THE JOURNAL OF EXPERIMENTAL ZOOLOGY 230:387-392 (1984) zy Does Vitamin A Act on Pattern Formation Via the Epidermis or the Mesoderm? M. MADEN zyxwvutsr National Institute for Medical Research, Mill zyxwv Hill, London, United Kingdom ABSTRACT To determine whether the effect of vitamin A on pattern for- mation in the regenerating axolotl limb was due to a change in the blastemal mesoderm or the epidermis, tissue grafts were exchanged between normal and retinoic acid-treated animals. The results were conclusive. The combination zy of treated epidermis and untreated mesoderm resulted in normal regenerates. The combination of treated mesoderm and untreated epidermis produced re- generates that were serially duplicated in the proximodistal axis just as the controls were. We may, therefore, conclude that vitamin A acts on the meso- dermal cells of the blastema in causing proximodistal duplications. Administration of vitamin A to the regen- erating axolotl limb causes serial duplication of elements in the proximodistal axis such that a complete limb can be regenerated from a distal amputation plane (Maden, '82). In other systems vitamin A also induces pat- tern abnormalities: in the chick limb dupli- cations in the anteroposterior axis appear (Tickle et al., '82; Summerbell and Harvey, '83), and in frog and toad limbs both axes are affected (Saxena and Niazi, '77; Niazi and Saxena,'78; Maden, '83a). One explanation of these effects is that vi- tamin A respecifies the positional informa- tion of the mesodermal cells of the limb. As a first step to the testing of such a hypothe- sis, the site of action of vitamin A needs to be identified. To this end, the cellular effects of vitamin A on the regenerating axolotl limb have recently been reported (Maden, '83b) and alterations in both epidermal and me- sodermal cells observed. According to the above hypothesis, the epidermal effects ought not to be significant in terms of pattern reg- ulation and a test of whether this is so is reported here. Epidermis of the regenerating limb has been exchanged between normal animals and those treated with retinoic acid (RA). The results show that treated epidermis does not induce pattern abnormalities in normal mes- oderm, whereas serial duplications do appear with the untreated epidermisltreated meso- derm combination. MATERIALS AND METHODS The experiments were performed on 80- to 100-mm axolotls, Ambystoma mexicanum of both black and white genotypes. A total of 38 animals had either their forearms amputated through the mid-radius and ulna or their hindlimbs amputated through the mid-tibia and fibula, under MS 222 anaesthesia. Those animals to be treated with vitamin A were placed for 6 days in a 0.12 mM (30 mgiliter) so- lution of retinoic acid (all trans, Type XX, Sigma) immediately after amputation. The solution was prepared by sonicating the com- pound in water, and no intermediate solvent was used. Untreated animals were placed in water immediately after amputation. The animals were divided into four series. Controls (4 animals): simple retinoic acid (RA) treatment for 6 days to confirm the effect on proximodistal duplication. Sham controls (6 animals): after RA treatment the skin over the lower arm (or the lower leg) and the epidermis over the amputation surface was removed all in one piece in the manner described below. It was then replaced on the same limb in the same orientation to investigate the detrimen- tal effects of the operation on proximodistal duplication. Experimentals: 14black animals were treated with RA after amputation and 14 white animals were not. After 6 days the skin Address reprint requests to Dr. Malcolm Maden, Division of Developmental Biology, National Institute for Medical Re- search, The Ridgeway, Mill Hill, London NW7 lAA, UK. zy 0 1984 ALAN R. LISS, INC.