Summary Solid tumours develop an acidic extracellular environment with high concentration of lactic acid, and lactic acid produced by glycolysis has been assumed to be the major cause of tumour acidity. Experiments using lactate dehydrogenase (LDH)-deficient rastransfected Chinese hamster ovarian cells have been undertaken to address directly the hypothesis that lactic acid production is responsible for tumour acidification. The variant cells produce negligible quantities of lactic acid and consume minimal amounts of glucose compared with parental cells. Lactate-producing parental cells acidified lightly-buffered medium but variant cells did not. Tumours derived from parental and variant cells implanted into nude mice were found to have mean values of extracellular pH (pHe) of 7.03 ± 0.03 and 7.03 ± 0.05, respectively, both of which were significantly lower than that of normal muscle (pHe = 7.43 ± 0.03; P < 0.001). Lactic acid concentration in variant tumours (450 ± 90 ,ug g-1 wet weight) was much lower than that in parental tumours (1880 ± 140 gg/g-1) and similar to that in serum (400 ± 35 ,ug/g-1).These data show discordance between mean levels of pHe and lactate content in tumours; the results support those of Newell et al (1993) and suggest that the production of lactic acid via glycolysis causes acidification of culture medium, but is not the only mechanism, and is probably not the major mechanism responsible for the development of an acidic environment within solid tumours.

Summary Mean values of extracellular pH (pH,) in tumours tend to be about 0.5 pH units lower than in normal tissues, whereas values of intracellular pH (pHi) in tumours and normal tissues are similar. Previous studies have shown that drugs that acidify cells at lower pHe such as nigericin, used alone or with agents that inhibit the regulation of pH,, have toxicity to cultured cells at pH,<6.5 in short-term exposure; these agents also lead to modest anti-tumour effects in mice when given acutely. To evaluate the long-term effects of these drugs at levels of pHe that might occur commonly in tumours, we exposed cells for up to 72 h at pHe 6.8 or 7.2 in vitro. Nigericin (0.033 gM) caused time-dependent cell killing of murine KHT and EMT-6 cells at pHe 6.8 (but not at pHe 7.2) with a surviving fraction approximately 5 x 10-3 after 72 h exposure. Cell killing was increased in the presence of 4,4-diisothiocyanstilbene 2,2-disulphonic acid (DIDS), an inhibitor of Na+-dependent HCO3-/Cl-exchange, and to a lesser extent in the presence of 5-(N-ethyl-N-isopropyl) amiloride (EIPA), an inhibitor of Na+/H+ exchange. Cell killing was exquisitely sensitive to the level of pHe. Osmotic pumps were used to obtain a 72 h continuous infusion of nigericin in mice; this led to dose-dependent killing of cells in KHT tumours with surviving fraction of approximately 0.1 at maximum tolerated doses. Hydralazine, which may cause tumour hypoxia and lower pHi as well as pH,, caused cytotoxity when given alone by chronic infusion, and enhanced the cytotoxicity due to nigericin. The addition of DIDS and/or EIPA (using two pumps) further enhanced anti-tumour toxicity, with a surviving fraction of approximately 0.002 at tolerated doses of the four drugs used to treat KHT tumours. The experiments demonstrate the activity of drugs that inhibit the regulation of pHi against murine tumours when delivered by chronic infusion.

Mandibuloacral dysplasia (MAD) is a rare autosomal recessive progeroid syndrome, characterized by mandibular hypoplasia, acroosteolysis affecting distal phalanges and clavicles, delayed closure of the cranial sutures, atrophic skin, and lipodystrophy. Recently, mutations in lamin A/C (LMNA) and zinc metalloprotease (ZMPSTE24), involved in post-translational processing of prelamin A to mature lamin A, have been identified in MAD kindreds. We now report novel compound heterozygous mutations in exon 1 (c.121C>T; p.Q41X) and exon 6 (c.743C>T; p.P248L) in ZMPSTE24 in two Japanese sisters, 7-and 3-year old, with severe MAD and characteristic facies and atrophic skin. The older sister had lipodystrophy affecting the chest and thighs but sparing abdomen. Their parents and a brother, who were healthy, had heterozygous mutations. The missense mutation, P248L, was not found in 100 normal subjects of Japanese origin. The mutant Q41X was inactive in a yeast halo assay; however, the mutant P248L retained near normal ZMPSTE24 activity. Immunoblots demonstrated accumulation of prelamin A in the patients' cell lysates from lymphoblasts. The lymphoblasts from the patients also revealed less intense staining for lamin A/C on immunofluorescence. We conclude that ZMPSTE24 deficiency results in accumulation of farnesylated prelamin A, which may be responsible for cellular toxicity and the MAD phenotype. Mutations in LMNA gene are also associated with other progeroid syndromes such as Hutchinson-Gilford progeria syndrome(HGPS), restrictive dermopathy (RD), progeriaassociated arthropathy, and atypical progeroid syndrome (also referred to as atypical Werner's syndrome) (13-16). Mutations in ZMPSTE24 gene are also associated with RD (17,18). There are many overlapping features among patients with MAD, HGPS, RD and atypical progeroid syndrome, although each presents with some unique phenotypic features. So far, 21 MAD patients have been reported to have LMNA mutations (6-10), but only 3 MAD patients have been reported with ZMPSTE24 mutations (11,19,20). Here, we report two Japanese siblings with MAD resulting from novel compound heterozygous mutations in ZMPSTE24. Keywords Materials and methodsThe studies were approved by the Institutional Review Boards of the Osaka University Hospital and the University of Texas Southwestern Medical Center at Dallas. Each participant or their legal guardian gave a written informed consent for all studies and photographs. Mutational analysis of the LMNA and ZMPSTE24 genesGenomic DNA was isolated from the subjects' peripheral whole blood cells using Dr. GenTLE (Takara Bio, Otsu, Shiga, Japan) according to the manufacturer's instructions. Direct sequencing of the entire coding region and the flanking exon-intron boundaries of the LMNA and ZMPSTE24 was first conducted in the proband, and the mutations were later confirmed in the younger sister. The primer sets that amplify each exon of LMNA and ZMPSTE24 genes were designed from published sequence information and exons were amplified by polymerase cha...