Calcium Reserve Assembly: A Basic Structural Unit of the Calcium Reserve System of the Hen Egg Shell

Calcium Reserve Assembly: A Basic Structural Unit of the Calcium Reserve System of the Hen Egg Shell J. W. DffiCKERT, M. C. DDECKERT, and C. R. CREGER Department of Plant Pathology and Microbiology and Department of Poultry Science, Texas A&M University, College Station, Texas 77843 (Received for publication October 12, 1988)

1989 Poultry Science 68:1569-1584 INTRODUCTION

It is well known that a calcium reserve is located in the avian egg shell (Johnston and Comar, 1955; Sajner, 1955; Simkiss, 1961), that the reserve is located in the mammillary layer (Sajner, 1955; Tyler and Simkiss, 1959; Terepka, 1963b; Schmidt, 1965), and that the chorioallantoic membrane of the embryo mobilizes the calcium reserve in the latter phases of incubation (Sajner, 1955; Tuan, 1980). In the domestic chicken, approximately 80% of the calcium in the hatchling comes from the egg shell (Simkiss, 1961). The normal egg shell is a highly ordered system of organic matrices stabilized with calcium carbonate. Major difficulties are encountered in interpreting work on egg shell structure due to the fragile nature of the regional matrical gels after the calcite is removed. For example, the gels may shrink or be partially or totally removed in processing. This frustrates attempts to accurately determine the distribution of matrix in demineralized structures; and, in the worst case, the true identity of a structure can be totally missed. By minimizing these problems, the present work brings into focus a well-defined proper subregion of the classical mammilla (mammillary knob), which is here proposed to be a

basic structural component of the calcium reserve system of the avian egg shell. This structural unit, denoted as the calcium reserve assembly, or assemblies (CRA), consists of a baseplate (BP) partially integrated into the outer shell membrane, a calcium reserve body (CRB) attached to the BP, and a distinct CRB cover (Figure 1). This model accommodates much of the data found in the extensive literature on the biology of the mammillary layer, provides a basis for understanding several anomalous egg shell structures (Dieckert et al., 1988a,b), offers a rationale for the existence of calcified knobs (instead of a more uniform sheet) as an inner calcified layer of the egg shell, and serves as an heuristic for modeling the organization of the calcium reserve in molecular terms. The purpose of this study was to describe the morphology of the CRA and identify the CRA in the normal egg shell. The time table for CRA development and the use of the CRA as a calcium source for the embryo were also studied. MATERIALS AND METHODS

Fifty white Leghorn hens were maintained, one to a cage, in a controlled environment. A

1569

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

ABSTRACT A well-defined proper subregion of the mammillary knob is described, consisting of baseplate (BP), calcium reserve body (CRB), and an integral but distinct CRB cover. The BP extends 14 to 20 um into the outer shell membrane and rises a few microns above its surface. The organic matrix of the BP is a composite of modified membrane fibers and adjacent components. The distinct CRB matrix is attached to the BP and extends outward in attenuated form, much like the fractal aggregates associated with filter "cakes." The three regions comprise a unit designated as the CRA (calcium reserve assembly, or assemblies). The CRA are fully developed and mineralized by 10.5 to 11 h postoviposition of the previous egg. By 13 h postoviposition a new structure, the crown, is evident. The classical mammillary knob consists of the CRA/ crown complex. By hatch time the CRA are partially decalcified, except in the air cell region. Consequently, a zone of separation develops in the affected CRA about midway through the CRB. Estimates of the calcium recovered from the egg shell and of the calcium found in the CRA, excluding the CRA in the air cell region, indicate that more than enough calcium is present in the CRA to meet the drawdown of calcium from the egg shell by the developing embryo. It is concluded that the CRA that are available to the action of the chorioallantoic membrane represent the calcium reserve of the hen egg shell. (Key words: Ca-reserve, egg shell, embryo, mammilla)

1570

DIECKERT ET AL.

standard laying ration (16.5% CP; 2,856 kcal ME/kg) was made available ad libitum. Eggs were obtained manually or surgically from the oviducts at specified time intervals, and shell membranes were subjected to various treatments after preliminary examination with a Reichert phase-contrast microscope (ReichertJung Div. of Cambridge Instruments CO., Nussloch, West Germany). Stains included Toluidine Blue (Fisher Scientific Co., Houston, TX), Alcian Blue (Fisher Scientific Co., Houston, TX), Alcian Blue (Bio-Rad Laboratories, Richmond, CA), and Stains-all (1-ethyl2-[3-( 1 -ethylnaptho[ 1,2d]thiazolin-2- ylidene)2-methylpropenyl]naptho[l,2d] thiazolium bromide; Eastman Chemicals, Rochester, NY). Observations were recorded by color photomicrography using Leica equipment (Ernst Leitz, Wetzlar, West Germany), and films and papers by Kodak (Eastman Kodak Co., Rochester, NY) or Ilford (Ciba-Geigy Co., Paramus, NJ). Procedure la. Decalcification of Folded Soft Egg Shell. In this technique slides were prepared with folded pieces of egg shell ( 5 x 5 mm) from soft eggs. Normal, phase contrast, or polarization optics were used to observe changes in individual groups of CRA on the folded edges during progressive decalcification with 5% EDTA or 3% acetic acid. Decalcifica-

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

FIGURE 1. Cross section of mature egg shell: scale diagram indicating structural components and their relationship to one another. CRA = Calcium reserve assembly; CRB = calcium reserve body; SM = shell membrane.

tion fluids, rinses, and stains (Toluidine Blue, Alcian Blue, Stains-all) were introduced at the edge of the coverslip, while surveillance was maintained. Procedure lb. Decalcification of Folded Soft Egg Shell, with Subsequent Drying. To study drying effects on the CRA, decalcified soft shell was allowed to dry in place on slides. This was later rehydrated and stained with Stains-all (Green et al, 1973). Procedure 2a. Fixation!Decalcification, with Double Embedment. To maintain structural integrity of the CRA and other egg shell matrices during sample preparation, pre-embedment in 2% SeaPlaque agarose (FMC Bioproducts, Rockland, ME) was used. This step was designed to protect the delicate matrices and original outlines of CRA against possible shrinkage by solvents used in dehydration. Cubes of agar-embedded tissue were fixed in 1% glutaraldehyde (Polysciences, Warrenton, PA) with 1% EDTA added as a decalcifying agent. Either .1 M phosphate buffer (pH 7.4) or sodium cacodylate/HCl buffer (pH 7.4) was used. Following fixation/ decalcification, the cubes were extracted with distilled water and dehydrated stepwise with acetone before infiltration and embedment with Spurr's (1969) low viscosity resin. Procedure 2b. Fixation!Decalcification and Double Embedment, with Added Rehydration Series. In a modification of Procedure 2a, a rehydration step was introduced after dehydration, followed by a second dehydration series. All other steps remained the same. Procedure 3. Sectioning and Slide Preparation for Light Microscopy. Cured blocks were cut with glass or diamond knife (E. I. Dupont de Nemours and Co., Wilmington, DE) in a plane normal to the surface of the egg shell, using an LKB Ultratome III (LKB Produkter AB, Bromma, Sweden). Sections were affixed to the slides and stained with Toluidine Blue (Hayat, 1970) or Stains-all (Green et al., 1973). Stains-all was found to yield bluer colors with crown and palisade matrices if the plastic was softened with potassium hydroxide in absolute alcohol (Hayat, 1970). Procedure 4. Preparation of Thick Sections of Fixed Egg Shell in Agarose. In this procedure 5 x 1 0 mm pieces of egg shell were embedded in 2% SeaPlaque, decalcified/fixed with 5% EDTA, 1% glutaraldehyde in sodium cacodylate buffer (pH 7.4), extracted with water, and re-embedded in 7% SeaPlaque. Sections 50 to 225 am thick were obtained

1571

CALCIUM RESERVE ASSEMBLY

postoviposition of the previous egg, maturing eggs were manually removed from die uteri of hens, using a technique developed by R. C. Fanguy (Department of Poultry Science, Texas A&M University, College Station, TX, personal communication). Eggs were taken at 9, 9.5, 10, 10.5, 11, 11.5, 12, 12.5, and 13 h postoviposition. RESULTS AND DISCUSSION

At 10.5 h postoviposition of the previous egg the shell membrane of a developing egg was covered by birefringent CRA, as shown in surface and lateral views in Figures 2a and b. A typical mineralized CRA is shown in side view in Figure 3a. The BP rose a few microns above the shell membrane in which it was embedded; the CRB was the bulbous distal part. Key dimensions of the CRA and their variability are given in Table 1. The principal regions of the CRA were clearly visible in ground radial sections of normal egg shell (Figures 3a and c). The dimensions of the CRA in ground radial sections of fully developed egg shells were quite similar to those from eggs at 5.5 h in utero (10.5 h postoviposition of previous egg) (Table 1). This result indicates that the CRA reached maximal size by 10 to 11 h postoviposition of the previous eggs. The CRA were proper subregions of the mammillae, and the CRA accounted for approximately 59.2% of the mammillary height (data not shown). The BP, which incorporated outer shell membrane substance, extended to a deptii of 15.5 um.

TABLE 1. Dimensions (x ± SEM) of calcium reserve assembly (CRA)-related structures in undecalcifled shells in lateral view1 n1

Time in uterus

62 83 94 56

(h) 5-5.5 5.5 -20.4 5 -20.4

Ha 53.6 53.0 53.4 55.6

Hb ± ± ± ±

1.5 1.8 1.7 2.8

18.1 ± 2.2 15.5 ± 1.4 11.8 ± 1.7

Hr

wr

(um) — 6.0 ± 0.6 49.0 62.3 63.5 8.1 ± 0.5 50.0 12.8 ± 1.6

w

± 4.0 ± 7.7 ± 12.6 ± 6.8

m

79.0 78.9 87.8 106.0

± 3.8 ± 6.8 ± 12.9 ± 8.9

H a = Height of CRA distal to shell membrane surface; Hj, = depth of CRA penetration into shell membrane; H r = height of rise of base plate (BP); W r = width of CRA at the rise of BP; W m = maximum width of CRB; n = number of structures measured. isolated CRA. 3 A CRA-ghost in SeaPlaque (FMC Bioproducts, Rockland, ME) Spurr (1969) double embedment (Procedure 2b). 4 Pooled data from Schmidt (1960), Figure 2a; Terepka (1963a), Figure 1; Schmidt (1965), Figure lb. Approximate time a normally laid egg remains in uterus. ^athusius (1882), Figure 98a, in Romanoff and Romanoff (1949).

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

using a TC-2 Tissue Sectioner (Ivan Sorvall, Inc., Norwalk, CT). Tissue was stained with Stains-all in place on slides. Procedure 5. Isolation and Decalcification of CRA. Mature CRA (10.5 h postoviposition) were rubbed off the shell membrane and placed on a slide in either 5% EDTA or 3% acetic acid. While maintaining constant surveillance at the Reichert microscope, photographic series were obtained of individual CRA undergoing decalcification. Procedure 6. Preparation of Baseplate Rings from Normal Egg Shell. After removing the inner membrane from normal infertile eggs, pieces of egg shell were soaked 30 min in 5% EDTA in tris(hydroxy methyl)aminomethane buffer (pH 6.0) and men rinsed in distilled water. Outer membrane layers were removed and studied at the microscope. Crosspolarization optics were used to illuminate the baseplate rings, which retained their birefringence after staining with Stains-all. Procedure 7. Preparation of Rings from Egg Shells of 18-day Fertile Incubated Eggs. Shell membranes from 18-day early hatchlings were easily removed from egg shells due to the natural separation that occurs as calcium is mobilized from the mammillary layer. Shell membranes were pulled free of the chorioallantoic membrane and placed on slides. Birefringent rings were visualized when cross-polarization optics were used. Staining with Stainsall did not destroy or alter the appearance of the rings. Procedure 8. Fanguy's Method of Removing Eggs from Uterus. At designated times

1572

DD3CKERT ET AL.

The CRA had a height of 53.3 yon above the surface of the shell membrane. Demineralization with dilute acids or chelating agents usually causes shrinkage of the egg shell matrices (Figure 3a and b). Estimates of shrinkage resulting from decalcification and other preparative procedures are given in Table 2. Shrinkage may reduce the height of the CRB matrix by as much as 60%. In most of

the procedures described, demineralization of CRA or normal egg shell with dilute acetic acid or EDTA resulted in shrinkage or other disruptions of the matrices. The degree of shrinkage during decalcification with .5 M acetic acid containing 1% Alcian Blue may be visualized by comparing Figures 3a and b. Quantitative estimates of shrinkage obtained with different protocols are given in Table 2.

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

FIGURE 2. Mineralized egg shell of egg removed manually from uterus at 10.5 h post-oviposition of the previous egg: a) birefringent calcium reserve assemblies (CRA) in surface view; b) lateral view of birefringent CRA on folded shell membrane. Bar = 20 um.

CALCIUM RESERVE ASSEMBLY

1573

Even when EDTA in aqueous media was applied alone, the native CRA matrix above the shell membrane shrunk to about 75 to 80% of its original height. In time-course studies on the decalcification of mature CRA on folded shell membrane, native organic matrix appeared concomitandy with the removal of calcium, but shrinkage progressed also. This indicates that an organic matrix permeates the CRA structure and that it condenses with time and exposure to the milieu. Further condensation occurred with time and staining with .05% Toluidine Blue in 1% sodium borate. The result of one such experiment, using Stains-all, is seen in Figure 4a. In this case the CRA matrix had shrunk to about 79% of its original height. The CRB matrix was concentrated on the BP surface and the matrix attenuated distally. The CRB matrix stained red with Stains-all. The bluish-purple cast was caused by the CRB cover, which stained blue with

Stains-all or Alcian Blue at pH 2.8. When the above preparations were dried at room temperature, rehydrated, and stained with Stains-all, the height of the CRA distal to the shell membrane was reduced to approximately 40% of the original height. In Procedure 2a (SeaPlaque agarose/glutaraldehyde-EDTA/Spurr) the CRA adorning the shell membrane appeared shrunken; however, a ghost of die calcified CRA remained in the agarose. Shrinkage was measured on each CRA (Table 2). An analysis of the results of Robinson and King (1968) with decalcified shells fixed with formaldehyde suggested that severe shrinkage occurred in material embedded in paraffin (see Table 2). The modified procedure that used an added dehydration step (Procedure 2b) gave good results with CRA on shell membranes and on mature shells having most of the membranes removed. A thick radial section of decalcified

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

FIGURE 3. Isolated calcium reserve assembly (CRA) compared with CRA in ground radial section: a) crosspolarized lateral view of isolated mineralized CRA, illustrating calcium reserve body (CRB) and baseplate (BP); b) the same isolated CRA after decalcification in .5 M acetic acid containing 1% Alcian Blue (Bio-Rad Laboratories, Richmond, CA), demonstrating shrunken matrix (M) of CRB; c) ground radial section of normal egg shell (Schmidt, 1960, Figure 2a), showing palisade region (P), crown layer (C), CRB, and BP. Bar = 20 fun.

1574

DIECKERT ET AL. TABLE 2. Dimensions (x ± SEM) of calcium reserve body (CRB)-related structures in decalcified shells in lateral view1

n1 72 4i

4

2 25 7

H.

(h) 5.5 5.5 5.5 5.5

52.1 53.6 56.0 56.0

b

-20.4 7 -20.4 -20.4 -20.4

Hi ± ± ± ±

1.7 2.0 1.9 1.9

52.1 30.4 42.0 36.4 21.4 36.0 54.9 22.7 25.6

H.

Hc ± ± ± ± ± ± ± ± ±

1.7 1.6 0.9 0.9 1.1 2.4 2.1 2.9 1.1

17.1 ± 2.8 18.3 ± 1.8 11.7 28.0 23.6 21.9 9.7

± ± ± ± ±

100 57.1 ± 4.4 75.0 ± 0.9 65.0 ± 0.5

0.3 2.4 1.6 1.9 1.9

Ha = Original height of CRB; H a = height of CRB after treatment; He = height of CRB matrix knot attached to base plate.; n = number of structures measured. 2 Double embedment with added rehydration series (Procedure 2b). Double embedment, no added rehydration step (Procedure 2a). Folded shell membrane, unfixed, unstained (Procedure la). Procedure la plus Toluidine Blue. folded dried membrane, rehydrated, Stains-all (Procedure lb). Approximate time a normally laid egg remains in uterus. "Ground radial sections. EDTA-decalcified, stained [Terepka (1963b), Figure 4.5]. 9 EDTA/formaldehyde simultaneous decalcification/fixation, paraffin embedment.

mature CRA (10.5 h postoviposition), prepared by Procedure 2b and stained with Toluidine Blue, showed little shrinkage (Figure 4b; Table 2). The CRB matrix appeared to be attenuated distally from a knot of CRB matrix on the BP, and it completely permeated the CRA above the surface of the shell membrane. These results confirm those obtained with unprotected CRA on folds of the shell membrane. Radial sections (225 nm) of 13-h egg shells embedded in SeaPlaque agarose, decalcified with EDTA in glutaraldehyde, and stained with Stains-all revealed differentially stained lensshaped bodies newly appearing in a thin sheet of matrix material over the CRA (Figure 5a). The authors have named these distinct structures "crowns." The crowns and CRA were easily separated in these sections without apparent damage to either matrix (Figures 5b and c), demonstrating weak bonding between the layers. The formation of the layer of crowns between 11 and 12 h postoviposition of the previous egg confirms that the growth of the CRA was completed by 10.5 h postoviposition, well before the crowns were initiated. Sections (2 |am) of decalcified normal egg shell prepared according to Procedure 2b were

stained with Toluidine Blue or Stains-all without removal of the plastic. As shown in Figure 6a, the CRB matrix stained differendy from the crown matrix, and the CRB-cover stained blue. Similarly, Stains-all differentially stained the CRB, CRB-cover, crown, and palisade matrices (Figure 6b). The CRB matrix stained red; the crown and palisade matrices stained purple. The plastic altered the staining response of the crown and palisade matrices, but not irreversibly. When the plastic was removed with KOH in absolute ethanol, both matrices stained blue with Stains-all. Though some shrinkage of the shell matrices was evident, the boundaries of the different regions of the mammillary knobs were recognizable by virtue of the unique optical properties peculiar to each component, as well as by their location, morphology, and staining characteristics. They are comparable to similar structures seen in Terepka's (1963b) distortion-free decalcified ground radial sections (Figure 6c). The CRA/crown system seen in Figure 6c measured 108.9 urn (± 3.5 um) in height. The CRA represented about 67.1% (± 1.5%) of the height of the mammillae, showing again that the CRA are proper subregions of the mammil-

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

73 58 5y 3y

HJ, x 100

Time in uterus

CALCIUM RESERVE ASSEMBLY

1575

TABLE 3. Dimensions (x ± SEM) of calcium reserve body matrix remnants on shell membrane released from egg shell with acids, EDTA, and natural agents (surface view) i1

Max PD 2

Min PD 3

Geometric mean4

Treatment

23 17 20 30 15 16 25 26

41.4 42.5 35.8 35.1 38.7 37.7 37.1 38.4

33.4 32.1 29.5 30.3 30.1 24.3 27.8 29.8

37.1 ± 36.4 ± 32.4 ± 31.6 ± 33.9 ± 30.1 ± 32.0 ± 33.7 ±

5% EDTA, pH 6.0, 30 min. Incubated 18-day fertile egg. 5% EDTA, pH 6.0, 30 min (rings remain). .5 M Acetic acid, 90 min (rings remain). Acid, Purkinje (1855) [from Sajner, 1955]. 3% Nitric acid, Sajner (1955). 5% EDTA, pH 8.5, Robinson and King (1963). 3% Nitric acid, Stemberger (1971).

± 1.7 ± 2.3 ± 2.0 ± 1.2 ± 1.6 ± 2.6 ± 2.0 ± 1.8

± 1.8 ± 2.4 ± 1.6 ± 1.2 ± 1.0 ± 1.4 ± 1.4 ± 1.4

1.7 2.0 1.7 1.2 1.0 1.7 1.6 1.5

4

Geometric mean = ^Max PD x Min PD .

lae. Based on measurements of similar structures in the mammillae, as recorded by von Nathusius' diagram of an undecalcified ground radial section of egg shell (1885, as cited in Romanoff and Romanoff, 1949), heights for the CRA/crown complex and CRA were 100.2 |im (± 4.0 n-m); and 67.4 urn (± 3.9 u.m), respectively. The CRA accounted for 67.2% (± 2.3%) of the height of the mammillae. These comparative dimensional analyses showed that in favorable preparations the position of the CRA and crown may be recognized in calcified sections by the optical properties peculiar to each region. The relationships between CRA subregions, between the shell membrane and chorioallantoic membrane, and

between the crown layer and palisade layer may be clarified by comparing the color photomicrographs with Figure 1. A grainy membrane was released from the calcified egg shell by dilute acids and chelators and is naturally released at hatching. The dimensions of the grains were similar for all such preparations (Table 3). The fact that the grains stain red with Stains-all (Figure 7a) indicates that they probably represent residual CRB matrix. Birefringent BP rings surround CRB residues both on naturally released membranes (Figure 7c) and on membranes released by treatment with 5% EDTA or 3% acetic acid (Figure 7a). Side views are seen in Figures 7b and d. Tables 3 and 4 present dimensions of CRB and rings as published by

TABLE 4. Dimensions (x ± SEM) in surface view of the birefringent rings and mammillary caps associated with the calcium reserve assemblies i1

Max PD 2

Min PD 3

Geometric mean4

18 16 20 30

64.3 66.1 52.8 52.8 61.4

46.7 48.3 39.1 39.8 45.1

54.7 56.4 45.3 45.7 52.5

23

± 3.2 ± 3.3 ± 2.7 ± 1.9 ± 4.7

±3.0 ± 2.8 ± 2.3 ± 1.6 ± 3.0

± 2.9 ± 2.8 ± 2.4 ± 1.7 ± 3.7

*n = Number of structures measured. Max PD = Maximum projected diameter. 3 Min PD = Minimum projected diameter. Tjeometric mean = ^Max PD x Min PD 2

Treatment Incubated 18 days. Hatched egg, Terepka (1963b) Figure 15. 5% EDTA, pH 6.0, 30 min. .5 M Acetic acid, 90 min. Plasma-etched egg shell, inner surface, Belyavin and Solomon (1986).

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

Max PD = Maximum projected diameter. Min PD = Minimum projected diameter.

1576

DfECKERT ET AL.

M

various researchers. In each case the diameter of the CRB residue was smaller than that of its corresponding ring. The dimensions of the rings were similar to those of the "mammillary caps" as published by Belyavin and Solomon (1986) and probably are the measure of the "basal cap" (Schmidt, 1960) at its junction with the shell membrane surface and the rise above it. The ring dimensions were smaller in artificially released samples than in naturally released ones. However, the dimensions of the CRB residues were the same. Ring birefringence was abolished by BAPTA (l,2-bis(oaminophenoxy)ethane-N,N,-N',N',-tetraacetic acid), a specific chelator for calcium ion, thus

demonstrating that calcium adducts are responsible for the birefringence. In Table 5 the mean heights and widths of CRB residues and rings from several preparations are given. The height of the CRB remnants and rings on artificially released membranes was half that of those on naturally released membranes. These effects reflect the physical removal of distal CRA matrix. It seems probable that all the grainy membrane preparations portrayed in the literature represent an incomplete CRB matrix. Because of their residual mineral, naturally derived rings are dimensionally stable; therefore, their mean height can serve as a structural

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

FIGURE 4. Egg shell of soft egg removed from uterus at 10.5 h postoviposition: a) decalcified calcium reserve assemblies (CRA); side view on folded membrane, stained with Stains-all (Eastman Chemicals, Rochester, NY); matrix (M) of calcium reserve body shrunken to 79% of original height; b) section of 10.5-h egg shell, showing knots of M attenuated distally; prepared by Procedure 2b [glutaraldehyde-fixed, EDTA-decalcified, agarose-Spurr (1969) double embedment with added rehydration step] and stained with Toluidine Blue (Fisher Scientific Co., Houston, TX). Bar = 20 ujn.

CALCIUM RESERVE ASSEMBLY

1577

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014 FIGURE 5. Stains-all (Eastman Chemicals, Rochester, NY) stained radial sections (225 Jim) of egg shell at 13 h posloviposition; prepared by Procedure 4 (agarose embedment, glutaraldehyde-fixed, EDTA-decalcified): a) lateral view of calcium reserve assemblies (CRA) and crowns (C) in place on shell membrane; b) layer C separated from CRA; c) CRA on shell membrane, separated from crown layer. Bar = 20 urn.

pointer (inside to outside) to the calcium reserve in the egg shell. The mean of the three estimates in Table 5 was 24.7 |xm, which

places the zone of separation just distal to the principal knot of CRB matrix. Terepka's (1963b) photomicrographs of ground radial

1578

DESCKERT ET AL.

TABLE 5. Dimensions (x ± SEM) of calcium reserve body residues on shell membranes released with acetic acid, EDTA, or activity of the chorioallantoic membrane (CAM) Birefringent bodies2

Stains-all red bodies1 n3

Hg

Wc

Hbi

22.4 ±

8

.9

38.5 ± 3.2

3

22.5 ± 2.2

5

25.5 ± 2 . 2

...

7

26.1 ± 2 . 2

...

6

11.9 ± 1.3

37.8 ± 1.7

7

11.0 ±

28.1 ± 1.1

.5

7

14.2 ± 1.0

Treatment

Wbi (um) 69.3 ± 5.0

55.1 ± 5.0

Lateral view on fold, incubated 18 days CAM/Shell membrane, ground radial sections, Terepka (1963b) Figure 16, 18. CAM/Shell membrane, ground radial sections, Schmidt (1965) Figure 3a. Infertile egg, fold, no rings, 5% EDTA, pH 6.0, 30 min Infertile egg, fold, rings, .5 M acetic acid 90 min.

'Stains-all (Eastman Chemicals, Rochester, NY); Hg = Height, W c = width. H b i = Height, W bi = width. 3 n = Number of structures measured.

2

sections of egg shell at hatch clearly show that the position of the crown is distal to the spent calcium reserve. This provides a second pointer to the calcium reserve (outside to inside). Both pointers indicate that the calcium reserve lies in the CRA. This also explains why the zone of separation between the shell membrane and shell leaves the knot of CRB matrix attached to the shell membrane.

As previously pointed out, the CRA were fully formed and calcified, reaching maximum dimensions by 10.5 h postoviposition of the previous egg, before the appearance of the crown layer between 11 and 12 h. Therefore, shells taken from the uterus at this time should contain enough calcium in the region exclusive of the air cell to meet the calcium drawdown from the reserve during embryonic growth.

TABLE 6. Estimated calcium recovery from the egg shell at hatching and calcium in calcium reserve assemblies (CRA) Calcium recovery

Equation

, ^ Recovered from shell at 21 days1 141.8 In CRA after 5.5 h in uterus 244 Available in CRA 222 Apparent efficiency

Whole embryo at 21 days = 129.3 Shell, 5 h in utero2

Residual yolk, 21 days + 44.5

=

+

[(217 Total in CRA = 244 142 x 100 222

Shell, 6 h

X

271) Fraction CRA available3

Yolk, initially 28.5 Interpolation factor x .5]

Albumen, initially -

3.5 Shell membrane - .18

.91

64%

toata and equation from Romanoff and Romanoff (1967), Tables 1, 23, 24. Data from Burmester (1940), Table 14. 3 Shell surface area estimated for a 60-g egg; air cell with diameter = 2.54 cm and height = .6 cm. 2

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

n

,, ^

CALCIUM RESERVE ASSEMBLY

It was concluded that the CRA is the primary source of calcium in the calcium reserve system of the hen egg shell for the following reasons: 1) the CRA is a welldefined structure; 2) there is more than enough calcium in the CRA exposed to the chorioallantoic membrane to account for the observed drawdown from the reserve; 3) the CRA on the shell membrane that are exposed to the chorioallantoic membrane are decalcified during the last half of embryonic development, whereas those not so exposed remain intact; 4) there is an insignificant amount of calcium in shell membrane free of CRA; 5) there is no evidence that the crown or palisade layers of the shell are eroded during embryonic development. It is apparent that the various matrices of the normal egg shell are stabilized with calcium carbonate. As a result, the substructures of the shell remain dimensionally stable as long as the calcium carbonate remains in place. Once the calcium carbonate is removed, the organic matrices are subject to shrinkage, dislocation, or loss of components. Such events can make it difficult to determine the true distribution of matrical components in the egg shell and, consequently, can cause difficul-

ties in the recognition and definition of the true substructures of the egg shell. It was discovered early that demineralization of calcified egg shell with dilute acids yields a membrane covered with small projections or grains. In 1868, von Nathusius described the nipple-like projections on the inner shell surface, calling them "mammillae" (cited by Tyler, 1964). Sajner (1955) credited Purkinje (1855) with the first illustration of the "grainy" membrane (reproduced by Sajner, 1955). Sajner (1955) concurred with von Nathusius that these granules are remnants of mammillae remaining after demineralization, representing points of attachment for the mammillae on the outer shell membrane. Other workers noted similar structures in preparations obtained after demineralization with EDTA and other agents (Robinson and King, 1963, 1968, 1970; Bellairs and Boyde, 1969; Stemberger, 1971; Fujii, 1974). As demonstrated above, these preparations represent only part of the organic matrix of the primary structure, the CRA. Part of the CRB matrix is missing, and the BP is cryptic. Simkiss and Tyler (1957) reported that the organic material of the mammilla is concentrated in the mammillary core, within the base of each mammillary knob. No photomicrographs were published, and no dimensional analysis was given. The mammillary core was diagrammed as a knot of protein on the surface of the outer shell membrane in the tip of the mammillary knob. No structure was described by Simkiss (1957) that corresponds to the CRA or crown. Tyler (1969) stated that fibers of the membrane pass through protuberances called "mammillary cores." In our model the modified fibers are an integral part of the BP and do not contribute to the CRB matrix. According to Terepka (1963b) the organic matrix core is attached to the outer surface of the outer shell membrane. In the authors' model, there is a knot of CRB matrix that extends in attenuated form into the distal reaches of the CRB. Terepka (1963a,b) also noted that the egg shell has a well-ordered protein matrix outside the organic matrix core distal to the mammillae (mammillary knobs). The first of these corresponds to the crown layer, which Terepka considered to be distal to the mammillae. Wybum et al. (1973) related the mammillary core to a fiber on the external shell membrane, rather than to the substance of the mammilla, interpreting its ordered arrangement as being in continuity with the fiber lattice pattern. Measurements made from mi-

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

The total calcium in the chick at hatch time is partitioned between the residual yolk and embryo proper. The quantity of calcium in the chick at 21 days incubation in excess of the amount in the original yolk and egg albumen is an estimate of the calcium derived from the calcium reserve in the egg shell. An analysis of the problem is oudined in Table 6. The quantity of calcium in the egg shell at 10.5 h postoviposition of the previous egg was estimated as the average of the values for eggs taken at 5 and 6 h in utero (Burmester, 1940), corresponding to 10 and 11 h postoviposition, respectively. The quantity of calcium in the shell membranes free of CRA was neglected, based on Burmester's (1940) findings. Using equations patterned after Romanoff and Romanoff (1949), the fraction of the CRA exclusive of the air cell region was estimated, based on a standard-size egg of 60 g with an air cell 2.54 cm in diameter and .6 cm in height. The analysis given in Table 6 showed that 222 mg of calcium is expected to be in the CRA exposed to the chorioallantoic membrane and that the embryo was expected to recover approximately 142 mg of calcium from the shell by time of hatch, for an apparent efficiency of recovery of 64%.

1579

1580

DIECKERT ET AL.

-M

-P

-C

-M

FIGURE 6. Differential staining properties of palisade (P), crowns (C), and matrices (M) of decalcified normal egg shell; comparison with ground section: a) egg shell prepared by Procedure 2b (glutaraldehyde-fixed, EDTA-decalcified, agarose-Spurr (1969) double embedment with added rehydration step) and stained with Toluidine Blue (Fisher Scientific Co., Houston, TX); b) egg shell prepared by same procedure (2b) and stained with Stains-all (Eastman Chemicals, Rochester, NY); c) ground radial section of normal decalcified egg shell (Terepka, 1963b, Figure 4). Bar = 20 urn.

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

-c

CALCIUM RESERVE ASSEMBLY

1581

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014 FIGURE 7. Residues of calcium reserve bodies (CRB) naturally or artificially released from normal egg shell; a) surface view of CRB residues on shell membrane released from shell by decalcification fluid (5% EDTA) and stained with Stains-all (Eastman Chemicals, Rochester, NY); b) lateral view of CRB residues on folded shell membrane decalcified with 5% EDTA and stained with Stains-all; c) surface view of CRA residues on shell membrane released naturally from egg shell at time of hatch; birefringent baseplate rings (R) revealed by cross polarization optics, d) lateral view of CRA residues on naturally released shell membrane; R illuminated by cross polarization; stained with Stains-all. Bar = 20 um.

1582

DIECKERT ET AL.

In our proposed model the CRA is a welldefined structure which, together with the crown, constitutes the classical mammilla described by von Nathusius in 1885 (as cited in Tyler, 1964). The BP overlaps, but is not congruent with, the basal caps of Schmidt (1960) and Tyler (1969), the mammillae of Wyburn et al. (1973), or the mammillary cores of Solomon (1983). The CRB includes some of the material in the grains of Sajner (1955), Purkinje (1855, as cited by Sajner, 1955), and von Nathusius (1868, as translated by Tyler, 1964) as well as the mammillary cores of Simkiss and Tyler (1957), Robinson and King (1963), and Fujii (1974). The "mammil-

lary core" of Robinson and King (1968) is probably the condensed CRA, minus the cover. The "CRB cover" in the present study includes the sialic acid-containing component of Robinson and King (1968). In each case the correspondence is not exact, due to imprecise definitions, overlap regions, and deletions associated with preparative techniques. The mammillary cores of Wyburn et al. (1973) were probably BP elements, including the "rise." The mammillary knob seen in Figure 17 of Bellairs and Boyde (1969) corresponds to part of the "knot of CRB matrix" on the BP, as described in the proposed model. The observation that layers of the egg shell are deposited and calcified in a definite temporal order provides a basis for understanding various anomalies that are seen in several different types of abnormal egg shells (Dieckert et al., 1988a,b). Aberrations have been noted that indicate misfunction of the oviductal gland cells responsible for certain of the deposited matrices, resulting in skipped phases in the normal order of deposition, or cessation of normal development, or both. Abnormalities are detected by a difference in appearance or texture, such as soft, papery, or powdery shells. It is apparent that when one component of the CRA fails to develop properly (e.g., the baseplate), subsequent depositions malfunction, resulting in the gross deformations apparent at the microscope. Also, it appears that the baseplate may serve as a collector for the calcifying CRB matrix. Under this model, the macromolecules of the CRB add to the BP macromolecules to form a CRB/BP interface. More CRB macromolecules are added to this interfacial layer until the CRB matrix is complete. If there is no baseplate, the CRB macromolecules combine with each other to form clusters. Calcification is concomitant with the aggregation process. Some anomalous soft egg shells exhibit relatively normal CRB with added calcified spherulites. The rounded "Type B bodies" reported by Solomon (1985) and Belyavin and Solomon (1986) may be calcified spherulites of CRB matrix, either attached to relatively small BP elements or having no detectable intermediate baseplate. The movements of the egg in the uterus and the directed flow of the plumping fluid may drive the aggregation process. The flow of plumping fluid is maximal when the egg is in the uterus for about 30 min, and the flow stops

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

crographs published by Wyburn et al. (1973) indicate that shrinkage of the egg-shell matrix occurred in their decalcified samples and that they included much of the CRB matrix with the palisade matrix. They did not report the existence of crown-like structures. Stemberger (1971) concluded that mammillary cores are initiated in the isthmus as centers for the deposition of the secretions of the red isthmus and that in the shell gland, calcite covers the mammillary cores to form the mammillae. No mention was made of a structure equivalent to the crown. In retrospect, it is clear that the oldest egg shell studied by Stemberger (1971) was taken at about 10 h postoviposition of the previous egg, before crown formation had been initiated. In a later paper Stemberger et al. (1977), using the same data base but changing the terminology, described "mammillae" as enmeshed in the fibers of the outer shell membrane and growing to confluence in the shell gland. No mention was made either of mammillary cores or of crown-like structures. Fujii (1974) described the mammillary core as the apex of the mammilla, which is partly embedded in the shell membrane. Fujii (1974) equated the mammillary core with the "basal cap" described by Schmidt (1960) and the "organic matrix core" described by Terepka (1963b). Solomon (1983) considered the chemically modified portions of the membrane fibers in the mammillary cap region to be the mammillary cores. Schmidt (1960) theorized that a primary spherite (granule) in the tip of the mammillary knob is the site where crystallization is initiated, extending outward to form the exospherite (cone plus palisade layer) and inward to form the eisospherite (basal cap), a hypothesis favored by Tyler (1969), Erben (1970), and Tyler and Fowler (1978). No layer corresponding to the observed crown layer was reported.

CALCIUM RESERVE ASSEMBLY

REFERENCES Bellairs, R., and A. Boyde, 1969. Scanning electron microscopy of the shell membranes of the hen's egg. Z. Zellforsch. Mikrosk. Anat. 96:237-249. Belyavin, C. G., and S. Solomon, 1986. Microscopic evaluation of egg shell quality. Misset Int. Poult. 2:60-63. Burmester, B. R., 1940. Study of the physical and chemical changes of the egg during its passage through the isthmus and uterus of the hen's oviduct. J. Exp. Zool. 84:445-500. Dieckert, J. W., M. C. Dieckert, and C. R. Creger, 1988a. Abnormal eggshell structure: Type I. Shells with intact calcium reserve assemblies and no palisade layer. Poultry Sci. 67(Suppl. 1):77. (Abstr.) Dieckert, M. C, J. W. Dieckert, and C. R. Creger, 1988b. Anomalous hen eggshells: Type II. Abnormal calcium reserve assemblies. Poultry Sci. 67(Suppl. 1):78. (Abstr.) Erben, H. K., 1970. Ultrastructuren und Mineralisation

rezenter und fossiler Eischalen bei Vogeln und Reptilien. Biominer. Forsch. 1:1-66. Fujii, S., 1974. Further morphological studies on the formation and structure of hen's eggshell by scanning electron microscopy. J. Fac. Fish. Anim. Husb. Hiroshima Univ. 13:29-56. Green, M. R., J. V. Pastewka, and A. C. Peacock, 1973. Differential staining of phosphoproteins on polyacrylamide gels with a cationic carbocyanine dye. Anal. Biochem. 56:43-51. Hayat, M. A., 1970. Principles and Techniques of Electron Microscopy. Biological Applications. Vol. 1. Van Nostrand Reinhold Co., New York, NY. Houi, D., and R. Lenormand, 1984. Particle deposition on a filter medium. Pages 173-176 in: Kinetics of Aggregation and Gelation. Elsevier Science Publishers B. V., Amsterdam, The Netherlands. Johnston, P. M., and C. L. Comar, 1955. Distribution and contribution of calcium from the albumen, yolk and shell to the chick embryo. Am. J. Physiol. 183: 305-370. Packard, G. C , and M. J. Packard, 1980. Evolution of the cleidoic egg among reptilian antecedents of birds. Am. Zool. 20:351-362. Packard, M. J., T. M. Short, B. C. Packard, and T. A. Gorell, 1984. Sources of calcium for embryonic development in eggs of the snapping turtle Chelydra serpentina. J. Exp. Zool. 230:81-87. Robinson, D. S., and N. R. King, 1963. Carbonic anhydrase and formation of the hen's egg shell. Nature 199: 497-498. Robinson, D. S., and N. R. King, 1968. Mucopolysaccharides of an avian egg shell membrane. J. Royal Microsc. Soc. 88:13-22. Robinson, D. S., and N. R. King, 1970. The structure of the organic mammillary cores in some weak egg shells. Br. Poult. Sci. 11:39-44. Romanoff, A. L., and A. J. Romanoff, 1949. The Avian Egg. John Wiley and Sons, New York, NY. Romanoff, A. L., and A. J. Romanoff, 1967. Biochemistry of the Avian Embryo. A Quantitative Analysis of Prenatal Development Interscience Publishers, John Wiley and Sons, New York, NY. Sajner, J., 1955. Uber die Mikroskopischen Veranderungen der Eischale der Vogel im Laufe der Inkubationszeit. Acta Anat. 25:141-159. Schmidt, W. J., 1960. Polarisationsoptik und Bau der Kalkschale des Huhnereies. Z. Zellforsch. Mikrosk. Anat. 52:715-729. Schmidt, W. J., 1965. Morphologie der Kalkresorption an der Ausgebruteten VOgel-Eischale. Z. Zellforsch. Mikrosk. Anat. 68:874-892. Simkiss, K., 1961. Calcium metabolism and avian reproduction. Biol. Rev. 36:321-367. Simkiss, K., and C. Tyler, 1957. A histochemical study of the organic matrix of hen egg-shells. Q. J. Microsc. Sci. 98:19-28. Solomon, S. E., 1983. Oviduct. Pages 379-419 in: Physiology and Biochemistry of the Domestic Fowl. Vol. 4. Academic Press, New York, NY. Solomon, S. E., 1985. Eggshell quality. A structural evaluation. Poult. Int. Oct:58-62. Spurr, A. R„ 1969. A low-viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. Res. 26:31-43. Stemberger, B. H., 1971. Microscopic examination of the avian shell membrane from the posterior oviduct to study the formation of mammillae. M. S. Thesis. Penn.

Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014

after 12 to 13 h postoviposition of the previous egg (interpretation of data from Burmester, 1940). The model for an analogous system that involves the deposition of particles on a filter medium in "cake" filtration may serve as an heuristic model for studying the process of aggregation/calcification in the egg shell. Houi and Lenormand (1984) created a computer simulation of such a system, based on concepts of fractal aggregation, in order to study the growth and properties of filter cake deposition under various combinations of ballistic (fluid flow) and diffusive motion (Brownian motion). Under certain conditions the morphology of the deposits were reminiscent of the distribution of CRB matrix in the CRA. In addition, the morphology of the fractal aggregates in the filter cake simulation suggests that one can have an organic matrix/calcite crystal aggregate without the complete encasement of one by the other. Structures similar to the mammillae in hen egg shells are a common feature of the egg shells of other avian species, crocodilians, certain turtles, and dinosaurs (see Erben, 1970). For example, knob-like structures about 127 u.m in height are found attached to the surface of the egg shell membrane of the snapping turtle, Chelydra serpentina (Packard and Packard, 1980). In this species of turtle about 56% of the calcium in the embryo and residual yolk at hatching comes from the pliable egg shell (Packard et al., 1984). It may be that these structures are homologues of the mammillae, with included CRA, of the hen egg shell.

1583

1584

DJECKERT ET AL.

State Univ., University Park, PA. Stemberger, B. H., W. J. Mueller, and R. M. Leach, Jr., 1977. Microscopic study of the initial stages of egg shell calcification. Poultry Sci. 56:537-543. Terepka, A. R., 1963a. Structure and calcification in avian egg shell. Exp. Cell Res. 30:171-182. Terepka, A. R., 1963b. Organic-inorganic interrelationships in avian eggshell. Exp. Cell Res. 30:183-192. Tuan, R. S., 1980. Calcium transport and related functions in the chorioallantoic membrane of cultured shell-less chick embryos. Dev. Biol. 74:196-204. Tyler, C , 1964. On Avian Eggshells, (translation of original paper by Wilhelm von Nathusius, 1868) Univ. of Reading, Reading, England, UK. Tyler, C , 1969. Avian egg shells: their structure and charac-

teristics. Pages 81-130 in: International Review of General and Experimental Zoology, Vol. 4. W.J.L. Felts and R. J. Harrison, ed. Academic Press, New York, NY. Tyler, C , and S. Fowler, 1978. The distribution of organic cores, cones, cone junctions and pores in the egg shells of wild birds. J. Zool. (Lond.) 186:1-14. Tyler, C , and K. Simkiss, 1959. Studies on egg shells. XII. Some changes in the shell during incubation. J. Sci. Food Agric. 10:611-615. Wyburn, G. M., H. S. Johnston, M. H. Draper, and M. R. Davidson, 1973. The ultrastructure of the shell forming region of the oviduct and the development of the shell olGallus domestic us. Q. J. Exp. Physiol. Cogn. Med. Sci. 58:143-151. Downloaded from http://ps.oxfordjournals.org/ at University of Melbourne on October 11, 2014