In the past 10 years many new associations between diabetes mellitus (DM) and the skin has been noted. Dermatologic problems are common in diabetes, with approximately 30% of patients experiencing some cutaneous involvement during the course of their illness1.
Diabetes mellitus is a heterogeneous disease characterized by micro vascular pathology leading to chronic complications2.
A therapy that normalizes the metabolic dysfunctions in diabetes should be expected to prevent, delay or substantially reduce the severity of diabetic long term microvascular complications, improving the quality of life3.
Chromium picolinate is a nutritional supplement that works to increase the efficiency of insulin to optimal levels. Gaining increased popularity in the United States, this supplement has been touted a miracle mineral, including weight loss, mood enhancement, energy promotion, increase in life span and even the prevention of acne4.
Chromium is a naturally-occurring mineral, trace amounts of which are found in a wide range of foods like egg yolk, meat, coffee, nuts, broccoli, yeast and wholegrain products5.
Growing evidence has been suggested that chromium supplementation, might improve insulin sensitivity and glucose metabolism in patients with glucose intolerance and type 1, type 2, gestational and steroid-induced diabetes and in some individuals without diabetes6.
The role of dietary manipulation, vitamins and minerals in the prevention and treatment of certain skin diseases was reviewed. Manipulation of nutrition by either diet restriction or supplementation could therefore affect cutaneous disorders such as skin cancer, wound healing, atopic dermatitis, psoriasis, diabetic dermatopathology and dermatitis herpetiform7.
So, the aim of the present study was to investigate the efficacy of chromium picolinate in ameliorating diabetesinduced histological skin alterations.
MATERIALS AND METHODS
Twenty five adult male albino rats with an average weight of 250 gm were housed five per cage with unlimited access to laboratory food and water.
The rats were divided into 2 groups:
Group I: Formed of 10 rats and further subdivided into 2 subgroups 5 animals each.
Subgroup Ia: Served as a control group.
Subgroup Ib: Formed of 5 rats that received chromium orally at a dose of 25ug/kg/day for 8 weeks6.
Group II: Formed of 15 streptozotocin (STZ) induced- diabetic rats. They were further subdivided into 3 subgroups 5 animals each.
Subgroup IIa: Formed of STZ-induced diabetic rats.
Subgroup IIb: Formed of STZ-diabetic rats which were subcutaneously injected with insulin at a dose 10u/kg/day for 8 weeks (the dose of insulin was calculated according to table of comparison between human and animals)8.
Subgroup IIc: Formed of STZ-diabetic rats that were subcutaneously injected with insulin at a dose 10u/kg/day. Animals were given oral chromium at a dose of 25ug/kg/day for 8 weeks.
Induction of diabetes:
The rats were intraperitonealy injected with 35mg/kg of streptozotocin in 0.9% sodium chloride solution. Streptozotocin (STZ) was available in powder form 1gm produced by Sigma Company.
Although, the range of the diabetogenic dose of STZ ranged from 40 to 9o mg/kg9. Yet, the present study selected 35mg/kg STZ as the minimum and safest dose for induction of diabetes (to avoid animal loss). Forty eight hours after the injection of STZ, animals exhibiting overt polydepsia, polyuria and polyphagia were considered diabetic, confirmed by measuring their fasting blood glucose levels.
Rats with fasting blood glucose levels over 250mg/dl were included in the study as diabetic rats.
Chromium was available in capsules containing 200ug chromium picolinate produced by MEPACO Company. The capsules were dissolved in distilled water and given at the previous dose to the animals orally by a gastric tube.
Insulin was available in vial form. Each vial contained 40u/ml produced by ALameriya Company Egypt.
The animals were housed in animal house in Bilharizial Institute of Researches, Faculty of Medicine, Ain Shams University.
Blood glucose level assay:
At the beginning of experiment, at start of polyuria and at the end of 8 weeks, blood samples were collected by taking a drop of blood from the tail. Fasting blood glucose level was analyzed using enzymatic colorimetric assay kits (spinreact).
At the end of experiment, the animals were sacrificed. All the animals were prepared by cutting the hair over an area 2 X 2 cm over the back. Specimens of thin dorsal skin from the prepared areas were dissected out. The specimens were prepared for the following:
- Electron microscopic study.
- Immunohistochemistry for CD34 of vascular endothelial cells.
For light microscopic study, samples were taken; after fixation in 10% formalin, they were dehydrated through alcohols, cleared in xylene and embedded in paraffin wax. Later on, 5um thick sections were stained with haematoxylin and eosin10.
For electron microscopy, other parts of the specimens were fixed in 3% phosphate buffered glutaraldehyde (ph 7.3). Then, post fixed in 1% osmium tetra oxide at 4c, dehydrated and embedded in epoxy resin. Semi thin sections were cut and stained with H&E10. Ultra thin sections were stained with uranyl acetate and lead citrate11 and examined and photographed by transmission electron microscope (Jeol-Ex1010 TEM) in Al Azhar University.
For immunohistochemical study, staining was performed for detection of the transmembrane protein CD34 on endothelial cells.
The hematopoietic cell antigen (CD34) is a 105-120 KD, transmembrane cell surface glycoprotein encoded by a gene located in the Iq31 region 1. It reacted with several non-hematopoietic tissues including vascular endothelium and fibroblasts in connective tissue. So it could be a valuable marker for vasculature12.
Primary antibodies monoclonal mouse anti-human CD34 class II antibody was purchased from Dako Carpenteria, Ca, USA. Sections were treated with 0.01 m citrate buffer (ph 6.0) for 10 minutes to unmask antigen. Sections were incubated in 0.3 % hydrogen peroxide for 30 min to abolish endogenous peroxidase activity. Blocking with serum blocking solution for 10 minutes was done. Slides were incubated with the primary antibody (1:500 monoclonal mouse anti-CD34) at room temperature for 2 hours. After washing, they were incubated with biotinylated secondary antibodies (ABC kit, 1:200) and then with avidin-biotin complex. Finally, sections were developed with 0.05% diaminobenzidine. Slides were counterstained with Mayer's haematoxylin, dehydration, clearing and mounting were done10.
Plasma membrane of endothelial cells and connective tissue fibroblasts containing CD34 appeared brown, while nuclei appeared blue in color.
For morphometric and statistical analysis, epidermal and dermal measurements were done. Five non overlapping fields from 5 different sections of five different rats were examined in each group at X 400 (number of readings were 25). The measurements were done by the image analyzer using Leica Q500 MC program installed on a PC. The PC was connected to a camera attached to Zeis universal microscope. Histology Department, Faculty of Medicine, Ain Shams University.
Total thickness of nucleated epidermal keratinocytes.
Area percentage occupied by CD34 transmembrane protein of vascular endothelial cells in:
- The papillary dermis.
- The reticular dermis.
For statistical analysis, all data were collected, revised and then subjected to statistical analysis using student's “t”test. The significance of the data was determined by P value, where P>0.05 non-significant (NS), P<0.05 significant (S) and P<0.001 highly significant (HS).
Group I: (Control group):
Examination of sections of thin skin of subgroups Ia & Ib appeared nearly similar. H&E stained sections showed the skin to be formed of 2 principal layers, the epidermis and its subjacent connective tissue layer, the dermis. Beneath the dermis there was a layer of loose connective tissue formed largely of adipose tissue, the hypodermis. The dermis of the skin was formed of an outer papillary layer and an inner reticular layer. The dermis contained sweat glands, hair follicles with the associated sebaceous glands (Fig. 1). The epidermis was formed of keratinocytes arranged in four layers. The basal cell layer (stratum basale) resting on basement membrane and formed of low columnar cells with oval nuclei. The prickle cell layer (stratum spinosum) contained polyhydral cells with rounded nuclei. The granular layer (stratum granulosum) was formed of spindle shaped cells which contained deeply basophilic granules. The horny layer (stratum corneum) consisted of many layers of very flat, homogenous, keratinized cells containing no nuclei (Fig. 2). In semi-thin sections, most of epidermal cells appeared with vesicular nuclei and narrow intercellular spaces (Fig. 3).
Immunohistochemical staining for detection of endothelial cells using CD 34, showed the plasma membrane of endothelial cells lining capillaries and blood vessels of the papillary and reticular dermis stained brown in color (Figs. 4-A,B).
Ultra-thin sections revealed the whole thickness of epidermis together with the underling dermis (Fig. 5). Basal cells appeared resting on the basal lamina, where the tonofilaments were condensed in form of hemidesmosomes and desmosomes (Fig. 6).
As the cells move on into the stratum granulare, coarser bundles of dense irregular keratohyaline granules appeared (Fig. 7).
Epidermal Langerhans cells were infrequently detected and recognized by their characteristic convoluted nuclei. The cytoplasm was of low electron density with absence of tonofilaments. It contained few mitochondria, lysosomes and the characteristic Birbeck granules which appeared rod like with a central linear density. No desmosomes could be seen on Langerhans cell surface (Figs. 8-A,B).
Subgroup IIa: (diabetic rats):
The epidermis of thin skin sections of animals of this subgroup showed significant reduction in the thickness of nucleated keratinocytes as compared to control (Table 1-A & Histogram 1). Most of epidermal cells appeared with deeply- stained shrunken irregular nuclei surrounded by pale clear cytoplasm (Fig. 9). In semi-thin sections, most of epidermal cells appeared with widened intercellular spaces (Fig. 10).
Immunohistochemical analysis revealed nearly complete absence of staining in some of the endothelial cells lining the papillary and reticular blood vessels (Figs. 11-A,B), together with significant reduction in CD34 area% (Tables 2-A,3-A & Histogram 2).
Ultra-thin sections revealed the basaly located cells at the dermo-epidermal junction with focal absence of hemidesmosomes (Fig. 12). Some cells showed perinuclear halos (Fig. 13). Extensive widening in the intercellular spaces was frequently detected (Figs. 13,14). Disruption of desmosomes with intercellular disjunctions was evident as well (Fig. 14).
Langerhans cells showed relative increase in the content of lysosomes and microvesicles compared to control (Figs. 15-A,B).
Subgroup IIb: (diabetic rats which were subcutaneously injected with insulin):
Thin skin sections of this subgroup appeared with significant decrease in the thickness of nucleated keratinocytes as compared to control, cytoplasmic vacuolation was also detected in some keratinocytes (Fig. 16, Table 1-A and Histogram 1). Semi-thin sections revealed that most of epidermal cells appeared with marked cytoplasmic vacuolations. However, more or less normal intercellular spaces between adjacent cells were frequently recognized (Fig. 17).
Examination of CD34 stained sections revealed significant reduction in the area % of the immune-reaction of endothelial cells lining the vascular network in the papillary and reticular layer when compared to control group (Figs. 18-A,B) (Tables 2-A,3-A & Histogram 2).
Ultra structurally, some keratinocytes appeared with large vacuoles occupying the whole cytoplasm and even indenting their nuclei. However, desmosomes between adjacent cells were frequently perfectly preserved (Figs. 19-A,B).
Subgroup IIc: (diabetic rats which were subcutaneously injected with insulin plus oral chromium):
Examination of histological sections of these rats revealed that most of epidermal cells preserved their normal characteristic picture (Fig. 20).
Moreover, significant increase in thickness of nucleated keratinocytes was also detected when compared to subgroup IIb (Table 1-B & Histogram 1). Semi-thin section, showed the keratinocytes of the different layers with their cytoplasmic projections extending across intercellular spaces to meet those of adjacent cells. However, slight widening of intercellular spaces could be recognized between some keratinocytes. occasional cells with a deeply-stained nucleui, pale cytoplasm and dendritic processes were seen extending among the cells of the stratum basale (Fig. 21) Langerhans cell could be predicted (Fig. 21).
Dermal blood vessels showed in-apparent change in immunoreactivity of CD34, as compared to control group (Figs. 22-A, 22-B), together with significant elevation in CD34 area% compared to subgroup IIb (Tables 2-B,3-B & Histogram 2).
Ultra-thin sections showed epidermal keratinocytes with numerous bundles of tonofilaments radiating from perinuclear region and ending in dense numerous desmosomes along the highly interdigitated cell boundaries. Slight widening of intercellular spaces was infrequently detected (Fig. 23).
Epidermal Langerhans cells were recognized by their characteristic convoluted nuclei which were irregular with prominent grooves and folds. It contained some lysosomes. The profile of their Birbeck granules occasionally had an expanded vesicular end, giving them a racquet- like appearance (Figs. 24-A, 24-B).
Student's t test:
There was no statistical significant difference between Ia and Ib as regards the mean thickness of nucleated epidermal keratinocytes P>0.05.
Subgroup IIa and IIb showed significantly lower mean when compared to subgroup Ia p<0.05.
There was no statistical significant difference between Ia and IIc as regards the mean thickness of nucleated epidermal keratinocytes P>0.05.
Subgroup IIc showed significantly higher mean when compared to subgroup IIb p<0.05.
Student's t test:
There was no statistical significant difference between Ia and Ib as regards the mean area % of CD 34 expression in papillary or reticular dermis P>0.05.
Subgroup IIa and IIb showed significantly lower mean when compared to subgroup Ia p<0.05.
There was no statistical significant difference between Ia and IIc as regards the mean area of CD 34 expression in papillary or reticular dermis % P>0.05.
Subgroup IIc showed significantly higher mean when compared to subgroup IIb p<0.05.
Student's t test:
There was no statistical significant difference between Ia and Ib as regards mean blood glucose level P>0.05.
Subgroup IIa, showed significantly higher mean when compared to subgroup Ia p<0.05.
There was no statistical significant difference between Ia and IIc as regards the mean blood glucose level P>0.05.
The results of the present study confirmed severe affection of thin skin of STZ-induced diabetic rats. This was demonstrated as significant reduction in the thickness of nucleated keratinocytes. Moreover, most of epidermal cells appeared with deeply- stained shrunken nuclei surrounded by clear cytoplasm. Immunohistochemical analysis revealed significant reduction in area % CD34 of vascular endothelial cells. Ultrastructurally, some epidermal cells showed perinuclear halos. Extensive widening in the intercellular spaces was frequently detected. Disruption of hemidesmosomes and intercellular disjunctions was evident as well. Langerhans cells showed features of increased activity represented by relative increase in amount of lysosomes and microvesicles. Similarly, many workers reported dermatologic manifestations of diabetes. They believed that, poor microcirculation, hypohydrosis, peripheral vascular disease, peripheral neuropathy and the decreased immune response could account for dermatological problems and poor wound healing observed in diabetics1,13,14.
On histopathological backgrounds, some authors proposed a relationship between various forms of blistering skin diseases and disruption of keratinocytes desmosomes15.
Dermal microvessels were qualitatively and quantitatively assessed by other workers. Endothelial cell proliferation was not noted in dermal diabetic microvessels. Microangiopathies were thus frequently recognized favoring the occurrence of diabetic dermopathy14,16.
The enhanced cellular activity of epidermal Langerhans cells was also observed by other investigators in cases of disruption of epidermal permeability barrier. They believed that, increased cellular activity of Langerhans might reflect their enhanced ability to capture exogenous substances17.
With the use of insulin in subgroup IIb, the structural changes were demonstrated to be improved but not completely. This was in accordance with other workers, who recorded that; tight glucose control in diabetics could prevent and minimize most of the skin diseases secondary to infection. However, the relationship between tight glucose control and prevention of noninfectious skin disorders is debatable1.
The current study revealed that oral intake of chromium was associated with insignificant increase in fasting blood glucose level compared to control group. Moreover, histological examinations revealed considerable reduction in the severity and incidence of changes observed in subgroups IIa & IIb. Most of epidermal cells regained their light and electron-microscopic characteristic picture. Significant increase in thickness of nucleated keratinocytes was detected as well.
Dermal blood vessels showed bright labeling of CD34 together with significant elevation in its area %.
Several studies concurred that chromium could decrease fasting blood glucose and LDL-cholesterol levels in STZ-induced diabetic rats. Therefore, chromium might considerably improve carbohydrate and lipid metabolism amongst diabetic patients9,18.
It had been also hypothesized that chromium might improve glucose tolerance by one or more of the following ways: increasing the number of insulin receptors; increasing binding of insulin to its receptors and increasing activation of the receptors in the presence of insulin19.
More recently, they realized that after chromium absorption in the gastrointestinal tract, chromium could most likely be transported to cells bound to the plasma protein transferrin. Chromium could initiate insulin transport into the cells through its binding to the oligopeptide apochromodulin. This oligopeptide combined with four chromium (III) atoms to form chromodulin, which is important for amplifying the insulin signalling effect. After binding to insulin-activated receptor, chromodulin increased tyrosine kinase activity by one order. Therefore, chromium deficiency was thought to participate in insulin resistance and hyperlipidaemia observed in diabetic patients20.
Another mechanism of action was also postulated, chromium could increase plasma membrane fluidity through its effect on cholesterol. Consequently, an increase in insulin sensitivity and its rate of internalization might be expected21,22.
On the other hand, the interaction between glycemic control, diabetic vascular complications and the influence on most of the free radicals and antioxidant indices was reported by many workers. Their data suggested that, chromium supplementation was an effective treatment strategy that could minimize increased oxidative stress in diabetes mellitus patients23–25. Many workers reported that, chromium supplementation offered a decrease in Tmor necrosis factor -alpha, interleukin-6 and C-reactive protein in diabetic rats26. Others suggested that chromium picolinate could induce apoptosis to peripheral blood lymphocytes in diabetics determined by TUNEL positivity and ultra structural characteristics such as nuclear condensation and formation of apoptotic bodies27. The role of chromium in lowering the risk of vascular inflammation in diabetes was therefore expected.
Moreover, the long tern safety of chromium picilonate had been investigated by many workers. They realized that chromium was found to have no hepatotoxic or nephrotoxic potential. It was rather found to improve renal and hepatic function and to reduce abnormalities associated with Streptozotocin diabetes9.
In summary, chromium picilonate was rather found to effectively improve and reduce dermal lesions associated with diabetic state in STZ- rats. Hence, this study helped to establish the efficacy of chromium picilonate in preventing long term diabetic complications.
Ahmed I and Goldstein B. (2006): Diabetes mellitus
. Clin. Dermatol. Jul-Aug;24(4):237-246.
DeFronzo RA. (1988):
Lilly lecture 1987. The triumvirate: Beta-cell, muscle, liver. A collusion responsible for NIDDM. Diabetes Jun;37(6):667-687.
Bailey CJ. (2000):
Potential new treatments for type 2 diabetes. Trends Pharmacol.Sci. Jul;21(7):259-265.
Reading SA. (1996): Chromium picolinate
. J. Fla. Med. Assoc. Jan;83(1):29-31.
Rezende Gomes M, Macedo Rogero M and Tirapegui J. (2005):
Considerations about chromium, insulin and physical exercise [Considerações sobre cromo, insulina e exercício físico]. Rev. Bras. Med. Esporte;11(5):246e-250e.
Cefalu WT and Hu FB. (2004):
Role of chromium in human health and in diabetes. Diabetes Care;27(11):2741-2751.
Rackett SC, Rothe MJ and Grant Kels JM. (1993):
Diet and dermatology. The role of dietary manipulation in the prevention and treatment of cutaneous disorders. J.Am.Acad.Dermatol. Sep;29(3):447-461.
Paget G and Bernes E. (1964):
Toxicity tests. In: Laurence DR, Bacharach AL, editors. Evaluation of drug activities. Pharmacometrics: Academic Press, London. p. 50.
Shinde UA and Goyal RK. (2003):
Effect of chromium picolinate
on histopathological alterations in STZ and neonatal STZ diabetic rats. J. Cell.Mol.Med. Jul-Sep;7(3):322-329.
Bancroft JD and Cook HC. (1994):
Manual of histological techniques and their diagnostic application. 1st
ed. Churchill Livingstone.
Glauret AM and Lewis PR. (1998):
Biophysical, specimen preparation for transmission electron microscopy. Parhand Press: London.
Van de Rijn M, Rouse RV. (1994):
CD34. A review. Appl. Immunohistochem.;2:71-80.
Goodfield MJ and Millard LG. (1988):
in diabetes mellitus
. Diabetologia Aug;31(8):567-575.
Fisher ER and Danowski TS. (1968):
Histologic, histochemical and electron microscopic features of the shin spots of diabetes mellitus
. Am.J.Clin.Pathol. Nov;50(5):547-554.
Klymkowsky MW and Parr B. (1995):
The body language of cells: The intimate connection between cell adhesion and behavior. Cell Oct 6;83(1):5-8.
Yasuda H, Taniguchi Y, Yamashita M, Kobayashi N, Sonobe M, Hatanaka I, Kikkawa R and Shigeta Y. (1990):
Morphological characteristics of dermal diabetic microangiopathy. Diabetes Res. Clin.Pract. May-Jun;9(2):187-194.
Falck B, Andersson A and Bartosik J. (1985):
Some new ultrastructural aspects on human epidermis and its Langerhans cells. Scand.J.Immunol. May;21(5):409-416.
Lai MH, Chen YY and Cheng HH. (2006):
Chromium yeast supplementation improves fasting plasma glucose and LDL-cholesterol in streptozotocin
-induced diabetic rats. Int.J.Vitam. Nutr.Res. Nov;76(6):391-397.
Vincent JB. (2000):
Quest for the molecular mechanism of chromium action and its relationship to diabetes. Nutr.Rev. Mar;58(3 Pt 1):67-72.
Racek J. (2003):
Chrom jako biogenni prvek. [Chromium as an essential element]. Cas.Lek.Cesk.;142(6):335-339.
Evans GW, Bowman TD. (1992): Chromium picolinate
increases membrane fluidity and rate of insulin internalization. J. Inorg.Biochem. Jun;46(4):243-250.
Horvath EM, Tackett L, McCarthy AM, Raman P, Brozinick JT and Elmendorf JS. (2008):
Antidiabetogenic effects of chromium mitigate hyperinsulinemia-induced cellular insulin resistance via correction of plasma membrane cholesterol imbalance. Mol. Endocrinol.;22(4):937-950.
Cheng HH, Lai MH, Hou WC and Huang CL. (2004):
Antioxidant effects of chromium supplementation with type 2 diabetes mellitus
and euglycemic subjects. J.Agric.Food Chem. Mar 10;52(5):1385-1389.
Komosinska Vassev K, Olczyk K, Olczyk P and Winsz Szczotka K. (2005):
Effects of metabolic control and vascular complications on indices of oxidative stress in type 2 diabetic patients. Diabetes Res.Clin.Pract. Jun;68(3):207-216.
Esen Gursel F, Tekeli SK. (2009):
The effects of feeding with different levels of zinc and chromium on plasma thiobarbituric acid reactive substances and antioxidant enzymes in rats. Pol.J.Vet.Sci.;12(1):35-39.
Jain SK, Rains JL and Croad JL. (2007):
Effect of chromium niacinate and chromium picolinate
supplementation on lipid peroxidation, TNF-α, IL-6, CRP, glycated hemoglobin, triglycerides and cholesterol levels in blood of streptozotocin
-treated diabetic rats. Free Radic.Biol.Med.;43(8):1124-1131.
Jana M, Rajaram A and Rajaram R. (2009): Chromium picolinate
induced apoptosis of lymphocytes and the signaling mechanisms thereof. Toxicol.Appl. Pharmacol.;237(3):331-344.
Keywords:© 2010 The Egyptian Journal of Histology
Chromium picolinate; skin; streptozotocin; diabetes mellitus; immunohistochemistry.