Click for next page ( 30


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 29
6 Current Knowledge and Estimation of Genetic Risk The estimation of the genetic effect of ionizing radiation on human popula- tions has been a matter of concern since World War II. The two main bodies involved are the United Nations' Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) and the U.S. National Academy of Sciences' Committee on the Biological Effects of Ionizing Radiations (BEIR). In addition, the Inter- national Commission on Radiological Protection (ICRP, Oftedal and Searle, 1 980J as well as the Nuclear Regulatory Commission (Nuclear Regulatory Commission, 1985) have published documents in which genetic risk estimates are included. All tend to give similar estimates because they all use basically the same set of data. In what follows it is important to understand three points: 1. The health effects resulting from mutations induced by ionizing radiation are indistinguishable from those resulting from other agents or that arise sponta- neously. 2. Even if a significant increase in some endpoint is shown statistically to be due to additional radiation exposure, no specific case can be proved to be ascribable to that exposure. 3. Finally, even high doses of radiation (greater than 2,000 mSv t200 rem]) will add only a small number of additional cases of genetic disorders to the rela- tively large number that are expected to occur as a result of spontaneous muta- tions, most of which have existed in the population for many generations. 29

OCR for page 29
30 AD VERSE REP ROD UCTI HE OUTCOMES BASIC ESTIMATION EQUATION Many of the estimates of the genetic impact of ionizing radiation on human populations have made use, in one form or another, of the following formula: I=Sx 1/DDxMCxD, where: (Equation 1) I is the increased number of cases (per generation) of genetic effects due to radiation, often called the induced burden, S is the number of cases (per generation) normally present in a population not exposed to additional radiation, the spontaneous burden, DD is the doubling dose (see below), MC is the mutation component (see below), and, D is the dose of additional radiation to which the population is exposed. The use of Equation 1, especially when applied as in the case of the Atomic Veterans, requires some explanation. Let B = the total burden to the population of some genetic disease or class of diseases, e.g. the total number of cases arising per generation; and let m = mutation rate. Now consider the equality: /\B/B = Am/m x (/\B/B) I (/\m/m). (Equation la) that is, the relative change in B equals the relative change in m times the relative change in B to the relative change in m. Suppose that in the dose range being considered, we may assume that muta- tion rate is a linear function of dose, for example, m = mO + bD, where mO is the spontaneous mutation rate. Then /\mlm= (m-mO)/mO=bD/mO. But, m0/b is the doubling dose exactly that dose that induces mO mutations. So we see that /\m/m = D/DD. Furthermore, (/\B/B) I (/\mlm) is the mutation component (Crow and Denniston, 1981), the relative change in the burden to the relative change in the mutation rate. Hence, we see that Equation la can be written /\B/B = (D/DD) x MC or since /\ B = I and B = S. we have I = S x (D/DD) x MC, which is Equation 1. Now, in the usual application of Equation 1, I ~ = /\B) applies to the change in the burden from just before a permanent change in the dose to the time the population reaches the new equilibrium between mutation and selection. But, in fact, the relevant time period is determined by how the mutation component (MC) is defined. The MC may be defined to apply to any number of generations after the change in radiation exposure. In particular, it may be defined to apply to the first generation after the increase in dose. If the increase in dose is perma

OCR for page 29
GENETIC PRINCIPLES 31 nent, I (= /\B) slowly increases from generation one to equilibrium; if the dose increase is temporary (e.g., a burst, as in the case of the atomic veterans), then I (= AB) increases in the first generation but then slowly decreases until the old equilibrium is reestablished. The simplest example is that of an autosomal dominant gene. At equilib- rium between mutation and selection, the frequency of the trait is 2m/s, where m is the mutation rate and s is the selection coefficient. If the mutation rate in- creases permanently from m to m(1 + k) then in generation n after the increase the frequency is: 2tm/s+km(1 -~1 -s)"/s)~. If the increase in mutation is only a burst, the frequency in generation n is: 2tm/s+km(1 -s)"-". The first frequency eventually rises to 2m (1 + k)/s while the second returns to 2m/s. In generation one, the two are identical. Using the definition of muta- tion component given above, the mutation component in generation n can be defined as MC,, = 1 - (1 - s)" in the case of a permanent change and as MCn = s(1 - s)"~ ~ in the case of the burst. The term "mutation component," without specification of the generation, conventionally refers to MC~. For more complicated traits, no simple formulas exist, but for threshold traits, such as congenital abnormalities, the mutation component in the first gen- eration is generally less than one or two percent. Equation 1 thus can be applied to the case of the Atomic Veterans by using the value of mutation component that applied to the first-generation effect. Subsequent generations (e.g., grand- children) would show even smaller effects. If the dose-response curve is not linear but concave upward, this use of the doubling dose in Equation 1 will tend to overestimate risk if the data from which doubling dose is estimated are obtained from high doses. The doubling dose has traditionally been estimated from experimental animal data, mostly the mouse, although an estimate is also provided by the extensive studies of the children of atomic bomb survivors from Hiroshima and Nagasaki. In summary, the Beir V report (NRC, 1990) states "Although the doubling dose method is based on equilibrium considerations, the method can be used to estimate the effects of an increase in the mutation rate on the first few generations by taking a proportion of the equilibrium dam- age. For example, for a permanent increase in the mutation rate the effect of a dominant mutation in the nth generation is 1 - (1 - s)" of the equilibrium dam- age, where (1 - s) is the fitness of carriers of the dominant gene." An alternative method of estimating genetic risk in the first generation is provided by the so-called "direct method" pioneered by Ehling and Selby (see Ehling, 1991~. A detailed description of this method is given in UNSCEAR 1993, Appendix G. Briefly, the method is based on the equation: lo, a,

OCR for page 29
32 where: ADVERSE REPRODUCTIVE OUTCOMES Risk per unit dose = Fit x M x N. Fat is the frequency of radiation-induced dominant mutations per unit dose, M is the reciprocal of the fraction of total mutations thought to affect the body systems under study, e.g., skeletal, cataracts, and N is the number of children born in the population under consideration. For example, the dominant cataract mutation frequency in the mouse was estimated to be 0.15-0.18 x 10-6 mutations per 0.01 Gy per gamete for low dose rate data. It was also estimated that approximately 2.7% of all serious dominant mutations are cataract causing mutations, i.e., M = 36.8. This gives a risk of 6-7 serious dominant disorders per 0.01 Gy of paternal exposure per 106 offspring. The estimate based on skeletal mutations in the mouse is similar. Returning to the discussion of the doubling dose method, the spontaneous burden S is estimated from human epidemiologic data. The mutation compo- nent, MC, is roughly that portion of the spontaneous burden expected to increase in proportion to the mutation rate (Crow and Denniston, 19811. Dose as used in Equation l usually refers to the average or common dose to the gonads of both sexes, unless a sex-specific effect is being estimated. As an example, the BEIR ~ committee (1990) estimated the induced burden of congenital abnormalities caused by radiation to be, after a new equilibrium is attained, 10 to 100 additional cases per million liveborn offspring per 10 mSv (1 rem) per generation (NRC, 19901. The calculations were as follows: S = 20,000-30,000 spontaneous cases of congenital abnormalities per mil- lion liveborn offspring. DD = 1 Sv (100 rem) for low dose or low dose rate estimated from a con- sideration of data from studies in mice and humans. MC = 0.05-0.35, at the new equilibrium. D = 0.01 Sv (1 rem) to each of the parents. Therefore, I = (20,000-30,000 cases) x (1/1 Sv) x (0.05 - 0.35) = 10-105 per million liveborn, at the new equilibrium. In the report this estimate was rounded to 10-100 per million liveborn, to avoid the appearance of false accu- racy. As a worst case, it was assumed that as much as 10% of this effect might manifest itself in the first generation after the increase in exposure. To estimate the effect of increased radiation exposure on the children of ex- posed parents, then, one must have estimates of the spontaneous burden of the endpoint of interest, its doubling dose, the dose itself, the mutation component of the endpoint, and finally, how much of the total effect is expected to appear in the first generation after exposure.

OCR for page 29
GENETIC PRINCIPLES 33 DATA FROM WlIICII RISK ESTIMATES HAVE BEEN MADE Mice Studies with the mouse have yielded two kinds of results: (1) a general qualitative and semiquantitative understanding of the nature of genetic radiation effects and (2) quantitative estimates of the doubling dose. Both have been summarized in detail by the National Research Council's BEIR V committee (NRC, 1990~. The qualitative conclusions were as follows: Radiation-induced mutation rates are higher in the mouse than in the fruit fly (this original finding stimulated much of the subsequent emphasis on mice because of its obvious greater relevance to estimating radiation risks in humans). 2. For mutations of specific loci (a locus is a point on a gene) induced in the spermatogonial stage, there is no significant change in the mutation rate with time after irradiation (i.e., the risk does not decrease with time after exposure). 3. Radiation-induced mutation rates differ markedly from gene to gene. 4. Mutations induced in spermatogonial and post-spermatogonial stages dif- fer with respect to absolute and relative frequencies among loci and by radiation quality. 5. A significant proportion of the mutations detected in the specific locus test have proved to be recessive lethals. 6. Some of the recessive lethal mutations have had a heterozygote effect dramatic enough to be identified in specific individuals. 7. Dominant effects on viability are demonstrable in the first-generation progeny of irradiated males. 8. Chronic irradiation is considerably less effective than acute radiation in inducing mutations in both spermatogonia and oocytes. This dose rate effect appears to be less in males than in females. 9. A significant proportion of radiation-induced mutations in the specific locus test are small deletions. 10. The immature mouse oocyte is highly sensitive to cell killing. Extensive literature on the mouse provides multiple endpoints from which to estimate genetic doubling doses. A detailed summary of the data can be found in Chapter 2 of the BEIR V report (NRC, 1990). The question of estimating dou- bling dose is discussed in the next section which includes a summary table of doubling doses for mice. The mouse is the only mammal for which substantial data on the mutagenic effects of ionizing radiation are available. These effects have been shown to de- pend on dose, dose rate, fractionation pattern, LET, cell stage, sex, age at expo- sure, and the test stock and gene loci used. Qualitatively, these conclusions

OCR for page 29
34 ADVERSE REPRODUCTIVE OUTCOMES probably apply to humans as well, but whether the specific quantitative relations observed in mice transfer to humans is much less certain Humans Two sets of human data have played the predominant role in estimating the risks of ionizing radiation to human populations: (1) data pertaining to the esti- mation of the spontaneous burden, S. in Equation 1 (Stevenson, 1959; Trimble and Doughty, 1974; Jacobs, 1975; Carr and Gedeon, 1977; Carter, 1977; Hook and Hamerton, 1977; Childs, 1981; Czeizel and Sankaranarayanan, 1984; Baird et al., 1988) and (2) data on the Hiroshima-Nagasaki atomic bomb survivors and their children used to estimate the doubling dose (Neel et al., 1953, 1974, 1990; Neel and Schull, 1956b, 1991; ABCC, 1975; Schull et al., 1981a, b; Neel and Lewis, 1990~. A useful compendium of the major articles on the Japanese stud- ies was provided by Neel and Schull (1991~. Summaries and discussions of these data may also be found in reports by Denniston (1982) and UNSCEAR (1986 and 1993) and in the BEIR V report (NRC, 1990~. Estimating the Spontaneous Burden The doubling dose approach uses the existing "normal" incidence of genetic disease as a yardstick against which to measure the effect of radiation. To do this one must know the approximate natural incidence of the endpoints under study. For example, in a sample of random births, approximately 3 percent are expected to have some kind of major congenital abnormality. Presumably, ex- posing the parents to additional radiation will produce additional cases over and above this spontaneous incidence. The major studies that have provided estimates of the spontaneous burden for a number of genetic categories are provided elsewhere (Stevenson, 1959; Trimble and Doughty, 1974; Jacobs, 1975; Carr and Gedeon, 1977; Carter, 1977; Hook and Hamerton, 1977; Childs, 1981; Czeizel and Sankaranarayanan, 1984~. A summary of findings is provided in Table 1 (Table 2-5 of BEIR [NRC, 1990]).

OCR for page 29
- c~ ~ - o of of - v) ~ - . - =: ~ en o ~ ~ ~ o o - . o o ~1 1 11 = - ^ v' o ~ `: ~ e - ~ e ~D.o ~ -O O O ct - ~ o o . l cr~ 0 0 ~C~l ~ ~ u) - :> - o o o s~ - m C~ o a' Ct ~: o S~ C~ 5 Ct _4 C~ m Ce ^ O ._ O ~ 0 0 ~ ~ .= ._ O ~ O ~O O O O ~ ~ ~C~ ~S,~ k~, ~ . ._ C~ V) C) O _ ~- O ~ _ ~ . ~ - = ~ X 00 0 . o o C~ ~ o 1 ~ .. ~ _ ~ O O _ O ~O O O O O O O O O _ _ _ _ ~ ~OooO oo O . .... O 0~00 ~O ( - ,_ . _ ~_ t_ O ~ Ct ~ ~N =~ C~ ~ (t Cr~ ~Z Z ~ rs] ~ ~ Z ~ ~ ~ ~ ~ ~ .N ~ ~ ~ ~ ~ O c~ ? ~ m E ~ ~ ~ m ~ ~ ~ ~ ~ m ~ ._ . _ c~ v' c. u: ._ =4 ~ 0 O > .c ~ .= v, ~ ~ o ~v o o x v' ~ ~ ,.= ~ r . c~ ~ 0 z ~ 0 ~ v' 00 . c~ c~ ~ ~ o ~ 0u cr~ _ ~ ~ - ~ u, - - ~ .- c.> ~i ~ 0 0 ~ ~: ~ 0 ~ c<; ct ~ ) ~ au ct ~ ct c' - ct ~ ~ D v, O - s: Ct ~ C) D ~r, ~ . O .= ' ~ ~ :; 0 - . oca ~ ~c ~ c v' ct =: - u, x ~ ~ 0 0 ~ ct - . _ ct c' v, 5D - . _ ._ ._ - c~ v, ;> ~ ct . _ ~ D ~ ._ 5) 3 ~ v' O c: ~ ._ O v' ~ O v~ c) ct ~ ~ ~ 0 t: O ~ _ ~ ~ . _ Ct s ~ . c: ~ ~ ._ ~ D .= O . _ ~ (V ~ C; :: 04 ce

OCR for page 29
36 ADVERSE REPRODUCTIVE OUTCOMES Again, it is important to stress that a host of genetic defects and heritable disorders will appear in any population in each generation whether or not the parents have been exposed to ionizing radiation. Radiation will tend to increase this number, but as will be seen below, the increased incidence ascribable to exposure to ionizing radiation is likely to be a very small proportion of the natu- rally occurring incidence. Estimating the Effect of Ionizing Radiation The cohort of atomic bomb survivors and their children from Hiroshima and Nagasaki is the main body of humans capable of providing estimates of the ef- fects of ionizing radiation on the incidence of genetic disorders. In November 1946, a presidential directive was issued at the request of the Secretary of the Navy, James T. Forrestal, giving authority to establish a Com- mittee on Atomic Bomb Casualties. The committee was formed in January 1947 (ABCC, 1975) and was the forerunner of the Atomic Bomb Casualty Commis- sion (ABCC), which was later transformed into the Radiation Effects Research Foundation (RERF). Its mission was "to undertake long range investigations of the effects on survivors of the bombs in Hiroshima and Nagasaki" (NRC, 1947~. The data structures and experimental designs used since the initiation of the genetic program of ABCC and RERF are described by Neel and Schull (1956a, b), Kato et al. (1966), Schull et al. (1981a), and Awa (1987~. The studies were divided into four substudies: _ \ 7 ~ 1. The clinical program, 1948-1954. This was a prospective study of the children of atomic bomb survivors and controls involving both questionnaires and physical examinations. Five endpoints were measured: sex ratio, congenital abnormalities, viability at birth, birth weight, and survival during the neonatal period. About 92% of the children were examined as neonates and 30% were reexamined at about 9 months of age. In addition, some 717 infants who were stillborn or died in the neonatal period were autopsied. The sample included 69,706 births, of which 12,401 were from parents who were proximally exposed (i.e., were within 2,000 m of the hypocenter at the time of the bombing tATB]~. 2. Fi mortality cohort, 1946-1985. In 1959, to increase the efficiency of the survival study, three cohorts were created from among the children born in the two cities since the bombings. The first cohort comprised all children born in the city where one or both of the parents were less than 2,000 m from the hy- pocenter ATB (proximally exposed). The second cohort comprised age-, sex-, and city-matched control births to parents who were more than 2,500 m from the hypocenter ATB (distally exposed). The third cohort comprised age-, sex-, and city-matched control births to parents who were not in the bombed cities ATB (not in city NICK. The proximal cohort contained 31,150 children, and the distal and the NIC groups numbered 41,066 children. These cohorts have been

OCR for page 29
GENETIC PRINCIPLES 37 followed through the years and form the basis not only of the Fat mortality study but the following studies as well. 3. Cytogenetic study, 1968-present. In a subset of the Fat cohort samples, X-chromosome anomalies and balanced structural rearrangements were looked for in the blood of the children of proximally exposed, distally exposed, and NIC parents. All children were at least 13 years of age when the samples were ob- tained. 4. Biochemical studies, 1975-1992. In a subset of the Fat mortality cohort, a direct search was made for new mutations by using a battery of 30 serum and erythrocyte proteins. Overall, eight health outcomes have been investigated: 1. Untoward pregnancy outcomes: congenital malformations, infant still- birth or death within the first 2 weeks after birth. 2. Fit mortality: death in children of exposed parents after 2 weeks, exclu- sive of cancer. 3. Malignancies in the Fat cohort: cancer arising in the children of survi- vors. Some cancers are the result of a combination of germinal and somatic mutations. Mutations induced by radiation might be detected by observing an increase in the incidence of such cancers. 4. Balanced structural rearrangements of chromosomes in children over age 13 years: because the children from whom samples were obtained had all reached at least the age of 13 years, only balanced rearrangements would be ex- pected. 5. Sex chromosome aneuploids in children over age 13 years: Individuals with the sex chromosome anomalies XXY, XYY, XO, and XXX are all viable, although some would not be expected to survive to age 13 years. 6. Mutations altering protein charge or function: this program centered on the detection of rare protein variants, in which case studies involving the family were carried out to determine whether the variant had been inherited or was the result of a mutation in the preceding generation. Collectively, 1,256,555 locus tests were done, and among these, seven apparent mutations were detected. Four of these occurred among the children of exposed parents and three occurred among the children of the controls. 7. Sex ratio in children of survivors: the proportion of male births among parents exposed to different amounts of ionizing radiation. 8. Growth and development of children of survivors: birth weight and weight, body length, head circumference, and chest circumference at 8-10 months of age. Overall, the studies of health outcomes in the Hiroshima-Nagasaki atomic bomb survivors and their children have revealed a small but statistically nonsig- nificant difference in health outcomes between the children of the atomic bomb survivors conceived subsequent to the bombing and the children of nonexposed

OCR for page 29
38 ADVERSE REPRODUCTIVE OUTCOMES parents. This increase, albeit small, is qualitatively and quantitatively consistent with the known mutagenicity of ionizing radiation in experimental animals and provides the best currently available basis for estimating the doubling dose from human data (Neel and Schull, 1991~. ESTIMATING TlIE DOUBLING DOSE To make use of Equation 1, it is necessary to estimate a genetic doubling dose. This has been done in a number of original studies and also by the various committees assigned the task of evaluating the genetic risks of radiation. The idea is straightforward. If one assumes that the relation between mutation rate and dose is linear, at least at low doses the model may be written as M = ml + m2 + bide + b2D2, (Equation 2) where 1 refers to males and 2 refers to females. Here me and m2 are the sponta- neous mutation rates in males and females, be and b2 are the induction rates in males and females, and Do and D2 are the doses applied to the two sexes. If applied to both sexes, the common dose would be M = 2(m~ + m2), therefore, the doubling dose (D), is ems + m2~/(b~ + b21. This is estimated by the regressing effect on the average dose to the two sexes, M = or + ED, and obtaining the doubling dose from the estimation equation DD = or / 0, where a is the intercept and ~ is regression. Mice Table 2 (Table 2-11 of BEIR V tNRC, 1990~) contains estimates of the doubling doses for chronic (low dose rate) ionizing radiation for a number of different endpoints. The ranges in parentheses were obtained by the BEIR V committee by multiplying acute doubling dose estimates by a correction factor range of 5-10. The figures in this table are based on a large number of studies of different sizes and reliabilities. The reader should refer to the original studies (references given in BEIR V) before making use of individual estimates. The overall median estimate is in the range of 1.0-1.14 Sv (100-114 rem) for chronic exposure. The median acute doubling dose estimate is about 0.30 Sv (30 rem). Humans The Japanese data have been used to estimate minimum and probable ge- netic doubling doses in humans (Neel et al., 1974, 1990; Neel and Lewis, 19901. Table 3, modified from Table 5 of Neel et al. (1990), contains the most re- cent estimates of minimum acute doubling dose on the basis of data from Hi

OCR for page 29
GENETIC PRINCIPLES 39 roshima-Nagasaki atomic bomb survivors. The numbers in the last three col- umns are the lower 99, 95, and 90 percent confidence limits of the doubling dose for five endpoints: untoward pregnancy outcome (UPO), Fat mortality, Fat cancer, sex chromosome aneuploids, and loci-encoding proteins. These lower 95 per- cent confidence limits range from 50 mSv (5 rem) to 2,270 mSv (227 rem). In addition, Neel et al. (1990) suggest a range of point estimates for acute doubling doses of from 1,690 mSv (169 rem) to 2,230 mSv (223 rem) and for chronic (low-dose-rate) ionizing radiation exposure of 3,380 mSv (338 rem) and 4,660 mSv (466 rem). TABLE 2. Estimated Doubling Doses for Chronic Radiation Exposure (primarily mouse) Genetic Endpoint and Sex Doubling Dose (rads)a Dominant lethal mutations, Both sexes Recessive lethal mutations, Both sexes Dominant visible mutations Male Skeletal Cataract Other Female Recessive visible mutations Postgonial, male Postgonial, female Gonial, male Reciprocal translocations Male Mouse Rhesus monkey Heritable translocations Male Female Congenital malformations Female, postgonial Male, postgonial Male, genial Aneuploidy (hyperhaploids) Female Preovulatory oocyte Less mature oocyte Median (mouse, all endpoints, both c Direct estimates Indirect estimates Overall 40-1 00 (150-300) (75-1 00) (200~00) 80 (40-1 60) 70-600 1 14 10-50 (20~0) (12-250) (50-1 on) (25-250) ( 125-1,250) (80-2,500) (15-250) (250-1,300) 70-80 (150) 100-1 14 " Values not in parentheses are based on the spontaneous rate divided by the induced rate/reds for the low dose rate; values in parentheses are based on the spontaneous rate divided by the induced rate/red at the high dose rate, multiplied by a factor of 5-10 to correct for the dose rate effect. It is important to note that these doubling dose estimates for humans and their lower limits refer to "conjoint" doubling doses, that is, the sum of the pa

OCR for page 29
40 AD VERSE REPRODUCTIVE OUTCOMES rental doses that is expected to double the genetic burden. The doubling dose estimate for the mouse given in Table 2 and the doubling dose calculated from Equation 1, as customarily used, refer to the common or average dose exposure to each of the two parents that is expected to double the genetic burden. For example, a common exposure of 10 mSv (1 rem) to each of the two parents cor- responds to a conjoint dose of 20 mSv (2 rem). Consequently, to compare the doubling doses for mice with the lower bounds and point estimates of doubling doses for humans, either divide the figures for human by 2 or multiply the fig- ures for mouse by 2. Both are perfectly valid doubling doses; they are simply scaled differently. Either can be used in Equation 1, so long as DD and D are used consistently. It appears that humans may be less sensitive to the mutagenic effects of ionizing radiation than mice. Neel and Lewis (1990) have recently attempted to resolve this difference by suggesting that, overall, the mouse estimates of dou- bling doses are actually higher than those suggested in Table 2. In sum, the general scientific consensus is that the overall doubling dose of mutation induction for low-LET, low-dose ionizing radiation is on the order of 100 rem, and it may, in fact, be larger.

OCR for page 29
- :> ~: a~ o .~ ct x LL1 a) m Ct C: C~ o ._ - ~S o a) Cd _I C~ a' - Cd C) C) 3 Cd _ C~ ~> ;~ o == ~ o =.- o .~ _ o ._ U) V) _ ~o o~ _ C~ ~ . o Ct ~ ~s o ~o Ct t_5 O ~ ,~ V] ~ D cd om Ct o\ O Ch o\ \ C~ _ O . . _ O 1 1 1 _ _ t_ _ _ ~ 00 0 ~ . . . . . 000- _ ,~ _ _ . . . O _ O I I I O _ ~0 0 . . . . . 000- 00 0 . . . ~ I 4 - ~ _~0 - ~ . . . . . 000 ~0 _ ~ ~_ ~ 00 - ] ~' O 0~gg g oOOo O O . . . +1 +1 +1 +1 +1 ~- = - o' o oo co - ~o co ~ o o o oooo o . . . ~ ~ oo c ~- ~o ~ - oo o oooo o oooo o . . . . . +1 +1 +1 +1 +1 ~ ~ oo ~- ~o ~ o ~o mooo o oooo o oooo o 1 1 ~ ~ o . . ~ ~ l oo g . . ~- ~ oo ~ o o o ~ - ~o '=oo o oooo o oooo o 04 ;^ ~.= . - o E ~ X . ~ o o C) 3 s~ =: Ct C) o ~C .= - ._ C,2 =^ o ~,o Ct Ct . - c: o .. C) o _ ~ o ~ ~ . O~ ~ P~ 3^ o s =, ~ s~ o ~ ~ o ~o