983

An SFG rickettsia was isolated from heparinised blood collected the fifth day and analysed by polymerase chain reaction and restriction endonuclease fragment length polymorphism with and oligonucleotide primer pairs for the 190 kDa and 120 kDa (BG1 and BG2)4 antigen genes of R rickettsii, and Rsal and Pstl restriction endonucleases. In our laboratory (Unite des Rickettsies) the use of these primer pairs and restriction endonucleases has enabled us to differentiate rapidly between the pathogenic species of SFG rickettsiae. Our isolate was different from R conorii and the other pathogenic SFG rickettsiae (Rickettsii, sibirica, australisjaponica, and akarz) and the Israeli and Thai tick typhus rickettsiae. It was, however, identical to six SFG rickettsia isolates from A hebraeum collected around Zimbabwe and to an SFG rickettsia isolated from A cohaerens in Ethiopia. The isolation of this pathogenic SFG rickettsia enables us to confirm that there are two SFG rickettsioses in southern Africa. One is boutonneuse fever caused by R conorii and transmitted by ticks from dogs, and the other is African tick-bite fever caused by an SFG rickettsia transmitted by Amblyomma spp. We propose that the new SFG rickettsia be named "Rickettsia africae". on

PATRICK KELLY LINDA MATTHEWMAN

Faculty of Veterinary Science, University of Zimbabwe, Harare Rickets Unit, Faculty of Medicine,

LORENZA BEATI DIDIER RAOULT

13385 Marseille, France

ovulation

was

pregnancy. Section of Reproductive Endocrinology and Fertility, Department of Obstetrics and Gynaecology,

Dijkzigt University Hospital, 3015 GD Rotterdam, and Scientific Development Group, Organon International BV, Netherlands

P. F. J. DONDERWINKEL D. C. SCHOOT H. J. T. COELINGH BENNINK B. C. J. M. FAUSER

JL, Matzuk MM, Otani T, et al. Expression of biologically active human follitropin in Chinese hamster ovary cells. J Biol Chem 1989; 264: 4769-75. 2. Mannaerts B, De Leeuw R, Geelen J, et al. Comparative in vitro and in vivo studies on the biological characteristics of recombinant human follicle-stimulating hormone. Endocrinology 1991; 129: 2623-30. 3. Mannaerts B, Shoham Z, Schoot B, et al. Single dose pharmacokinetics and pharmacodynamics of recombinant human follicle-stimulating hormone in gonadotropin deficient volunteers. Fertil Steril (in press). 4. Schoot DC, Coelingh Bennink HJT, Mannaerts BMJL, et al. Human recombinant follicle-stimulating hormone induces growth of preovulatory follicles without concomitant increase in androgen and estrogen biosynthesis in a woman with isolated gonadotropin deficiency. J Clin Endocrinol Metab 1992; 74: 1471-73. 5. Devroey P, van Steirteghem A, Mannaerts B, Coelingh Bennink H. Successful in-vitro fertilisation and embryo transfer after treatment with recombinant human 1. Keene

FSH. Lancet 1992; 339: 1170-71. M, Dessole S, Senn A, et al. Successful in-vitro fertilisation and embryo transfer after treatment with recombinant human FSH. Lancet 1992; 339: 1170.

6. Germond

Faculty of Medicine, University of Zimbabwe

PETER MASON

Chiredzi

MARK DREARY ROBERT MAKOMBE

Consulting Rooms,

Chiredzi, Zimbabwe

Dau H. Etude expérimentale comparée de fièvre boutonneuse et de la tick-bite fever. Arch Institut Pasteur Tunis 1936; 25: 388-401. 2. Gear JHS. Other spotted fever group nckettsioses: clinical signs, symptoms, and pathology. In: Walker D, ed. Biology of rickettsial diseases, vol 1. Florida: CRC Press, 1988: 101-15. 3. Regnery RL, Spruil CL, Plikaytis BD. Genotypic identification of rickettsiae and estimation of intraspecies divergence for portions of two rickettsial genes. J Bacteriol 1991; 173: 1576-89. 4 Gilmore RD, Joste N, McDonald GA. Cloning expression and sequence analysis of the gene encoding the 120 kD surface exposed protein of Rickettsia rickettsii. Mol Microbiol 1989; 3: 1579-86.

1. Pijper A,

Pregnancy after induction of ovulation with recombinant human FSH in polycystic ovary syndrome SIR,—Transfection of human follicle-stimulating hormone subunit genes into Chinese hamster ovary cells results in secretion of the intact FSH dimer.1 Bioactivity and pharmacokinetic properties of recombinant human FSH (rhFSH) are similar to natural FSH in animals2 and man.3 Moreover, follicle development can be induced in a hypogonadotropic woman with rhFSH only, and luteinising hormone (LH) is mandatory for adequate oestrogen production.4 Pregnancies have now been reported after exogenous rhFSH in in-vitro fertilisation programmes.5,6 Induction of ovulation in patients with polycystic ovary sydrome (PCOS) resistant to clomiphene citrate with human menopausal gonadotropins (hMG) is associated with an increased risk of ovarian hyperstimulation and multiple pregnancies. A 27-year-old woman (body mass index 25 kg/m2) had been infertile for 3 years due to chronic clomiphene-resistant anovulation. PCOS diagnosis was based on: oligomenorrhoea (cycle length 8-52 weeks), increased serum LH (14-5 IU/1), hyperandrogenaemia (testosterone 5-1 nmol/1, and dehydroepiandrosterone sulphate 10-1 [1-2-10] polycystic appearance of ovaries by transvaginal sonography. Previous induction of ovulation with hMG and adjuvant gonadotropin-releasing hormone agonist resulted in six ovulatory cycles, but no pregnancy was achieved. As part of a multicentre trial (approved by the local ethics committees) rhFSH (Org 32489) was started on the third day of progestagen withdrawal bleeding with daily intramuscular administration of 75 IU. On day 10 ultrasound revealed the development of a single follicle (12 mm diameter). On day 13 the follicle reached 18 mm and 10 000 IU human chorionic

(FSH)

was given intramuscularly. 2 days later confirmed by sonography. No luteal support was given. 16 days after hCG, a pregnancy test was positive and 2 weeks later ultrasound revealed an intact intrauterine pregnancy. This case shows the capacity of rhFSH to induce monofollicular development and adequate luteal function resulting in a viable

gonadotropin (hCG)

Simple method for cystic fibrosis carrier screening cystic fibrosis (CF) carrier screening in the general population is controversial, previous trials have shown widespread acceptability and a positive psychological impact among prospective parents.1,2 It has been suggested that general screening would be too expensive and inefficient until the method had a high detection rate (85-95 %).3 In several North European countries, this detection rate of couples at risk can be achieved by testing only the most common CF mutation, AF508. With a two-step approach, screening primarily the AF508 in couples and more mutations only in the partners of AF508 carriers, over 85% of couples at risk could be identified in any community in which the AF508 mutation is present in more than 65% of the CF chromosomes.’ Here we describe a simple and cheap method for population carrier screening of AF508 with small pieces of blood spots on filter papers as a template for the allele specific polymerase chain reaction (PCR) amplification’ and a pooling strategy. 300 health workers of Hospital Nino Jesus without a family history of CF volunteered for the study. A small drop of blood from each individual taken from skin puncture was placed on a Guthrie card under sterile conditions. Filter paper pieces of about 1-2 mm2 with dried blood from 20 different individuals were mixed in the same reaction tube of each containing (100 total volume) PCR buffer, 200 deoxynucleotide triphosphate, and 30 pmol of each primer. The mixture was overlaid with 60 paraffin oil and placed in a thermocycler. After a first DNA liberation and denaturation step at 97°C for 8 min, 2-5 U Taq polymerase was added per tube during the first annealing step at 64°C for 2 min. After another 2 min at 72°C, 34 additional amplification cycles (94°C for 1 min, 64°C for 45 s, and 72°C for 2 min) followed with a final extension of 10 min at 72°C. Of the 15 lanes with all the 300 samples run in a single non-denaturing 8% polyacrylamide gel, heteroduplexes could be identified in 3 by staining with ethidium bromide. A subsequent pooling step divided the 60 samples of possible carriers in 6 reaction tubes, and so on. With a total of 41 reactions, 3 carriers for the AF508 mutation were identified in this population. Individual analysis of each of the 300 samples confirmed the 3 AF508 heterozygotes and did not show any false-negatives. Although the occurrence of a mutation in an amplification-resistant sample might not be detected by this method, we did not find any. We have not seen contamination problems with Guthrie spots.

984

We also verified sensitivity by mixing the sample from a known AF508 heterozygote with 4,9,14, and 19 samples from non-AF508 controls. In all the assays with multiple samples, the heteroduplexes formed by cross-hybridisation of the 95 and 98 base pair (bp) PCR products could be detected, although the 95 bp fragment was not visible. AF508 homozygotes could be detected, since heteroduplexes between the 95 bp patient’s product with the 98 bp fragments from other samples also appeared, giving a false

heterozygous reading. The theoretically optimum number of samples to be pooled initially depends on the expected prevalence of AF508 heterozygositY’ (19 if carrier probability is 1/28 or 31 if carrier proability is 1/45). However, pooling more than 20 samples in the same assay would require a higher reaction volume and may increase the risk of false negatives. The advantage of this procedure is simplicity and reduction in the cost and work required for AF508 screening. Hundreds of samples can be analysed in a day. The use of blood spots on filter paper, avoiding the DNA isolation step, decreases the risk of sample contamination and may increase uptake by facilitating the collection of samples, which could be done by individuals at home and mailed to the laboratory in sterile plastic bags. The pooling strategy reduces 5 to 7 times the number of assays needed. This procedure would also be suitable for neonatal CF screening since both AF508 homozygotes and heterozygotes can be detected.

Supported in part by Cilag (Johnson & Johnson) Laboratories. Department of Paediatrics, Hospital Infantil Niño Jesús, 28009 Madrid, Spain

J. C. MOLINA CABAÑERO L. A. PEREZ JURADO

Division of Genetics, Vanderbilt University,

Nashville, Tennessee, USA

S. RASKIN

Histocompatibility Unit,

J. L. VICARIO A. BALAS PEREZ M. D. GARCIA NOVO

Transfusion Centre, Madrid

ME, Gilfillan A, Compton M, et al. Prenatal screening for cystic fibrosis Lancet 1992; 340: 214-16. Watson EK, Mayall ES, Lamb J, et al. Psychological and social consequences of community carrier screening programme for cystic fibrosis. Lancet 1992; 340: 217-20. Statement from the National Institutes of Health workshop on population screening for the cystic fibrosis gene N Engl J Med 1990; 322: 70-71. Beaudet A, O’Brien W. Advantages of a two-step laboratory approach for cystic fibrosis carrier screening. Am J Hum Genet 1992; 50: 439-40 Keren BS, Rommens JM, Buchanan JA, et al. Identification of the cystic fibrosis gene: genetic analysis. Science 1989; 245: 1073-80. Gille Ch, Grade K, Coutelle Ch. A pooling strategy for heterozygote screening of the &Dgr;F508 cystic fibrosis mutation. Hum Genet 1991; 86: 289-91.

1. Mennie 2.

3. 4. 5. 6.

Screening for cystic fibrosis carriers SIR,-Dr Super and colleagues (Aug 22, p 490) draw attention to the apparent merits of cascade testing for carriers of the cystic fibrosis (CF) gene. They claim that they can identify more heterozygotes with less effort by testing relatives of affected probands than my colleagues and I can through the type of population screening programme we described (July 25, p 214). This is self-evident. However, there are limits to how far cascade testing can penetrate the population. Additionally, whether cascade identification of a carrier is of any benefit is not clear, since carrier status is only important in the context of reproduction. We began to offer cascade testing in Scotland as soon as the CF gene was cloned in 1989. The take-up rate was fast at first, but then slowed, and we are now approaching a plateau at less than 1 % of the estimated 200 000 Scottish CF carriers. Even in a population as homogeneous as this one, to identify more than a minute proportion of carriers by family extension studies will not be possible. The proportion will obviously be lower in communities that have seen much immigration. The argument for population screening is thus strengthened by the failure of the cascade approach. Either we ignore the 99% of carriers who do not have an affected relative, which is what Super seems to be saying, or we try and reach them by more broad-based techniques. We have presented evidence that carrier screening for

CF during pregnancy is probably the most efficient way of doing this. We concede that it does cause transient anxiety among a few of those identified as carriers. But we believe that this is an acceptable price to pay for a form of preventative medicine that lacks the nepotistic features of cascade testing. Human Genetics Unit, University of Edinburgh, Western General Hospital, Edinburgh EH4 2XU, UK

D.

J. H. BROCK

Cystic fibrosis mice with disease-related changes in lung and reproductive tract SIR,-In your Sept 19 editorial you comment on mice homozygous for cftr gene disruption. These mice were produced concurrently and independently by three groups using gene targeting in embryonal stem cellos. 1-3 You report that most mutant mice die perinatally or at weaning because of intestinal obstruction, and that, in contrast to cystic fibrosis in man, no excess mucus is seen in the lung and the male reproductive tract is normal. You

highlight the failure of the mice to mimic important hallmarks of the disease (notably lung and gonadal involvement) and suggest that the usefulness of these laboratory animals will be increased if the intestinal disorder can be alleviated. To describe the homozygous cftr mutant mice from all three groups as having a single phenotype is misleading. The distinct phenotype of our micemerits further comment.

Firstly, our cystic fibrosis mice do not die perinatally. Most of our homozygous mutant mice (cflcj)survive weaning, and cannot be distinguished visually from their wild type (+ / +) and heterozygote (cfl + )littermates at 30 days. However, postmortem examination of 4 of 6 c :fl cf mice showed overt gastrointestinal pathological changes, including abnormal mucus accumulation. Mild dilation of the salivary ducts was also seen. Most importantly, 1 mouse showed clear evidence of mucin in the lung, causing mild focal atelectasis consistent with presymptomatic lung disease. 1 of 2 male mice examined showed gross accumulation of mucin in the vas deferens. We have suggested two possible explanations for the phenotypic differences between our homozygous mutant mice and those described by the Chapel HilP and Cambridge2 groups. The first is that our mutation was crossed onto an outbred genetic background whereas the cftr mutations in the other two studies were crossed onto inbred strains. It is noteworthy that comparable cystic fibrosis mutations arising in ethnically closed human populations are severe whereas those arising in ethnically mixed populations are typically mild. A second possiblity is that a trace amount of residual normal cftr protein may result from aberrant splicing of our insertional targeted mutation, whereas this could not occur in a cftr gene disrupted by targeted gene replacement. A low level of wild-type protein in our insertional efl ef mice might explain why they do not show the extreme degree of intestinal disorder that is seen in both the Chapel Hill and Cambridge replacement mutant mice. Any wild-type protein present must be at a very low level, because the expected chloride channel defect is demonstrable by in-vivo potential differences in all our insertional cflcf mutant mice.3 Further experiments are necessary to resolve this issue. In any event, the enhanced viablity of our cystic fibrosis mouse model uniquely allows the development of aspects of the disease more relevant to the human condition. Thus our insertional cftr mutant mice constitute a valuable model to explore the pathophysiology of the disease, the relation between genotype and phenotype, and the development and testing of therapies relevant to the treatment of

patients. 3.4 MRC Human Genetics Unit, Western General Hospital, Edinburgh EH4 2XU, UK

JULIA R. DORIN DAVID J. PORTEOUS

1. Snouwaert JN, Brigman KK, Latour AM, et al. An animal model for cystic fibrosis made by gene targeting. Science 1992; 257: 1083-88. 2. Colledge WH, Ratcliff R, Foster D, Evans MJ. Cystic fibrosis mouse with intestinal obstruction. Lancet 1992, 340: 680. 3. Dorin JR, Dickinson P, Alton EWFW, et al. Cystic fibrosis in the mouse by targeted insertional mutagenesis. Nature 1992; 359: 211-15. 4. Wilson JM, Collins FS. More from the modellers. Nature 1992; 359: 195-96