EXPERIMENTAL AND MOLECULAR PATHOLOGY 42, II-77 (1985) Mosaicism in Female Hybrid Hares Heterozygous for Glucose-6-phosphate Dehydrogenase VII. Evidence for Selective Advantage of One Phenotype over the Other in Ditypic Samples from Aortas of Hares Fed Cholesterol Oxidation Products’ K. T. LEE, K. JANAKIDEVI, M. KROMS, J. SCHMEE,* AND W. A. THOMAS Department of Pathology, Albany Medical College, Albany, New York 12208, and *Statistical Unit, Union College, Schenectady, New York 12308 Received August 17, 1984, and in revised form October 5, 1984 In a previous study the authors demonstrated monotypic foci, all of timidus type, in the atherosclerotic lesions and normal aortic tissue of glucose-6-phosphate dehydrogenase (G- 6-PD) mosaic hares fed 25hydroxycholesteroI or trio1 in addition to a high-cholesterol diet. Other hares fed the high-cholesterol diet without the cholesterol oxidation product additives did not develop monotypic foci. In the current study the question asked was whether ditypic samples from lesions or aortic mediae of the same hares fed the additives would show a higher percentage of timidus type G-6-PD than the hares fed the cholesterol alone. The results showed a significantly higher percentage of timidus type in both lesions and mediae in the groups fed either 25-hydroxycholesterol or trio1 than in the group fed cholesterol alone. The observations suggest that the change from ditypism to monotypism reported earlier with these cholesterol oxidation products was the result of progressive elimination of the europaeus type to leave only the timidus G-6-PD. This is consistent with the “selec- tive advantage of one phenotype over the other” hypothesis for explaining the observed monotypism as opposed to the “genetic transformation” hypothesis. o 1985 Academic Press, Inc. INTRODUCTION We reported in an earlier paper that monotypism was produced in atheroscle- rotic and/or normal aortic tissues of 4 of 10 glucosed-phosphate dehydrogenase (G-dPD) mosaic hybrid hares fed small amounts of cholesterol oxidation products for 6-21 months in addition to a high-cholesterol diet. Fourteen hares fed the high-cholesterol diet without additives for a similar period did not develop mono- typism. For the background regarding the mosaicism and the significance of the findings in relation to atherosclerosis the reader is referred to the earlier paper (Janakidevi et al., 1984). In the current investigation we have carried out additional studies on the same 24 hybrid hares. In the previous study we used samples from lesion tissue and from grossly normal intima-media. In the current study we took punch biopsies of lesions along with underlying media. The punch samples were separated into lesion and normal-appearing medial tissue using a dissecting microscope. Portions were used for histologic measurement of thickness and other portions for deter- mination of the relative percentages of the two G-6-PD phenotypes (T for the timidus parent and E for europaeus) in the aortic lesion and medial tissue. This permitted direct comparison of G-6-PD ratios in lesions with those in adjacent underlying media. Emphasis in this study was on the nonmonotypic (ditypic) samples. i Supported by PHS Grant HL-20993. 71 0014-4800/85 $3.00 Copyt’kbt B 1985 by Academic Press, Inc. All rights of reproduction in any form reserved.
Transcript
EXPERIMENTAL AND MOLECULAR PATHOLOGY 42, II-77 (1985)
Mosaicism in Female Hybrid Hares Heterozygous for Glucose-6-phosphate Dehydrogenase
VII. Evidence for Selective Advantage of One Phenotype over the Other in Ditypic Samples from Aortas of Hares Fed Cholesterol Oxidation Products’
K. T. LEE, K. JANAKIDEVI, M. KROMS, J. SCHMEE,* AND W. A. THOMAS
Department of Pathology, Albany Medical College, Albany, New York 12208, and *Statistical Unit, Union College, Schenectady, New York 12308
Received August 17, 1984, and in revised form October 5, 1984
In a previous study the authors demonstrated monotypic foci, all of timidus type, in the atherosclerotic lesions and normal aortic tissue of glucose-6-phosphate dehydrogenase (G- 6-PD) mosaic hares fed 25hydroxycholesteroI or trio1 in addition to a high-cholesterol diet. Other hares fed the high-cholesterol diet without the cholesterol oxidation product additives did not develop monotypic foci. In the current study the question asked was whether ditypic samples from lesions or aortic mediae of the same hares fed the additives would show a higher percentage of timidus type G-6-PD than the hares fed the cholesterol alone. The results showed a significantly higher percentage of timidus type in both lesions and mediae in the groups fed either 25-hydroxycholesterol or trio1 than in the group fed cholesterol alone. The observations suggest that the change from ditypism to monotypism reported earlier with these cholesterol oxidation products was the result of progressive elimination of the europaeus type to leave only the timidus G-6-PD. This is consistent with the “selec- tive advantage of one phenotype over the other” hypothesis for explaining the observed monotypism as opposed to the “genetic transformation” hypothesis. o 1985 Academic
Press, Inc.
INTRODUCTION
We reported in an earlier paper that monotypism was produced in atheroscle- rotic and/or normal aortic tissues of 4 of 10 glucosed-phosphate dehydrogenase (G-dPD) mosaic hybrid hares fed small amounts of cholesterol oxidation products for 6-21 months in addition to a high-cholesterol diet. Fourteen hares fed the high-cholesterol diet without additives for a similar period did not develop mono- typism. For the background regarding the mosaicism and the significance of the findings in relation to atherosclerosis the reader is referred to the earlier paper (Janakidevi et al., 1984).
In the current investigation we have carried out additional studies on the same 24 hybrid hares. In the previous study we used samples from lesion tissue and from grossly normal intima-media. In the current study we took punch biopsies of lesions along with underlying media. The punch samples were separated into lesion and normal-appearing medial tissue using a dissecting microscope. Portions were used for histologic measurement of thickness and other portions for deter- mination of the relative percentages of the two G-6-PD phenotypes (T for the timidus parent and E for europaeus) in the aortic lesion and medial tissue. This permitted direct comparison of G-6-PD ratios in lesions with those in adjacent underlying media. Emphasis in this study was on the nonmonotypic (ditypic) samples.
i Supported by PHS Grant HL-20993.
71
0014-4800/85 $3.00 Copyt’kbt B 1985 by Academic Press, Inc. All rights of reproduction in any form reserved.
72 LEE ET AL.
The principal question being asked was whether there was evidence for change in the T:E ratios in the ditypic samples of aortic media or lesions of the hares fed cholesterol plus cholesterol oxidation products as compared to those in the hares fed cholesterol alone that would suggest shifts toward the monotypic state (which was the T monotypic in all cases reported earlier). This would provide support for the “selective advantage of one phenotype over the other” hypothesis for explaining the observed monotypism as opposed to the “genetic transfor- mation” hypothesis. An additional question was in regard to similarity (or lack of it) of the G-6-PD T:E ratios in the lesions vs the underlying media which would permit us to characterize the mosaic hare model further.
MATERIALS AND METHODS
Basic data on the 24 mosaic hares fed either the standard high-cholesterol diet alone or additional oxidation products and analytical procedures for G-6-PD were presented in the earlier publication (Janakidevi et al., 1984). Only the additional procedures required for the study presented in this report are given here.
Sampling methods. Samples of the full thickness of the aorta (excluding ad- ventitia) were taken with a 3-mm skin punch. One-third of the sample cut through the entire thickness was taken for measuring thicknesses of lesion and media. The remaining two-thirds of the sample was divided into four equal parts under a dissecting microscope and each part was subdivided into atherosclerotic lesion and medial portions. This sampling was done in such a fashion that each subdi- vided lesion sample could be paired with its underlying medial sample. If either member of the pair was monotypic that pair was not included in the current study, which was aimed only at ditypic samples. Three or more lesions were sampled for each of the aortas. They were selected in most cases from lesions in the upper, middle, and distal portions of the aorta. Each sample was analyzed for G-6-PD phenotypes and results are expressed as percentage timidus (%T).
Statistical methods. Data were analyzed by Student’s t test, using the paired t test where indicated. Analyses were also made by the GLM procedure of SAS and a triple-nested design was used in order to identify the sources of variation in the present T values.
The daily dose of cholesterol was 2 g in 200 g of the diet, 25hydroxycholesterol (5Cholesten-3B,25-diol) 0.5 mg, and trio1 (cholestan-3B, 5a, 6B-triol) 0.25 mg.
RESULTS
Results for the group fed diets supplemented with cholesterol without added cholesterol oxidation products are summarized in Table I. In 14 of the 42 lesions values for %T in the lesion and underlying media were significantly different. In 11 of the 14 the value for the media was greater than that for the lesion. There was no significant correlation between lesion thickness and %T.
Results for the group fed diets supplemented with cholesterol plus 25-hydrox- ycholesterol are presented in Table II. In 14 of the 25 lesions values for %T in lesions and underlying media were significantly different. In 13 of the 14 the value for the media was greater than that for the lesion. There was no significant cor- relation between lesion thickness and %T.
In Table III results are presented for the group whose diet was supplemented with cholesterol plus triol. In 8 of the 15 lesions there were significant differences between %T in lesions and underlying media. In 7 of the 8 the medial values were
TABLE I G-6-PD Patterns and Thicknesses of Lesion and Medial Tissues from 14 Hares
Fed Cholesterol without Additives
Hare No. Media Months on diet Serum chol. Lesion Thickness (mg/dl) No. (km) %T (range)
- 7A 1
6 2 8.50 3
22 1 7 2
485 3
69 1 9 2
485 3
126 1 11 2
482 3
129 1 11 2
505 3
133 1 11 2
455 3
142 1 11 2
430 3
1lA 1 12 2
1047 3
30 1 14 2
786 3
131 1 14 2
742 3
132 1 14 2
500 3
31 1 15 2
751 3
36 1 16 2
764 3
41 1 17 2
699 3
Mean for 14 hares
270 585 522
122 207 437
243 270 113
338 563 338
270 423 383
360 338 419
486 180 270
324 180 216
788 338 275
581 171 221
342 720 113
270 135 221
450 432 450
225 243 216
297 239 320
230 513 99
360 473 495
207 221 243
540 518 257
630 729 203
162 378 419
378 347 167
572 576 270
572 527 153
185 162 279
351 675 675
473 360 495
40 (32-49) 48 (41-56) 53 (51-57)
68 (65-72) 62 (57-68) 57 (37-72)
74 (71-80) 72 (64-81) 77 (73-82)
73 (71-77) 69 (56-79) 69 (64-73)
71 (60-78) 79 (76-83) 79 (71-85)
73 (64-79) 59 (54-65) 61 (40-74)
67 (56-74) 70 (65-79) 68 (60-74)
74 (56-83) 73 (67-81) 79 (76-82)
63 (51-79) 61 (56-70) 63 (55-68)
76 (72-79) 63 (58-69) 69 (61-76)
75 (66-80) 77 (72-83) 77 (73-79)
67 (63-71) 74 (71-79) 67 (64-69)
73 (62-82) 52 (44-68) 65 (57-68)
62 (50-66) 63 (58-76) 66 (61-75)
509 360 158
43 (33-60) 47 (33-59) 39 (30-44)
69 (64-74) 64 (57-70) 52 (38-72)
75 (73-78) 68 (60-77) 66 (61-71)
52 (46-62) 50 (39-58) 55 (42-61)
71 (59-80) 80 (75-87) 77 (71-82)
73 (67-79) 66 (53-76) 64 (52-72)
55 (46-63) 65 (59-70) 53 (40-65)
63 (50-75) 69 (63-75) 77 (75-80)
71 (64-75) 63 (56-74) 62 (53-71)
78 (71-84) 71 (60-79) 66 (62-71)
65 (54-72) 73 (67-78) 70 (66-76)
68 (64-73) 71 (66-76) 66 (60-73)
73 (63-79) 58 (42-69) 62 (53-78)
66 (60-73) 64 (59-70) 66 (55-72)
12 months 642 ” 50 mg/dlb n = 42 375 2 24 67.4 f 1.4 335 f 26 64.8 f 1.5
Lesion
Thickness (w-4 %T (range)
Significance of paired
differences
NS NS
0.03
NS NS NS
NS NS
<O.OOOl
<O.oool <O.OOOl <O.OOOl
NS NS NS
NS 0.009 NS
0.008 NS
<O.OOOl
<O.OOOl 0.04 NS
0.03 NS NS
NS 0.003 NS
0.0002 NS
0.004
NS NS NS
NS NS NS
NS NS NS
0 Values are averages of six samples analyzed for each lesion in this and subsequent tables. b SEM.
74 LEE ET AL.
TABLE II
G-6-PD Patterns and Thicknesses of Lesion and Medial Tissues from 6 Hares Fed Cholesterol Plus 25Hyroxycholesterol
Hare No. Months on diet
Serum chol. Lesion
(mg/dll No.
214 1 6 2
768 3 4
67” 1
7 2 653 3
4 5
62 1
9 2 395 3
4
105 1 14 2
657 3
4
103 1 21 2
706 3 4
109 1 21 2
802 3 4
Mean for 6 hares
13 months
663 zt 59 mg/dlb n = 25
Media Lesion Significance
Thickness
(km0
330 660
550 660
550
270 275
330 660
1100
550 440
440
440 330 220
220
550 275
220 330
220 220 770
330
%T (range) Thickness
(t-d %T (range)
80 (74-87) 550 85 (74-91) 880
81 (73-89) 990 82 (76-89) 880
78 (67-83) 550
78 (74-83) 330 79 (72-85) 330
77 (74-80) 440 78 (73-82) 330
53 (45-64) 275
48 (46-52) 220 57 (49-66) 220 55 (41-68) 495
76 (68-80) 330 72 (65-77) 440 71 (59-81) 550
68 (56-78) 440
74 (71-77) 1110
75 (70-78) 1320
76 (72-81) 550 75 (71-79) 550
80 (74-83) 1430
76 (63-83) 880 80 (68-86) 550 78 (74-85) 440
71 (67-74) 75 (65-85)
71 (67-80) 75 (70-79)
74 (69-84)
73 (67-79) 73 (63-781
67 (61-75) 72 (67-761
50 (42-70)
52 (40-68) 51 (38-60) 44 (35-55)
76 (74-82)
75 (62-84) 74 (67-80)
73 (69-79)
68 (59-79)
73 (69-84) 74 (70-77)
74 (69-76)
76 (72-78)
70 (53-81) 71 (61-77) 74 (68-79)
438 k 43 73.1 2 1.9 603 4 67 69.1 t 1.8
of paired differences
co.02 co.02
co.003 co.05
co.05 <0.02
eo.02 ‘zo.003
<0.005
NS
NS NS
co.05
NS NS
NS co.03
NS
NS NS
NS
co.02 co.03
co.03 NS
0 Monotypic samples found but not included here.
b SEM.
greater. There was again no significant correlation between lesion thicknesses and %T.
In Table IV mean values for %T in the lesions and mediae of the cholesterol- only group are compared with corresponding values for the two groups that also received cholesterol oxidation products. For each comparison the %T in the groups supplemented with cholesterol oxidation products is significantly higher than that for the cholesterol-only group.
An additional observation made was in regard to the thicknesses of lesions among the three groups. The mean thickness of the lesion in the 25hydroxycho- lesterol group was 603 ? 67 pm; in the cholesterol-only group, 335 + 26 pm;
MOSAICISM IN FEMALE HYBRID HARES
TABLE III G-6-PD Patterns and Thicknesses of Lesion and Medial Tissues from 4 Hares
Fed Cholesterol Plus Trio1
75
Hare No. Media Lesion Months on diet Significance Serum chol. Lesion Thickness Thickness of paired (m&l) No. (v-4 %T (range) (wm) %T (range) differences
a Monotypic samples found but not included here. * SEM.
and in the trio1 group, 350 + 59 pm. The difference between the 25hydroxycho- lesterol group and the cholesterol-only group was statistically significant. How- ever, the difference between the trio1 group and the cholesterol-only group was not significant.
TABLE IV Comparisons of Differences of Means between Cholesterol Only and Cholesterol Plus Oxidation
Product Groups in Lesion and Media Samples from the Aortas
Chol. only
25-Hydroxycholesterol group
Difference of
Chol. + 25 OH means
Significance of
difference (P)
Lesion %T Media %T
64.8 IT 1.5 69.1 2 1.8 4.3 <0.02 67.4 + 1.4 73.1 f 1.9 5.7 <0.02
Trio1 group
Lesion %T Chol. only Chol. + Trio1 64.8 f 1.5 70.9 f 1.2 6.1 co.01
Media %T 67.4 k 1.4 74.5 2 1.3 7.1 <o.oos
76 LEE ET AL.
DISCUSSION
The principal conclusion to be drawn from the current study is that the diet containing cholesterol oxidation products (in addition to high cholesterol) results in shifts of G-6-PD T:E ratios toward the T phenotype in both ditypic lesion and medial aortic tissue as compared to corresponding values for the hares fed cho- lesterol without the additives. This supports the hypothesis that selective advan- tage of one phenotype over the other accounts for the T-type monotypism re- ported previously for the same hares. .
In comparing T:E ratios in lesions with those in underlying media we found that in general the percentage of T phenotype was as great or greater in the media as in the lesions. This is consistent with the hypothesis that the cholesterol oxi- dation products are producing their effect on the G-dPD ratios by selectively killing off the E phenotype rather than by selective stimulation of the T phenotype to multiply faster than the E phenotype. We arrive at this conclusion because the cells in the lesions are presumably multiplying faster than those in the media, which should result in any phenomenon related to rapid multiplication being more prominent in the lesions than in the media.
In the cell culture situation we reported previously that the T phenotype was more susceptible to the lethal effects of cholesterol oxidation products than the E phenotype (Murray et al., 1981). The in viva results suggest the opposite- namely that the E phenotype is more susceptible to the toxic effects than the T phenotype. A possible explanation for this paradox is that the genetic character- istics that give an advantage in one type of environment (in vitro for example) may give a disadvantage in another (in viva).
Another feature of interest was that we obtained relatively few samples in the 80 and 90% T classes even in the four hares that yielded some samples that were 100% T (monotypic). With the selective-advantage hypothesis viewed simplisti- cally we would rather expect to find an accumulation of high values particularly in those hares with some 100% T values. One possibility relates to a “critical level” of some putative toxic material. Perhaps such a critical level is reached in a particular time site and all of the cells of sensitive phenotype die out. Another possibility relates to the sensitivity of the G-6-PD phenotype detection method. We occasionally obtained values of 95% or more for the predominant phenotype and assumed that the sensitivity is in the range of 95%. However, it is possible that the sensitivity is not that high in all cases.
In human G-6-PD mosaics monotypism is frequent in advanced atherosclerotic lesions and seldom seen in the media (Benditt and Benditt, 1973; Pearson et al., 1978; Thomas et al., 1979). In the hybrid-hare model the monotypism was as frequent (or more so) in the media as in the lesion. Consistent with this in the current study was the fact that the G-6-PD T:E ratios were generally closer to monotypism (T type a higher percentage) in the media than in the lesions. The reason for this result is not clear. The underlying basis perhaps has to do with the source of the putative toxic products. In the human situation cholesterol oxidation products have been found in advanced lesions presumably formed in situ (Smith and Pandya, 1973). In this situation the adjacent viable lesion tissue would probably be exposed to greater concentration than would the media. In the hares the lesions in general had not yet reached the advanced necrotic phase which when present perhaps facilitates cholesterol oxidation; the source of the putative harmful cholesterol oxidation products is probably mainly from the diet.
MOSAICISM IN FEMALE HYBRID HARES 77
The harmful substances arriving from the diet by way of the blood stream would presumably reach the cells in the media as readily (or more so) as those in the intima.
REFERENCES BENDITT, E. I?, and BENDITT, J. M. (1973). Evidence for a monoclonal origin of human atherosclerosis
plaques. Proc. Natl. Acad. Sci. USA 70, 1753-1756. JANAKIDEVI, K., LEE, K. T., KROMS, M., IMAI, H., and THOMAS, W. A. (1984). Mosaicism in female
hybrid hares heterozygous for glucose-6phosphate dehydrogenase (G-6-PD). VI. Production of monotypism in the aortas of 4 of 10 mosaic hares fed cholesterol oxidation products. Exp. Mol. Pathol. 41, 354-362.
MURRAY, C. D., JANAKIDEVI, K., THOMAS, W. A., REINER, J. M., and LEE, K. T. (1981). Mosaicism in female hybrid hares heterozygous for glucose&phosphate dehydrogenase (G-6-PD). III. Studies of effects of 25-hydroxycholesterol exposure on cultured skin Bbroblasts from hybrid hares hetero- zygous for G-6-PD. Exp. Mol. Pathol. 34, 209-215.
PEARSON, T. A., DILLMAN, J. M., SOLEZ, K., and HE~INSTALL, R. H. (1978). Clonal characteristics in layers of human atherosclerotic plaques. Amer. J. Pathol. 93 (l), 93-102.
SMITH, L. L., and PANDYA, N. L. (1973). Sterol metabolism. Atherosclerosis 17, 21-30. THOMAS, W. A., REINER, J. M., JANAKIDEVI, K., FLORENTIN, R. A., and LEE, K. T. (1979). Population
dynamics of arterial cells during atherogenesis. X. Study of monotypism in atherosclerotic lesions of black women heterozygous for glucose-dphosphate dehydrogenase (G-6-PD). Exp. Mol. Pathol. 31, 367-386.
