Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Metabolism of Arsenic PLANTS Among the many chemical combinations in which arsenic may exist, some readily enter plants and translocate, some in the symplast and some in the apoplast (nonliving cell-wall phase). Plants take up relatively small amounts of arsenic from soils in their natural state and exhibit no symptoms of arsenic injury. Around smelters, high-arsenic soils may be rendered completely sterile and bare of higher plants. Natural waters are usually low in arsenic, and plants reflect this in their arsenic contents. However, some natural waters are very high in arsenic and may thus cause problems for plants (as well as animals). Arsenic in the air in gaseous form has not been known to cause injury to plants. Particles from smelter fumes and smoke may settle out on plants; these may prove toxic to animals or to man, and they may harm plants through the soil. Absorption of Arsenic Compounds Arsenic rarely occurs naturally in the topsoil in quantities toxic to plants. Indeed, all soils contain stance arsenic, and plants have evolved 80
Metabolism of Arsenic 81 in its presence. In outcroppings of ores high in arsenic, small areas may contain arsenic in toxic quantities. Arsenic uptake by plants is of concern principally around smelters, where arsenic trioxide dust has spread over large areas, rendering the topsoil sterile. With the intro- duction of modern methods of arsenic recovery, cases of soil steriliza- tion by smelter dusts have practically disappeared. Sulfur dioxide, which often accompanies the arsenic trioxide in smelter effluent, has caused more widespread damage. The uptake of inorganic arsenic by plants affects the arsenic cycle. The arsenic concentrates in leaves, which fall to the ground. It is returned to the surface of the soil-at least in some organic, pentavalent form-after the leaves decay. Arsenic is known to be fixed in soils.837 Thus, what arsenic occurs in a mobile form is absorbed out of a deep soil profile, is deposited on the surface, is fixed to varying degrees, and, after oxidation, is subject to leaching by rainwater. Eventually, some of the arsenic originally present in the soil is returned by leaching and takes its original position in the soil and ground water. Some of it also moves in the ground water into springs and streams and is eventually carried into the ocean. Studies of arsenic fixation in soils have indicated that arsenic ab- sorbed by plants must be mainly in the soil solution. Thus, arsenic in water should be available for uptake by microorganisms, algae, and the roots of higher plants. Except for locations around smelters or where the natural arsenic content is unusually high, the arsenic taken up is distributed throughout the plant body in nontoxic amounts, e.g., arse- nic trioxide below 0.02% (dry weight) of foliage or 0.0003% (dry weight) of roots in the case of bindweeds (see Appendix A). These values are very high, compared with those reported by Woolson for the arsenic contents of a wide variety of plants~.O3~.5 ppm. Natural arsenic absorption from the air is negligible. Smelter fumes and dusts may deposit on plant leaves, but there is no evidence that arsenic from this source is taken into plants. Sulfur dioxide in the atmosphere, and more recently constituents of smog, may have de- leterious effects on plants, but not arsenic. Although arsenic may be present in volcanic emission and in some soils, it is rapidly oxidized in air to a particulate form that will not penetrate plant cuticle. Translocation of Arsenic Compounds The application of dilute sodium arsenite solution to bindweed foliage in 1917 resulted in the killing of root tissues and showed that it was possible for toxic compounds to be translocated in plants.306
82 ARSENIC Two vascular systems in plants are responsible for translocation: the xylem, which transports water and salts absorbed by roots from the soil to the top of a plant; and the phloem, which carries elaborated food materials from green to nongreen parts of the plant. Under conditions of high transpiration, the rate of intake of water by roots, even from an abundant supply, may fall so far behind the rate of evaporation from leaves that a reduced or subatmospheric pressure is built up not only within the xylem conduits, but within all living cells. Under these conditions, if the xylem is cut or broken under a solution, the liquid will be forced into the conduits until the internal pressure equals that of the atmosphere. If the plant is growing in soil in which the moisture is below field capacity, solution will be forced into the cut stem and move throughout the roots, the liquid originally present in them moving osmotically back into the soil. This is a reversal of the normal flow of the transpiration stream. If a solution of a strong acid is applied to the leaves so that they die rapidly, the sap that the cells contain can be forced into the xylem and down into the roots. If arsenic in sufficient quantity is included in such a solution, it too will be injected into the roots, where its toxic action is expressed, and the roots will die. This is the mechanism responsible for translocation of arsenic in the acid-arsenical method. When the tops of plants were cut off at the ground level under sodium arsenite solution, the roots were killed to depths of 7 It (2.1 m) and even more. And if normal plants were sprayed with the acid- arsenical solution and resprayed with water often enough to keep the foliage wet for an hour, depth of penetration of the arsenic was increased. On the basis of the anatomy and physiology of bindweed plants and their distribution in soils, it was explained that the translo- cated acid-arsenical spray could not be used for eradication, because roots at the edges of infestations have available water and hence do not absorb arsenic solution. i79 The authors stressed the influence of tem- perature, incident radiation, humidity, and air velocity on evaporation from leaves. They recommended delaying application until after dark. In years of deficient rainfall, when the soil is moistened to only 3 It (0.9 m) or less, the acid-arsenical method is not effective for killing bindweed, because arsenic does not penetrate below the current sea- son's roots, which are confined to the moist soil layer. The acid- arsenical method might prove useful in eliminating old deep-rooted perennial weeds before the application of a soil sterilization treatment. Sodium arsenite, long used as a general contact herbicide, was not considered to be translocated, probably because, as a contact treat- ment, it was used at a concentration that rapidly destroyed the foliage.
Metabolism of Arsenic 83 This prevented food movement with which arsenic movement is as- sociated. At lower concentrations, it was combined with acid to hasten penetration; this also rapidly killed the foliage. Rumberg et al.693 compared DSMA at 100 mg/ml with sodium arsenite (4 parts arsenic trioxide to 3 parts sodium hydroxide) at 50 mg/ml; both DSMA and sodium arsenite were labeled with arsenic-76. With soybean plants, they found evidence of translocation of sodium arsenite, but the organic form of arsenic, DSMA, proved to be more mobile. Sodium arsenite injury appears first as loss of turgidity, indicating possible effects on membrane integrity, whereas the organic arsenicals usually cause slowly developing chlorosis with little or no wilting; different enzyme systems may be affected. Rumb erg et al. 693 suggested that the rapid injury from sodium arsenite treatment may be responsible for the lesser transport; the arsenite usually produces injury symptoms within a few hours of treatment, whereas DSMA requires many hours or even days to produce chlorosis. Cacodylic acid is considered to be a general contact toxicant. It is applied in solution to cuts around the base of trees or on foliage. Apparently, its only translocation is apoplastic. DSMA and dodecyl (octyl~ammonium methanearsonate (amine methanearsonate, AMA- equal parts of tetraoctylammonium and tetradodecylammonium methanearsonates) were both shown by Long and Holt to be effective for controlling purple nutsedge ¢Cyperus rotundas) in Bermuda grass (Cynodon dactylon) turf.483 This result can be explained only by translocation from foliage to tubers, a movement that occurs via the phloem. In a later paper, Long, Allen, and Holt showed that those organic arsenicals kill nutsedge tubers if they are applied several times over a 2-year period.482 Long and Holt483 showed amine methanearso- nate to be somewhat superior to DSMA at equivalent rates of application-a result that seems logical in view of the mechanism of herbicide activation described by Crafts and Reiber.~° In further studies on purple nutsedge, Holt et al., 366 using single and repeated applications of amine methanearsonate to shoots of single tubers and shoots of terminal tubers on chains of tubers, found that arsenic was translocated laterally into tubers separated from the treated shoots by up to four tubers. The tuber at the opposite end of the chain from the treated shoots tended to have a higher arsenic content than the tubers between; translocated arsenic tended to be higher in tubers in which active growth was taking place. This is in accord with the source-to-sink nature of food movement; actively growing tubers are active sinks. There was no apparent relationship between the arsenic content of tubers and their ability to produce new shoots, and
84 ARSENIC there was a tendency for tubers to produce more than one shoot in the regrowth after treatment. Evidently, arsenic treatment altered the apical dominance in tubers that received arsenic. The writers con- cluded that death of tubers after repeated treatments was due to depletion of food reserves, rather than to the concentration of arsenic in the tubers. The variability in arsenic content in killed tubers and the variability in the number of treatments required to kill tubers suggested that failure to sprout is not related to the overall arsenic content of the tuber; some viable tubers contained more arsenic than some dead ones. Interruption of normal oxidative phosphorylation and exhaustion of the food supply resulting from increased sprouting may also have been involved in the ultimate death of tubers. In the repeated-application tests, the arsenic content decreased in tubers from which new rhizomes, tubers, and shoots developed. This indicated retransloca- tion, a property common to phosphate, amitrole (3-amino- 1,2,4- triazole), and dalapon (a sodium salt of 2,2-dichloropropionic acid) in plants. The high arsenic content of terminal tubers and the appearance of chlorosis in untreated shoots confirm this interpretation. Roots and newly developing shoots were not analyzed, and these might account for the loss of arsenic after the initial influx into a sprouting tuber. Translocation of the methanearsonates is indicated in many field studies. For example, Johnsongrass greenhouse studies by Sckerl et al. 7~9 showed rhizome deterioration when 12-in. (30.5-cm) plants were sprayed with DSMA. Kempen et al.4~9 showed that a single 6-lb/acre (6.7-kg/ha) treatment gave better rhizome and regrowth control than 3 lb/ acre (3.4 kg/ha). However, McWhorter found no evidence of rhizome deterioration in his many trials; he ascribed control to top-kill only.535 OnPaspalum conjugatum (sourgrass), Headford suggested that fail- ure of sequential treatments of DSMA to give good results may have been due to relatively inferior transport to the shoot apices.340 On purple nutsedge, Holt et al. 366 showed arsenic accumulation in tubers after foliar sprays. On hardstem bulrush, Hempen indicated that rhizomes showed symptoms. 4~8 Lange's trials on deciduous trees indicated translocation, in that symptoms appeared in new growth and in untreated growth on the trees (personal communication, 1968~. Studies on cotton by Ehman and Baket et al. indicated that arsenic derived from methanearsonates applied to the foliage may ultimately be present in cottonseed.46 229 Rumberg et al. did comparative translocation studies with t76As] DSMA and F76As~sodium arsenite on soybean at 85 F (29.4 C) and found translocation to be greater with DSMA.693 But he was able to recover only 30~0~o of the radioactivity from the DSMA in the foliage,
Metabolism of Arsenic 85 whereas 85% of the arsenite was recovered. He assumed that the DSMA not recovered may have gone into the roots, which he did not measure. With crabgrass, Rumberg et al.693 found translocation and toxicity with DSMA greater at 85 F (29.4 C) than at 60 F (15.6 C) on crab- grass and soybean; on the latter plant, movement was more basipetal (to the basal stem) than acropetal (to the opposite primary leaf and trifoliate leaf). The toxicity of sodium arsenite and cacodylic acid was not affected by the temperature difference. Sckerl and Frans indicated that t~4C]methanearsonic acid (~14C]MAA) was absorbed by Johnsongrass roots in significant amounts in 1 h and moved throughout the plant in 8 h.7~7 Spot applications of the acid formulation to foliage resulted in acropetal movement within 8 h and movement throughout the plant within 24 h. Duble et al. studied the translocation of two organic arsenicals in purple nutsedge.2~6 DSMA and AMA were the forms used, and tracers containing carbon-14 were applied to greenhouse plants. Chromato- graphic tests on extracts from t~4C]DsMA-treated plants indicated that the compound was not readily degraded; the i4C-As bond ap- peared to remain intact, although some t~4C]carbon dioxide was found several days after treatment. Comparison of the retardation factors (Rf) of plant extract-DsMA with that of standard [~4C]DsMA suggested that a plant extract-DsMA conjugate might have been formed; the values for extract and standard solution were 0.59 and 0.66, re- spectively; only one spot was found in each case. Over 85% of the material applied to the plant remained in the treated shoots. DSMA moved both acropetally and basipetally in single leaves, and such movement was not influenced by relative age of the leaf. The writers reported that both DSMA and AMA are moved symplastically and apoplastically a property that they share with amitrole. Carbon- 14 distribution in an untreated shoot appeared to be very similar to that reported by Andersen~5 for amitrole distribution in the same plant. It seems apparent from those results that translocation in nutsedge follows a source-to-sink pattern and that the amount of arsenic moved into a tuber depends very much on the sink activity of that tuber. Duble et al. found that actively growing terminal tubers in a chain accumu- lated arsenic, whereas intermediate and dormant tubers did not.2~6 Thus, the arsenic content of a tuber may not serve as an index of the lethality of a treatment; the effects of the initial impact of arsenic on later growth activity may be the critical factor in lethality and continu- ing transport of arsenic to growing roots. Other tubers and shoots may mask the effects of the arsenic content, as determined by analysis at any time.
86 ARSENIC McWhorter found no evidence of translocation of DSMA in Johnson- grass in field and greenhouse tests.535 Sckerl and Frans, by contrast, found that FI4C]MAA was both xylem- and phloem-mobile in Johnson- grass and cotton.7~8 Root uptake of this labeled arsenical by Johnson- grass from nutrient solution was rapid, and translocation into all portions of the plant took place within 4 h. Apoplastic movement in both plants was more rapid than symplastic movement; symplastic movement was more rapid in Johnsongrass than in cotton. 7i 6 As Duble et al. found, chromatography of extracts from Johnsongrass revealed values that differed from those of treated plants and standard EI4C]MAA solutions; Sckerl and Frans suggested complexing with sugars, organic acids, or both. When amino acid fractions were prepared from methanol extracts of both plants, an MAA metabolite with a positive ninhydrin reaction was found in the Johnsongrass fraction. Comparing Rf values, the authors suggested a complex with histidine or one of its analogues. Amino acid accumulation was noted in Johnsongrass as a result of MAA treatment, and the authors suggested that the MAA metabolite may block protein synthesis or some other biosynthetic pathway. Wilkinson and Hardcastle produced radioactive arsenic (arsenic -76) by neutron activation and determined concentrations of this radioisotope in new (unsprayed) and old (MsMA-sprayed) cotton leaves and in the soil from beneath the plants; MSMA treatment rate was 2.24 kg/ha.86, By using long counting periods, they were able to extend the lower limit of detection to 5 ng of arsenic. Table 4-1 presents the results TABLE 4-1 Arsenic Content of Leaf and Soil Samples Taken from Field-Grown Cotton Treated with MSMA at 2.24 kg/haa Arsenic Content, ,ug/g New Leaves Old Leaves No. Treatments (Unsprayed) (Sprayed) Soil 0 0.05 0.17 3.75 1 0.13 0.45 8.25 2 0.95 1.05 8.35 3 0.23 o.go 7.50 4 3.15 17.30 10. 15 5 4.80 41.10 4.00 6 10.30 30.25 1 1.20 "Derived from Wilkinson and Hardcastle.H6'
Metabolism of Arsenic 87 of the analyses. Differences in arsenic content of young untreated leaves from the two increments of MSMA application (i.e., one to three and four to six applications) may be due to dilution of arsenic content by continued growth. The increasing arsenic content in new unsprayed leaves also indicates the presence of translocatable arsenic from MSMA- treated leaves. Kempen has made a study of MSMA in plants, principally Johnson- grass, using both detached leaves and whole plants.4~7 He found that relatively high temperature, 35 C? and light, 2,800 ft-c (30,128 lx), produced SOYo necrosis of rhizomatous Johnsongrass foliage in less than a day, whereas lower temperature, 15 C, and light, 320 ft-c (3,443.2 lx), required 12 days to give the same results. Regrowth from rhizomes showed similar trends, indicating that the arsenic had trans- located into the rhizomes. Translocation of the label from leaf-applied ti4C]MsMA was primarily acropetal in the xylem, but small amounts also moved basipetally, proving that this organic arsenical is phloem-mobile. Within a week, the arsenic was transported from a treated mature leaf into the leaf base and sheath, to meristematic regions, and to roots and rhizomes; this indicates symplastic move- ment. Presence of the methanearsonate was detected by autoradiog- raphy and by counting. Sachs and Michael found cacodylic acid and MSMA to be transported about equally from the leaves to the terminal buds and expanding leaves of bean plants.695 There was no indication that either was demethylated or reduced to a trivalent compound. With ti4C]MsMA, Sachs and Michael found that about 40~o of the carbon-14 and arsenic recovered was bound to another molecule to form a ninhydrin-positive complex.695 Sckerl and Frans7~8 reported this ninhydrin-positive com- plex, as well as a sugar complex, and postulated that such a complex may block a specific biosynthetic pathway and that this blockage may account for the herbicidal activity. Duble et al. noted that DSMA was almost completely complexed with some plant component and that the complex was the mobile form of the herbicide.2~7 They also found that less than O.l~o of the carbon-14 applied as DSMA was metabolized and given off as volatile carbon-14 10 days after treatment. They con- cluded that the carbon-arsenic bond was stable in plants. However, von Endt et al. 827 found this carbon-arsenic bond to be rapidly broken in soils, which suggested that the bond is subject to attack by some biologic systems. In studies with ~ 4C-labeled MSMA, MAA, and DSMA, Keeley and Thullen found little translocation of these herbicides from cotyledons to developing leaves of co.tton seedlings.4~4 An exception was noted
88 ARSENIC where [14C]MsMA was applied at 13 C; there was little contact injury of the treated cotyledons, and the terminal leaves were well labeled. With the same three herbicides in studies on purple and yellow nutsedge, Keeley and Thullen found the yellow species to be more susceptible; the yellow species absorbed and translocated more of the ~4C-labeled tracers than did the purple.4~34~5 In chromatography of plant extracts and standards, there was less than 5% variation; nut- sedge plants did not readily metabolize these arsenicals in 72 h. Surfactants have been found to increase the penetration and translo- cation of the organic arsenical herbicides. The principal manufacturer of the methanearsonates formulates mixtures containing tested surfac- tants. Apparently, anionic and nonionic surfactants are satisfactory with these materials. A summary of the above discussion is presented in Table 4-2. In almost all cases, the arsenicals are translocated either upward or downward. The rate and direction of movement vary according to plant species and chemical. The degree of response may be temperature- dependent. A metabolite may be formed in susceptible species, but the formation is apparently not necessary for efficacy. MICROORGANISMS The sources of arsenic available to animals and man are naturals (such as water, plants, and animal tissue) and synthetical (such as agricultural chemicals, industrial wastes, and drugs). The natural sources of arsenic are modified by bacteria, molds, and algae, and they will be discussed from this point of view. Bacteria It has recently been shown527 that Methanobacterium (MoH strain) can reduce and methylate arsenate to dimethylarsine (cacodyl hydride). The medium must contain methylcobalamin as a methyl donor and adenosine triphosphate in a hydrogen atmosphere. The formation of the alkylarsine was detected by its odor and by aerating into 2 N nitric acid to oxidize it to cacodylic acid. With t~4CH3]cobalamin and t74As~sodium arsenite, a double-tagged compound was isolated whose ratio of ti4C]methyl to arsenic-74 was 2.0: 1. When cacodylic acid was used as a substrate, the alkylarsine was formed without the need of a methyl donor. The presence of an excess of sodium arsenite inhibited the production of the alkylarsine, the reaction producing
Metabolism of Arsenic 89 methanearsonic acid. The use of formal oxidation numbers for ar- senic in Figure 3 of Wood's article796 can be misleading, in that there is no actual change in the valence of arsenic, as explained by Zingaro.892 It should be noted that dimethylarsine was not positively identified, because both cacodyl (tetramethyldiarsine) and cacodyl oxide, which have a strong garlic odor, could have been formed and would have the same methyl: arsenic ratio. The recent sugges- tion796 that such a reaction in stream sediments could be hazardous to fish is without foundation; methylarsine decomposes to methane- arsonic acid in the presence of oxygen, 658 and the latter compound is less toxic than sodium arsenite. The necessity for an anaerobic atmosphere greatly limits the possibil- ity of this reaction in soil. Such conditions may be present in the rumen of the cow, and there is evidence that both arsenate and arsenite can be methylated by rumen flora. The reverse procedure, the demethylation of MSMA in soil, has been reported.827 When t~4C]MsMA was added to loam, 1.7-10% of the compound was degraded, yielding t~4C]carbon dioxide and arsenates. This did not occur in soil sterilized by heat. A fungus and two actino- mycetes isolated from the soil degraded 3, 13, and To of MSMA added to a substrate. The bacterial species similarly isolated degraded 20~o of the MSMA to arsenate; this changes the arsenical to a more toxic form. Fungi and Molds As early as 1815, cases of arsenic poisoning were reported to have been caused by wallpaper containing such arsenic compounds as Scheele's green (cupric arsenite) and Schweinfurt green (copper acetoarsenite, Paris green). The mechanism was first thought to be the ingestion of particulate material from the paper; but, when poisoning occurred with fresh paper, that theory wars abandoned. Gmelin (cited in Challenger et al. }39) was the first to report that rooms where symptoms occurred had a garlic odor, and he ascribed it to a volatile arsenic compound produced by molds on damp arsenic-pigmented wallpaper. Numerous investigations attempting to identify the chemical nature of this volatile arsenical have been reviewed by Cha'lenger,~35 i36 whose own research established the chemical structure of the compound. Gosio used pure cultures of bacteria and fungi on a potato medium and found that, although no bacteria produced a garlic odor, a mold, Penicillium brevicaule (formerly Scopulariopsis brevicaulis), was very active. He analyzed the gas formed and concluded that it was an alkyl
at: no ~ - o o ~9 · CL c) ~ c) o ~ c) o o ) co o ~ cq: 'e ~ o ~ - - ~ ~ an ~ oo ~ ~ ~ ~ - ~ ~ ~ ~ 0a~c~------~D---CJ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Go o of ~. - E 3 I o X . ~U. ~Cat Cat c' 3 ~ 3 ~ ~ ~ o ~o ' o o ~I, - ·6 3 ~ . u, ~ ~ O ~ O O Ct ·C) ~ ~ ~ ~ ~ ~ ~ ·- ~ ~ ~ .= m~ z z" ~ o~ ~ ;:<D ~- a z z z z~ZZZZ z zz o ._ Ct o - Ct ._ C~ ._ - C~ ·~: C~ e v ° E ~5 e, O 0 c O u ~ 0 c a O ~ct ¢ E & ~ U ~[_ ~ ~ ~ ~ ~ ~ ~ ~ o ~ ~ ~ Z ~Z Z Z Z Z Z>> ZZ>> Z mm =^ ~ ~ ~ ~ ~ ~ ~ Z ~ ~ ~ ~ ~ ~ ~ X ~ ~ ~ ~^ U~ U ~U~ ~ U~ C~ Cq C~ U~ U~ C~ C~ U~ U~ >> z~> 90
~ ,r ~ ~ XID ON UP ~ ~ ~ ~ ~ ~ ~ ~ O ~ 00 ~ -} \0 ~ ~ ~ ~ 00 AN =~ ~ ~ ~ ~ 00 (~\ ~\ \0 ~0 ~ ~ ~ Cut 06 _ ~ ~.O Cut ~Ct C ~ ~ ° ~ ~O a,, O ~ ~ ~' E Amman ~ to A,= ~ u, co t- a ~ ~ % Z Z a a ~ >~ z Z Z a a ~ a o ~ ~ ~ ~ ~ a ~ ~ z z zz>> >> z z z z z z a ~ ~ ~ ~ o~ ~o-^ Pa z~X ~ z Zeus ~ z us us us cut cn ~ cut 0 ~ c' c, ~ ~ 0 ~Z>~> >>Z>~> ~z c~ U3 ct oo E ~ ~ e E c c O O ~9 D O E ~0 D ~O O ~ ~O ~ ·- O C¢ ~ ~ O ~ ~ ~ ~, V~ ~ ~ C ) ~ ~ ~ ._ Ct C) - ;^ . ~ - C~> ~0 Ct C ._ S 3 c~ ._ "o _ _ c 33, 0 .= G) u _ ·CL 11 D C ~ .^ ·- (L) O CD C: ~ - C ~ E . 11 ~ 3 . ,._ 0 ~ c~ ·~ ._ ;~ co 0 == ct ct ~ q) ._ ~ ~ .= :^ 11 X cFq c ~ ._ ;~. .o c c li 11 11 z 91
92 ARSENIC arsine (actually, trimethylarsine, called Gosio gas).300 Sodium cacody- late also produced garlic odor in cultures of S. brevicaulis134 and Monilia sitophila Saccardo. 6 0 In a series of studies begun in 1931, Challenger et al. identified the volatile substance produced from breadcrumb cultures of four strains of S. brevicaulis as trimethylarsine.~39 Bird et al. reported that trimethylarsine was produced from sodium arsenite, sodium methanearsonate, and sodium cacodylate.77 In 1932, Thom and Raper784 isolated from soil several strains of fungi that were active in producing trimethylarsine, including strains of Aspergillus, Fusarium, and Penicillia. They also found that the strains were active with a wide variety of arsenicals used commercially and suggested that any arsenical could probably be acted on by fungi. Several fungi isolated from sewages ~75 can reduce arsenic com- pounds to trimethylarsine (TMA), as identified by gas chromatography and mass spectroscopy. Candida humicola was the only organism producing TMA from arsenate and arsenite. C. humicola, Penicillium, and Gliocladium produced TMA from arsenates. It is apparent that a wide variety of fungi, particularly those found in soil, can methylate both organic and inorganic arsenic compounds to the highly volatile TMA, which could thereby be lost to the air. The environmental fate of TMA iS unknown, because there have been no studies to place this process in the normal arsenic cycle; it may play an important role where arsenic concentrations are high as a result of the use of arsenical pesticides. Algae The review of Vinogradov823 on marine organisms includes a table of the arsenic contents of the principal varieties of marine algae reported by workers from 1902 to 1948. The values vary from 0. l to 95.0 ppm, without apparent relation between a species and its arsenic content. The marine algae had higher contents than freshwater species. This difference may be due to the lower concentration of arsenic in fresh water; it was found in New Zealand that algae grown in lakes fed by hot springs with arsenic contents as high as 0.1 ppm had arsenic concentra- tions between 20 and 1,450 ppm on a dry-matter basis.448 These concentrations prevented the use of the dried algae for animal food, because the arsenic concentrations produced in edible organs were above tolerance. The relationship between the water concentration and the uptake of arsenic by algae has been studied experimentally in aquariums contain
Metabolism of Arsenic 93 ing several species of algae, fish, and soil. The addition of t74As~sodium arsenate to such a system resulted in high concentrations of arsenic in all species of algae within 2 h. The arsenic was easily washed off with dilute hydrochloric acid and was considered to be loosely held on the surface, rather than absorbed into the tissues.47 Isensee et al.383 studied the distribution of cacodylic acid and di- methylarsine in nontoxic concentrations in an ecosystem including algae, Daphnia magna, snails, and fish. They found (Tables 4-3 and 44) that the lower food-chain organisms, algae and Daphnia, accumu- lated more cacodylic acid and dimethylarsine than the snails and fish and that there was no buildup throughout the food chain. The gradual loss of cacodylic acid and dimethylarsine from the water phase of the ecosystem could be accounted for by the increasing mass of the growing algae. It was suggested that adsorption, and not biomagnif~ca- tion, was important in the distribution pattern, inasmuch as algae and Daphnia have larger surface: area ratios than snails and fish. Lunde498 has shown that absorption takes place when marine and fresh-water algae are grown in aquariums in the presence of trivalent and pentavalent inorganic arsenic. Algae of both marine and fresh- water origin can synthesize both fat-soluble and water-soluble organic arsenic compounds, as shown in Table 4-5. It was suggested that algae TABLE 4-3* Tissue Content (Ppm) of ~4C-Cacodylic Acid in Algae, Daphnia, Fish, and Snails after Various Exposure Times, Rates, and Treatmentsa Treatment Fish Fish w/o Snails Snails ppmb Algae'Daphnia' w Daphnia Daphnia w algae w/o algae 0.1 4.5 (45Y3 9 (39) 0 09 (0 9) 0.14 (1.4) 0.9 (9) 2.0 (20) 1.0 17.0(17)41.6(42) 0.36(0.4) D.92(0.9) 2.3 (2.3) 8.5 (68.5)t 10.0 71.4 (7)254.0 (25) 6.71 (0.7) 11.20 (1.1) 7.3 (0.7) 68.3 (6.8) aAverage of 3 replications. bSolution concentrations of '4C-labeled cacodylic acid. ''Samples taken after two-day exposure. dW Daphnia-fish placed in untreated solution containing CA-treated Daphnia. Without Daphnia, fish placed in CA-treated solution not containing Daphnia. All fish harvested after two days. CW Algae-snails placed in untreated solution containing CA-treated algae. Without algae, snails placed in CA-treated solution not containing algae. All snails harvested after 7 days. 'Bioaccumulation ratios given in parentheses. *Reprinted with permission from Isensee et al.383 tA typographic error in the original publication; should be 8.5 (8.5).
Id o o · - 4 - 'e au - · - c~ 50 'e · - cd · - ;^ o 'e ~ Id ~ . - 'e ~ - ~· - - ~ of o ~ o ~ ~ In c;, - c) 'e of * >l ~ - m '°- ~ .~ O O O O Cal . . A: - . C¢ V) O _ Ct at C: O C: O . O V' _ Cal Ct Cal Ct Cal Ct 04 U' Cal rat Ct _ - S At' +1 ~ +1 +1 c ~+1 _ . ~ O lo O ~ +1 ,~+1 O{~N _ O~ O 8_ o~-X X ° l_ ~ ~o _ _~ _t_ ', oo _oo o~ o~ o +1 +1+1 +1+I tl o \o~o - o- o- o ooo~ oo c`, o(~! o~ o tl tltl tltl tl ON ~o~C) ~cN _ ~ ~ _~ _ _ - o tl tl _ ax o o tl tl tl tl oo ~o _ ~_ ~, _ ~o o °. °. _ mn~ m~ m~ : C,,} ·_ ·_ Ce ~ ~ S: >` ~ ~o -= 0 Ct .~ .S ~ 94 - U. et - U) , C~ o ._ U) CO Ct ;^ Ct ~ - , 3 O ·= go au ~ f: C~ ~ o O cI4._ X ca 3 ~ - O ..C Cn U. .~ ~ O 00 Ct ~ S & _ C,, ~ C, ;^ ~ Ct ~ C~ C ~ ,u, O U) ~ _ ~ C~ ~ Ct 00 . Ct =~ ~ O o~o Ct ~D ~ ~ Ce ~ x 'e - o ·- `' .> c o l c) { 'e o :o (~ - ~ - c\S ~ o cts ~ - o o ~ e~ ~ _ ~ 3 o ~ E3 ~.~.o ~ m ~ m u, ~ .0 ~ ~ C~ - C} C, - o O ._ C'3 .= ·3 C: .c *
Metabolism of Arsenic 95 TABLE 4-5* Accumulation of Arsenica in Algae as Arseno Organic Compounds in the Lipid Phase and in the Aqueous Phase, Respectively Culturing Media Salt Water Fresh Water Lipid Aqueous Lipid Aqueous Phase Phase Phase Phase Phaeodactylum tricornutum 2900 2000 2800 1800 Chlorella ovalis 1600 1300 Chlorella pyrenoidosa 400 190 Oscillatoria rubescens 540 240 Skeletonema costatum 1100 710 aThe calculation is based on the ratio between organic bound As-74 in the lipid phase and inorganic As-74 in the medium, and correspondingly in the aqueous phase and in the medium. *Reprinted with permission from Lunde.498 are an important source of the organic arsenic compounds found in organisms higher in the food chain. The total arsenic in seaweed collected from the fjords of Norway, where there is minimal pollution, varies with the species from 0.15 to 109 ppm on a dry-matter basis,495 as shown in Tables 4-6 and 4-7. If the oil from seaweed is saponified, the major part of the arsenic is found in the unsaponified lipid, rather than in the fatty-acid fraction, as shown inTable4-8.49~ TABLE 4-6 Arsenic in Seaweed (Reine in Lofotenja Date of Ash, % of Arsenic Sample and Location Collection dry matter Content, ppm Pelvetia canaliculata, Reine March 1951 23.6 22 Pelvetia canaliculata, Reine June 1951 17.6 21 Fucus serrates, Reine March 1951 27.6 47 Fucus serrates, Reine June 1951 23.8 40 Fucus spiralis, Reine March 1951 25.3 34 Fucus spiralis, Reine June l95I 21.6 26 Fucus vesiculosus, ovre Reine March 1951 23.2 65 Fucus vesiculosus, ovre Reine June 1951 20.6 26 Laminaria digitata lamirca, Reine April 1952 36.2 73 aDerived from Lunde.495
96 TABLE 4-7 Arsenic in Seaweed (Trondheimsfjordja ARSENIC Sample and Location Date of Collection dry matter Ash, IS of Arsenic, Content, ppm Laminaria digitata lamina, Munkholmen October 1956 27.6 109 Laminaria digitata lamina, Flakk April 1957 38.0 107 Laminaria hyperbor~a lamina, Munkholmen February 1957 37.0 69 Laminaria hyp`~rb`'r~ a lamina, Munkholmen January 1962 36.9 55 Laminaria hyp`~rbor`~a stifles, Munkholmen June 1957 36.8 94 Asc`'phyllum mod`'sum, Flakk September 1968 19.9 22 Gigartina man~ill`'sa, Flakk April 1952 30.5 10 Rh<'dymenia palmata, Flakk March 1952 31.5 13 Fetus v`,si~ul`'sus, Flakk March 1952 20.0 39 Fucus v`~si~ul`'sus, Flakk September 1968 17.9 24 Fucus serrates, Flakk September 1968 21.9 28 Fucus spiralis, Flakk September 1968 24.8 15 Pelv`~tia canali~ulata, Flakk September 1968 23.7 15 aDenved from Lunge." Plankton The term "plankton" includes both phytoplankton and zooplankton. Both are primary sources of food for marine and freshwater animals and are the starting point for the biologic phase of the recycling of arsenic. In spite of its primary importance, there is little specific information about the metabolism of arsenic by marine or freshwater plankton. In 1942, Ellis et al., 233 who were interested in the source of arsenic in the oil of fresh-water fish, analyzed the zooplankton of a pond and found that the ether-extracted remainders contained arsenic at 3.0-51 ppm, and the oil fraction, 4.0-25.7 ppm. Dupree, 223 in 1960, treated ponds with sufficient sodium arsenite to give a water concentration of 4.2 ppm. This concentration decreased slowly, to 2.0 ppm at 30 days and 0.2-1.7 ppm at 78 days. The arsenic in the plankton rose from an initial value of 5.9-10.6 ppm to a peak of 6,955 ppm at 27 days and then fell to 2,172 ppm at 78 days. Ball and Hooper47 used sufficient t74Asisodium arsenite on a pond and in aquariums to be able to follow
Metabolism of Arsenic TABLE 4-8 Arsenic in Oil and Fatty Acids Extracted from Seaweeda Arsenic Content, ppm Sampleb Oil Fatty acid Laminaria digitata 221 36 Laminaria saccharina 155 7.5 Laminaria hyperborea 197 16 Ascophyllum nodosum (1968) 7.8 5.2 Ascophyllum nodosum(l969) 49 21 Fucus vesiculosus 35 5.1 Fucus serrates 27 6.1 Fucus spiralis 5.7 5.0 Pelvetia canaliculata 10.8 7.3 aDer~ved from Lunde.495 Samples collected off the west coast of Norway. 97 its concentration in a complete ecosystem of fish, water plants, plankton, and soil over an extended period. The initial water concen- tration, 200 psi/ml, declined at an exponential rate to 50 psi/ml during a 60-day period. The bottom soil concentration increased during this period from 0 to 1,900 pCi/g. The plants reached 1,100 pCi/g in 2 h. Leaves that dropped off early in the death of the plants retained their shape, but showed counts as high as 15,000 pCi/g. The low-arsenic tissue was probably lost early, leaving tissue having the highest reten- tivity for arsenic. The early rapid uptake by the leaves at a time when the soil concentration was low indicates that the leaves, not the roots, are the primary path of arsenic uptake. The plant chara was not killed, but did accumulate high concentrations of arsenic. Phytoplankton were reduced by 50~o in a day and recovered slowly over 4 days. Zoo- plankton were greatly reduced and recovered even more slowly. The fish showed little or no radioactivity, indicating that they take up arsenic from the water not directly, but mainly from arsenic-containing food. These results indicate that water plants take up arsenic from the water rapidly and that, when they die, their arsenic is recycled, except for that eaten by animals. It would be interesting to determine the chemical composition of the arsenic compounds involved at each step of this ecosystem.
98 MOLLUSKS AND CRUSTACEANS ARSENIC There are no reliable data on the arsenic content of freshwater mollusks and crustaceans, but it is generally considered to be much lower (by a factor of 10) than that in marine species, as stated by Vinogradov.823 It is interesting to compare the work reported between 1919 and 1933, as compiled by Vinogradov in Table 4-9, and later work of Costa and Da Fonseca, i67 Del Vecchio et al., 202 Coulson et al., i7i and Lunde,492 as shown in Table 4-10. Although there are wide variations between studies and within the same species in a single series, there is no evidence of a trend in arsenic concentration with time, nor are there significant geographic differences. The arsenic compounds present in mollusks and crustaceans have never been characterized chemically, but studies in which shrimp were fed to rats and humans clearly indicated that the compounds are less toxic than arsenic trioxide and, although absorbed from the gastroin- testinal tract, rapidly excreted in the urine in both rats and humans. Coulson et al. i7i were probably correct in postulating that the arsenic was found in a complex organic molecule, whose characteristics de- serve study by modern biochemical methods. Concern over the high arsenic content of seafood makes such study imperative. -TABLE 4-9 Arsenic in Shelled Mollusksa Arsenic Concentration, ppm Organism (wet wt) Reference Date Oyster 4.1 143 1926 Oyster 3.7 176 1925 Oyster 2.0 853 1933 Oyster 1.1 356 1919 Oyster 34.0 143 1926 Mussel 68.0 143 1926 Cockle 19.0 143 1926 Whelk 18.0 143 1926 Sea snail 21.0 143 1926 Land snail 0.3 143 1926 Softshell clam 2.0 853 1933 aData from Vinogradov.823
Metabolism of Arsenic TABLE 4-10 99 Arsenic in Mollusks and Crustaceans Arsenic Concentration, Organism ppm (wet wt) Reference and Date Location Gulf shrimp 1.94 171, 1935 Texas Bay shrimp 2.44 171. 1935 Texas Bay shrimp 15.10 171, 1935 Georgia Bay shrimp 9.10 171. 1935 Georgia Bay shrimp 1.27 171, 1935 Alabama Bay shrimp 18.80 171, 1935 Louisiana Bay shrimp 3.83 171, 1935 S. Carolina Deep-sea shrimp 17.30 171, 1935 S. Carolina Deep-sea shrimp 36.60 171. 1935 S. Carolina Deep-sea shrimp 41.60 171, 1935 S. Carolina Deep-sea shrimp 5.94 171, 1935 S. Carolina Deep-sea shrimp 15.40 171. 1935 S. Carolina Deep-sea shrimp 30.70 171, 1935 S. Carolina Cockle 3.7-6.6 167, 1967 Portugal Cockle 1.3-2.4 167, 1967 Portugal Sea snail 3.6-63 167, 1967 Portugal Rockshell 14.6-26.4 167, 1967 Portugal Rockshell 1.8-3.7 167. 1967 Portugal Cuttlefish 6.2-11.5 167. 1967 Portugal Mussel 0.7-2.8 167, 1967 Portugal Oyster 1.2-3.6 167. 1967 Portugal Octopus 2.6-40.3 167, 1967 Portugal Shrimp 4.4-19.6 167, 1967 Portugal Shrimp 1.3-38.2 167, 1967 Portugal Crab 2.7-7.0 167, 1967 Portugal Crab 2.1-33.4 167, 1967 Portugal Lobster 12.0-54.5 167. 1967 Portugal Lobster 7.9-19.4 167, 1967 Portugal Lobster 10.8-17.2 167. 1967 Portugal Mussel 3.0-3.3 202. 1962 Firmer Italy Mussel 2.6-2.9 202' 1962 Civitavecchia. Italy Mussel 3.5-3.7 202, 1962 Taranto. Italy Lobster 2.2 492, 1970 Norway Mussel 8.0 492, 1970 Norway Clam 11.6 492, 1970 Norway Oyster 7.6 492, 1970 Norway Squid 6.5 492, 1970 Norway
100 FISH ARSENIC The presence of relatively high arsenic concentrations in marine fish has been known since 1875, and commissions were set up in England and Sweden in 1900 to study reported cases of arsenic poisoning due to the eating of fish. Vinogradov823 summarized the early reports and included data on both marine and freshwater species. There was no clear difference between marine and freshwater species, with respect to arsenic concentration in the entire fish it varied between 0.2 and 3.6 ppm. However, the arsenic content of the liver oils was much higher in freshwater than in marine fish. Ellis et al.233 analyzed 15 species of freshwater fish and found that the arsenic concentration of the whole fish varied from 0.30 to 1.3 ppm, whereas that of the liver oils was 0.9-101.0 ppm. He attributed these concentrations to the arsenic in the diet of these fish, inasmuch as he found that the livers of amphipods, isopods, and crustaceans had total arsenic con- centrations of 3.0-5.0 ppm and arsenic at 4.5-26.0 ppm in their oils. Lovern486 had shown that the oil of freshwater fish had a composition that resembled the lipid fraction of the zooplankton on which the fish fed. Sadolin697 in 1928 found total arsenic concentrations in codfish and herring of 1.3 and 0.4~.8 ppm, respectively. He found liver to have a higher arsenic concentration than muscle, owing to its larger oil con- tent. The oil that had an arsenic content of 3.0~.0 ppm was studied extensively by him, but he was unable to isolate the compound to which the arsenic was bound. Although fish-oil concentrations are high, most of the total arsenic in fish is in the muscle. Lunde492 has studied the water extract (N-liquor) and pressed cake of fish muscle and found that the water fraction had most of the arsenic and selenium (Table 4-1 11. Attempts to fractionate the N-liquor with Sephadex did not yield conclusive results, but indicated that more than one organic arsenic compound was present with a molecular weight below 5,000. He had previously shown that the arsenic was tightly bound and did not exchange with either t74As~arse- nate or t74As~arsenite.500 Lunde49~ has attempted to isolate the lipid fraction containing the lipid-soluble organic arsenic compound in the oil of fish and other marine animals. He found that there are two types of compounds, an arsenic-containing acid that follows the fatty acids during saponifi- cation and a type that is converted to a water-soluble compound by this process. More recently, Lunde496 has analyzed fish from various seas around Norway for organic and inorganic arsenic. He distilled the
Metabolism of Arsenic TABLE 4-11 Arsenic and Selenium in Dehydrated Raw Material and N-Liquor of Marine Organisms 101 Raw Material N-Liquor Arsenic, Selenium, Arsenic, Selenium, Sample ppm ppm ppm ppm Cod fillet 2.2 1.2 13 2.6 Cod bone 0.9 0.7 11 1.6 Cod liver 9.8 3.7 37 4.6 Cod skin 3.5 8.6 6.1 1.4 Hemng fillet 3.8 1.0 24 2.4 Hemng bone l.9 0.8 n.d. n.d. Hemng skin 7.2 0.8 n.d. n.d. Mackerel fillet (mature) 3.5 1.3 n.d. 2.9 Mackerel fillet (immature) 3.0 1.4 17 3.9 Norway haddock fillet 3.3 1.5 23 4.5 Lobster fillet 5.3 1.5 14 2.9 Mussel 8.0 3.9 9.7 1.1 Clam 11.6 2.6 18 1.0 Oyster 7.6 n.d. 9.8 0.9 Squid 6.5 3.0 17 0.6 Whale meat 0.36 0.5 0.9 0.4 aDenved from Lunde.492 n.d. = no detectable amount. inorganic arsenic from 6.6 N hydrochloric acid as arsenic bichloride and considered the remainder to be bound to organic molecules. He also recognized that the arsenic bichloride could in part have come from the breakdown of easily decomposed organic arsenic compounds. As shown in Table 4-12, most of the arsenic in marine fish is present in the organic form. When freshwater ponds are treated with sodium arsenite to control water weeds, the arsenic concentrations in the water are reflected in the fish. Gilderhus289 compared water and fish concentrations; the re- sults are shown in Table 4-13. BIRDS The diet of birds, excluding predators, is composed of seeds, grass, fruit, and insects. The concentration of arsenic in the diet is quite low-around 0.1 ppm, according to Schroeder and Balassa.709 With a
in - lo; 'e v) · - 'e c) · - in in · - 'e 04 hi o 'e o m ._ Ct 0£ o m 3 5~ o - c~ lo: A: ._ U. Cal C) 1 (V o _ m ~ Cal C) ._ 0 Cal - - c~ o Cal O ~ ~ ON ~ ~ ~ at ~ ~ aN _ ~ ~ ~ c: ~ ~ X ~ ~ ~ ~ ~ ~ ~ \0 em Cal 0 at 0 0 at ~ Do Go ~ ~ 0 00 ~ ~ ~ Us ~ ~ at Go _ ~ _ ~ _ -} ~ Cal oo - 0N ~ ~ r~ O ~ O ~ ~ ~ ~ oO ~ O - - o - - - ~ ~ o o o - - I _~ E°o Y 2 ., ;. ~ ~ ~ ~ ~ ~ y .c e · e ,,' D s~ .> .> ~ _ _ =, =" ~ ~.O E ~ ~, ~ ,3~, ~ ~ ~ 5 -~, ~ ~ E E ~ ~ 102
Metabolism of Arsenic TABLE 4-13 Arsenic Residues in Water, Bottom Soils, and Fish from Pools Treated for 8 Weeks in 1962a 103 Arsenic Application Rate Arsenic Residue after 8 Weeks, ppm in Herbicide, Pool ppm Water Fish Flesh Soil 1 2.31 yearly 1.01 0.38 92.1 0.69 yearly 0.056 0.35 37.3 3 0.23 yearly 0.024 1.02b 10.7 4 0.69 monthly 0.43 0.17 38.1 5 0.23 monthly 0.12 0.02 22.5 6 control n.d. n.d. n.d. 7 0.69 weekly 4.81 3.88b 44.9 8 0.23 weekly 0.98 0.786 36.7 9 0.023 weekly 0.12 0.09 6.5 aDenved from Gilderhus.289 Small fish. n.d. = no detectable amount. method that determines organic (methanearsonate) and inorganic arse- nic simultaneously, Lakso et al.447 found both Johnsongrass and cot- tonseed to contain methanearsonic acid at 0.05~.10 ppm and only traces of inorganic arsenic. The tissues and excrete of birds should reflect this intake, but no such studies have been made. Quantitative studies have been made to determine the arsenic resi- dues in birds that were fed growth-promoting drugs. In 1948, Ducoff220 injected t76As~sodium arsenite into chickens and other animals and found that it was excreted faster by the chicken than by any other animal. The chicken retained only To at 60 h, compared with 90~o for the rat. . The chemical nature of the arsenic normally found in bird tissues is not known, but analyses of eggshells with the helium-arc method suggest that the high concentrations of methanearsonic acid and cacodylic acid present must have been derived from the bird's tissues (R. S. Braman, personal communication, 1974) (Table 4-141. MAMMALS Forms of Arsenic in the Diet and Water Normally, the intake of arsenic from water is not significant, except in areas of abnormally high concentration, such as Fallon, Nevada, and
104 ARSENIC TABLE 4-14 Analysis of Various Samples for Methylarsenic Compoundsa Arsenic Concentration, ppb Sample Methanearsonic Acid Cacodylic Acid Chicken eggshell 1.3 4.8 Bobwhite eggshells C 7.3 14.3 Scrub jay eggshells 1.7 7.9 1.3 5.2 Seashell (unidentified) 1,300 7,000 Phosphate rock 0.005 0.005 Local source) aData from R. S. Braman (personal communication. Accuracy is within logo. Eggshells collected at the Archbold Research Station, Florida. CPhenlarsonic acid tentatively identified in this sample. dAlso contained As(III) at < 20 ppb and As(V) at 3,140 ppb. Lane County, Oregon.294 The arsenic compounds present presumably are pentavalent inorganic, but proof has not been reported. It would be interesting to determine the selenium: arsenic ratio in areas where the high arsenic concentrations appear to be innocuous to animals, be- cause these elements are known antagonists. Plants Arsenic concentrations in plant foods are presented in Chapter 3 and Appendix A. The finding that the arsenic in Johnsongrass and cotton- seed is almost entirely in the form of methanearsonate,447 a compound of low toxicity, should be considered in evaluating the importance of arsenic derived from plant sources. Animal Tissues The work of Coulson et al. i7i was discussed earlier. "Shrimp arsenic" is probably an organic compound of low molecular weight that is easily absorbed and rapidly excreted by rats and humans. A compound with similar properties is found in the liver of swine that are fed arsanilic acid.6~5 When the dried liver was fed to rats, 97% of the arsenic was excreted in 7 days, whereas only 30~o of the equivalent amount of arsenic fed as arsenic trioxide was excreted in the same time.
Metabolism of Arsenic 105 The normal sources of arsenic available to man and animals are probably complex organic molecules, whose nature offers a necessary and interesting field of research. The abnormal sources of arsenic that can enter the diet from plants or animals include arsenical pesticides, such as lead arsenate, arsenic acid, and sodium arsenite. Physiologic Aspects of Arsenic Absorption from the Gastrointestinal Tract Compounds that enter the gastrointestinal tract are subjected to the action of bacteria and enzymes and, after absorption into the portal venous system, must pass through the liver before reaching the general circulation. This process could alter the chemical form of such com- pounds before they reach other organs. It has recently been shown that arsenates and arsenites are altered in cows and dogs.446 After a control period, cows were fed sodium arsenate and potassium arsenite daily for 5 days and then fed the control diet for 7 days. Urine samples were collected twice a day and analyzed for inorganic arsenic and methanearsonates.629 As shown in Figure 4-1, the concentrations were low in the control period, in- creased during the feeding of the arsenicals, and returned to normal within 5-7 days after return to control feed. It is clear that more than So-so of both the trivalent and pentavalent inorganic arsenic was methylated. Because the rumen is anaerobic and is a site of great bacterial activity, it was thought that the action was similar to that of methanobacteria. However, when the experiment was repeated on dogs (Figure 4-2), a similar degree of methylation occurred, which cast doubt on bacterial action as the sole mechanism. Inasmuch as the process results in a less toxic compound of arsenic, it is truly detoxify- ing. The rapid excretion of the relatively large doses of arsenic suggests that no substantial portion of the added arsenic is retained. The mechanism of intestinal absorption of organic arsenicals in rats has been studied.378 Solutions of carbarsone, tryparsamide, and sodium cacodylate were injected into isolated loops of small intestine in anesthetized rats, and the arsenic remaining in the loops was determined periodically. The results indicated that the process was simple diffusion and was not an active transport mechanism. The rate of diffusion was not related to the molecular size of those compounds, and, inasmuch as the compounds had a high trichloromethane: water ratio, they probably passed through the lipid portion of the cellular membranes.
i o z - - o ~ ~ (, A ~ z ~ ~ z ~ lo o z ~ o co ~ 11 1 . 1 . o us o ~ - - o - - ~ --~ ,~ ~--~L f~ - - - - .~ NOIJ3IW d3d SldUd OI" O ~ O ,` I ~ <:E O o < ~ ~ Ct ~^ ~ ~- --~ '`) ~ 1,,,,,,, , 1,,,,,,, , · 1 ` O ~O o O O o C~ O ~ z 0 CL. x 1 ~_ ~, t ~_ O O ° O O ~_ 1 1 1 I`J - x tD o ~r ~ C) _ n. _ C~ _ _ _ o o NOI13IW d3d SldUd ~ 106 ~ ~o q ~ · - .= 3 C~ '_ O .< ~ Cd 4_ c: ~ ._ ~o ~ a, ~: CO Ct ._ C~ Cd ._ o Ct Ct ~: C~ o U) o - Ct - o os C~ I C~ ~ ._ S~ ~ 0 ~ .= , ~a ~ ~ 0 e ^= C) Ct ~ ~ so 0 0 = C', ~ ·= £ o o Ct ~ o Ct U o ~ ·~3 O ~ ~ ,-, 4 - . - i_ O ~ os O _ . - _ - . ._ 04 ~ oo Cd 3 _ ._ . C~ C~ - Ct ~ o Cd _ . a' U, et O O o
gore ~m +- ~ 45~<and Hi- A) Ott~ ~ _ ~ ~'a ~ _ _ 1 , 1 . 1.,,, , - _ o ~° o o ° o O _ I, _~ ~ ~ A Z ~',, C) ~ LO ~ ,,,, Z ~\ \ ~ 8 To - A' in CE WAS ~ t ~ N013~ IS H3d Sled a': _~ 1,,.,,,,, I,,,'~l,,,,,,,,_ o ~°. ~o o o ° Z ~n ~ 0 Cl. {D o ~n c o NOI1lIW H3d Sl~Vd ~ 107 U) ~ ~ c 3- . o ~Z ·c ~ ~: ~, ~,.. Ct . ·- C) _ · ~ o _ oo C.5 s~ ~ o -. a~ o Ct ~ ~ Ct - et t8= o O ·O C`, ._ Ct Ct =.O ~ u°, Ck O _ Cq, · _ ~ ~ C`3 ._ _ ._ .O ~ 0 UO' ~ ·~ _ ~ Ct Ct ~ ~ cl,, ~ 0 ~ O C) ~ oo ~ .o ~ ~ . 5, ~ ^ _ ~ ~ ~ ° _` . ~ · ·U3 _ ~ U, O Ct a~ ~ ~ Ct 04 P~ ~ 0 ~ U) - ° 6 ~ - n~
108 ARSENIC Distribution of Arsenic in the Body The most complete study of the distribution of arsenic in animals and man resulted from studies with radioactive isotopes of arsenic. In 1942 Hunter et al. 376 injected [74AsUpotassium arsenite subcutane- ously into rats, guinea pigs, rabbits, chimpanzees, a baboon, and leukemic humans and analyzed their tissues and body fluids. With the exception of the rat, in which the arsenic concentrates in the red l food cells, the arsenic is generally distributed to all tissues, with the largest total amount going to the muscles. Most of the arsenic was excreted by the kidneys; excretion was essentially complete in 6 days, with only a trace appearing in the feces. Arsenic does not pass into the spinal fluid of humans, but small amounts were found in apes. Lowry et al. 488 fractionated various tissues from this experiment and studied the acid-soluble lipid and protein fractions for arsenic and phosphorus content. They also separated liver protein into different fractions, including nucleoprotein. Most of the arsenic was in the protein fraction, a small amount in the acid-soluble portion, and a trace in the lipid fraction. The nucleoproteins did not take up more arsenic than other proteins, and there was no evidence that phosphorus was displaced by arsenic in tissues. Duc off et al. 220 used t76As~sodium arsenite on rats, rabbits, mice, and man, studying their excretion rates and tissue distribution patterns. The rat retained most of the arsenic in the red blood cells, with smaller concentrations in the spleen, heart, lungs, kidneys, and liver. In rabbits, the arsenic was lowest in the blood and highest in liver, kidneys, and lungs. The most interesting results are shown in Figures 4-3 and 4-4, which compare the relative excretion rates in rat, man, and rabbit and the blood concentrations of arsenic in rat, man, rabbit, and chicken. Lanz et al.452 studied the absorption, distribution, and excretion of arsenic-74 injected intramuscularly in the rat, dog, cat, chick, guinea pig, and rabbit. The accumulation of arsenic in the tissues after 48 h showed that less than 0.27% was stored in the organs studied in all species except the rat and cat, which stored 79~O and 5.6%, respectively, in the blood. They studied the distribution of arsenic in rat blood and found that none was bound in the plasma proteins, a trace in cellular ``ghosts," and most of it in hemoglobin from which it could not be removed by dialysis. Arsenic was rapidly excreted in the urine as inorganic compounds, with lO-15% of the arsenate being reduced to the trivalent form. Their analytic method required the precipitation of
Metabolism of Arsenic 111 O 100 c, L1J ~ 80 A LO o Z 60 c: 111 u' z 40 o - LL x 20 UJ - Rabbit - ! l l - - _ t / ! _- _ Patient GR _ ._ _ ~ - -I ~~ ~ Katie, - Rats 0 24 48 72 ]6 HOU RS AFTE R I NJECTION FIGURE 4-3 Excretion of arsenic-76. Cumulative excretion from the time of injection is expressed as percentage of the adminis- tered dose. Reprinted with permission from Ducoff et al. 220 109 the arsenate as magnesium ammonium arsenate; the filtrate was as- sumed to be trivalent arsenic (which is highly questionable). Mealey et al. 536 studied the distribution and turnover of radioactive arsenic after intravenous injection into man. The isotope used, arsenic-74, was 90% in the trivalent form and was given as the sodium salt in a dosage of 2.3 mCi/70 kg. The different rates of urinary excretion indicated that the arsenic was distributed into three com- partments, as shown in Figure 4-5. Clearance rates for each were calculated as the percentage of the total dose per hour, giving 25%/h for compartment I, 2.5%/h for II, and 0.3%/h for III. The proportion of pentavalent arsenic in the urine showed a steady increase until the fourth day, when it remained constant at 75%.
110 ARSENIC 100 L So 20 1 0 a 2 At LO 1 1 _ ~ _~ _ RAT win 4c ~-\ MAN \ \ N"~" ~ " RABBIT '\ " _ HICKEN 24, 48 72 96 TIME IN HOURS FIGURE 44 Arsenic-76 in whole blood. Concen- tration of arsenic-76 per gram of blood at a particular time is expressed as percentage of the administered dose per gram of body weight. Reprinted with per- mission from Ducoff et al. 220 The low rate of excretion of arsenic in the rat is probably due to fixation of 80-90~o of it in the hemoglobin of the red cell, which must break down before the arsenic can be released. The high content in blood also makes it difficult to get true tissue values in such organs as the spleen and liver. In an attempt to find another small-animal model for man, Peoples625 studied the distribution of arsenic in rats, guinea pigs, rabbits, and hamsters that were fed arsenic trioxide; the results are shown in Table 4-15. The hamster appears the animal of choice, with the rabbit also a good possibility. The relationship between arsenic ingestion and urinary arsenic has led to its use in measuring industrial exposure. The importance of /
Metabolism of Arsenic 111 checking the diet for the intake of seafood has been reported.708 It is interesting that a wide variety of fish, crustaceans, and mollusks caused increases of up to 10 times the normal content lasting for over 20 h. No symptoms were reported at concentrations that would have caused serious poisoning if the substance ingested had been sodium arsenate. The chemical nature of the arsenic compound is unknown, but it is probably organic. With the helium-arc method, 99 the urine of four subjects was analyzed for As(III), As(V), methanearsonic acid, and cacodylic acid; the results are shown in Table 4-16. The average excretion of 66~o of the arsenic as DMA is striking, especially because the subjects were from Florida and likely to eat seafood. The relationship of industrial exposure to seafood intake should be analyzed with this method. The distribution of arsenic in the cow has been studied with a variety of compounds in nontoxic doses.625 The feeding of arsenic acid to cows at 0.05-1.25 mg/l~g of body weight for 8 weeks did not increase the arsenic content of the milk, but it did increase the arsenic in the tissues. Urine was the main pathway of excretion, which was very rapid-the urine was free of arsenic 2 days after arsenic administration was 100 10 In o ~5 TO - UJ ~ 1.0 at o LL a: c' o.1 x .- L.L. -J.G. v -R.P. O -E.H. O - D.D. v o _ art v 0 0 0 0 0 0 0 - _ 0 01 ) 20 40 60 80 100 120 140 160 180 200 220 240 260 280 TIME AFTER INJECTION (hours) 0 0 0 · 0 1 1 1 1 ~ I 1 1 1 FIGURE 4-5 Rate of urinary excretion of radioarsenic in five subjects. Reprinted with permission from Mealey et al. 536
o · - x o . ~ v · - o~ · - ·'e 4 - s: o · · au o c~ ;^ cd - s~ u, a' c~ c~ . - · - v · - c~ - o . - 'e - c) c~ .o u, o o o o o - ~ o ~ o ~ u~ o ~ o - o ~ m _ <-, .= ~ o. o. 1 o. ~ °. °. V) O ~ O O ~ 0 X 4 - Cd LL4 L~ Ct - ~ 0 0 0 - ~ ~ O . .. .. .. . O '}- ~O -- O _ ~ ~: ·Ct ~0000 0 000 _.... . ... - au - ~ O O O O r ~oo ~ . . . . . . . . O r~ O ~ O O - ~D o O oo 0 ~ O ~ O ~ O O O O O O _ ~ ~ 0 0 0 0 ~O O V) O O O ~ O O O ~ O O ._ - ~O - O - O ~ ~ 0 0 0 0 ~O O . . ~ ~O O O O O ; - ~ O O O O O 0 0 0 .> O O O - O - O _ O O O O ~. ~ 4_ ~ O ~ O ~ O ~ O ~ o~ o~ o~ o -0 .s 3 ~ X 112 ·~3 ~D U, C) _ . s~ o ~ C) o os o ~ C) . c) o
Metabolism of Arsenic TABLE 4-16 Analysis of Human Urine Samples for Arsenic Compoundsa 113 Methane arsonic Caca As(lII) As(V) Acid Acid Total, Subjectppb 56 ppb To ppb 5! ppb To ppb Ma age 28<0.1 0.84 8.1 0.61 5.9 8.9 86.5 10.4 M age 275.1 20 7.9 30 2.5 9.7 10.4 40.2 25.9 M, age 42<0.5 - 2.4 7.8 2.4 8.1 25.2 84.0 30.0 F. age402.4 10 4.3 18 1.8 7.6 15.5 64.5 24.0 Average1.9 - 3.9 - 1.8 - 15.0 22.6 aDenved from Braman.~9 The results are given in parts per billion (ppb) as arsenic. The precision of individual runs is + logo relative. or +0.1 ppb for small sample sizes. Four other samples not completely analyzed gave: methanearsonic acid. 3.6 + 2.4 (S.D.) ppb; and cacodylic acid, 15.5 + 6.8 (S.D.) ppb. discontinued. The valence of urinary arsenic was determined; pentava- lent arsenic was the only form found. A similar study with sodium arsenite and cacodylic acid yielded essentially the same results (Peoples, unpublished data). When MSMA was fed to lactating cows, the blood concentration of arsenic rose, but the milk concentration remained low.627 These results support the conclusion that there is a blood-mammary barrier to arsenic. The most likely explanation is that an active trans- port mechanism is saturated at normal plasma concentrations. Alipha- tic organic and inorganic arsenic must use the same transport mecha- nism. Toxic doses of arsenic do not break down the barrier to arsenic. Marshall gave Jersey heifers lead arsenate in their feed at 12.95 mg/lOO lb (0.285 mg/kg) of body weight for 126 days and noted no values in the milk above the control value of 0.05 ppm. The same results were obtained by Fitch,254 who fed a heifer 0.3428 g of arsenic trioxide daily for 3 days without increasing the arsenic content of the milk. In these acute studies, the skin and hair were found to have low concentrations, indicating slow uptake. Dubois et al.2~9 showed that exposure to arsenic in atmospheric dust can give fallacious results up to 243 ppm. Washing in detergent reduced all values to 3.0 + l .O ppm. Lander et al. 449 examined hair in arsenic poisoning cases and found a wide range of concentrations, 3.0-26.0 ppm. These values were reached a short time after exposure, and the concentration in hair near the scalp was often no higher than that in hair near the tip. The
114 ARSENIC difference was thought to be due to contamination from sweat. The same situation was found in the nails. Perkons and Jervis632 used neutron-activation analysis on human hair and reported the frequency distribution of 12 elements, including arsenic, as shown in Figure 4-6. The range of arsenic concentration of 1-5.5 ppm is in accord with the concentrations in washed hair found by Dubois et al. 2~9 Apparently, the value of hair as a reliable indicator of chronic arsenic poisoning is open to serious question (see Chapter 61. The valence of arsenic in water, food, and body tissues is largely unknown, not because of its lack of importance, but because of the technical difficulties involved. The method of Crawford and Storey,~84 in which ethyl xanthate is used to extract trivalent arsenic, has worked well in the hands of Ginsburg,29i who used it in renal clearance studies of arsenite and arsenate in dogs. Ginsburg found that arsenate is reabsorbed in the proximal renal tubule and is reduced in part to arsenite, which then appears in both urine and blood. This is the only evidence of a reduction in valence in a living animal. Winkler868 analyzed the livers of rats that had been fed sodium arsenite and found that most of the arsenic was pentavalent. Livers of rats fed sodium arsenate and sodium arsanilate contained only pentavalent arsenic. With Winkler's method, Peoples625 found only pentavalent arsenic in the urine of cows fed arsenic acid. The arsenic content of neoplasms has received scant attention, although arsenic is often listed as a carcinogen, particularly in books on dermatology (see Chapter 6~. Domonkos209 reviewed studies on the arsenic content of normal skin, skin that was pigmented by exposure to arsenic, and keratoses. Most of the reports indicated normal or low concentrations of arsenic in the lesions, compared with nearby normal skin. He used neutron activation to determine the arsenic in skin and epitheliomas in humans with and without a history of arsenic ingestion. The arsenic content of skin samples varied from 0.15 to 95 ppm when taken from the same patient, and that of four epitheliomas in one patient, from 0.22 to 4.1 ppm. The values for skin vary from 0.1 to 5.0 ppm, regardless of a history of arsenic ingestion. He concluded that arsenic determination in skin lesions was of no value. There are some data on other types of tumors. Ducoff et al.220 administered sodium arsenite to a person with a parotid tumor; the tumor took up less arsenic than the liver and kidney and the same amount as most of the other organs. Ducoff et al. also studied the uptake in mice inoculated with Jackson-Brues embryoma and lym- phoma. The concentrations in the tumors and organs were too variable
o at c ~ = CD Z m c cut 0 0 ED `ouanba`' 0 0 0 cut m ~ ° 1 L '''- 1 o A: o . Cat . as 3 so cut _ _ . l 0 0 0 . ~ 0 E NAZI i' r L o o o ~uanba` L 1 Cat o C to - o ~Couenbm~ 115 . . o o ~ Cat /ouanb~ ° C£ an _ ~ 0 0 ce u, . - ct ct at: 0 ._ C`7 au cTV cat Q_, .sO~ Cat o O ._ C~ ._ ~C) C~4 CO . C~ s_' ·- - =\ ~ ~ ~: O ~ ~ (L)
116 ARSENIC to show a significant trend. The liver: spleen: kidney arsenic ratios were different in normal and tumorous mice. Hunter et al.385 injected L74As~potassium arsenite into two leukemic humans and found that the tissue distribution resembled that in the guinea pig. There is no evidence from these data that neoplasms have any peculiar ability to store arsenic. Dermal Absorption of Cacodylic Acid in Man In 1972, the forestry workers in north central Washington applied MSMA and cacodylic acid into the cambium layers of trees.78~ The purpose was to thin the trees and thereby ensure a better stand. The technique resulted in wetting of the workers' clothing, and many developed a garlic breath odor. Urinary arsenic studies were made on a six-man crew over a period of 9 weeks, samples being taken on Monday and Friday of each week. Each man used a different proce- dure, as follows: (1) a control, hack squirt tool with (2) MSMA or (3) cacodylic acid, injection hatchet with (4) MSMA or (5) cacodylic acid, and (6) MSMA with injector tool. The difference between the two chemical means and the method means was not significant at the 5% level. The control was significantly lower than all chemical treatments, with Monday values being significantly lower than Friday values. The values plateaued or dropped during the study, indicating that the chemicals did not accumulate. There were no health problems related to the compounds used. Further studies were made by Wagner and Weswig on forestry workers who were exposed to cacodylic acid over a 2-month period.832 An attempt was made to correlate blood and urinary arsenic concentra- tions with degree of exposure without much success. The urinary values were markedly increased, but did not reach the heights of the previously mentioned study. They attributed the odor to arsine; it is highly unlikely that arsine caused the odor, because it is so toxic. When cacodylic acid was commonly used as a tonic, the odor on the breath was reported to be caused by tetramethyldiarsine (cacodyl).750 Suff~- cient amounts of arsine to give the breath an odor would be lethal. As would be expected, none of the workers complained of ill health, be- cause the amounts of cacodylic acid absorbed were less than the dose formerly prescribed as a drug, or 30 mg/day.
