aortic endothelial and smooth muscle histamine...

Aortic Endothelial and Smooth MuscleHistamine Metabolism

Relationship to Aortic 125l-Albumin Accumulationin Experimental Diabetes

Theodore M. Hollis, Stephen G. Gallik, Alicia Orlidge, and Jeffrey C. Yost

We studied rat aortic endothelial and smooth muscle cell de novo histamine synthe-sis mediated by histidine decarboxylase (HD) and the effects of its inhibition by alpha-hydrazinohistidine on the intracellular histamine content and intraaortic albumin ac-cumulation in streptozotocin-induced diabetes. Diabetes was induced by a singlejugular vein injection of streptozotocin (60 mg/kg, pH 4.5, ether anesthesia), withanimals held 4 weeks following the overt manifestation of diabetes. Additional diabet-ic and nondiabetic rats received alpha-hydrazinohistidine (25 mg/kg, i.p. every 12hours) during the last week; this had no effect on the severity of diabetes in any animalreceiving streptozotocin. Data indicate that the aortic endothelial (EC) HD activity wasincreased more than 130% in the untreated diabetic group but was similar to controlvalues in the diabetic group receiving alpha-hydrazinohistidine; similarily, the EChistamine content from diabetic aortas increased 127% over control values, but in ECfrom diabetic animals receiving alpha-hydrazinohistidine it was comparable to controlvalues. Similar trends were observed for the subjacent aortic smooth muscle. Inuntreated diabetic animals the aortic 125l-albumin mass transfer rate was increased60% over control values, while in diabetic animals receiving alpha-hydrazinohistidinethe 125l-albumin mass transfer rate was essentially identical to controls. These dataindicate that in streptozotocin diabetes there is an expansion of the inducible aortichistamine pool, and that this expansion is intimately related to the increased aorticalbumin accumulation. (Arteriosclerosis 3:599-606, November/December 1983)

It is widely recognized that diabetes mellitus consti-tutes an important independent risk factor of

atherosclerosis.1"4 However, mechanisms responsi-ble for this diabetic atherogenicity are essentially un-known. In part, this stems from our lack of under-standing of the mechanisms of atherogenesis.Equally important, however, is the lack of studies ofalterations in artery wall metabolism associated withdiabetes that are likewise associated with athero-genesis. One notable exception is the study of Wo-

From the Department of Biology, 208 Mueller Laboratory, Penn-sylvania State University, University Park, Pennsylvania.

This work was supported by NIH Grant HL 20460.Stephen G. Gallik is now at the Department of Physiology and

Biophysics, University of Louisville, Louisville, Kentucky.Address for reprints: Dr. Theodore M. Hollis, 208 Mueller Labo-

ratory, Department of Biology, Pennsylvania State University,University Park, Pennsylvania 16802.

Received December 13, 1982; revision accepted June 20,1983.

linsky et al.5 who reported that aortic smooth muscleacyl-cholesteryl esterase activity is reduced in strep-tozotocin-induced diabetes. An obvious possibilityarising from this finding is that the capacity of arterialsmooth muscle to clear infiltrated lipid is impaired ata time when the total blood lipid concentration iselevated. This, combined with altered vessel wallpermeability to blood-borne macromolecules, couldthen result in an increased presentation of lipidssuch as low density lipoproteins (LDL) to underlyingmuscle. A reduced capacity to metabolize or other-wise clear these lipids could then lead to intravascu-lar lipid accumulation.

For some time we have been examining metabolicmechanisms involved in the mediation of abnormalarterial permeability alterations associated with ath-erogenic risk factors. We have specifically been fo-cusing our attention on the role of arterial wall endo-thelial and smooth muscle histamine metabolism; aconsistent finding has been that aortic histamine

599

by guest on July 16, 2018http://atvb.ahajournals.org/

Dow

nloaded from

http://atvb.ahajournals.org/

600 ARTERIOSCLEROSIS VOL 3, No 6, NOVEMBER/DECEMBER 1983

synthesis increases in experimental hypertension,dietary-induced hypercholesteremia, exposure ofaortas to elevated shear stresses, and in regions ofthe normal aorta that show spontaneous vascularinjury (see reference 6 for a summary of references).Most recently we have reported that in streptozoto-cin-induced diabetes, increased histamine synthe-sis, decreased histamine catabolism, and a corre-sponding increase in the intracellular histaminecontent occur in both endothelial and smooth musclecells. These alterations are completely reversible byinsulin treatment.7 These findings have suggestedthat the aortic inducible histamine pool may be ex-panded in experimental diabetes, and that factorsassociated with this expanded inducible pool may beintimately associated with altered aortic macromole-cule permeability.

The present study has been designed to deter-mine: 1) if our previous speculation that the aorticinducible histamine pool is expanded in diabetes iscorrect; and 2) if this expanded pool is directly impli-cated in mediating increased aortic transmural per-meability, as assessed by the measurement of aortic125l-albumin flux and the aortic 125l-albumin masstransfer rate. These were examined by using alpha-hydrazinohistidine, the only relatively specific, non-toxic inhibitor of histidine decarboxylase (HD), whichhas again become available and which produced amarked reduction in the severity of atherosclerosis incholesterol-fed rabbits.8

MethodsAnimals

Male Wistar rats having initial body weights be-tween 160-180 g were used in this study. All surgicalprocedures were performed under anesthesia and inaccordance with established institutional policies.Diabetes was induced by a jugular vein injection ofstreptozotocin (60 mg/kg, Sigma Chemical Com-pany, St. Louis, Missouri) under ether anesthesia.Animals were held in metabolic cages, with food andwater intake as well as urine output being continu-ously monitored. Blood samples were obtained bytail clipping under ether anesthesia before streptozo-tocin injection, 3 days after injection and at leastweekly thereafter. The blood glucose concentrationwas measured using a Yellow Springs 23A glucoseautoanalyzer; animals having a nonfasting blood glu-cose concentration more than 230 mg/dl and exhibit-ing glycosuria were deemed diabetic. Controls con-sisted of both untreated and saline-injected animals.Three weeks following diagnosis of diabetes, sub-sets of animals in each group received alpha-hydra-zinohistidine (Regis Chemical Company, Chicago,Illinois) at a dose of 25 mg/kg i.p. every 12 hours forthe next 7 days. This dose has previously beenshown to reduce albumin accumulation and athero-sclerotic plaque severity in cholesterol-fed animals.8

Aortic Endothelial and Smooth MuscleHistamine Metabolism

Four weeks after the initial diabetes diagnosis,eight animals in each treatment group were killed bydecapitation under ether anesthesia. The thoracicand abdominal cavities were opened via a ventrallaparotomy and thoracotomy, the heart was ex-posed, the right atrium was cut, and the circulatorysystem was perfused with ice-cold phosphate-buf-fered saline (PBS, pH 7.4). The entire thoracic aortawas then excised, placed in fresh ice-cold PBS (pH7.4) and cleared of its periadventitial fat. After beingopened longitudinally, aortic endothelial cells weremechanically removed using cotton-tipped applica-tor sticks as described by Gospodarowitz et al.9 andby us.7 These cells were collected in approximately 2ml of ice-cold PBS containing 0.1 % Triton X-100 (pH7.9), and homogenized using a ground-glass pestle.Smooth muscle cells were mechanically removed instrips and subjected to the same procedures de-scribed for the endothelial cells. The homogenateswere centrifuged at 10,000 g for 20 minutes (4° C)and the supernatant was collected.

Verification of cell types involved both ultrastruc-tural examination of cell isolates and phase contrastexamination of primary cultures established fromthese isolates, as previously described.7

A 0.9 ml aliquot of each supernatant was markedwith 1 mg blue dextran and applied to 13 x 130 mmSephadex G-25 columns (grain size 50-150 m; Sig-ma Chemical Company, St. Louis, Missouri). Ice-cold 0.01 M PBS with 0.1% Triton X-100 (pH 7.9)was then added as the elutant, the blue fraction be-ing collected at 4° C and stored in liquid nitrogen for asubsequent determination of histidine decarboxyl-ase activity and protein concentration (see below).The remaining supernatant was divided into 0.3 mlaliquots and used to determine the intracellular hista-mine content.

The histamine content and histidine decarboxyl-ase activity of both endothelial cells and subjacentsmooth muscle cells were determined by a modifica-tion of the double isotopic microassay procedure ofTaylor and Snyder10 as previously described.7 In thisprocedure, histamine methyltransferase (HMT) pre-pared from guinea pig brain transfers a 14C-methylgroup from 14C-S-adenosylmethionine (14C-SAMe)to histamine, forming 14C-methylhistamine. Traceamounts of 3H-labeled histamine were added to cor-rect for the variation in histamine methylation amongsamples. An extraction of 14C-methylhistamine froman alkaline, salt-saturated solution into chloroformseparated out unreacted 3H-histamine, 14C-SAMe,and histamine. The ratio of ^C-dpm^H-dpm wasused to determine the amount of histamine per sam-ple. As the amount of cold histamine increased, the^C^H also increased due to the increased formationof 14C-methylhistamine in the reaction. All assayswere performed in quadruplicate. Samples werecounted using a Packard Tricarb 460CD liquid scin-tillation counter (Packard Instrument Company, Chi-


Dow

nloaded from


AORTIC HISTAMINE AND ALBUMIN UPTAKE IN DIABETES Hollis et al. 601

cago, Illinois) and standardized on the basis ofprotein.11

Aortic Albumin Accumulation

These studies used additional rats, with six ani-mals in each group. At the end of the 4-week holdingperiod, each animal was anesthetized with a 2:1 mix-ture of Ketamine HCI (100 mg/ml, Bristol Laborato-ries, Syracuse, New York) and xylazine (20 mg/ml,Haver-Lockhart, Shawnee, Kansas) at a final dose of120 mg/kg, i.p. The right jugular vein was exposed,and 50 /^Ci (0.5 ml) of 125l-bovine serum albumin(New England Nuclear, Boston, Massachusetts)was injected. This albumin had previously been puri-fied by means of an AgCI ion-exchange resin, andwas 95% pure (personal communication, New Eng-land Nuclear Corporation). At 3 minutes and 12 min-utes following injection, a 0.1 ml blood sample wasobtained from the jugular vein for determination ofplasma 125l-albumin activity. Exactly 19 minutes and30 seconds following injection, the thoracic and ab-dominal cavities were opened, the heart was isolat-ed, and the right atrium was cut. At 20 minutes thecirculatory system was gently perfused with ice-coldPBS (pH 7.4).

Following perfusion the thoracic aorta was isolat-ed, and both the aortic-left subclavian and aortic-diaphragmatic junctions were located and markedwith pins. The distance between these points wasmeasured. The aorta was then excised, cleared of itsperiadventitial fat, opened longitudinally, and re-stretched to its original in vivo length. Three aorticplugs (5 mm in diameter) were taken from the de-scending thoracic aorta using a dermal punch. In-tima-media preparations were made from theseplugs using the luminal stripping technique de-scribed by Wolinsky and Daly.12 Using methods de-scribed by Bratzler et al.,13 the intima-media seg-ments were homogenized in 2 ml of 10%trichloroacetic acid (TCA) and centrifuged at 1500 gfor 15 minutes. The supernatant was removed anddiscarded, the pellet was washed three times with10% TCA, and the final pellet was counted in a Mi-

cromedic 4/200 gamma counter (Micromedia Sys-tems, Horsham, Pennsylvania).

All plasma samples were diluted 1:1000 with iso-tonic saline. To increase the bulk of the precipitate,15 .̂l of 6% bovine serum albumin in physiologicalsaline were added to 1001 of the diluted plasma. Theplasma was then treated exactly like the tissue sam-ples.

Aortic albumin accumulation was determined asdescribed by Bratzler et al.13 Initially, the aortic 1 2 5 I -albumin flux was calculated as:

J = (CT) (A)"1 (t)"1

where CT is the tissue activity of the 125l-albumin incounts per minute cpm, A is the aortic surface area(cm2), and t is the circulation time in seconds. Fromthis, the aortic intima-media 125l-albumin masstransfer rate (P, cm/sec) was then calculated as J/CPO, where CPO is the initial, 3 minute plasma radio-activity. This calculated value is equivalent to theaortic albumin accumulation parameter described byOwens and Hollis.6

Statistical Analyses

For comparisons of differences between groups,we used a variance analysis followed by applicationof Duncan's multiple range test.14 The results wereconsidered significant at the 95% level of confidence(p < 0.05).

Results

Characteristics of Animal Groups

The overall characteristics of the animals fromeach treatment group are given in Table 1. Whilethere is no significant difference between groups ininitial body weights, animals from both diabeticgroups had mean final body weights approximately30% lower (p < 0.05), as well as persistent glycos-uria and polyuria. Alpha-hydrazinohistidine adminis-tration had no apparent effect on any of the diabeticcharacteristics monitored.

Table 1. Characteristics

Treatment group

ControlNondiabetic + a-HHDiabeticDiabetic + a-HH

of the Animal Groups

Body weight (g)

Initial

202 ±3.5201 ±5.8196 + 5.2199 ±7.3

Final

367 ±5.7347 + 6.7248 ±8.0*247 ±15.4*

Nonfastingblood

glucose(mg/dl)

110 + 1.4108±1.9299 ±7.4*284 ±8.9*

Urineoutput

(ml/24 hr)

5.7±0.66.3 + 0.6

85.0±3.1*89.0 ±4.4*

Urineglucose(mg/dl)

00

>150*>150*

All values represent means ± mean standard errors (SE). a-HH = alpha-hydrazinohistidine (25 mg/kg,i.p. at 12 hours for Days 21-28).

'Significant difference from control (p < 0.05), determined by variance analysis followed by Duncan'smultiple range test.


Dow

nloaded from



Cell Type Verification

The ultrastructural characteristics of cell isolatesderived from aortic luminal scrapings included nu-merous endocytotic vesicles, a low density of mito-chondria, prominent ovoid nuclei, and typical inter-endothelial junctions.15"17 The characteristics ofsubendothelial cell isolates were typical of smoothmuscle cells. Phase contrast examination of primarycultures established from these isolates further con-firmed that these isolates were viable and indeedwere endothelial and smooth muscle cells. All obser-vations made in this study were identical to those inphotomicrographs of cells isolated by identical pro-cedures and previously published.7

Endothelial and Smooth Muscle HlstamlneMetabolism

The histamine content and histidine decarboxyl-ase (HD) activity, together with standard errors, ofthoracic aortic endothelial and smooth muscle cellsof all rats from each group are presented in histo-gram form in Figure 1.

I IControl

[^Diabetic

H Nondlabetlc-aHH

Dlabetlc-aHH

Figure 1. Aortic endothelial and smooth muscle hista-mine content and histidine decarboxylase activity in strep-tozotocin-induced diabetes with and without a-hydrazino-histidine treatment. Diabetes was induced by i.v.administration of streptozotocin (60 mg/kg) in acidified ci-trated saline (pH 4.5). Alpha-hydrazinohistidine (a-HH, 25mg/kg, i.p. (5,12 h) was given for the last 7 days of a 28-dayperiod to both nondiabetic and diabetic animals. Eachgroup contained 6 rats (n = 6). EC = endothelial cells;SMC = smooth muscle cells; HD = histidine decarboxyl-ase activity. Results are standardized on the basis of cellprotein, -fr = significant difference from control (p < 0.05).* = significant difference from untreated diabetic group(p < 0.05).

For endothelial cells, the mean histamine contentfrom the aortas of the untreated diabetic group was127% greater than that of the corresponding con-trols; alpha-hydrazinohistidine administration for thelast week resulted in a 78% decrease in the endothe-lial histamine content over that of the untreated dia-betic group. No significant differences in endothelialhistamine content existed between control, nondia-betic, alpha-hydrazinohistidine, and diabetic, alpha-hydrazinohistidine groups. The HD activity of thesesame cells was 136% greater in the untreated dia-betic animals than in cells from control animals. Al-pha-hydrazinohistidine administration to the diabeticgroup resulted in a 67% decrease in HD activity fromthe corresponding untreated diabetic group. Thesedifferences are significant (p < 0.05). No significantdifferences in HD activity were evident between con-trol and nondiabetic, alpha-hydrazinohistidine treat-ed animals.

For smooth muscle cells, the histamine contentfrom untreated diabetic animals was 85% greaterthan that of the smooth muscle cells from the controlgroup. In the diabetic group given alpha-hydrazino-histidine, the smooth muscle histamine content de-creased 50% (p < 0.005) over that of the cells fromthe untreated diabetic group. With respect to the HDactivity in these same smooth muscle cells, thosefrom diabetic aortas had an HD activity more than150% greater than that of cells from the controlgroups. In the diabetic, alpha-hydrazinohistidinegroup, the HD activity decreased 70% with respect tothat of cells in the untreated diabetic animals. Inter-estingly, the HD activity of smooth muscle cells fromthe nondiabetic, alpha-hydrazinohistidine group was43% lower than that of cells from the control group.All differences are significant (p < 0.05).

Linear regression analyses of the histamine con-tent against the HD activity yielded significant corre-lations in the case of smooth muscle cells from thenondiabetic, alpha-hydrazinohistidine group (r =0.81, p < 0.05), and in endothelial cells of the un-treated diabetic group (r = 0.68, p < 0.05). A highlysignificant correlation (r = 0.86, p < 0.005) wasobtained for smooth muscle cells when both diabeticgroups were included (Figure 2).

Aortic Albumin Accumulation

The 3- and 12-minute postinjection plasma activi-ties of 125l-albumin from all rats used in this aspect ofthis study are presented in Table 2. No significantdifferences exist in either the 3-minute or the 12-minute 125l-albumin activities among the four treat-ment groups. Overall, the 12-minute postinjection125l-albumin activity represents a 6% decrease overthat of the 3-minute plasma samples.

The thoracic aortic intima-media 125l-albumin fluxand mass transfer rates are presented in Table 3. Asindicated, the mean albumin mass transfer rate ofthe untreated diabetic groups was 5.56 ± 0.59 x10~7 cm/sec, while that of the control was 3.45 ±0.48 x 1O~7 cm/sec. This 61% difference between


Dow

nloaded from



Table 2. Plasma Albumin Activity (cpm) at 3- and 12-Minutes following Injection of (12Sl)-BovlneSerum Albumin

Mean ±SEM

Control

3 min

933009462090097943729644996770

94268991

12 min

879179131185260880029120290195

88981962

Nondiabetic-aHH

3 min

933199472994017913109675793515

93941731

12 min

894008866289115863199095587920

88729633

Diabetic

3 min

977639544195211960059137396097

95315869

12 min

900989091289645917608502088097

89255985

Diabetic-aHH

3 min

925989967394211895439376094110

939831343

12 min

862129372789514832728637990010

881861496

Diabetes was induced by a single jugular vein injection of streptozotocin (60 mg/kg) in acidified, citratedsaline (pH 4.5). aHH = alpha-hydrazinohistidine (25 mg/kg, i.p. at 12 hours for Days 21-28 of the 28-dayperiod).

2 3 4 5 6 7HISTAMINE CONTENT(nM/mg)

Figure 2. Relationship between aortic smooth musclehistamine content and histidine decarboxylase (HD) activ-ity of untreated and alpha-hydrazinohistidine-treated dia-betic rats.

the two groups is highly significant (p < 0.01). Themean aortic intima-media albumin mass transfer rateof the untreated diabetic group was 44% greaterthan that of aortas from the diabetic, alpha-hydra-zinohistidine-treated animals, a difference which isalso highly significant (p < 0.025). No significantdifferences in the mean albumin mass transfer ratesof aortic intima-media preparations existed betweenthe alphahydrazinohistidine-treated diabetic ani-mals, the controls, and the alpha-hydrazinohistidine-treated nondiabetic animals.

Discussion

The albumin accumulation studies involved threebasic assumptions: 1) that the 125l-activity was con-tained within the albumin fraction of the ii5l-albuminas opposed to its degradation products; 2) that theinjected 125l-albumin was not subject to any signifi-cant plasma degradation during its circulation; 3) thatthe aortic 125l-activity reflected primarily that of accu-

Table 3. Thoracic Aortic (125l)-Albumin Flux and Mass Transfer Rate in Experimental Diabetes

Mean ±SEM

Control

J(X10"2)

2.994.762.062.304.622.86

3.270.47

J/CpO(x10-7 )

3.215.032.292.444.792.96

3.450.48

Nondiabetic-aHH

J(X10"2)

1.842.283.912.864.942.51

3.060.47

J/CpO(x10-7 )

1.982.414.163.145.112.69

3.250.48

Diabetic

J(x10-2 )

4.616.535.004.353.577.40

5.24*0.59

J/CpO(x10"7)

4.986.555.314.863.817.87

5.56*0.59

Diabetic-aHH

J(x10"2)

5.354.413.432.922.333.62

3.68t0.43

J/CpO(x10"7)

5.484.623.613.022.553.77

3.85t0.43

Diabetes was induced by a single jugular vein injection of streptozotocin (60 mg/kg) in acidified, citratedsaline (pH 4.5). aHH = alpha-hydrazinohistidine (25 mg/kg, i.p. at 12 hours for Days 21-28 of the 28-dayperiod). J = aortic albumin flux (cpm/cm2/sec); J/CpO = aortic albumin mass transfer rate (cm/sec),where CpO is the 3-minute postinjection plasma (125l)-albumin activity (from Table 2).

'Difference from control significant (p < 0.01, variance analysis followed by Duncan's multiple rangetest).

tDifference from diabetic group significant (p < 0.025, variance analysis followed by Duncan's multiplerange test).


Dow

nloaded from



mulated 125l-albumin as opposed to its degradationproducts or significant amounts of free 1 2 5 I . With re-spect to the first assumption, less than 2% of the totalradioactivity was contained within the TCA solublefraction. Furthermore, information supplied by themanufacturer indicated that the albumin was morethan 95% pure as quantified by AgCI ion exchangechromatography. Undoubtedly, there are some albu-min degradation products present, but these areminimal with respect to the total albumin used. Theassumption that the injected 125l-albumin was notsubject to any significant degradation is based on theshort circulation time (20 minutes) used in this study,and upon data of Bratzler et al.13 which indicate thatover a period of 100 minutes the nonprecipitableplasma 125l-activity is approximately 1% of the totalradioactivity. While we did not directly assess thedistribution of 125I between the circulating albuminand other plasma proteins, on the basis of the shortcirculation time we believe that no significant ex-change occurred between the various classes ofplasma proteins. The assumption that the 125l-activ-ity of the aorta reflects that of accumulated 125l-albu-min is also based on the data of Bratzler et al.13 whoshowed that in regions of the media corresponding tothe thickness of our intima-media preparations, asignificant component of the total 125l-activity is TCA-precipitable. While they do show that some dissocia-tion of the label occurs within the aorta during a 30-minute period, the aortic label accumulation doesserve as an overall estimate of the influx of the 1 2 5 I -albumin. Finally, the similarity of our data with thoseof others (see below) serves as an additional indica-tor that the methods used in this study are valid indi-cators of aortic albumin accumulation as assessedhere.

Levine and coworkers18'19 performed a series ofstudies showing that alpha-hydrazinohistidine is arelatively specific inhibitor of HD, and that a single100 mg/kg dose results in lowering the histaminecontent in such organs as the heart (30%) and stom-ach (50%). Taylor and Snyder20 achieved a 50% re-duction in histamine content of the mouse brain afterinjection of a single 300 mg/kg dose. Schwartz etal.21 showed that alpha-hydrazinohistidine almostcompletely inhibited histamine formation in rat brainextracts at a concentration of 10~4 M, and that therewas a close parallel between HD activity and thehistamine level in most brain regions. Levine andNoll19 have also shown that repeated administrationof alpha-hydrazinohistidine in a 50 mg/kg dose re-sults in a significant lowering of histamine levels invarious tissues, values in each case being lowerthan those obtained by a single dosage. Thus, alpha-hydrazinohistidine offers an excellent pharmacologi-cal tool for examining the role of the aortic HD systemin the regulation of aortic histamine pools in experi-mental diabetes as well as in experimental athero-sclerosis.

Kahlson and Rosengren22 and Levine and Noll19

have described three distinct pools of histamine

present in mammalian tissues. One is the mast cellpool which can be depleted by detergents such asCompound 48/80. The second pool has not beenfully characterized and its depletion has not yet beenachieved. Both pools are characterized by low ratesof histamine synthesis and catabolism and, there-fore, have slow turnover rates. The third pool, re-ferred to as either the inducible or nascent pool, hasa high turnover rate, is depleted by inhibition of HD,and the histamine content of this pool is intimatelyrelated to the balance between its rate of HD-mediat-ed synthesis and enzymatic catabolism. This poolhas been implicated in the physiological processesassociated with intrinsic regulation of microcircula-tory flow23'24 and neurotransmission.25 This induciblepool is also associated with pathological processes,including the prolonged phase of inflammation28 andabnormal increases in arterial wall transmural per-meabililty during atherogenesis.68

Orlidge and Hollis7 showed that in experimentaldiabetes there was a large increase in aortic endo-thelial and smooth muscle histamine content. Thisincrease showed excellent correlations with meas-ured HD and histaminase activities for both endothe-lial cells (r2 = 0.94) and aortic smooth muscle cells(r2 = 0.90). In each case the strongest predictor wasthe HD activity. These authors also reported thatinsulin administration to diabetic rats for the lastweek of a 4-week diabetic period produced completereversal in the histamine content, HD, and histamin-ase activities in both cell types. On the basis of theseobservations, they postulated that the aortic induc-ible histamine pool is expanded in experimental dia-betes, and that because insulin reverses this expan-sion, it may be one modulator of aortic HD activity.

Data from the present study clearly show that inthe dosage and dose regimen used, under controlconditions alpha-hydrazinohistidine had no effect onthe endothelial histamine content or HD activity,while in smooth muscle it produced a 61% reductionin histamine content and a 43% reduction in HD ac-tivity. The lack of change in endothelial cells is notsurprising, since under normal conditions endotheli-al histamine is derived from several sources includ-ing plasma27 and the normal contribution of the endo-thelial HD system to the intraendothelial histaminecontent appears to be minimal.7 However, these dataindicate that in the diabetic state increases in bothendothelial and smooth muscle histamine contentand HD activity are completely reversed in the pres-ence of alpha-hydrazinohistidine, despite the persis-tence of uncontrolled diabetes. Since this expansionof histamine is blocked by inhibition of the increasedHD activity, these data clearly indicate that in strep-tozotocin-induced diabetes there is a marked expan-sion of the aortic inducible histamine pool in bothendothelial and smooth muscle cells, confirming ear-lier speculations.7 Considered in the light of previousfindings that insulin treatment also produces com-plete reversal of diabetic-induced increased HD ac-tivity and histamine content of these same cells,7


Dow

nloaded from



these data also offer additional evidence that insulinmay modulate the aortic inducible histamine poolthrough its actions on HD, at least under the condi-tions of these experiments.

Bratzler et al.13 offer quantitative data indicatingthat albumin enters the media of the thoracic aorticwall of rabbits via two routes, the intimal endotheliallining and the adventitial vasa vasorum. Thus, as-suming a constant albumin space, the rate of albu-min uptake by the media depends on the permeabil-ity of the barriers presented by both routes, andconversely, the permeability of each barrier influ-ences the medial uptake of albumin. A major con-tribution of this study is the demonstration thatbecause of a two-barrier, double-route entry, mea-suring the aortic albumin activity without also mea-suring the albumin distribution interferes with the es-timation of aortic albumin permeability by any singleroute. For this reason, the 20-minute aortic intima-media albumin uptake studies performed here havebeen based on a measurement of the aortic albuminmass transfer rate as defined by Bratzler et al.13 Ourstudies cannot serve as a quantitative estimate ofthe permeability of any one route of albumin entry,nor are they intended to do so. They do serve, how-ever, as estimates of total endothelial permeabilityby both routes combined.

The albumin mass transfer values from normalanimals in the present study are comparable to thoseof Bell et al.28 These researchers used a longitudinalsection technique on intima-media preparationsfrom pig aortic arch and abdominal aorta to estimatealbumin influx which ranged from 9.0 ^g/cm2/hr to 27^g/cm2/hr. When expressed as mass transfer rates,these values ranged from 8.5-27 x 10~8 cm/sec.Our values are also comparable with the data ofBratzler et al.,13 who obtained mass transfer rates inrabbit descending thoracic aorta of 2.4 x 10~8 cm/sec at a 125l-albumin circulation time of 30 minutesand with the data of Owens and Hollis6 who in thedog aorta obtained rates from 1.48 x 10~7 cm/sec to58 x 10~7 cm/sec, depending on the location of theaorta.

The results of the present study indicate that instreptozotocin-induced diabetes, there is a 61% in-crease in the albumin mass transfer rate in the tho-racic aorta. While the aortic albumin space was notexamined here, this finding is consistent with theincreases in permeability within the microvascula-ture commonly associated with the diabetic micro-angiopathy.15

The most significant finding of the present study isthat alpha-hydrazinohistidine administration pre-vents increases in intima-media albumin mass trans-fer rates in experimental diabetes, despite the per-sistence of overt diabetic symptoms and chronichypoinsulinemia. This finding gives direct experi-mental evidence that the accelerated aortic de novohistamine synthesis, which occurs in experimentaldiabetes67 and which can be maintained at normallevels using this inhibitor, is a significant mediator of

the increased aortic albumin accumulation in the dia-betic state. It is likewise consistent with the initialstudy by Owens and Hollis8 which showed that al-pha-hydrazinohistidine administration to cholesterol-fed rabbits resulted in a 38% inhibition of aortic hista-mine synthesis, a significant decrease in aorticalbumin uptake, and a 50% reduction in the severityof atherosclerotic plaques as assessed by the per-centage of surface area showing plaque develop-ment. Thus, a common finding between two differentatherogenic situations is that prevention of the in-crease in aortic histamine synthesis that occurs inboth these conditions can likewise prevent the in-creased aortic albumin accumulation.

The other interesting finding of this study is thelack of any significant differences in aortic albuminaccumulation between the control and alpha-hydra-zinohistidine-treated nondiabetic animals. This re-sult is consistent with the data of Owens and Hollis6

who found no difference in albumin accumulationvalues between control rabbits and noncholesterol-fed rabbits who had been given alpha-hydrazinohis-tidine in the same dose and dose regimen used in thepresent study. Both studies taken together give ex-tremely strong evidence that the role of the induciblehistamine pool in the regulation of arterial macromol-ecule permeability in experimental diabetes may bea pathological phenomenon indicative of a subtleform of vascular injury.

It is obvious that a number of additional studiesmust be undertaken in order to clarify the involve-ment of the inducible histamine pool in mediatingaccelerated aortic macromolecule accumulation inboth diabetes and other atherogenic states. Suchstudies should include an examination of the in-traaortic albumin space, the dose-albumin accumu-lation relationships, the role of histamine H r andhistamine H2-receptors and their agonists and an-tagonists in this process, the mechanisms that mod-ulate the aortic inducible histamine pool, and mostimportantly, the mechanisms whereby aortic hista-mine mediates increased albumin uptake. However,the data from the present study clearly indicate thatthe inhibition of accelerated endothelial and smoothmuscle histamine synthesis, and thus the preventionof the expansion of the aortic inducible histaminepool prevents an increase in aortic albumin accumu-lation in streptozotocin-induced diabetes.

AcknowledgmentsThe authors thank William Carroll, William Laughlin, Michael

Shetzline, and S. Adam Strickberger for their technical assis-tance, and Mary Alice Shea for manuscript typing.

References1. Brownless M, Cahill GF. Diabetic control and vascular com-

plications. In: Paoletti RW, Gotto AM, eds., AtherosclerosisRev, vol 4, New York: Raven Press 1979:29-70


Dow

nloaded from


606 ARTERIOSCLEROSIS VOL. 3, No 6, NOVEMBER/DECEMBER 1983

2. Jarrett RJ, Keen H. Diabetes and atherosclerosis. In: KeenH, Jarrett RJ, eds. Complications of diabetes. London: Ed-ward Arnold 1975:179-205

3. Knowles HD. The problem of the relation of the control ofdiabetes to the development of vascular disease. Trans AmClin Assoc 1964;76:142-147

4. Robertson WB, Strong JP. Atherosclerosis in persons withhypertension and diabetes mellitus. Lab Invest 1968:18:538-553

5. Wolinsky H, Goldfischer S, Capron L, Capron F, Coltoff-Schiller B, Kosak L. Mydrolase activities in the aorta. I. Ef-fects of diabetes mellitus and insulin treatment. Circ Res1978;42:821-831

6. Owens GK, Hollis TM. Local aortic histamine metabolismand albumin accumulation: Differences between blue andwhite areas. Arteriosclerosis 1981; 1:265-272

7. Orlidge A, Hollis TM. Aortic endothelial and smooth musclehistamine metabolism in experimental diabetes. Arterioscle-rosis 1982;2:142-150

8. Owens GK, Hollis TM. Relationship between inhibition ofaortic histamine formation, aortic albumin permeability andatherosclerosis. Atherosclerosis 1979;34:365-373

9. Gospodarowitz D, Moran D, Brown D, Birdwell C. Clonalgrowth of bovine endothelial cells: fibroblast growth factor asa survival agent. Proc Natl Acad Sci 1976:73:4120-4124

10. Taylor KM, Snyder SH. Isotopic microassay of histamine,histidine decarboxylase and histamine methyltransferase inbrain tissue. J Neurochem 1972;19:1343-1358

11. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Proteinmeasurement with the folin phenol reagent. J Biol Chem1954:193:265-275

12. Wolinsky H, Daly MM. A method for the isolation of intima-media samples from arteries. Proc Soc Exp Biol Med1970:135:364-368

13. Bratzler RL, Chisolm GM, Colton CK, Smith KA, Zilver-smit DB, Lees RS. The distribution of labeled albumin acrossthe rabbit thoracic aorta in vivo. Circ Res 1977:40:182-190

14. Snedecor GW, Cochran WG. Statistical Methods. 7th ed.Iowa: Iowa State University Press 1980

15. Florey L. The endothelial cell. Br Med J 1966;2:487-490

16. French JE. Endothelial structure and function. In: Jones RJ,ed. Evolution of the atherosclerotic plaque. Chicago: Univer-sity of Chicago Press 1964; 15-28

17. Majno G. Ultrastructure of the vascular membrane. In: Hamil-ton WF, Dow P, eds. Handbook of physiology. Sect 2: Circu-lation, vol 3. Am Physiol Soc, 1965:2293-2375

18. Levine RJ, Sato TL, Sjoerdsma A. Inhibition of histaminesynthesis in the rat by a-hydrazino analog of histidine and 4-bromo-3-hydroxy-benzyloxyamine. Biochem Pharmacol1965;14:139-149

19. Levine RJ, Noll WW. Histidine decarboxylase and its inhibi-tion. Ann NY Acad Sci 1969; 166:235-256

20. Taylor KM, Snyder SH. Dynamics of regulation of histaminelevels in mouse brain. J Neurochem 1972;19:341-354

21. Schwartz JC, Lampart C, Rose C. Properties and regionaldistribution of histidine decarboxylase in rat brain. J Neuro-chem 1970:17:1527-1534

22. Kahlson G, Rosengren E. New approaches to the physiol-ogy of histamine. Physiol Rev 1968;48:155-196

23. Schayer RW. Biogenic amines and microcirculatory homeo-stasis. In: Blum JJ, ed. Biogenic amines as physiologicalregulators. Englewood Cliffs, New Jersey: Prentice Hall1970:237-251

24. Schayer RW. Evidence that induced histamine is an intrinsicregulator of the microcirculatory system. Am J Physiol1962:202:66-72

25. Green JP, Johnson CL, Weinstein H. Histamine as a neuro-transmitter. In: Lipton M, Dimascio A, Killman K, eds. Psycho-pharmacology— A generation of progress. New York: RavenPress 1978:319-332

26. Schayer RW. Significance of induced synthesis of histaminein physiology and pathology. Chemotherapia 1961;3:128-136

27. Foldes A, Mead RJ, De La Lande TS. Endogenous andexogenous histamine in rabbit thoracic aorta. Aust J Exp BiolMed Sci 1976:55:89-102

28. Bell FP, Adamson I, Schwartz CJ. Aortic endothelial perme-ability to albumin: local and regional patterns of uptake andtransmural distribution of 131l-albumin in the young pig. ExpMol Pathol 1974:20:57-68

Index Terms: aortic histamine metabolism • atherogenesis • atherosclerosis • diabetes mellitusaortic albumin uptake • aortic histamine pools • histamine and albumin accumulation


Dow

nloaded from


T M Hollis, S G Gallik, A Orlidge and J C Yost125I-albumin accumulation in experimental diabetes.

Aortic endothelial and smooth muscle histamine metabolism. Relationship to aortic

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 1983 American Heart Association, Inc. All rights reserved.

Avenue, Dallas, TX 75231is published by the American Heart Association, 7272 GreenvilleArteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/01.ATV.3.6.5991983;3:599-606Arterioscler Thromb Vasc Biol.

http://atvb.ahajournals.org/content/3/6/599World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

http://www.lww.com/reprints

Information about reprints can be found online at: Reprints:

document.Permissions and Rights Question and AnswerFurther information about this process is available in theis being requested is located, click Request Permissions in the middle column of the Web page under Services.Clearance Center, not the Editorial Office. Once the online version of the published article for which permission

can be obtained via RightsLink, a service of the CopyrightArteriosclerosis, Thrombosis, and Vascular Biology Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:


Dow

nloaded from

http://atvb.ahajournals.org/content/3/6/599

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://atvb.ahajournals.org//subscriptions/


aortic endothelial and smooth muscle histamine...

Documents