Decreased glucose effectiveness but not insulin resistance in glucose-tolerant offspring of Japanese non-insulin-dependent diabetic patients: a minimal-model analysis

The aim of the study was to estimate insulin sensitivity (SI), insulin secretion, and glucose effectiveness (SG) in 10 subjects with normal glucose tolerance (eight men and two women) with a family history of non-insulin-dependent diabetes mellitus (NIDDM offspring). Ten glucose-tolerant subjects (eight men and two women) without a family history of NIDDM served as control subjects. All subjects were Japanese. They underwent a modified frequently sampled intravenous glucose tolerance test (FSIGT): glucose (300 mg/kg body weight) was administered, and insulin (20 mU/kg over 5 minutes) was infused from 20 to 25 minutes after glucose. SI and SG were estimated by Bergman's minimal-model method. No significant difference was observed in body mass index (22.6 +/- 1.5 v 21.5 +/- 0.6 kg/m2) and fasting glucose (5.1 +/- 0.1 v 5.2 +/- 0.1 mol/L) and insulin (40.7 +/- 6.3 v 42.6 +/- 6.7 pmol/L). SI was not different between the two groups (0.83 +/- 0.11 v 0.94 +/- 0.15 x 10(-1).min-1.pmol/ L-1, P > .05). The acute insulin response to glucose (AIRglucose) estimated by intravenous glucose tolerance testing was significantly lower in the offspring than in the normal controls (2,139 +/- 265 v 3,438 +/- 318 pmol/L.min, P < .05). The glucose disappearance rate (KG) and SG were significantly diminished in the offspring versus normal controls (KG, 1.50 +/- 0.22 v 2.10 +/- 0.15 min-1, P < .05; SG, 0.016 +/- 0.003 v 0.023 +/- 0.002 min-1, P < .05). Thus, glucose-tolerant Japanese NIDDM offspring with normal insulin sensitivity are characterized by a reduced AIRglucose and diminished SG. This is the first report that glucose resistance but not insulin resistance already exists in glucose-tolerant Japanese NIDDM offspring.

Cause and repair of flap necrosis over cochlear implant

OBJECTIVE

The aim of this report was to study the cause and treatment of flap-related complications over cochlear implant.

SETTING

The study was performed in an academic tertiary referral center.

PATIENT

A 53-year-old Japanese male patient had had two retro-auricular skin incisions for tympanomastoidectomy in the postauricular region before implantation of a cochlear implant. He also had worn a helmet daily during work in his factory.

INTERVENTIONS

Corrective surgery was performed for reformation of the local blood supply to the overlying skin flap and reinforcement of the tissue overlying the implant by use of a superiorly based temporal muscle and fascia flap that were sutured with an inferiorly based muscle and fascia flap.

RESULTS

We successfully transposed the flap covering the implant without explantation of the implant.

CONCLUSIONS

The delayed development of flap necrosis was thought to be due to pressure necrosis produced by the band in the patient's helmet lying on the skin over the implant and to poor local blood supply in the postauricular area stemming from the two previous skin incisions for mastoid surgery performed for cholesteatoma. In those patients who wear helmets, it is important to position the cochlear implant sufficiently behind the ear and to avoid the use of a helmet.

Effects of absorption enhancers on the transport of model compounds in Caco-2 cell monolayers: assessment by confocal laser scanning microscopy

Three typical absorption enhancers, i.e., sodium caprate (Cap-Na), sodium deoxycholate (Deo-Na), and dipotassium glycyrrhizinate (Grz-K), were compared in terms of their permeability-enhancing effects on hydrophilic and hydrophobic model compounds in Caco-2 cell monolayers. The transepithelial electrical resistance (TEER) of the monolayers was reduced concentration-dependently by treatment with Cap-Na and Deo-Na, while treatment with Grz-K increased the TEER. Two patterns of TEER reduction were observed: one pattern indicated that Cap-Na had a rapid reducing effect, and another indicated that Deo-Na had a delayed reducing effect. These reductions in the TEER were accompanied by the increased transepithelial transport of two hydrophilic model compounds, sodium fluorescein (Flu-Na; MW = 376, log P = -1.52) and fluorescein isothiocyanate-dextran 4000 (FD-4; MW = 4400, log P = -2.0), and one hydrophobic model compound, rhodamine 123 hydrate (Rh123; MW = 381, log P = 1.13). The transport-enhancing effects of Cap-Na and Deo-Na on these model compounds decreased in the following order: FD-4 > Rh123 > Flu-Na, while Grz-K was found to have no effect on the transport of any of these model compounds. Confocal laser scanning microscopy (CLSM) of Caco-2 cell monolayers revealed that Cap-Na and Deo-Na enhanced the transepithelial transport of the hydrophilic model compounds via the paracellular route and that of the hydrophobic model compound via both paracellular and transcellular routes. Semiquantitative visual information obtained from CLSM images reflected the results of the transport experiment.

Identification of macrophage migration inhibitory factor in adipose tissue and its induction by tumor necrosis factor-alpha

Macrophage migration inhibitory factor (MIF) has been rediscovered as a proinflammatory cytokine, pituitary hormone, and glucocorticoid-induced immunoregulator. A survey of tissue distribution revealed that MIF expression is not limited to T lymphocytes, but exists in several other tissues; however, its presence in adipose tissue has never been investigated. In this study, we examined the expression of MIF in adipose tissue using the rat epididymal fat pad and murine 3T3-L1 adipocytes. Northern and Western blot analyses revealed the expression of MIF mRNA and MIF protein, respectively, in both the fat pad and the adipocyte cell line. In immunohistochemistry, a positive staining reaction with an anti-rat MIF antibody was detected largely in the cytosol of adipocytes of the epididymal fat pad. To examine the production and release of MIF by adipocytes, we examined its content in the culture medium of the 3T3-L1 adipocytes. The results showed that MIF content was 1.6 +/- 0.48 ng/ml (mean +/- SD) after 24 hr culture, and the content was increased up to 9.7 +/- 2.8 ng/ml by stimulation with TNF-alpha (50 nM). Since TNF-alpha produced in adipocytes is known to induce insulin resistance, the results suggest the possibility that MIF plays an important role in the mechanism of insulin resistance often observed in obesity and diabetes via regulation of TNF-alpha expression.

Emergency pulmonary embolectomy with percutaneous cardiopulmonary bypass

BACKGROUND

The management of patients with acute pulmonary embolism remains difficult, particularly when cardiogenic shock is involved. The preoperative incidence of cardiac arrest compromises the results of emergency pulmonary embolectomy. In an attempt to reduce the operative mortality rate, we applied percutaneous cardiopulmonary bypass support to restore vital organ perfusion before the surgical intervention.

METHODS

Percutaneous cardiopulmonary bypass support was preoperatively instituted in 3 patients with acute cardiopulmonary collapse caused by massive pulmonary embolism. In each patient, cardiac massage and endotracheal intubation were necessary due to loss of consciousness, hypotension, and severe cyanosis. Transesophageal echocardiography was performed to detect any evidence of thrombus in the main pulmonary artery, and each patient underwent the emergency pulmonary embolectomy using conventional cardiopulmonary bypass through a median sternotomy.

RESULTS

Percutaneous cardiopulmonary bypass support immediately provided effective cardiopulmonary resuscitation. Transesophageal echocardiography clearly demonstrated any evidence of thrombus located in the pulmonary artery. Each patient was discharged from the hospital without any postoperative complication.

CONCLUSIONS

The use of percutaneous cardiopulmonary bypass support immediately resuscitated and stabilized the cardiopulmonary function and allowed for successful emergency pulmonary embolectomy. In each patient, transesophageal echocardiography was useful for prompt and noninvasive diagnosis.

Comment on autogenic training and hypertension

We comment on a report by Watanabe, et al. regarding the effects of autogenic training on hypertension. Using previous reports in the United States, we mention methodological problems on how to evaluate the effects of autogenic training and express our hope that they would provide further research to clarify the effects of autogenic training on hypertension.

Immunostimulating effect of growth hormone: in-vivo administration of growth hormone in rainbow trout enhances resistance to Vibrio anguillarum infection

In-vivo administration of purified (nGH) and recombinant (rGH) chum salmon, Oncorhynchus keta, growth hormone enhanced the survival of rainbow trout, O. mykiss, against virulent Vibrio anguillarum. Macrophages obtained from rainbow trout that were treated in vivo with either GH preparation showed enhanced stimulation of the chemiluminescent responses induced by opsonized V. anguillarum cells and increased phagocytic activities. However, the serum bactericidal activity against V. anguillarum was not enhanced in fish injected with nGH or rGH. These findings show that in-vivo administration of either GH preparation can effectively prime macrophages and increase the resistance to V. anguillarum in rainbow trout.

Characterization of nociceptin hyperalgesia and allodynia in conscious mice

1. Intrathecal (i.t.) administration of nociceptin and high doses of morphine induced allodynia in response to innocuous tactile stimuli, and i.t. nociceptin evoked hyperalgesia in response to noxious thermal stimuli in conscious mice. Here we have characterized the nociceptin-induced allodynia and compared it with the morphine-induced allodynia and the nociceptin-evoked hyperalgesia. 2. Nociceptin-induced allodynia was evoked by the first stimulus 5 min after i.t. injection, reached a maximum at 10 min, and continued for a 50 min experimental period. Dose-dependency of the allodynia showed a bell-shaped pattern from 50 pg to 5 ng kg-1, and the maximum effect was observed at 2.5 ng kg-1. 3. Morphine-induced allodynia reached the maximum effect at 15 min and declined progressively until cessation by 40-50 min. The dose-response curve showed a bell-shaped pattern, similar to that induced by nociceptin, with a maximum effect at 0.5 mg kg-1, five orders of magnitude higher than that of nociceptin. 4. The allodynia evoked by nociceptin and morphine were dose-dependently blocked by glycine, D(-)-2-amino-5-phosphonovaleric acid (D-AP5, an N-methyl-D-aspartate (NMDA) receptor antagonist), gamma-D-glutamylaminomethyl sulphonic acid (GAMS, a non-NMDA receptor antagonist) and methylene blue (a soluble guanylate cyclase inhibitor), but were not affected by muscimol (a gamma-aminobutyric acidA (GABAA) receptor agonist) and baclofen (a GABAB receptor agonist). 5. Morphine did not inhibit forskolin-stimulated cyclicAMP formation in cultured cells expressing the nociceptin receptor. 6. Nociceptin-induced hyperalgesia was evoked 10-15 min after i.t. injection. Nociceptin produced a monophasic hyperalgesic action over a wide range of doses from 5 fg to 50 ng kg-1. The nociceptin-induced hyperalgesia was blocked by glycine only among the agents examined. 7. None of the pain responses evoked by nociceptin and morphine were blocked by naloxone. 8. These results demonstrate that, whereas the mechanisms of the nociceptin-induced allodynia and hyperalgesia are evidently distinct, they involve a common neurochemical event beginning with the disinhibition of the inhibitory glycinergic response. Morphine may induce allodynia through a pathway common to nociceptin, but the nociceptin receptor does not mediate the action of high doses of morphine.

[Extended Kalman filtering trained neural networks and multicomponent analysis of amino acids]

The principle of neural networks was briefly described. Backpropagation (BP) is a widely used algorithm to train NN with slow convergence and local optima. In order to overcome the above weakness of BP, a novel learning algorithm, extanded Kalman filtering (EKF, EF), was proposed with high performance. EFNN was used for simultaneous multi-component analysis with good results and no separation.

Tyrosine hydroxylase (TH)-only-immunoreactive non-catecholaminergic neurons in the brain of wild mice or the human TH transgenic mice do not contain GTP cyclohydrolase I

We previously reported the presence of transiently tyrosine hydroxylase (TH)-only-immunoreactive (ir), non-catecholaminergic (non-CAnergic) neurons in some brain regions of postnatal mice; anterior olfactory nucleus, medial geniculate nucleus, and spinal trigeminal nucleus, where CAnergic terminals but not cell bodies are present in the adult wild mouse. These transiently TH-ir brain regions of the postnatal wild mouse showed stable TH-immunoreactivity in the adult brain of the human TH transgenic (hTHTg) mice. TH expression was also observed in the nucleus parabigeminalis of the hTHTg mice. Using the antiserum against GTP cyclohydrolase I (GCH), first rate-limiting enzyme of the biosynthesis of tetrahydrobiopterin (BH4), the cofactor for TH, we proved that these TH-only-ir neurons in the wild mice and in the hTHTg mice were not stained with the antiserum against GCH. The results indicate that these TH-only-ir neurons which do not synthesize the BH4 cofactor do not produce dihydroxyphenylalanine, suggesting a new unknown function of TH in these neurons.

Production of heparin binding epidermal growth factor-like growth factor in the early phase of regeneration after acute renal injury. Isolation and localization of bioactive molecules

We have recently reported that heparin-binding epidermal growth factor-like growth factor (HB-EGF) mRNA is induced in the rat kidney after acute ischemic injury. The present studies were designed to investigate whether bioactive HB-EGF protein is also produced in response to renal injury induced by either ischemia/reperfusion or aminoglycosides. Heparin-binding proteins were purified from kidney homogenates by heparin affinity column chromatography using elution with a 0.2-2.0 M gradient of NaCl. A single peak of proteins that eluted at 1.0-1.2 M NaCl was detected in the postischemic kidney within 6 h of injury. This eluate fraction stimulated DNA synthesis in quiescent Balb/c3T3, RIE, and NRK-52E cell lines, all of which are responsive to the epidermal growth factor family of mitogenic proteins. The EGF receptor of A431 cells was also tyrosine phosphorylated by this eluate peak. Furthermore, immunoblotting with a polyclonal antibody against rat HB-EGF indicated that the eluate peak contained immunoreactive proteins of 22 and 29 kD mol wt, consistent with the reported sizes of the secreted form and membrane anchored form of HB-EGF, respectively. Immunohistochemical studies revealed that HB-EGF was produced predominantly in distal tubules in kidneys injured either by ischemia/reperfusion or aminoglycoside administration. We also found that during metanephric development immunoreactive HB-EGF was detected in the ureteric bud as early as E14.5 and persisted in structures arising from the ureteric bud throughout embryogenesis. These results suggest that in response to acute injury, HB-EGF is produced predominantly in distal tubules and that endogenous HB-EGF may be an important growth factor involved in renal epithelial cell repair, proliferation, and regeneration in the early stages of recovery after acute renal injury, as well as in nephrogenesis.

Intravenous glucose tolerance test-derived glucose effectiveness in bulimia nervosa

The aim of the present study was to estimate insulin secretion, insulin sensitivity (SI), and glucose effectiveness at basal insulin (SG) in subjects with bulimia nervosa. Eight bulimic patients and eight age-, body mass index-, and sex-matched healthy control subjects without a family history of diabetes were studied. The subjects all had normal glucose tolerance. They underwent a modified frequently sampled intravenous glucose tolerance test; glucose (300 mg/kg body weight) was administered, and insulin (4 mU/kg body weight/min) was infused from 20 to 25 minutes after administration of glucose. SI and SG were estimated by Bergman's minimal model method. Basal insulin (27 +/- 3 v 45 +/- 3 pmol/L) was significantly lower in bulimic patients than in normal controls (P < .05), but basal glucose was similar between the two groups (4.5 +/- 0.1 v 4.9 +/- 0.1 mmol/L, P > .05). The glucose disappearance rate (KG) and acute insulin response to glucose estimated by the intravenous glucose tolerance test (AIR(glucose)) were similar between the two groups (KG, 1.35 +/- 0.29 v 2.20 +/- 0.21 min(-1), P > .05; AIR(glucose), 2,920 +/- 547 v 2,368 +/- 367 pmol/L x min, P > .05). No significant difference was observed in SI between the two groups (1.34 +/- 0.18 v 1.25 +/- 0.20 x 10(-4) x min(-1) x pmol/L(-1), P > .05). On the other hand, glucose effectiveness at basal (SG) and zero (GEZI) insulin was significantly diminished in comparison to normal controls (SG, 0.011 +/- 0.002 v 0.024 +/- 0.002 min(-1), P < .01; GEZI, 0.008 +/- 0.002 v 0.017 +/- 0.003 min(-1), P < .01). Thus, bulimic patients with normal glucose tolerance without a family history of diabetes were characterized by normal insulin secretion, normal SI, and reduced SG and GEZI.

p53 mutation and protein overexpression in the early stages of esophageal tumorigenesis utilizing endoscopically obtained biopsy specimens

It is unclear whether p53 abnormality in the early esophageal tumorigenesis causes clonal expansion with tumor growth. In this study, we analyzed p53 abnormalities by PCR-SSCP and immunohistochemistry in 86 esophageal endoscopic biopsy specimens. Eleven of 27 specimens (39%) of mild dysplasias showed p53 mutations. Six moderate or mild dysplasias were followed by endoscopy (average 73 weeks) and continually exhibited the same p53 mutation, but none of them showed apparent tumor growth and subepithelial epithelial invasion. These results suggest that p53 mutations occur very early in the esophageal tumorigenesis and contribute to cell proliferation, but cannot be related with malignant phenotype directly.

Expression of endothelial nitric oxide synthase in human eccrine clear cells

Nitric oxide is generated from L-arginine by nitric oxide synthase (NOS), which has at least three isoforms; endothelial-type NOS (eNOS) and brain-type NOS (bNOS) are constitutive enzymes, and inducible-type NOS (iNOS) is expressed after stimulation. Studies by the avidin-biotin immunocomplex method, revealed eNOS immunoreactivity exclusively in the human eccrine clear cells. No eNOS immunoreactivity was observed in the eccrine dark cells or myoepithelial cells. No staining of iNOS or bNOS was observed in the eccrine gland. These findings indicate that NO plays a physiological part in the production and/or excretion of sweat in the human skin eccrine gland.

Immunohistochemical colocalization of GTP cyclohydrolase I in the nigrostriatal system with tyrosine hydroxylase

Immunohistochemical colocalization of GTP cyclohydrolase I (GCH) in the mouse nigrostriatal system with tyrosine hydroxylase or aromatic L-amino acid decarboxylase in the somata and terminals of GCH-positive catecholaminergic neurons are proved for the first time by a double-labeling immunofluorescence method with a confocal laser-scanning microscope. GCH-immunoreactive somata in the mouse substantia nigra have synaptic contacts with monoaminergic and non-monoaminergic terminals.

Application of autogenic training for anxiety disorders: a clinical study in a psychiatric setting

The effects of autogenic training for anxiety disorders were investigated in a psychiatric setting of a medical school hospital and the predictors of this treatment outcome were identified. Fifty-five patients who meet the DSM-III-R criteria for anxiety disorders were treated individually with autogenic training by the author from October 1981 to October 1995. The medical records of the patients were investigated retrospectively. The results showed that the autogenic training was successful. Twenty-eight patients (51%) were cured, fourteen (25%) much improved, eight (15%) improved and five (9%) unchanged at the end of the treatment. Forty-two patients (76%) were assessed as having had successful treatment. Pretreatment variables, such as patient's clinical characteristics, did not provide a useful guide to the outcome. Four treatment variables did have a bearing on outcome. First, practicing the second standard autogenic training exercise was a satisfactory predictor of a better outcome. Second, practicing generalization training also was a useful predictor. Third, the application of other behavioral treatment techniques was found to be positively associated with outcome. Fourth, longer treatment periods were associated with a better outcome. These findings suggested that autogenic training could be of significant benefit for the treatment of anxiety disorders.

[Twenty cases of superior septal approach for mitral valve operation]

We experienced 20 cases of superior septal approach (SSA) for mitral valve surgery from June 1992 to December 1995. From pre-operative coronary angiography, sinus node arteries (SNAs) were grouped into three types, branched from right coronary artery (RCA: 45%), from left circumflex (LCx: 35%), and from both RCA and LCx (Bilateral: 20%). And, in diameter, they were grouped into two groups, above 1 mm (80%) and below 1 mm (20%). There were no peri-operative deaths and complications associated with SSA. In acute post operative phase, atrial fibrillation, junctional rhythm, atrial flutter, and sino-atrial block were observed, especially in patients whose SNA branched from RCA, and were above 1 mm in diameter. In late follow-up, all patients' rhythm returned to pre-operative rhythm, although some cases became bradycardia less than 60 per-minute. Bradycardia was observed in patients whose SNA branched from RCA, and were above 1 mm in diameter. There may be relationship between post operative arrhythmia and division of large SNAs originated from RCA. We concluded that SSA is safe and excellent approach for mitral valve surgery, especially in patients who require reoperations and in patients with small LA, although we should consider the advantages of SSA versus the risks of post operative arrhythmia and bradycardia.

Developmental sequence of human fetal body movements in the second half of pregnancy

To describe fetal body movements precisely and to clarify their developmental sequence in the second half of pregnancy, fetal body movements were observed in 31 normal fetuses between 20 and 41 weeks of pregnancy by using multiple ultrasonic machines. All movements of the whole trunk were able to be classified into six fundamental movements: flexion, stretch, rolling, startle, stepping, and writhing. Startle was always observed as a single movement, but others were more frequently observed as a part of combination rather than as a single movements. The incidence of flexion and stretch peaked at 28-31 weeks, whereas the incidence of startle and stepping lowered at the period. The incidence of rolling and upper half trunk movements increased at 40-41 weeks. Limb movements were frequent and did not exhibit characteristic change in the incidence. In conclusion, we classified fetal body movements into several fundamental movements instead of using the expression of "general movements" by focusing principally on movements of the trunk. In particular, some of the fundamental movements seem to appear in characteristic ways as related to gestational age. In view of that fetal behaviors are closely related to the neural activity, the present findings may contribute to an understanding of the developmental process of the central nervous system.

Rat maf related genes: specific expression in chondrocytes, lens and spinal cord

maf is a family of oncogenes originally identified from avian oncogenic retrovirus, AS42, encoding a nuclear bZip transcription factor. We have isolated two maf related cDNA clones, maf-1 and maf-2, from a rat liver cDNA library. Comparison of the sequence homologies of the proteins encoded by maf-1 and maf-2 with those of c-maf and chicken mafB indicated that maf-1 and maf-2 are the rat homologues of mafB and c-maf, respectively. Both genes are expressed at low levels in a wide variety of rat tissues, including spleen, kidney, muscle and liver. Immunohistochemical studies and in situ hybridization analyses show that maf-1 and maf-2 are strongly expressed in the late stages of chondrocyte development in the femur epiphysis and the rib and limb cartilage of 15 day old (E15) embryo in rat. Cartilage cells, induced by subcutaneous implantation of bone morphogenic protein, also expressed maf-1 and maf-2. In situ hybridization analyses of E15 embryos show that both genes are expressed in the eye lens and the spinal cord as well as the cartilage. However, the expression patterns of maf-1 and maf-2 in lens and spinal cord are different.

Clinical trial of prophylactic endoscopic variceal ligation for esophageal varices

Endoscopic variceal ligation is an effective therapy for variceal bleeding, and use of the method has recently been increasing. We evaluated the clinical usefulness of prophylactic endoscopic variceal ligation. Twenty-two patients with enlarged, tortuous varices and "red color signs" were selected. These patients were treated with ligation therapy alone and the varices were eradicated, i.e., reduced to small, straight varices without red color signs. Ligation therapy was withdrawn if the general condition of the patient worsened or if the varices could not be removed by suction. Follow-up endoscopy was performed every 4 months, and another ligation was performed if there were recurrent varices or variceal bleeding. The total reduction rate was 86.4%, and eradication required two sessions of therapy and 30 days of hospitalization on average. Complications included esophageal injury in 1 patient and treatment-induced bleeding in 1 patient; both complications were easily controlled. No variceal bleeding occurred after the eradication. There was no mortality due to gastrointestinal bleeding during the median follow-up period of 346 days. Prophylactic endoscopic variceal ligation made it possible to prevent fatal variceal bleeding with a minimum risk of complications, suggesting that this could be an alternative method for the prevention of first-time variceal bleeding.

Immunohistochemical localization of nitric oxide synthase in normal human skin: expression of endothelial-type and inducible-type nitric oxide synthase in keratinocytes

Nitric oxide (NO) is a critical mediator of various biological functions. NO is generated from L-arginine by nitric oxide synthase (NOS), which has three isoforms; endothelial-type NOS (eNOS) and brain-type NOS (bNOS) are constitutive enzymes, and inducible-type NOS (iNOS) is expressed after stimulation. We investigated the expression of NOS in normal human skin by an immunohistochemical technique and western blotting analysis. In human skin, epidermal keratinocytes and the outer root sheath were labeled with not only eNOS antibody but also with iNOS antibody. Both eNOS and iNOS protein in epidermal keratinocytes were confirmed by western blotting. eNOS immunoreactivity was observed in endothelial cells, fibroblasts, the arrector pili muscle, apocrine secretory gland, eccrine coiled duct, and eccrine secretory gland. bNOS immunoreactivity was observed in mast cells. No staining with anti-bNOS antibody was observed in any other cell type. Our present findings suggest that epidermal keratinocytes in normal human skin contain both eNOS and iNOS.