Proc. NatL Acad. Sci. USA Vol. 79, pp. 879-883, February 1982 Genetics Evolution of alkaline phosphatases in primates (gene duplication/gene expression/placenta/lung/inhibition) DAVID J. GOLDSTEIN, CAPRICE ROGERS, AND HARRY HARRIS Department of Human Genetics, University of Pennsylvania School of Medicine, 195 Med Labs/G3, Philadelphia, Pennsylvania 19104 Contributed by Harry Harris, October 13, 1981 ABSTRACT Alkaline phosphatase [orthophosphoric-monoes- New World monkeys. Nor is an ALPase with the same inhi- ter phosphohydrolase (alkaline optimum), EC 3.1.3.1] in placenta, bition characteristics present in human lung. We have found, intestine, liver, kidney, bone, and lung from a variety of primate however, that human lung contains trace amounts ofan ALPase species has been characterized by quantitative inhibition, ther- with the characteristic features of human placental ALPase. It mostability, and immunological studies. Characteristic.human pla- represents on average about 3.5% oftotal lung ALPase activity. cental-type alkaline phosphatase occurs in placentas ofgreat apes These findings, taken as awhole, pose intriguing questions con- (chimpanzee and orangutan) but not in placentas of other pri- cerning gene duplication and the regulation ofgene expression mates, including gibbon. It is also present in trace amounts in hu- man lung but not in lung or other tissues of various Old and New in multilocus enzyme systems. World monkeys. However, a distinctive alkaline phosphatase re- sembling it occurs in substantial amounts in lungs from Old World MATERIALS AND METHODS monkeys but not New World monkeys. It appears that duplication Tissue samples from a variety ofprimates were obtained at au- of alkaline phosphatase genes and mutations of genetic elements topsy, except for placentas, which were collected at delivery. controlling their tissue expression have occurred relatively re- The specimens were stored at -200C until extracted. The var- cently in mammalian evolution. ious analytical methods used have been described: tissue ex- At least three gene loci code for the various forms of human traction (1), ALPase assay (1), thermostability studies (1, 8), in- alkaline phosphatase [ALPase; orthophosphoric-monoester hibition studies (5, 8), and Ouchterlony double-diffusion studies phosphohydrolase (alkaline optimum), EC 3.1.3.1]: one for the and staining ofprecipitin lines (10). Rabbit anti-human placental polymorphic placental forms, at least one for the intestinal forms ALPase was a gift from Clive A. Slaughter. In the studies to (adult and fetal), and at least one for the liver, bone, and kidney determine the characteristics of heat-stable ALPase in various forms (1-4). In a recent investigation ofALPases in various lab- tissues, the tissue extract was preheated at 650C for 1 hr and oratory and domestic animals, evidence for only two such loci, then centrifuged prior to analysis of the supernatant. When one coding for liver/bone/kidney ALPase and the other for in- heat-stable residual ALPase activity was very low, the super- testinal ALPase, was obtained (5). The ALPase expressed in natant obtained after heating and centrifugation was concen- placenta in these species was quite different from human pla- trated 10-fold or more by using Minicon concentrators (Amicon). cental ALPase and corresponded to the liver/bone/kidney ALPase in each species and also in man. This result led to the RESULTS hypothesis that the expression ofhuman placental-type ALPase Placental ALPase appeared relatively late in mammalian evolution, subsequent to the divergence of the evolutionary lineage leading to man Thermostability Studies. Human placental ALPase is re- from the lineages leading to these mammalian species. This markably thermostable. It may be heated at 650C for an hour hypothesis prompted the present study, in which we have ex- or more without loss ofactivity, in contrast to liver, bone, kid- amined ALPases from a variety of tissues in a diverse series of ney, and intestinal ALPases, which are inactivated under these primates. Our findings indicate that the appearance of human conditions (11, 12). We found that the ALPase in chimpanzee placental-type ALPase in placenta was indeed a late event in placenta is as thermostable as human placental ALPase. Ther- mammalian evolution. It apparently occurred subsequent to the mostability studies on the ALPase in two orangutan placentas divergence of the lineages leading to man and the great apes suggested that two major components are present (Fig. 1). One, (chimpanzee and orangutan) from the lineages leading to lower accounting for 25-35% of the total ALPase activity, is ther- primates, including gibbon, which is classified with man and mostable like human and chimpanzee placental ALPase. The the great apes in the Hominoideae, but in a separate taxonomic other is thermolabile like liver/bone/kidney ALPase. The category. Others have reported the occurrence ofa heat-stable ALPase in placentas from other primates (gibbon, rhesus, pig- human placental-type ALPase in placentas of chimpanzee and tailed macaque, squirrel monkey, and owl monkey) are all very orangutan (6) and also somewhat unexpectedly in baboon lung thermolabile and indistinguishable in this respect from human (7). We have found that the heat-stable ALPase in baboon lung liver/bone/kidney ALPase and the ALPase from liver, bone, also occurs, in substantial amounts, in lungs from other Old and kidney from a variety of primate species. World monkeys. It resembles human and chimpanzee placental Immunological Studies. Antiserum raised in rabbits against ALPase immunologically but differs quite markedly in certain purified human placental ALPase crossreacts with intestinal of its inhibition characteristics, indicating significant structural ALPase (13, 14). However, when intestinal ALPase is tested differences at least at the inhibitor binding sites. We have not against placental ALPase in Ouchterlony double-diffusion found this ALPase in detectable amounts in lung from several plates the crossreaction is seen to be one ofonly partial identity. The crossreacting antibodies can be adsorbed out with intestinal The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertise- Abbreviations: ALPase, alkaline phosphatase; Har, L-homoarginine; ment" in accordance with 18 U. S. C. §1734 solely to indicate this fact. Leva, levamisole. 879 Downloaded by guest on September 24, 2021 880 - Genetics: Goldstein et-aL Proc. NatL Acad. Sci. USA 79 (1982) log % % 0 100 2.0 5 @2 5 \ @2 40 50 1.7- 4i,0 4 *3 25 1.4- FIG. 2. Composite diagram of Ouchterlony double-diffusion plates showing precipitin lines obtained with antiserum raised in rabbit against purified human placental ALPase and with the same anti- serum after it had been absorbed with human intestinal ALPase. Cen- 10 1.0' .ter wells: a, unabsorbedantiserum; b, absorbed antiserum. Outer wells: 1, human placental ALPase; 2, chimpanzee or heat-stable orangutan placental ALPase; 3, human intestinal ALPase; 4, intestinal ALPase from various primates (baboon, chimpanzee, rhesus monkey, etc.; see 5 0.7 Table 1j § footnote); 5, placental ALPases from various primates other than chimpanzee and orangutan (see Table 1) and also liver ALPase from man and various primates. Time, min any crossreaction with primate liver ALPase. (iii) With the un- FIG. 1. Thermostability studies at 650C on ALPase from chimpan- absorbed antisera the various primate intestinal ALPases, in- zee, orangutan, and rhesus monkey (Macaca mulatta) placentas. Each cluding those from chimpanzee and orangutan, showed cross- .point representuthe log % ofthe original activity remaining after heat- ing for the particular time. reactions of partial immunological identity with human, chimpanzee, and heat-stable orangutan placental ALPase, but apparently complete immunological identity with one another. ALPase, leaving antiserum reacting only with placental They did not crossreact with the absorbed antiserum. ALPase. These antisera do not crossreact with liver, bone, or Inhibition Studies. It is possible to discriminate sharply be- kidney ALPases, so one may infer that intestinal ALPase is more tween human placental, intestinal, and liver/bone/kidney closely related immunologically to placental ALPase than is ALPase by inhibition studies using various amino acids, pep- liver/bone/kidney ALPase. We tested such absorbed and un- tides, and other low molecular weight substances as inhibitors adsorbed antisera against various primate ALPases by the (1, 9, 15-21). We have used L-phenylalanine (Phe), L-homoar- Ouchterlony double-diffusion technique (Fig. 2). The results ginine (Har), L-phenylalanylglycylglycine (Phe-Gly-Gly), lev- can be summarized as follows: (i) With both antisera, chimpan- amisole (Leva), and L-leucine (Leu) as inhibitors and have de- zee placental ALPase and the heat-stable orangutan placental -termined for each case the concentration of inhibitor required ALPase gave reactions of apparent complete immuri6logical to give 50% inhibition, [I50], under standardized conditions (8). identity with human placental ALPase and with one another. Table 1 summarizes the findings on placental, intestinal, and (ii) Other primate placental ALPases, including gibbon ALPase, liver/bone/kidney ALPases from man and various primates. In showed no crossreaction with either antiserum, nor was there man, the liver, bone, and kidney ALPases give the same [I50]s Table 1. Inhibition of ALPase