophy and reduced proliferation of cells (Li et al, 2007). Mechanical stress has also been shown to cause a decrease in the size of the podocyte cell body as well as inducing reversible reorganization of the cell's actin cytoskeleton (Endlich et al, 2001 as cited in Lopes de Faria et al, 2010). Stretch exposure is also thought to initiate a protective response in podocytes whereby they enter a state of intermediate adhesion (Endlich et al, 2006 as cited in Lopes de Faria et al, 2010). But this response may have detrimental consequences later on, as the podocytes will be able to detach from the GBM more easily, leading to the podocytopenia seen in DN.

Another factor affecting podocyte adhesion to the GBM is the integrin protein α3β1 (Lopes de Faria et al, 2010). It has been shown that mechanical forces and TGF-β1 cause a downregulation of this integrin and other ECM substrates (Dessapt et al, 2009). This may also explain the podocyte loss seen in patients with DN, where glomerular hypertension rather than experimental stretch acts as one of the initial instigators of renal injury.

Exposure of podocytes to mechanical stretch and glucose is also thought to activate an intracellular rennin-angiotensin system (Durvasula et al, 2004, 2008 as cited in de Faria et al, 2010). This results in apoptosis of cells and decreased expression of nephrin (Miceli et al, 2010). This protein is an important constituent of the slit diaphragm (structure which connects neighbouring podocyte foot processes) and is essential for restricting protein filtration (de Faria et al, 2010). Thus a decrease in its expression explains the hallmark feature of DN; proteinuria. Such findings have led to trialling angiotensin-II receptor antagonists such as irbesartan in models of diabetes and hypertension which restores the deficiency in glomerular nephrin (Bonnet et al, 2001 as cited in de Faria et al, 2010). This in turn has resulted in the widespread use of such drugs to treat hypertension and diabetes.

Patients with type 1 and 2 diabetes often suffer from co-existing hypertension REF? These two conditions are known to interact REF ultimately leading to the development of numerous complications with DN being one of the most severe.

DN is ...define

The elevated blood pressure that often accompanies diabetes mellitus is transmitted to the glomerulus due to dilation of the afferent (precapillary) glomerular arteriole (de Faria et al, 2010). This causes raised intraglomerular pressure, the effects of which are only enhanced in the setting of diabetes. Hyperglycaemia disrupts the autoregulatory mechanism of the glomerular microcirculation (de Faria et al, 2010) resulting in...

This has the knock on effect of stretching of glomerular structures, particulary the mesangial cells. In fact mesangial expansion, typically measured as mesangial fractional volume [Vv(mes/glom)] has been proposed as the major structural feature of type 1 and 2 diabetes that is most closely correlated with glomerular filtration rate (GFR), proteinuria and hypertension (Dalla et al, 2001). The ultimate effect of such expansion is increased ECM and cytokine production which are the characteristic renal structural lesions of DN (Fioretto et al, 2007).

Clinical features

DN is categorized into two stages according to urinary albumin excretion (UAE) as either microalbuminuria (UAE >20µg/min and ≤199µg/min) or macroalbuminuria (UAE ≥200µg/min). The signs of DN thus depend on the stage of the disease and UAE levels of the affected individual (Table 3.1). The early symptoms of nephropathy are generally non-specific and include fatigue, malaise, nausea and vomiting and decreased appetite. In the late stage, oedema and weight gain often present due to fluid retention and proteinuria causes urine to appear foamy or frothy REF.

Table 1.1: Showing stages of diabetic nephropathy, UAE cutoff values for diagnosis and core clinical features

Prevention and treatment (Marshall et al)


Primary prevention


Novel therapies

1.3 Nogo-B (Reticulon 4)

Reticulon proteins

Nogo belongs to the reticulon family of membrane-bound proteins. As yet four reticulon genes (rtn1, rtn2, rtn3, rtn4) have been identified in mammalian genomes that encode a vast range of gene products (Yan et al, 2005). There has been increasing interest in the role of RTN4 or Nogo in the field of axonal regeneration (Raines, 2004). This protein has also been shown to play a key role in the regulation of vascular haemostasis and remodelling (Acevedo et al, 2004) suggesting that it may act as a 'brake on vascular lesions' (Raines, 2004).

There are 3 isoforms of RTN4 namely Nogo A, B and C. Nogo-A and C are predominantly expressed in the central nervous system whereas Nogo-B is thought to be present in most tissues (GrandPre et al, 2000, Chen et al, 2000 as cited in Acevedo et al, 2004). Recently researchers have identified Nogo-B in epithelial cells of the renal tubules and suggested its potential role as a marker of renal injury in mice and humans (Marin et al, 2010).

Role of Nogo in vascular remodelling

Acevedo et al, 2004 studied the role of Nogo in vascular remodelling by injuring the femoral arteries of mice and examining Nogo expression using immunofluorescence microscopy. They found a noticeable reduction in Nogo expression in injured vessels, 10 days after injury (Fig. 1.7). And Nogo expression was virtually abolished 21days post injury.

Fig 1.7: Immunostaining of uninjured vessels with α-Nogo show a marked decrease in Nogo expression at 10 and 21days after vessel injury (Adapted from Acevedo et al, 2004)

These findings suggest that blood vessel damage is associated with a reduction in Nogo expression which may be linked to the neointima formation that typifies vascular remodelling. This is highlighted in Figs 1.8a, b which display a significant (*) decrease in lumen size in Nogo A/B deficient mice (-/-) 3 weeks after injury, when compared to wild-type controls (+/+).

Fig. 1.8a: (Adapted from Acevedo et al, 2004) Fig. 1.8b:

The idea that Nogo can act as a 'brake on vascular lesions' (Raines, 2004) is supported by the work of Acevedo et al, 2004. They generated an adenoviral construct expressing Nogo-B (Ad-Nogo-B). This adenovirus along with a control expressing β-Galactosidase (Ad-β-Gal) was applied to the vessel walls of Nogo A/B deficient mice directly after wire injury (Acevedo et al, 2004). Injured vessels from mice transduced with Ad-β-Gal displayed a significant reduction in lumen size when compared to those given Ad-Nogo-B (Figs. 1.9a,b). These findings imply that endogenous Nogo-B expression can restore levels of the protein in deficient mice with the ultimate effect of limiting the extent of vascular injury in the murine model.

Role of Nogo-B in the kidney

Research carried out by Marin et al, 2010 has identified Nogo-B in the kidney. Mice expressing a Nogo-promoter lacZ reporter gene were used to identify Nogo expression in the kidney. Kidney sections from these mice were found to display intense X-gal staining in the renal papilla compared to other regions of the kidney. (Fig.1.10). Kidney sections from wild-type control mice were negative for X-gal staining. A particularly important finding from this study was the identification of a lesser degree of Nogo-B expression in the renal glomeruli (Marin et al, 2010).

Fig 1.12: Kidney sections from frozen tissue blocks

a) No non-specific staining present in wild-type kidney

b) Intense staining in inner medulla and papilla of reporter kidney

Following on from the study by Acevedo et al, 2004 which suggested a role for Nogo in preventing vascular injury, Marin et al, 2010 proposed a similar role for Nogo-B in limiting the extent of renal injury. This theory was tested by investigating changes in gene expression in Nogo lacZ reporter mice after an episode of renal injury induced by unilateral ureteral obstruction (UUO). When comparing the obstructed kidney (obstr) with the contralateral unobstructed kidney (CLU) or control, X-gal staining was significantly increased in the cortex of obstr compared to CLU (Fig. 1.11a).This suggests that renal injury induces a rise in Nogo-B expression. These findings were confirmed by quantitative PCR analysis (qPCR) on RNA extracted from the kidneys. Nogo-B mRNA levels were found to be four times higher in obstr when compared to wild type controls (Fig. 1.11b).

Fig. 1.11: Effect of UUO on Nogo-B expression b) Graph of qPCR findings showing marked

in mouse kidney a) Noticeable increase in Nogo- increase in Nogo-B expression in obstr vs

B expression (as shown by enhanced X-gal CLU. This is not the case for RTN3

Staining) in obstr kidney compared to CLU control confirming it is specific for Nogo-B (RTN4)

[Add Western, IHC and immunofluorescence]


a)The alterations in Nogo-B expression following UUO were subsequently confirmed to be present in other types of renal injury. Unilateral ischaemia/reperfusion (I/R) injury procedures were carried out on mice. Kidney sections from these mice showed greater X-gal staining in kidney exposed to I/R injury in comparison to the contralateral mock operated kidney (Fig 1.12a). This confirms that Nogo expression is upregulated in two different forms of renal injury in mice.

Fig. 1.12 a) b)

This was also found to be true within human kidneys. Immunofluorescence imaging of human renal biopsy tissue displayed significantly increased Nogo-B expression in the tubular epitheilal cells in samples from kidneys affected by acute tubular necrosis (ATN) when compared with normal control tissues. This suggests that Nogo-B expression is increased in the human kidney during disease.

This finding is of particular importance due to its potential clinical applications. As there are currently no biomarkers that can detect renal injury REF, if further research reveals a consistent increase in Nogo-B expression following various forms of kidney insults in human tissue, it may allow Nogo-B to be introduced as a novel marker of renal injury.

In addition the fact that Nogo-B levels are raised in renal injury, may indicate that it plays a role in the prevention of disease and is upregualated in an attempt to protect against tissue damage. Therefore finding a way to increase Nogo-B expression or prevent its initial down-regulation may serve to halt or limit the progression of renal injury. However further work is needed before any conclusive findings can be made and translated into realistic therapeutic targets.

1.4 Angiopoietin (Ang)-2

What is Ang-2?

Ang-2/Tie-2 receptor expression in diabetic nephropathy

Ang-2 interaction with Nogo-B

1.5 Aims and hypothesis

In order to understand how DN affects the GFB of the kidney we focused our attention on the 2 main cell types it is composed of; MCs and podocytes. We examined the effects of high glucose and mechanical stretch in vitro in HMCs and podocytes. The former i.e. hyperglycaemia is a feature of both DM and DN, while the latter is used to mimic the effects of glomerular hypertension is characteristic of DN. The above conditions were used to study the role of the new reticulon protein (RTN4) Nogo-B in DN and fulfil the following aims:

Determine the location of Nogo-B in mouse kidney sections

Examine the effect of mechanical stretch, high glucose and Ang-2 on Nogo-B gene expression

Examine the effect of the aforementioned variables on Nogo-B expression at the protein level

Previous research into this area has revealed that the overexpression of Ang-2 in podocytes is correlated with a ~7 fold downregulation of glomerular Nogo-B in comparison to controls (preliminary unpublished data). Thus it is hypothesised that Nogo-B expression and/or activity is downregulated by Ang-2 in glomerular diseases.

If this is found to be the case, then by inhibiting Ang-2 expression and the cascade of vascular damage that accompanies it, Nogo-B may serve as a potential therapeutic target in the treatment and/or prevention of diabetic renal disease.


2.1 An overview


Fig. 2.1: Flow diagram summarising various steps in the protocol

Cell culture and treatment of cells (blue)

Protein assay and Western blotting (red)

Immunohistochemistry (green)

RNA extraction and expression (purple)



Immunohistochemistry for Nogo-B

Stripping and re-blotting for endogenous control (actin)

Mouse kidney sections

Band quantification using Image J

Western blotting for Nogo-B:



Protein assay to determine protein concentration

Protein extraction

Exposure to Stretch/High glucose or Ang-2

Cell culture (Human podocytes and human mesangial cells)

Thawing cells2.2 Materials and products

All lab consumables were purchased from Appleton Woods Ltd (Birmingham, UK) and chemicals from Sigma Chemical Co. (Poole, UK) unless otherwise stated.

2.3 Cell culture

Culture of primary human mesangial cells (HMCs) from kidney tissue

HMCs were obtained from normal renal cortices of donor nephrectomies that were deemed unsuitable for transplantation due to abnormal vasculature. Intact glomeruli were obtained by homegenization and serial sieving of the cortices. HMCs were dissociated from the glomeruli by digestion with Collagenase IV as follows:

Trypsinize cells as normal and pellet the cells by centrifugation. Remove supernatant.

Add an equal volume of collagenase and leave for 5-10 mins at 37°C. Tap the tube occasionally to mix the solution and aid digestion.

When cells appear adequately dissociated dilute out the collagenase with media/PBS. It does not require deactivating with FBS.

Filter the cells using sieves to remove any remaining clumped cells.

(Any remaining collagenase can be re-frozen)

Cells were then seeded in 25cm2 culture flasks. Once HMC stated growing, the glomeruli were removed by washing with PBS and the cells cultured in RPMI-1640 medium, supplemented with Insulin-transferrin-selenium (ITS) (Sigma I-3146), L-glutamine, 20% heat-inactivated FCS, 7mM glucose, 100 units/ml penicillin and 100µg/ml streptomycin (ICN, Hampshire, UK). Cells were incubated in a humidified incubator at 37°C with 5% CO2.

Culture of conditionally immortalized human podocytes

Human podocytes were obtained from… (gift from Moin Saleem, Bristol). Cells were left at room temperature for approximately 1 minute then transferred to a 37°C water bath for 1-2 minutes until they were fully thawed. This process was done quickly to minimise any damage to the cell membranes that could occur as a result of toxic effects of DMSO (dimethyl sulfoxide) which the cells are stored in, along with FBS. The cells were then slowly pipetted into a 75cm2 flask containing pre-warmed medium in order to achieve the correct volume for the flask. Cells were cultured in complete media (RPMI-1640, Sigma R-8758) supplemented with ITS, FBS and Pen/Strep. They were incubated at 33°C with 5% CO2 with the medium being refreshed every 2-3 days.


When the cells were almost confluent there were split 1/8. The cells were first washed with PBS to wash away FBS as it is known to inhibit the actions of trypsin. A low concentration of trypsin/EDTA was exposed to the cells for a short a time as possible in order to minimise cell damage as a result of trypsin digesting protein within the cells.

Trypsinisation of cells

When the cells had almost formed a confluent monolayer they were split from a 25cm2 flask to two larger 75cm2 flasks containing new medium. This was done to limit cell death and allow a larger population of cells to be grown. This process is known as splitting or trypsinisation and uses trypsin, a serine protease or digestive enzyme, used to re-suspend HMCs adherent to the flask surface.

The cells were first washed with PBS to remove divalent ions such as Ca2+ and Mg2+ present within the medium, which would hinder the functioning of trypsin. PBS was not directly applied to the cells in order to limit any physical damage. Subsequently, 1.5mls of trypsin containing ethylenediaminetetraacetic acid (EDTA) was added to the cells in the flask which was incubated at 37°C for 5mins. EDTA, a calcium chelator was included in the solution in order to enhance the functioning of trypsin. The flask was then tapped in order to ensure as many cells had re-suspended as possible. The cells were checked under a microscope to confirm the majority of cells had detached from the flask; cells in suspension would be seen as bright round cells, whereas cells attached to the flask would be seen as dark, spindle shaped cells. In order to stop the reaction and prevent cell damage, complete media was added and pipetted up and down the flask. Being a protease, trypsin also digests the protein and polypeptides in the cells so by adding more media, trypsin begins to digest the proteins in the medium becoming neutralised and thus inert. The re-suspended cells in the fresh medium and neutralised trypsin were pipetted into the two T75 flasks. Before discarding the T25 flask, the new passage number was recorded on the T75 flasks. The two T75 flasks were then sprayed with 70% ethanol and replaced in the incubator. These cells were subcultured into several T75 flasks before being plated into well plates for treatment.

2.4 Treatment of cells

Application of mechanical stretch

Mesangial cells were seeded in equal number into six-well type I collagen-coated silicon elastomer-base culture plates (Flex I plates) and control plates (Flex II plates). After insulin and serum deprivation for 24 h, cells were subjected to repeated stretch/relaxation cycles by mechanical deformation using a Flexercell Strain Unit (FX3000; Flexcell Int.) (Fig 2.2). The stress unit is a modification of the unit initially described by Banes et al.and consists of a vacuum unit and a base plate. A vacuum was cyclically applied (60 cycles/min) to the rubber base plates via the base plate, which was placed in a humidified incubator with 5% CO2 at 37°C. Cells were exposed to an average 10% uniaxial elongation, which mimics that present in vivo in glomeruli exposed to supernormal pressure levels. Control cells were grown in nondeformable but otherwise identical plates (Flex II plates). We applied a cyclical mechanical stretch on the evidence that, in the normal glomerulus, capillary pressure is pulsatile and that, in situations such as diabetes, this pulsatility may be enhanced because of defective autoregulation. A rate of 60 cycles/min which approximates the pulse frequency, has been used in previous studies on mesangial cells exposed to stretch. A subset of experiments were performed on human mesangial cells that were exposed to high glucose concentrations for 48 h. Media that contained normal (5.5 mM) and high (25 mM) glucose concentrations were made iso-osmolar with the addition of mannitol. Stretch was applied during the last 4 h of the glucose incubation period.

Fig. 2.2: Picture of Flexercell Strain Unit used to apply mechanical stretch to HMCs and podocytes to simulate the biological strain conditions of glomerular hypertension which is a key feature of DN

Application of high glucose

A subset of experiments was performed on Hpods and Hmcs that were exposed to high glucose concentrations for 48h. Media that contained normal (5.5 mM) and high (25 mM) glucose concentrations were made iso-osmolar with the addition of mannitol. Stretch was applied during the last 4 h of the glucose incubation period.

Addition of Ang-2

When podocytes were almost confluent, the growth medium was replaced with 1% serum media. Cells were incubated at …°C overnight. The following day four different concentrations of angiopoietin (0, 100, 200, 400 ng/ml) were added to the cells in 1% serum media for the following times- 1hr, 6hrs, 24hrs and 48hrs. At the indicated time points, cells were washed with PBS and lysed using 50µl of RIPA. The lysis buffer also contained PMSF (…) for… and protease and phosphatase inhibitors in order to prevent… The cell lysates were then frozen at -40°C until required for experimental work.

Fig. 2.3: Diagram showing layout of 24-well plate used for addition of Ang-2 to human podocytes. Incubation times shown on top (1hr, 6hrs, 24hrs, and 48hrs), while Ang-2 concentrations displayed as they were within each well (Vehicle/0, 100, 200 and 400ng/ml).

2.5 Extraction of protein lysates for western blotting

Each well was washed with ice-cold PBS twice while the plates were kept on ice. The plates were then removed from the ice and 150µl of sodium dodecyl sulphate (SDS) lysis buffer was added to the middle of each well. The plates were swirled in order to ensure the SDS buffer covered the whole well. The lysates were then transferred from each well into eppendorfs labelled in the same way as shown in figure 2.1 and were stored at -20⁰C.

The location of Nogo-B protein in mouse kidney sections was determined by immunohistochemical staining. Nogo was found to be present in the renal tubular epithelial cells and glomeruli (Fig. 3.1). This finding is consistent with those of Marin et al, 2010 who also found significantly positive staining in tubules and mild staining for Nogo in the glomeruli.

RT (reverse transcriptase)-PCR was carried out in order to determine if Nogo-B is expressed in vitro in Hpods and Hmcs at the DNA level, as there is no evidence for this in the literature to date. Nogo-B expression was detected in both cell types exposed to stretch/no stretch.

Findings from western blot analysis showed an overall increase in Nogo protein expression in Hpods exposed to long and short-term stretch/glucose and in Hmcs exposed to long-term stretch/glucose. However, Hmcs exposed to short-term stretch/glucose displayed a decrease in Nogo expression. Based on the findings of Marin et al, 2010 where renal injury in the form of UUO caused increased Nogo expression, one would expect high glucose and stretch conditions which simulate the biological effects of diabetic nephropathy to also increase Nogo expression. Therefore the increase seen in Hpods and Hmcs in the long-term experiment is plausible, while the decrease in Hmcs exposed to ST glucose/stretch was unexpected. This can perhaps be explained by a cell-type specific response.

The addition of Ang-2 to Hpods for increasing lengths of time caused a progressive rise in Nogo-B expression. However at the longest exposure (48hrs) increasing concentrations of Ang-2 caused a stepwise decrease in Nogo expression. This was an interesting finding being consistent with previous research by Gnudi et al, 2007 (Preliminary unpublished data) where chronic Ang-2 over expression caused an approximate 6-fold down regulation of Nogo-B expression.

As Ang-2 over expression in podocytes has been shown to cause proteinuria and glomerular endothelial apoptosis (Davis et al, 2007) the findings from my study are in line with the idea that renal injury causes Nogo expression to be up regulated in an attempt to protect the against further tissue damage. This may also explain why the greatest incubation time (48hrs) with Ang-2 caused a decrease in Nogo-B expression. It is possible that long-term Ang-2 exposure tips the balance resulting in the loss of Nogo expression which causes in renal damage. But despite the significance of these findings, they are based on only 2 sets of experiments, so more work is needed before any valid conclusions can be reached.

4.2 Variations in experimental results

Despite the general trend of increased Nogo-B expression following stretch and high glucose exposure in Hpods the decrease seen in Hmcs was unexpected. The increase in Nogo expression seen following Ang-2 addition was also an interesting finding with the final stepwise decrease at 48hrs incubation being unexpected but perhaps physiologically plausible. These variations in the results may in fact be valid or they may be the result of human error or other problems in the experimental technique discussed below.

Some very small volumes had to be pipetted when preparing reagents and solutions, this may have caused inaccuracies in concentrations, and it is possible that cells may have received a different concentration of treatment condition to what was required. This could also have contributed to the variations seen in the results. Furthermore, samples loaded onto gels for electrophoresis were also very small in volume, thus difficult to pipette. This could have resulted in an inaccurate amount of protein loaded onto the sample; however, this was reflected in the thickness of the α-actin bands.

On some occasions, the transfer of proteins from the gels to nitrocellulose membranes did not occur properly. It is possible that this was due to bubbles between the layers, although the layer were rolled and compressed in order to prevent bubbles forming between layers. Incomplete transfer would lead to a decreased value in densitometric analysis; however, this again would have been corrected for by a reduced thickness of the α-actin bands, thus it is thought that values of the ratios were not affected due to this in this study. It is also possible that background marks on the membranes may have affected densitometric analysis of the bands if they were very close to the bands; it was not possible to get rid of any background using the Image J software.

2 bands-

4.3 Implications of results

IHC findings

RT PCR findings

Short-term mechanical stretch and high glucose exposure to mimic the glomerular HTN and hyperglycaemia of DN causes increased Nogo-B expression in Hpods. This is consistent with previous work by Marin et al, 2010 where unilateral ureteral obstruction (UUO) was the form of renal injury performed, which resulted in increased Nogo expression in experimental mice in comparison to wild type mice. However in my study, a decrease in Nogo expression was observed in Hmcs following short-term stretch and high glucose exposure. This finding was unexpected and may suggest… The results from the long-term stretch and high glucose experiments were roughly consistent in both cell types in that there was a general increase in Nogo-B expression. It is possible that long-term exposure to experimental stimuli is necessary to simulate the biological effects of DN which in human patients presents as a complication of diabetes of long duration (Marshall, 2004).

4.4 Future directions

Due to time constraints in this project, both the short and long-term stretch/glucose experiments were only carried out on one set of samples. Ideally these should have been repeated on two or more samples to produce sufficient data for statistical analysis. More time would also allow further investigation of the two bands produced from western blotting. As mentioned, the 67kDa band was concluded to represent Nogo-B due to its consistent appearance with the use of three different primary antibodies. However in order to confirm that this band was not the result of apoptosis REF of Nogo a caspase inhibitor could be used.

As the band was of a larger molecular weight than the expected 49kDa band, it is possible that the protein of interest was affected by post-translational modification, for example glycosylation or phosphorylation. In order to check whether the former was occurring, PNGase/EngoGase could be used, while a phosphatase enzyme could be used to eliminate the latter effect. In addition, the protein being detected may have been a splice variant produced due to alternative splicing which creates different sized proteins from the same gene. On the other hand, it could have been a multimer formed from the dimerisation of a protein. Although this is usually prevented in reducing conditions, strong interactions can result in the appearance of a higher band. Finally to confirm that the 67kDa was in fact representative of Nogo-B, a control experiment could be carried out using the blocking peptide of the Nogo primary antibody.

IHC to determine effects of stretch, high glucose and Ang-2 on location of Nogo-B expression and to visualise any changes in renal structure e.g. ECM accumulation… is there any evidence to show that these even occur with stretch and high glucose i.e. are these even effective at simulating phys and biolog effects of DN- if not may need more sound model of DN

Try in vivo work i.e. mice/rats with induced DN (streptozocin), look for changes in Nogo expression at protein level compared to WT controls, is this possible?!

If nogo is upregulated in renal injury- how? Need to find mechanisms responsible/involved in its upregulation because if it has preventative role in kidney injury maybe we can understand how to target its precursors to maintain its expression in long term kidney disease/DN?

qPCR to look at effects of stretch, glucose and Ang-2 on Nogo expression at DNA level

4.5 Conclusions

The role of Nogo-B in the kidney is not yet fully elucidated; however this study supports the hypothesis that Nogo expression is up regulated in human in vitro models of renal injury. This study was the first to suggest a potential protective role of RTN-4 in diabetic nephropathy. Marin et al, 2010 were the only other group to describe an increase in Nogo expression in murine models of acute renal injury in the form of unilateral ureteral obstruction (UUO). The findings from this study open up a whole range of possible future research to both consolidate these findings and investigate whether Nogo-B up regulation is a general feature of kidney injury. If so, Nogo-B may be introduced as a novel marker of renal impairment, as suggested by Marin et al, 2010 which would undoubtedly be a breakthrough in the field of nephrology. Furthermore if Nogo-B expression is found to be up regulated in renal disease, this will confirm its potential protective properties against tissue damage in the same way as it has been shown to limit the progression of vascular damage following injury (Raines et al, 2004). However a great deal of research will be needed before such results can be translated into therapeutic targets and offer any clinical implications.