Clinica Chimica Acta, 147 (1985) 1-5
`Elsevier
`
`CCA 03117
`
`Glycosylation of nail in diabetics: possible marker of
`long-term hyperglycemia
`
`Ebubekir Bakan * and Nuri Bakan
`Department of Biochemistry, Medical School of Ataturk University, Erzurum (Turkey)
`
`(Received September 4th, 1984; revision November 22nd, 1984)
`
`Key words: Glycosylation; Nail; Diabetes mellitus
`
`Summary
`
`Fingernail samples from 32 patients with diabetes mellitus and from 26 non-di(cid:173)
`abetics were analyzed in order to determine the protein glycosylation rate in nail.
`Nail glycosylation was assayed by the thiobarbituric acid reaction. Blood was taken
`from both diabetics and non-diabetics at the same time for measurement of
`hemoglobin glycosylation. In non-diabetics, the protein glycosylation in nail and
`glycosylated hemoglobin were found to be 8.35 ± 2.7 nmol fructosamine/mg nail
`and 2.24 ± 0.45 µmol fructosamine/g hemoglobin, respectively. In diabetics, how(cid:173)
`ever, there was an extremely high glycosylation in both nail protein and hemoglobin:
`16.0 ± 7.35 nmol fructosamine/mg nail and 5.17 ± 1.17 µmol fructosamine/g hemo(cid:173)
`globin ( p < 0.001 for both). A significant correlation was found between nail
`glycosylation and glycosylated hemoglobin in diabetics (r = 0.923, p < 0.001). Also,
`there was a correlation between diabetic fasting blood glucose and protein glycosyla(cid:173)
`tion in nail (r = 0.947, p < 0.001). Our findings show that it might be useful in the
`investigation of microvascular complications of diabetes mellitus to evaluate the
`possibility of nail glycosylation providing a stable long-term measure of tissue
`glycosylation.
`
`Introduction
`
`Glycosylated proteins, especially hemoglobins, have gained acceptance as an
`accurate index of long-term blood glucose control in diabetic patients [1-4]. The
`discovery of increased non-enzymatically glycosylated hemoglobin concentrations in
`disease has led to intensive research into similar excess glycosylation of other tissue
`
`* Correspondence to: E. Bakan, Atatiirk Universitesi, Tip Pak. Biyokimya Anabilim Dali, Erzurum,
`Turkey.
`
`0009-8981/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
`
`Page 1
`
`Anacor Exhibit 2042
`Flatwing Pharmaceuticals, Inc. v. Anacor Pharmaceuticals, Inc
`IPR2018-00168
`
`

`

`2
`
`proteins [5-9], because there may be a link between glycosylation and chronic
`complications of diabetes mellitus. In assessing diabetic control, measurement of
`percentage concentration of glycosylated hemoglobin has proved useful, but tissue
`collagen, being susceptible to functional changes from excessive glycosylation, has a
`much slower turnover rate than hemoglobin. We have conducted such a study
`because it is of interest to investigate possible longer term measures of chronic
`hyperglycemia other than hemoglobin.
`
`Subjects and methods
`
`The patients comprised 32 insulin-dependent diabetics (age range 20-76 yr, 21
`females). Twenty-six non-diabetic subjects (age range 17-50 yr, 10 females) served
`as controls. Blood samples were taken from both groups for measurement of
`glycosylated hemoglobin, whole blood hemoglobin, and fasting blood glucose con(cid:173)
`centrations at the time of nail sampling. Fingernail samples, about 20 mg, were
`obtained from healthy ones by paring away finely at the surface, but not the tip, of
`them. We avoided dirt on the nails.
`Hemoglobin was assayed by the cyanmethemoglobin method, and glucose by the
`glucose oxidase method (Boehringer, GOD perid method).
`Glycosylated proteins were measured by the colorimetric method of Parker and
`co-workers [10]. The method was modified in accordance with our experimental
`conditions. All assays were duplicated and conducted in one series. Another series
`was undertaken to determine the interassay coefficient of variation.
`
`Measurement of nail glycosy/ation
`Fingernail samples pared away finely were carefully weighed out into 5 mg
`portions and incubated for two hours at 120°C in an autoclave (Marsh Inst. Co.,
`Leave, Spacey, 0-150°C, 0-4 lbs/in2 ) with 2 ml of oxalic acid solution (0.5 mol/1).
`When cool 2 ml of 40% trichloroacetic acid was added to each tube, the tubes
`centrifuged, and two 1.5-ml samples of supernatant taken. One was incubated with
`0.5 ml of thiobarbituric acid solution (0.05 mol/1), and the other with 0.5 ml distilled
`water for 30 min at 40°C. After cooling to room temperature, the color produced
`was measured by spectrophotometry at 443 nm, the distilled water incubate serving
`as a sample blank. Serial dilutions of 200 µmol fructose and 5-hydroxymethylfur(cid:173)
`fural (5-HMF) per liter to 10-80 µmol/1 were tested to obtain a standard curve for
`each assay series (Fig. 1 ). The intra-assay coefficient of variation was found to be
`2.8% for control and 4.3% for diabetic samples, and that of interassay 4.9% and
`8.2%, respectively.
`
`Statistics
`Results were evaluated using least mean squares regression analysis for cdrrela(cid:173)
`tions, with analysis of variance and paired and unpaired t tests for difference
`between groups; for the difference between glycosylation of normal and diabetic
`nail, an unpaired t test allowing for dissimilar variances was used.
`
`Page 2
`
`Anacor Exhibit 2042
`Flatwing Pharmaceuticals, Inc. v. Anacor Pharmaceuticals, Inc
`IPR2018-00168
`
`

`

`0.40
`
`3
`
`C:
`"l
`
`D.30
`
`( o J ; 5-HMF
`( •) : Fructose
`
`--E
`....
`....
`..... .. g
`~ 0 .a
`
`0.20
`
`,q:
`
`0.10
`
`10 20 30
`
`,o 50 60 70
`,µmo/IL
`
`80 90 100
`
`Fig. 1. Standard curves obtained from autoclaved fructose and 5-HMF.
`
`Results
`
`The ranges of glycosylation of nail and hemoglobin, fasting blood glucose, and
`hemoglobin are shown in Table I. The mean nail glycosylation, as well as glyco(cid:173)
`sylated hemoglobin in the diabetics, was significantly higher than in the control
`subjects (p < 0.001). There was no alteration in the range of nail glycosylation with
`age and sex in either group of subjects. Nail from diabetics was found to be
`glycosylated in proportion to the hemoglobin glycosylation, that is, ,there was a good
`correlation (r = 0.923, p < 0.001) (Fig. 2). Hemoglobin values in diabetics and
`controls had no correlation with nail and hemoglobin glycosylation, while the
`
`TABLE I
`
`Mean values of parameters in non-diabetics and diabetics
`
`Hemoglobin (g/dl)
`Fasting blood glucose (mg/di)
`Glycosylated hemoglobin
`(µmol fructosamine/g Hb)
`Glycosylated nail
`(nmol fructosamine/mg nail)
`
`ns, Not significant.
`
`Non-diabetics
`(n = 26)
`14.5 ±1.15
`79.6 ±8.4
`
`Diabetics
`(n = 32)
`14.1 ± 1.26
`222.2 ±34.7
`
`2.24±0.45
`
`5.71± 1.17
`
`8.35±2.7
`
`16.0 ± 7.35
`
`p
`values
`
`ns
`< 0.001
`
`< 0.001
`
`< 0.001
`
`Page 3
`
`Anacor Exhibit 2042
`Flatwing Pharmaceuticals, Inc. v. Anacor Pharmaceuticals, Inc
`IPR2018-00168
`
`

`

`4
`
`450
`,o.o
`
`35-
`
`C:
`
`-0
`.....
`C: r 30.0
`-~,
`... " 25-0
`.!2 C:
`:,.,,·,
`"'E Ot, 20-0
`u"' ~o
`Ol°t - :::,
`. , ~
`:c:-0
`E
`.....
`C:
`
`0 ....
`
`15.0
`
`10-0
`
`5.0
`
`•
`
`.
`. •
`• . .
`. •
`. • . .
`•
`.. . ••
`
`•
`
`0
`
`1.0 2.0 3.0 4-0 s.o 6.o 1.0 s.o 9.o ,ao
`Glycosylated hemoglobin
`(JAmol fructosamine/g hemoglobin)
`Fig. 2. Relation between glycosylation of nail and glycosylated hemoglobin values in diabetics ( n = 32,
`r = 0.923, y = -16.8 + 57.6x ).
`
`correlation of the fasting blood glucose of diabetic group with nail and hemoglobin
`glycosylation wasp < 0.001 for both.
`The simultaneous test of both standards, fructose and 5-HMF, showed that the
`latter was destroyed during autoclave incubation at a rate of about 12% and that
`fructose was to be used as a standard in order to obtain a standard curve. We
`evaluated our data according to the fructose curve.
`
`Discussion
`
`The general nature of the reaction between aldose sugars and protein amino
`groups has led to studies involving the non-enzymatic glycosylation of proteins other
`than hemoglobin. Whether a causal relationship can be unravelled between the
`glycosylation of structural proteins and the development of macro- or microvascular
`complications of diabetes remains to be elucidated. Our findings show that glycosy(cid:173)
`lation of nail proteins occurs and is increased in diabetics when compared with
`non-diabetics. There was a significant correlation between nail glycosylation and
`simultaneously measured glycosylated hemoglobin values. This closely related to
`high blood glucose concentrations in the preceding several weeks. However, it seems
`likely that nail proteins are glycosylated with different dynamics from hemoglobin,
`an important factor when evaluating possible links between tissue glycosylation and
`diabetic microvascular complications.
`
`Page 4
`
`Anacor Exhibit 2042
`Flatwing Pharmaceuticals, Inc. v. Anacor Pharmaceuticals, Inc
`IPR2018-00168
`
`

`

`Further investigation into the chemical nature of the glycosylation in nail is
`needed. If more studies confirm our findings, the test may be of use forensically,
`because of the stability of fructosamine in nail. Such a test could also be of value in
`population studies, where a large number of samples might be taken quickly and
`stored easily.
`
`5
`
`References
`
`1 Javonic L, Peterson CM. Clinical utility of glycosylated hemoglobin. Am J Med 1981; 70: 331-338.
`2 McDonald JM, Davis JE. Glycosylated hemoglobins and diabetes mellitus. Human Pathol 1979; 10:
`279-291.
`3 Gonen B, Rubenstein AH. Hemoglobin A 1 and diabetes mellitus. Diabetologia 1978; 15: 1-8.
`4 Gabbay KH, Hasty K, Breslow JL, Ellison RC, Bunn HF, Gallop PM. Glycosylated hemoglobin and
`long-term blood glucose control in diabetes mellitus. J Clin Endocrinol Metab 1977; 44: 859-864.
`5 Trivelli LA, Ranney HM, Hong Tiens L. Hemoglobin components in patients with diabetes mellitus.
`N Eng J Med 1971; 284: 353-357.
`6 Koenig RJ, Peterson CH, Jones RL, Saudele C, Lehrman M, Cerami A. Correlation of glucose
`regulation and hemoglobin A1c in diabetes mellitus. N Eng J Med 1976; 295: 417-420.
`7 Fluckier R, Berger W, Winterhalter KH. Hemoglobin A 1c, a reliable index of diabetic control.
`Diabetologia 1978; 13: 393.
`8 Bunn HF, Gabbay KH, Gallop PM. The glycosylation of Hb: relevance to diabetes mellitus. Science
`1978; 200: 21-27.
`9 Paisey RB, Clamp JR, Kent MJC, Light ND, Hapton M, Hartog M. Glycosylation of hair: possible
`measure of chronic hyperglycemia. Br Med J 1984; 288: 669-671.
`10 Parker KM, England JD, Da Costa J, Hess RL, Goldstein DE. Improved colorimetric assay for
`glycosylated hemoglobin. Clin Chem 1981; 27: 669-672.
`
`Page 5
`
`Anacor Exhibit 2042
`Flatwing Pharmaceuticals, Inc. v. Anacor Pharmaceuticals, Inc
`IPR2018-00168
`
`