Electron Microscopic Studies of Normal and Abnormal Elastic ...

THE JOURNAL OF INVEST1OATIVE DERMATOLOGY

Copyright 1987 by The Williams & Wilkins Co.Vol. 48, No. 5

Printed in U.S.A.

ELECTRON MICROSCOPIC STUDIES OF NORMAL ANDABNORMAL ELASTIC FIBERS OF THE SKIN*

KEN HASHIMOTO, M.D. AND RICHARD J. DiBELLA, M.D.

With the advent of the electron microscope,investigators of elastic fiber have used twomajor approaches: first, the method of teasingout tissue fragments with or without addi-tional shadow casting technics (replicationmethod) (1—7), and second, that of using tis-sue sections (8—14). Several workers in eachgroup have shown interest in the elastic tis-sue of the skin, either normal (2—4, 6, 10),or abnormal (9, 11—13).

The replication method held a great ad-vantage over that of the tissue sections forstereographic analysis. Thus the group whichadopted the former technic had long ago es-tablished the fact that elastic fibers were madeup of elementary filaments or fibrils (2, 3),whereas the tissue section group consideredthe fine structural image of the elastic fiberto be an amorphous, relatively electron-lucidsubstance until quite recently (8, 10). Thereplication method is, however, much toosuperficial and too limited to elucidate ex-actly what is contained inside the elasticfibers. Furthermore, in teased tissues it wasalways difficult to know exactly what areaof the lesion had been prepared. These handi-caps have occasionally yielded serious misin-formation, as, for example, an investigationreporting the absence of elastic fibers in thelesions of pscudoxanthoma clasticum (3). Inthis series of investigations a technic em-ploying osmic acid-fixation, Araldite embed-ding, and uranyl acetate-lead citrate doublestaining has been used throughout in orderto provide a uniform basis for the comparisonof various pathological changes.

This investigation was supported by ResearchGrant GM-10299 from the National Institute ofArthritis and Metabolic Diseases, United StatesPublic Health Service.

Received for publication June 17, 1966.* From the Department of Dermatology, Tufts

University School of Medicine, and the Derma-tology Research Laboratories, New England Medi-cal Center Hospitals, and the Boston City Hospi-tal, Boston, Massachusetts.

PART I. FORMATION OF THE NORMALELAsTIc FIBER

Materials and MethodsIn selecting skin specimens for biopsy, care

was taken to choose a younger age group ofnormal individuals and to avoid sun-exposed areas.Thus, specimens were taken by punch biopsyunder local procain anesthesia from the back,upper arms and legs of white males and femalesbetween 26 and 46 years of age. Tissues were im-mediately cut into small pieces ( 1 x 1 mm across),fixed in a 1 per cent osmic acid solution bufferedto pH 8.0 with veronal buffer for 1 and ½ hours,dehydrated through graded concentrations ofethyl alcohol and propylene oxide, and embeddedin Araldite. Thin sections of 400—600 A were cuton an LKB Ultratome and were stained first for15 minutes in 1% uranyl acetate dissolved in 50%ethanol, then rinsed in plain 50% alcohol. Beforethe sections were completely dried, they wererestained in Reynolds' lead citrate for 10 minutes,rinsed briefly in distilled water, dried, and ex-amined with an RCA EMU-3G electron micro-scope. For the sake of comparison, the aorta of a500 gram rabbit was dissected, processed, stained,and observed under the same conditions as wereused for the human specimens.

Results

Initial stages of formation —In foci of ac-tive elastic fiber formation, it was observedthat an increased number of fibroblasts eitherlay very close to, were in definite contactwith, or encircled young elastic fibers (Figs.in, ib). Such fibroblasts demonstrated clustersof secretory vesicles merged with plasma mem-branes and very well developed rough-surfacedendoplasmic reticulum. Thus they could notbe differentiated from those fibroblasts en-gaged in collagen production. In fact, they werefrequently found producing elastic fibers onone side, and collagen fibers on the other(Figs. 3b, 5). Along the peripheral cytoplasmof these fibroblasts condensed aggregates offine cytoplasmic filaments with diameters of50 to 150 A were often observed (Figs. 2a,2b, 2c, 3a, 3b). Some of these aggregationswere highly electron-dense since, in additionto the filaments, dense substances had also been

405

406 THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

FIGs. la and lb. Elastic fiber (E) is being formed by a fibroblast (F) (or an elastoblast)in each picture. These elastic fibers could either be confined within cisternae of individualfibroblasts or simply be surrounded by them. Small vesicles (thin arrows) and other cellularorganelles (thick arrows) are embedded in the dense granular substances which fill thespaces between skeleton fibrils (s). C: collagen fibrils. er: rough-surfaced endoplasmic re-ticulum. m: mitochondria. Fig. la: X 14,600. Fig. lb: X 21,900.

0 __

__

, rj C

.-*

Jr

i.

S

— - -

N-

C:)

4

&

r -

;I .

a,2t

: !

'1

L-- I.

7. 'It

a;

•:

.. -a

. A

"•

S

IM

- -q

L

)Eis

: a

- -S r

' -.

w.

s4

2r

-

1

ULTRASTRUCTURE OF THE ELASTIC FIBER 407

Fies. 2a—2c. Aggregates (thick arrows) of fine cytoplasmic filaments (f) are seen alongthe periphery of each fibroblast (F). The aggregates in Fig. 2a and Fig. 2b are dense. Theplasma membranes of these fibroblasts are obscure and some filaments appear to be contin-uous with extracellularly located elastic fibers (E) and with the protofilaments (thin arrows)surrounding them: C: collagen fibrils with periodical striations. G: glycogen particles infibroblasts. S: skeleton fibril. >( 43,875.

Fro. 2d. Protofikiments (thin arrows) of an elastic fiber (E) are present in close proximityto or in contact with an elastic fiber (E). These filaments have similar dimensions to thoseof the fine cytoplasmic filaments shown in Figs. 2a—2c, and are at various stages of aggrega-tion. The aggregate demonstrates a beading pattern (thick arrow). X 43,875.

• ALd

S

F

Fios. 3a and 3b. Fine cytoplasmic filaments (f) and extracellular protofilaments (p) ofelastic fibers (F1, E2, E2) appear to be continuous through ill-defined plasma membranes(thick arrows). The fibroblast in Fig. 3b must have been cut tangentially near the surfacebecause a small portion of elastic fiber (e) is left inside the cytoplasm with surroundingplasma membranes. Note that within the aggregates of protofilaments of the elastic fiber,precollagen fibrils (c) are also formed. Dense granular substances surround individual skele-ton fibrils (s) and appear condensed along the peripherly of F1 and F3 (thin arrows). r:rihosome or glycogen of fibroblast.

408

,.

'a

S.

S;. ñ $

t 41_II- ,

I-F-

A'

I

1.--. -n -'tt

W

- ,-:e'c .-44

I

4 ,

I':1-.9

V0

S

II.e9

ULTRASTEUCTTJEE OF THE ELASTIC FIBEE 409

deposited (Figs. 2a, 2b). A few extracellularelastic fibers located in the vicinity of thesefibroblasts appeared to be continuous with theintracytoplasmie thin filaments when plasmamembranes were not distinct (Figs. 2a—2e) ortangentially cut (Fig. 3b).

In the majority of instances the formation ofthe dermal elastic fiber proceeded in the ex-traeellular spaces as follows (Figs. 4a—4d) : (1)precipitation of very thin proto filaments inthe extracellular spaces; (2) aggregation andaccretion of individual protofilaments; and(3) deposition of dense, amorphous or slightlygranular substances between and upon the ag-gregating filaments. The initial precipitationof protofilaments was seen either at a shortdistance from the fibroblast or in contact withit (Figs. 3b, 4a). When definition of the plasmamembrane was poor because of the angle atwhich the plasma membrane had been cut,and/or because of failure of preservationtechnic, these protofilaments appeared to becontinuous with the similar cytoplasmie fila-ments (Figs. 2a, 2b, 3b). The protofilamentsalso measured from 50 to 150 A, and thuswere equivalent in size to both the tropocol-lagen filaments and the eytoplasmie filaments(15, 16). Some protofilaments demonstratedprimitive headings of 150 to 200 A repeat(Fig. 4e); and in such instances, it was impos-sible to distinguish them from similarly sizedand beaded filaments of tropoeollagen. This factmade it even more difficult to differentiatebetween the two types of filaments morphologi-eally (Figs. 3a, 3b, 4b, 4d, 5, 6).

While the aggregation of these protofilamentswas taking place, dense matrix substances,which were either amorphous or finely granu-lar, were deposited on and between the ag-gregating fibrils (Figs. 4b, 4d). Thesevariously sized fibrils resembled precollagenfibrils of corresponding sizes (Figs. 4b, 4d),but they could usually be differentiated be-cause the aggregated protofilaments of theelastic fiber were (1) much denser than thoseof the precollagen fibrils (Figs. 3b, 4b, 4d),(2) irregular in width (Figs. 3b, 4b, 4e, 4d),(3) surrounded by finely granular dense matrixsubstance (Fig. 3b) which continued tothicken them, and (4) did not show well-de-fined, periodical eross-striations, though rather

irregular headings were occasionally present(Fig. 2d).

Maturation processes—Individual fibrils,formed from the aggregation of protofilamentsand increased in size by the incorporationof dense matrix substances, provided the majorfibers. Since these fibrils were to constitutethe skeleton-work for mature elastic fibersf,they will be referred to as skeleton fibrils inthis report. These fibrils varied widely inthickness from 150 to 800 A, and became eventhicker after lateral coalescence of individualfibrils (Figs. 4b, 4e, 4d). The individual skele-ton fibrils were surrounded by dense granularsubstances (Figs. la, ib, 3b, 4, 5, 6), whichwere essentially the same as the dense matrixsubstance, but were occasionally coarser. Whenthe skeleton fibrils coalesced, they became con-densed (Fig. 5). Often found mixed withthese coarsely granular substances were roundvesieles, dense granules, and even parts of thecytoplasm of near-by fibroblasts (Figs. la, ib,5). It was therefore assumed that dense granu-lar substances also derived from the fibro-blasts. In the mature elastic fiber, individualskeleton fibrils were clearly visible only alongthe periphery and at frayed ends (Fig. 6).Matured skeleton fibrils became relativelylight faseieles which were separated by thecondensed dense granular substances.

PART II. P5EUOOXANTHOMA ELA5TICUM

Since Darier (17) offered the name ofpseudoxanthoma elasticum to this condition in1896, his concept that the dermal elastic tis-sue is the main site of involvement has notbeen seriously questioned until fairly recently.Beginning with the report of Tunbridge etal. (2) who, in 1952, found no ill-definedstrands resembling elastic tissue in any ofthree specimens examined with the electronmicroscope (replication method), opinionshave become split concerning the true natureof the elastic stain-positive material in pseu-doxanthoma elastieum. Findlay (18), Moran

t In describing the thin fibrous components ofelastic fibers, the phrase elastic fit rum has some-times been used. But "fibril", as used in this re-port, denotes elements thicker than filaments, butthinner than fibers. So that elastic fibrils may referto thinner varieties of whole elastic fibers and notto their component fibrils. The term skeletonfibrils, therefore, has been adopted to denote thespecial fibrous components within each fiber.

FIGS. 4a 4d. The formation of the dermal elastic fiber proceeds in the extracellular spaceswith the initial precipitation of protofilaments (80—100 A, arrow 1), aggregation of these pro-tofilaments and incorporation of dense matrix substances forming a thin variety of skeletonfibrils (150—200 A, arrow 2) which grow thicker (350—400 A, arrow 3) and thicker (400—800 A,arrow 4). These thicker varieties become laterally coalesced (Figs. 4c and 4d). Collagen fi-brils (c) in the vicinity have similar dimensions (350—400 A) to those of thicker varieties(e.g., arrow 4) of skeleton fibrils, but they show regular periodicity. D: dense matrix sub-stances surrounding skeleton fibrils. F: fibroblast. r: ribosome or glycogen. v: secretoryvesicle or caveola. X 42,805.

Fia. 4e. Beading patterns on protofilaments. e: a part of an elastic fiber. X 42,805.

410

- - '- -."ic'

4-

y —•

•——••( .••• 1,....•

-

•',1jI•

•'w

4

p

p

C'7¼ - * a'----€

'It'1

a,..%'j t .—--I-,• y.

- 1


FIG. 5. Formation of both elastic and collagen fibers is taking place simultaneously neara fibroblast whose cytoplasm has been cut into two portions (F1, F1). Spaces between indi-vidual collagen fibrils (c) are filled with intercollagenous filaments, while individual skeletonfibrils (s) are cemented together by dense granular substances (d). Within these dense gran-ular substances small vesicles (arrows) are mixed. p: protofilaments of elastic fiber. t:tropocollagen filaments. v: secretory vesicles or caveola. >< 3,168.

-FF

'I1*l

if,.4

4..

t I

4, 3

tif

ML'

-I

if

P

.7 .

.4

P

•9 It

412 THE JOURNAL OF INVESTiGATIVE DERMATOLOGY

FIG. 6. Mature elastic fiber is composed of coalesced fascicles of skeleton fibrils (s) andcondensed dense granular substances between them forming dense granular septa (d). Whenthese dense septa are followed to forking points of the skeleton fibrils which they cementtogether, gradual processes of condensation of dense granular substances (which flank coa-lescing skeleton fibrils) into dense granular septa are seen (thin arrows). There are numerousthin filaments present which connect both with collagen fibrils (c) and with skeleton fibrils(thick arrows). >< 37,740.

C

_i

S $

iq

ULTRASTRUCTTJRE OF THE ELASTIC FIBER 413

and Lansing (19), Fisher, Rodnan and Lans-ing (20), Lobitz and Osterberg (21), andLoria et al. (22) defended the view that pseu-doxanthoma elasticum represents a degenera-tion of elastic fibers, basing their argumentsupon the finding that all the material stain-ing like elastic fibers was removed upon elas-tase digestion (18, 19); on the autofluorescence(20) and microincineration studies (21) ofthe affected skin; and on the electron micro-scopic identification of the elastic fibers inthe lesion (19, 22). Hannay (23), on the otherhand, could not identify elastic tissue in thelesion with the electron microscope, and Mc-Kusiek (24), who once believed that theorcein-positive material was produced by an"elastotic degeneration of collagen", later aban-doned this view and agreed with the theorythat calcification of the elastic fibers is theprimary change (25, 26). The histoehemicalfinding of Goodman et al. (25) that thedeposition of calcium on elastic fibers, whichare otherwise normal in appearance, is theearliest detectable change by the light micro-scope is particularly significant because theresults of this study have agreed with it atthe ultrastructural level.

Materials and Methods

Three patients who had clinically and histo-logically typical lesions of pseudoxanthoma elasti-cum were studied. The first two patients were Cau-casian women, sisters, 44 and 50 years of age, whoboth had, in addition to the skin lesions, centralscotoma due to the development of angioidstreaks in both eyes. The third patient was a 45year old Caucasian woman who also suffered fromimpaired vision. Each patient had noted the onsetof the disease during their high school years. Typi-cal skin lesions were observed in the areas mostfrequently subject to fiexion and extension, suchas the antecubital fossae of the elbows, the axil-lary fossae, the femoral-thigh regions, and bothsides of the neck. The lesions in the axillae andthighs were chosen for biopsy removal becausethese areas are not usually exposed to the sun,and particularly in these patients, for cosmeticreasons, thus avoiding the possible complication ofactinic (senile) elastosis.

Two lesions were taken from each patient andprepared for electron microscopic studies by thesame method as described in Part I. Part of eachspecimen was also embedded in paraffin and see-

tions stained with chloranilic acid for the detectionof calcium salts (21). This new method stainscalcium specifically, forming calcium chloranilatewhich is dark-red to brownish-black and hirefrin-gent (27), whereas the conventional Von Kossastain is specific for phosphates and carbonates (27,28).

Results

Elastic fiber.—The number of elastic fiberswas greatly increased in the reticular dermis,most of them showing severe degenerativechanges (Fig. 7a). The individual fibers werenot slender as in normal skin (Figs. 4c, 4d,5), but multiformed: stubby, folded, clumped,or star-shaped, for example (Figs. 7a, 8a). Mostof the elastic fibers in the lesion were heavilyinfiltrated by dense material which, in viewof a positive chloranilic acid stain (Fig. 7b)and strong birefringence of the stained prod-uct (Fig. 7c), as well as by its ultrastruc-tural morphology, was interpreted as being acalcium deposit. In comparison with the elas-tic fiber in the normal condition (Figs. 5, 6),these calcified elastic fibers showed abnormallydense areas, granular textures and manyempty holes of various sizes and configurations(Fig. 8a), interpreted as heavily calcified por-tions which dropped out when thin sections werecut. In severely damaged fibers, normal-lookingelastin was seen only along their peripheries(Figs. Sa, 8b, 10), and the rest of the fiberwas completely degenerated into a light, finelyparticulate substance (Fig. 9). In such degener-ated portions, there were criss-crosses of denselinear calcium deposits (Figs. 8a, 9). Calci-fication was found to occur primarily on theelastic fibers and in very limited areas ofthe connective tissue.

Collagen fiber—Collagen fibers appeared nor-mal in most of the areas examined (Figs. 7a,9), showing the usual striation pattern, theirdiameters being within the normal range of1,000 to 1,300 A (Fig. 9). In some instances,however, thick, clumped collagen fibers wereobserved (Figs. Sa, 10), whereas in others, anadmixture of degenerated fibers of collagen anda dense non-fibrous component of the elasticfiber, the so-called matrix substance (see PartI), derived from degraded elastic fibers, wasfound (Fig. 10). These changes are often pres-ent non-specifically in actinic (senile) elasto-sis (29, 30), as well as in many other disorders

t All three patients were made available forthis study through the courtesy of Dr. John L.Fromer from the Lahey Clinic, Boston, Massa-chusetts.


Fie. 7a. A scanning view of the skin lesion. Most of the elastic fibers show severe degen-erative changes due to heavy calcification (C). Calcium which has dropped out has leftclean, punched out holes (arrows) within degenerated fibers. Collagen fibers and fibroblasts(F) appear normal. X 4,198.

FIG. 7b. Chioranilic acid stain for calcium is positive (dark red) in fibrous componentsof the lesion, which together with the electron microscopic findings, were interpreted as cal-cium infiltrated elastic fibers. X 73.

Fia. 7c. Birefringence of the stained product with chioranilic acid. The same section asseen in Fig. lb demonstrates a strong birefringence when examined with polarized light.)< 73.

Fia. 7d. The same tissue seen by phase contrast microscopy. X 182

•'5r•

-—


FIG. Sa. An affected elastic fiber in the center is abnormally large and clumped. Calcifica-tion appears as dense areas (d), granulated portions (g), and as areas of light particulatetexture or of ground glass appearance, (p). Many punched out clear holes (h) representingareas from which calcium salts have dropped out are also seen. A criss-cross of dense lines(thin arrows) represents areas of heavier calcium deposition (cf. Fig. 4). The peripheries ofthis elastic fiber (thick arrows) and of small (young) elastic fibers (e) were either sparedfrom calcification or were less affected; they demonstrate, however, fibrous components, atleast. Collagen bundles (C) appear normal except for some thick clomped ones (c). X 10,585.

Fia. Sb. A completely degenerated elastic fiber shows fine particulate texture with sprin-kled crystals of calcium salts. Along the periphery of this fiber normal-appearing denseelastin remains intact (arrows). >< 10,585.

•'1':r-- -

4- 4,-.

.. ,4- —

S

'a4,/ *

S

'-'p.

SI 'tar)a


Fia. 9. High magnification of a degenerated elastic fiber reveals particulate deposits ofcalcium salts which are often accumulated in higher concentration producing dense granularlines of various sizes and densities (arrows) within the elastic fiber. Elastin (E) at the pe-riphery of this fiber appears normal (cf. Fig. 2b for normal elastin at a comparable magni-fication). C: normal collagen fiber with regular striations. X 43290.

sr

ti'

• :

• 7'Jr •, •.

• • A.

.1

ii'* .4%

1•I


FIG. 10. Higher magnification of elastin-collagen admixture. The dense amorphous com-ponent (A) of degenerated fibers and partially degenerated collagen (C) fibers are mixed.Some collagen fibers are thick, crooked and clumped (c). Ca: calcium salts. E: unaffectedperiphery of elastic fibers. X 42,120.

of connective tissue such as elastosis perforans tic fibers, fibroblasts with a well-developedserpiginosa (31). ergastoplasm (Fig. ha) and tropocollagen-

Fibroblasts and protofilaments of the elastic like thin filaments (Fig. lib) were frequentlyfiber—In the vicinity of the degenerated elas- observed. The tropocollagen-like filaments were

S

aS

S

.4

Ta a

t

4


Fia. ha. A fibroblast with well-developed rough-surfaced endoplasmic reticulum (re) inthe lesion seems to be producing both an elastic fiber (E), which has already been partiallydegenerated, and tropocollagen filaments (arrows) for a collagen fiber. Note the simi'aritybetween these tropocollagen filaments and the protofilaments shown in Fig. 7b. X 10,730.

FIG. lib. Fibrob lasts (F) and protofilaments (p) of elastic fibers in the lesion appear nor-mal. C'ear empty spaces (H) correspond to the areas where calcium salts have dropped out.Ca: calcium salts. E: unaffected part of the elastic fiber. >< 14,800.

4-

- -

*• e

0

•t •r'1,.• H

I,,

•-1

4tHrH

'fa

S

C:'-

"S -.4I.

1• a

7.,

4-a _____

C.

a

/ I

a

I- 0•-


Fm. 12. Amorphous substances fill in the spaces between collagen fibers (C) and degen-erated elastic fibers (E). There is only a faint indication that they may be fibrous also(arrow). X 10,585.

identical to the protofilaments produced byfibroblasts at the site of normal elastic fiberformation (Figs. 2—4). These findings suggestthat an active formation of elastic fibers wasunderway while the maturing elastic fiberswere undergoing degenerative changes.

The initial stages of elastic fiber formationseemed to proceed normally. Both the aggrega-tion of protofilaments and incorporation ofdense matrix substance were seen in manyfoci as in normal elastogenesis and thus thevery young elastic fibers in the lesions ofpseudoxanthoma elasticum were either freefrom calcium deposition, or were calcifiedonly in limited areas (Figs. 7a, 8a).

Others—In some areas of the lesion, amoderately electron-dense, amorphous or slightlyfibrillar substance was deposited between col-lagen bundles and between the degeneratedelastic fibers (Fig. 12). Thus the elastic fibersappeored to be embedded in this substance

which is unknown in normal connective tis-sue of the dermis. The fibrillar component,however, was too scanty, and its fibrous com-ponent too fuzzy and wavy for this substanceto be amyloid (32, 33). Neither was it fibrinnor tropocollagen because there was no periodi-cal striation. It might possibly be, however,some abnormally increased ground substance.

BI5CU55ION

Elostic fiber and colic qen fiber.—The initialstages of the dermal elastic fiber formation asdescribed above are very similar to those ofthe dermal collagen formation (15, 16). In bothinstances, the precipitation of thin filaments,occurring mainly in the extracellular spaces,was the initial event observed with the elec-tron microscope. These filaments, called pro-tofilaments and tropocollagen filaments (16, 34)respectively, showed primitive headings ofsimilar repeating intervals. Although the lat-

- — aa4

4--:.

-%wt.

S.A,i tit'*I e')1'

I

4%


eral aggregation of these filaments formsthicker fihrils in each case, the skeleton fibrilswhich formed the elastic fibers were impreg-nated with dense matrix substances and didnot develop periodicity, while the correspond-ing precollagen fibrils developed clear-cutstriations as they became mature collagenfibers. As the maturation progressed, individ-ual skeleton fibrils became irregularly thickerand coalesced laterally by means of the cement-ing action of the dense granular substanceswhich formed septa between them. The iden-tity of the skeleton fibrils of mature fibers waspreserved only by the presence of the densegranular substances which separated them. Inmature collagen fibers, on the other hand, theregularly thickened fibers, though bundled to-gether, never were coalesced.

The association between elastic and collagenfibers has been repeatedly confirmed by theelectron microscope (6, 16). This study, aswell, has demonstrated such close associationin practically every picture shown. This as-sociation is a fact easily understood becausethe elastic fibers are composed of essentiallythe same basic building materials as collagenfibers in their formative stages.

Fibroblast and "elostoblast".—As in the der-mal collagen fiber formation, the fibroblastsseemed to provide all the building materialsfor elastic fiber formation. The fibroblasts en-gaged in elastogenesis demonstrated abundantwell-developed rough-surfaced endoplasmic re-ticulum and small clustered vesicles attachedto plasma membranes. Morphologically, there-fore, the fibroblasts found in the foci ofcollagen synthesis (15, 16, 35) were no differ-ent from those found in the foci of elasto-genesis. It is possible, however, that theproteins synthesized by this endoplasmicreticulum arc qualitatively different. Thus thedense matrix substances and dense granularsubstances, which surrounded and thickenedthe individual fibrils, and filled the spacesbetween skeleton fibrils, were not found at thescene of collagen formation. Therefore, if anydistinction should be made between fibroblastand "elastoblast," it must be that the latterspecially synthesizes the dense matrix sub-stances whose deposition or incorporationseems to differentiate the fibrils destined tobecome elastic fibers from those which will

become collagen fibers. These dense matrix sub-stances are, in all probability, secreted fromthe "elastoblast" through clusters of vcsicles.Small vesicles and other cellular organellcsfound in these dense granular substances maypossibly be supplied by the secretion whichhas been called ecdysis in fibroblasts engagedin active collagen production (36).

Regarding the previous description of thedense granular substances, Charles (10) de-scribed dense elements with a beaded appear-ance and conjectured that they were eitherfibrillar, or deposits or accretions of a lipidsubstance. He also reported that some elasticfibers had an electron-dense membranoussheath (10) which was not present in elasticfibers examined in this study. It is assumedfrom the examination of his electron micro-graphs that the condensed dense granular sub-stances surrounding the periphery of thpelastic fiber (see Figs. 2d, 3a, 3b) appearedto be the dense, continuous sheath at lowmagnification.

Elasticity of the elastic fiber—Since theskeleton fibrils are the fibrous components ofthe elastic fiber from the initial stages of theirformation, i.e., from protofilaments, and sincethe dense granular septa are formed from amixture of dense granular substances and cel-lular components, it is assumed that the skele-ton fibrils are the responsible element for anyresilience ascribed to the elastic fiber. Thedense granular substances which form septa,then, may serve as a buffer or lubricant, re-ducing the amount of friction between adjoin-ing skeleton fibrils during the act of stretching.

Pat ho genesis of pseudoxanthoma elasticum.—The component of connective tissue primarilyinvolved in pseudoxanthoma clasticum was theelastic fiber, not the collagen fibrils. Thepathological changes observed in collagenoustissue were minor in extent, and non-specific.Because the fibroblasts in the lesion wereactively producing elastic fibers, the numberof elastic fibers, though many were severelydamaged, was markedly increased. The totalamount of normal elastin, therefore, might alsobe assumed to have increased, a fact whichhas been biochemically assayed by Smith etal. (37).

Morphologically, the fibroblasts, collagen, andthe initial stages of elastic fiber formation,

ULTRASTRTJCTURE OF THE ELASTIC FIBER 421

except for the increase in number, all appearednormal. The deposition of calcium salts seemedto be the primary event, the initiator of tbedegenerative changes which followed, sincethere had been no previous morphological ab-normalities in the elastic fibers. Furthermore,non-calcified portions of the involved fibersappeared completely normal. Goodman et al.(25) performed alizarin stains for calcium onelastic fibers in the lesion which appearednormal with routine elastic tissue stains, andfound that even normal-looking elastic fibershad already been impregnated with calcium.They then concluded that the calcium infiltra-tion of elastic fibers is the earliest demonstra-ble histologic change seen by the lightmicroscope, and proposed that the calcium dep-osition might lead to the brittleness andfracture of the affected elastic fibers. Concern-ing other possible etiologic factors such as anincrease of clastase, elastic fibers digested withpancreatic elastase do not resemble the de-generated fibers at all (31), thus ruling outthe effect of this enzyme, whether local orsystemic.

It is postulated, therefore, that some geneticfactor (or factors) selectively induces the dep-osition of calcium on the elastic fibers, par-ticularly on those in locations which aresubject to frequent traumas by fiexion andextension (locus minoris rcsi.stensiae). Such agenetic factor could either cause an abnormalmetabolism of calcium, or influence the pro-duction of defective building materials in theelastic fibers which attract the calcium ion.These macromolecular lesions, undetectableeven by the electron microscope, could well bethe initial site of calcium deposition. It isinteresting to note, however, that even nor-mal elastic fibers contain a small amount ofcalcium (38, 39). In experimental calcification,Anderson (40) demonstrated that the initialdeposit of calcium salts occurs on thin collagenfibrils. In calcifying cpithelioma the initialsite of calcium deposition seems to be onthin precollagen fibrils (41). If the calciumdeposition occurs preferentially on thin fibrouselements, the abnormal protofilaments or thethin variety of skeleton fibrils could elicitsuch deposition. McKusick (26), on the otherhand, suggests that there is a deficiency ofvarious cnzymcs which arc important to the

metabolism and incorporation of the elasticfiber proteins, and thus essential to the integ-rity of the mature fiber. He believes thatrecessive inheritance, which is the case withpseudoxanthoma clasticum, suggests the en-zyme defect, whereas in dominant disorderssuch as Marfan's syndrome of Ehlers-Danlossyndrome, for example, a change in a struc-tural protein should be sought. Progress con-cerning the normal metabolism of elastic tis-sue is expected to shed more light on thisproblem. The present investigation, meanwhile, seems to have confirmed the facts thatthe elastic fibers are the prime site of in-volvement in the development of the lesionsof pseudoxanthoma elasticum, that the amountof clastin is actually increased, and finally,that the calcification of elastic fibers is in-deed the initial change in the formation ofthe skin lesion.

SUMMARY

1. Formation of the dermal elastic fiberproceeded in much the same way as the col-lagen fiber in its early developmental stages.At the outset, proto filaments, equivalent indimension to tropocollagen, were precipitatedextracellularly by the fibroblasts. The subse-quent aggregation of these protofilaments anddeposition of the dense matrix substances madethem thicker, denser fibrils, called skeletonfibrils. Individual skeleton fibrils were sur-rounded by a dense granular substance andother cytoplasmic organellcs which derivedfrom near-by fibroblasts by means of secretionand ecdysis. When these skeleton fibrils ag-gregated to form mature elastic fibers, the densegranular substances were compressed betweenthem and formed dense granular septa.

2. In the mature dermal elastic fiber, indi-vidual skeleton fibrils had coalesced with eachother and were visible only at the peripheryand frayed ends. In the body of the matureelastic fibers, they were discernible as moder-ately dense fasicles of irregular widths,cemented by denser granular substances.

3. It was postulated that the skeleton fibrilsarc the main resilient component of theelastic fiber, the dense granular septa servingas buffers between them.

4. Electron microscopic studies have defi-nitely established the fact that the majority


of the elastic-stain positive material in pscu-doxanthoma clasticum is composed of an in-creased number of elastic fibers, though mostof them were severely damaged by calcifica-tion.

5. An admixture of degenerated collagen anddegraded elastin (as seen in actinic elastosis)was also observed, but to a minor degree.

6. Fibroblasts, collagen, and initial stages ofelastic fiber formation appeared normal. Smallnewly-formed, elastic fibers did not show agreat deposit of calcium salts; abnormalitiesincreased as they grew into large, stubby orbizarre-shaped elastic fibers.

7. The only visible abnormality in theseaffected fibers was the calcification, which waseither very dense, crystalline, lamellac-shapcd,or granular. Normal-appearing elastin was al-ways found along the periphery of affectedfibers.

S. It was suggested that an abnormal geneticfactor influences the abnormal metabolism ofcalcium, the production of the calcinophilicbuilding material of the elastic fibers, and/orthe deficiency of enzymes essential to the in-corporation and integrity of elastic fibers.

REFERENCES

1. Wolpers, C.: Zur ElektronenmikroskopischenDarstellung elastischen Gewebeselemente.Kim. Wschr., 23: 169, 1944.

2. Gross, J.: The structure of elastic tissue asstudied with the electron microscope. J. Exp.Med., 89: 699, 1949.

3. Tunbridge, R. E., Tattersall, R. N. and Hall,D. A.: The fibrous structure of normal andabnormal human skin. Clin. Sci., 11: 315,1952.

4. Schwartz, W. and Dettner, N.: Elektronenmi-kroskopische Tintersuchung des elastischenGewebes und der Media der MenschlichenAorta. Virchow Arch. Path. Anat., 323: 243,1953.

5. Hall, D. A., Reed, R. and Tunbridge, R. E.:Electron microscopic studies of elastic tissue.Exp. Cell Res., 8: 35, 1955.

6. Hall, D. A., Keech, M. K., Reed, R., Saxl, H.,Tunbridge, R. E. and Wood, M. J.: Collagenand elastin in connective tissue. J. Geront.,10: 388, 1955.

7. Chase, W. H.: Distribution and fine structureof elastic fibers in mouse lung. Exp. CellRes., 17: 121, 1959.

8. Rhodin, J. and Dalbam, T.: Electron micros-copy of collagen and elastin in laminapropria of the trachial mucosa of the rat.Exp. Cell Res., 9: 371, 1955.

9. Cohen, A. S. and Hashimoto, K.: Electronmicroscopic observations on the lesion ofelastosis perforans serpiginosa. J. Invest.Derm., 35: 15, 1960.

10. Charles, A.: Human elastic fibers, Brit. J.Derm., 73: 57, 1961.

11. Banfield, W. G. and Brindley, D. C.: Pre-liminary observations on senile elastosisusing the electron microscope. J. Invest.Derm., 41: 9, 1963.

12. Niebauer, G. und Stockinger, L.: Uber diesenile Elastosis. Histochemische und elek-tronenmikroskopische Tintersuchungen. Arch.KIm. Exp. Derm., 221: 122, 1965.

13. Caulfield, J. B.: Comparison of the ultrastruc-ture of normal and diabetic elastic tissue. Fed.Proc., 24: 433, 1965.

14. Greenlee, T. K., Jr. Jr., Ross, R. and Hartman,J. L.: The fine structure of elastic fibers. Fed.Proc., 25: 293, 1966.

15. Goldberg, B. and Green, H.: An analysis ofcollagen secretion by established mouse fibro-blast lines. J. Cell Biol., 22: 227, 1964.

16. Hashimoto, K.: Fibroblast, collagen and elastin.In: Ultrastructure of the Normal and Ab-normal Skin, ed. by Zelickson, A. S., Phila-delphia, Lea and Febeger, in press.

17. Darner, J.: Pseudoxanthoma elasticum. Mb.Prakt. Derm., 23: 609, 1896.

18. Findlay, G. H.: On elastase and the elasticdystrophy of the skin, Brit. J. Derm., 66: 16,1954.

19. Moran, T. J. and Lansing, A. I.: Studies on thenature of the abnormal fibers in pseudoxan-thoma elasticum. Arch. Path. (Chicago), 65:688, 1958.

20. Fisher, B. R., Rodnan, C. P. and Lansing, A.I.: Identification of the anatomic defect inpseudoxanthoma elasticum. Amer. J. Path.,34: 977, 1958.

21. Lobitz, W. C., Jr. and Osterberg, A. E.: Pseu-doxanthoma elasticum: Microincineration. J.Invest. Derm., 15: 297, 1950.

22. Loria, P. R., Kennedy, C. B., Freeman, J. A.and Henington, V. M.: Pseudoxanthomaelasticum (Gronblad-Strandberg syndrome):A clinical, light and electron microscopicstudy, Arch. Derm. (Chicago), 76: 609, 1957.

23. Hannay, F. W.: Some clinical and histopatho-logical notes on pseudoxanthoma elasticum,Brit. J. Derm., 63: 92, 1951.

24. McKusick, V. A.: Heritable disorders of con-nective tissue. VI. Pseudoxanthoma elasti-cum. J. Chronic. Dis:, 3: 263, 1965.

25. Goodman, R. M., Smith, E. W., Paton, D.,Bergman, R. A., Siegel, C. L., Otteson, 0.,Shelley, W. M., Puseh, A. L. and McKusick,V. A.: Pseudoxanthoma elasticum. A clinicaland histopathological study. Medicine, 42:1963.

26. McKusick, V. A.: Personal communication inBoston, Massachusetts, May 16, 1966.

27. Eisenstein, R., Werner, M., Papajiannis, S.,Konetzki, W. and Laing, I.: Chloraniiie acidas histochemical reagent for calcium, J.Histoehem. Cytochem., 9: 154, 1961.

28. Pearse, A. 9. E.: Histochemistry. Theoreticaland Applied, p. 934, Boston, Little, Brownand Co., 1960.

29. Niebauer, G. and Stockinger, L.: Uber diesenile elastosis. Histoehemisehe und elektro-nenmikroskopische Untersuchungen. Arch.Kim. Exp. Derm., 221: 122, 1965.

30. Hashimoto, K. and DiBella, R. J.: Electronmicroscopic studies of actinic (senile) elasto-sis. In preparation.

ULTRASTRUCTURE OF TIlE ELASTIC FIBER 423

31. Hashimoto, K.: Elastosis perforans serpiginosa.Further electron microscopic studies. Inpreparation.

32. Cohen, A. S. and Calkins, E.: Electron micro-scopic observations on a fibrous componentin amyloid of diverse origins. Nature (Lon-don), 183: 1202, 1959.

33. Hashimoto, K., Gross, B. G. and Lever, W.F.: Lichen amyloidosis. Histochemical andelectron microscopic studies. J. Invest. Derm.,45: 204, 1965.

34. Gross, J., Highberger, J. H. and Schmitt, F. 0.:Collagen structure considered as states ofaggregation of a kinetic unit. The tropo-collagen particle. Proc. Nat. Acad. Sci. U.S.A.,40: 679, 1954.

35. Zelickson, A. S.: Electron Microscopy of Skinand Mucous Membrane, p. 132, Springfield,Ill., Charles C Thomas, 1963.

36. Porter, K. R.: Cell fine structure and biosyn-thesis of intercellular macromolecules. In:

Connective Tissue: Intercellular Macro-molecules, p. 167—196, Boston, Little, Brownand Co., 1965.

37. Smith, J. G., Jr., Sams, W. M., Jr., Davidson,E. A. and Clark, R. D.: Pseudoxanthomaelasticum. Histochemical and biochemicalalteration Arch. Derm. (Chicago), 86: 741,1962.

38. Brown, H.: The mineral content of humanskin. J. Biol. Chem., 75: 789, 1927.

39. MacCardle, R. C., Engman, M. G., Jr. andEngman, M. F., Sr.: Spectographic analysisof neurodermatitic lesions. Arch. Derm.(Chicago), 44: 429, 1951.

40. Anderson, H. C.: Electron microscopy ofheterotopic bone formation induced by FLcells. Electron Microscopy, Vol. II., p. 573.ed., Uycda, R., Tokyo, Maruzen Co., 1966.

41. Hashimoto, K.: Calcifying epithelioma ofMalherbe. Further observations, in prepara-tion.

Electron Microscopic Studies of Normal and Abnormal Elastic ...

Documents

Transcript of Electron Microscopic Studies of Normal and Abnormal Elastic ...