Fate, transport, and interactions of metals.

A joint United States-Mexico conference, "Fate, Transport, and Interactions of Metals" was held in Tuscon, Arizona, USA, 13-16 April 1993. The conference was hosted by the University of Arizona Center for Toxicology and co-sponsored by the NIEHS Superfund Basic Research Program, the National University of Mexico Program for the Environment, and the Pan American Health Organization. The purpose of the conference was to pro? mote an exchange of scientific information on potentially toxic metals that may be present in hazardous waste sites. By shar? ing this technology, the United States and Mexico are better able to define the extent

Low grade non-Hodgkin's lymphoma (NHL) is as yet an incurable malignancy (Kalter et al., 1987;Schein et al., 1975;Horning & Rosenberg, 1984). Initial therapy may produce a complete remission but the high rate of recurrence indicates minimal disease persists. In at least 80% of the low grade NHLs encompassed by categories B and C of the working formulation, rearrangement of the bc1-2 proto-oncogene ( Figure 1) as part of the 14;18 translocation (Yunis et al., 1982) generates a suitable target for enzymatic amplification using the polymerase chain reaction (PCR). The breakpoints on chromosome 18 involving bcl-2 are clustered mainly at two sites. Fifty to sixty per cent are found in the major breakpoint region (MBR) which lies in the 3' untranslated portion of the gene while a further 25-40% occur in the minor cluster region (MCR) located in an intron 20 kb down stream (Cleary et al., 1986a;Weiss et al., 1987). The rearrangement brings bcl-2 into conjunction with one of the joining (JH) gens of the immunoglobulin heavy chain locus on chromosome 14. The resulting bcl-2/JH DNA sequence is unique to the malignant clone. The use of this sequence as a target for amplification by PCR in the detection of minimal residual disease has been demonstrated (Lee et al., 1987;Crescenzi et al., 1988;Stetler-Stevenson et al., 1988;Cunningham et al., 1989). In this report we have evaluated the utility of PCR for monitoring disease status in low grade NHL by comparing conventional bone marrow histology with PCR analysis of bone marrow and peripheral blood.

Methods
Bone marrow aspirates, trephines and contemporaneous peripheral blood were available from 30 patients with a diagnosis of working formulation category B or C non-Hodgkin's lymphoma. Infiltration of the bone marrow trephines was determined by routine histology.

DNA preparation
The extreme sensitivity of PCR coupled with the high molarity of 'positive' PCR product necessitates strict separation in the stages in the process of sample analysis from sample collection to Southern blotting, so as to prevent contamination between samples and the consequential false positive results. To this end the following procedure was adhered to: all glass and plastic ware and solutions were autoclaved prior to use; in room 1, in a pre-cleaned laminar flow cabinet, reaction ingredients were aliquoted and dispensed using pipettes that were never brought into contact with DNA; a washed pipette was used to dispense DNA and a 'dummy' addition of DNA to the first and last tubes was performed; thermocycling and gel electrophoresis were conducted in room 2, pipettes used in this room never enter room 1. Mononuclear cells were extracted from bone marrow aspirates (usually 2 ml) and peripheral blood (15 ml) by centrifugation at 400 g for 30 min on LymphoprepTM (Nycomed AS, Norway). The mononuclear cells were washed twice in ice cold PBS then in red cell lysis buffer (0.32 M sucrose, 10 mM Tris-HCl pH 7.5, 5 mM MgCl2 1% Triton X) and pelleted. DNA was isolated either by phenol-chloroform extraction and ethanol precipitation or by resuspension in digestion mix (1 x PCR buffer, 0.25% Tween 20, 0.6 l1 of 10 mg ml-' proteinase K per 100 ,lI) to give a cell count of 2 x I05 cells gl -' and incubated at 55'C for 3 h or overnight. The proteinase K was inactivated by heating at 95'C for 15 min.  (Cleary et al., 1986b), for the MCR; 0.1 g DNA from a low grade NHL for which we have sequenced the breakpoint. After an initial denaturation step at 95°C for 2 min, 45 cycles were performed with the following parameters: for the MBR; 94°C 1 min; 50'C 30 s, 72'C 1 min 30 s, for the MCR; 94°CF 1 min 550C 2 min, 72°C 3 min (Ngan et al., 1989). Fifteen gl PCR product was then electrophoresed on an ethidium bromide stained 1.5% agarose gel; viewed under ultraviolet light and then Southern blotted onto nylon and fixed under UV light. Filters were hybridised to the relevant 5' labelled a32 ATP internal oligonucleotide probe (sequences shown in Figure 1). The autoradiographs were exposed overnight at -700C. Films were then developed and the filter re-exposed over 5 days at -70°C. When PCR was negative, the quality of the DNA for PCR was checked by amplification of the B-globin gene using primers PCO3 and T.F. HICKISH et al.
PCO4 described elsewhere (Saiki et al., 1988). In addition to increasing sensitivity, the use of an internal oligonucleotide probe demonstrates that the PCR product is authentic and results from amplification across a bcl-2/JH junction. Authenticity was also demonstrated when necessary by direct sequencing using the dideoxy chain termination method. Briefly, DNA was purified from the PCR product by phenolchloriform extraction, passage down a Sephadex G50 column, ethanol precipitation and suspension in distilled water. Oligonucleotides used for probing (Figure 1) served as sequencing primers for Sequenase (USB, Cleveland, Ohio). The presence of a blc-2 rearrangement was also verified by repeating the PCR using a primer directed upstream on bcl-2 (either MBR or MCR) of the blc-2 primer used in the first reaction along with the JH primer. A PCR product with a size difference commensurate with the shift in priming location on bcl-2 results from specific amplification.

Results
Sensitivity of PCR Figure 2 shows the sensitivity of PCR for detecting cells bearing the bcl-2 rearrangementin this case with the rearrangement through the MBR.
Comparison of bone marrow histology with PCR analysis of blood and marrow  Total 31 *All patients with PCR positive peripheral blood also were PCR positive in the bone marrow. marrow was positive in 12 of the HPs and in seven of the HNs. PCR blood was positive in nine of the HPs and nine of the HNs. Primary tumour was unavailable in two of HP patients and two of the HN patients who were negative on PCR analysis. It is therefore not known whether these patients, in fact, had a bcl-2 rearrangement involving the MBR or MCR. Nine patients were PCR positive (blood and/or marrow) and HN negative. Two of these were in complete remission (CR) and the remaining six receiving chemotherapy (cyclophoshamide, vincristine and prednisolone [CVP] or chlorambucil) and were entering CR. Figure 3 demonstrates the application of molecular monitoring of response to treatment for a patient with a stage IIIB low grade NHL receiving CVP chemotherapy. After three courses of chemotherapy the patient had a good response with resolution of all lymphadenopathy confirmed by CT scanning although in November the marrow histology was positive. Note negative PCR blood in early January. In late January PCR blood was again positive and by early February the patient had clinically relapsed. (The difference in the blood and marrow signals for November probably reflects the number of lymphoma cells sampled).

Discussion
In this study we demonstrate how the application of molecular monitoring can add to the assessment of a patient's disease status.
When bone marrow histology was negative, PCR demonstrated malignant cells in approximately 50% of both bone marrow and peripheral blood samples. Using PCR it was possible to follow a patient's disease into remission and then detect the presence of subclinical diseasea 'molecular relapse'. PCR analysis of blood enables a high rate of disease detection. This indicates it is an effective means for monitoring response to treatment and could be used as an adjunct to analysis of the bone marrow thereby reducing the need for Figure 2 An autoradiograph of a Southern blot of PCR product probed for bcl-2 rearrangement through the MBR. SUDHL4 cells were serially diluted in peripheral blood lymphocytes (PBLs) from a healthy donor. After 18 h exposure bands of the expected size are seen at a dilution of 20 SUDHL4 cells in a background of 106 PBLs. After 5 days exposure a band was clearly visible at a dilution of 2 SUDHL4 cells/106 PBLs (autoradiograph not shown). M = marker. N = negative control, No DNA. P = placenta. PBL = 106 peripheral blood lymphocytes. 2,20,2 x 102, 2 x l03 = number of SUDHL4 cells/106 PBLs. S = 1 iLg SUDHL4 DNA. Patient K repeated bone marrow trephine and aspirate: clearly advantageous to the patient. When the marrow histology was positive, PCR analysis of the marrow was positive in 80% and PCR analysis of peripheral blood was positive in 60%. Of the three patients for whom bone marrow histology was positive and PCR of either bone marrow or blood was negative primary tumour was unavailable in two patients and so it is unknown if their tumour carries a bcl-2 rearrangement involving the MBR or MCR. This highlights the need for other targets for amplification in low grade NHL since some 10-15% of tumours do not have a bcl-2 rearrangement involving the MBR or MCR. One target that may prove suitable is the rearranged variable gene of the immunoglobulin heavy chain locus (Deane & Norton, 1990;Yamada et al., 1990).
With current therapies relapse in low grade NHL is almost inevitable. We and others are conducting studies on the role of maintenance therapy in prolonging remission in low grade NHL. Intriguingly, recently PCR has been used to detect circulating lymphoma cells in seven of eight patients in continuous clinical remission for more than 10 years after presenting with advanced follicular low grade NHL (Price et al., 1991). The explanation for this finding is uncertain (Sklar, 1991) but in the small group of patients who are long-term survivors with this malignancy detection of minimal residual disease may have little clinical value. In this context molecular monitoring should clarify the natural history of patients in complete remission but PCR positive and reveal the effectiveness of maintenance therapies on subclinical disease. This information is likely to influence decisions about treatment strategies, for example the use of biological therapies which may be most effective in minimal disease.
The detection of minimal disease is possible whenever there is a distinctive DNA/RNA target for PCR (Morgan et al., 1989;Shiramizu & Magrath, 1990). The implementation of molecular monitoring therefore offers a general approach for improving treatment strategies.