Expansion of B Lymphocytes with an Unusual Immunoglobulin Rearrangement Associated with Atypical Lymphocytosis and Cigarette Smoking Marcia A. Chan, Stephen H. Benedict, Kelvin C. Carstairs, William H. Francombe, and Erwin W. Gelfand Division of Basic Sciences and the Raymond and Beverly Sadder Foundation Laboratory, Department of Pediatrics, National Jewish Center for Immunology and Respiratory Medicine, Denver, Colorado and the Hematology Laboratory and Department of Medicine, Toronto General Hospital, Toronto, Ontario
Persistent polyclonal lymphocytosis has been described in a group of female patients who all have the HLA-DR7 antigen in common and who are all heavy cigarette smokers. Immunoglobulin heavy chain gene rearrangement was analyzed by hybridization with specific immunoglobulin heavy chain genes to restriction enzyme-digested genotnic DNA samples. The results in two of these patients showed that the lymphocytosis was associated with an expanded subpopulation of B-lineage cells represented by the presence of an unusual immunoglobulin gene rearrangement pattern. Expansion of this subpopulation of B cells appeared to be linked to cigarette smoking since the intensity of the cell population harboring the rearranged gene was much stronger in patients who were smoking heavily compared with the same patients who were temporarily not smoking.
Seven patients have now been described with persistent polyclonal lymphocytosis of B lymphocytes (1, 2). All patients were female, expressed the HLA-DR7 antigen, and were heavy cigarette smokers. The lymphocytosis was characterized by raised serum IgM levels and by the presence of atypical lymphocytes, including small numbers (1 to 2 %) of unusual binucleate forms. The lymphocytosis appeared to be polyclonal in nature, based on surface and cytoplasmic immunoglobulin staining of lymphocytes which revealed an equal distribution of kappa and lambda positive cells. In the present report, we extend our previous findings in two of the patients and demonstrate that the lymphocytosis is characterized by the presence of an unusual DHJ H rearranged immunoglobulin band. The intensity of this rearranged band was stronger when the patient was smoking heavily and was only just distinguishable after several months of reduced smoking.
Materials and Methods Peripheral blood mononuclear cells and T and non-T subpopulations were obtained following Ficoll-Hypaque gra-
Key Words: smoking, B lymphocytes, immunoglobulin, gene rearrangement (Received in original form June 29, 1989 and in revised form January 11, 1990)
Addresscorrespondence to: Dr. Marcia A. Chan, Department of Pediatrics, National Jewish Center for Immunology and Respiratory Medicine, 1400 Jackson Street, Denver, CO 80206. Am. J. Respir. Cell Mol. BioI. Vol. 2. pp. 549-552, 1990
dient centrifugation as previously described (3). High molecular weight DNA was obtained by standard methods, digested with specific restriction endonucleases, separated by 0.6% agarose gel electrophoresis, and blotted to nylon membranes (4). Filters were hybridized to DNA probes, as described (4), that had been radiolabeled with 32P-dCTP by the random primer technique (5). Hybridizing bands were detected by autoradiography. Fragment sizes were determined by reference to bacteriophage A DNA digested with Hind III. The DNA probes described were a 1.3 kilobase (kb) Eco RI CJl DNA gene fragment (6) and a 6.0 kb Hind III-Bam HI J HDNA gene fragment (6).
Results As B lymphocytes differentiate into different antibody-secreting cells, a series of changes take place at the DNA level. One such change occurs at the immunoglobulin heavy chain gene locus. This locus comprises a variable (V) region gene element, a diversity (D) gene element, a joining (J) gene element, and constant region genes encoding the various isotypes. As the B cell matures, rearrangements of the DNA encoding these genes occur. These rearrangements are accompanied by deletion of intervening DNA, resulting in the creation of a new and unique segment of DNA. This new region contains a V gene, a D gene, and a J gene, each selected at random from a large pool. Because the newly created DNA segment is unique, and because each cell rearranges the segments differently, the newly created DNA segment is specific for and characteristic of a single cell. The only way
550
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 2 1990
1
3
2
* 19kb-
19kb
*
Figure 1. Hybridization of Bam HI-digested DNA with a 1.3 kilobase (kb) Eco RI Cj1. DNA gene fragment. Lane 1: DNA pattern of two patients at a time when they were smoking heavily. Lane 2: DNA pattern of one patient taken at a time when she severely curtailed smoking. Lane 3: DNA pattern of the same patient in lane 2 taken when she resumed smoking. The germline band is 19.0 kb. The rearranged band is indicated by asterisks.
a single one of these rearranged genetic loci can be identified in a mixed population of peripheral lymphocytes is if the specific cell is expanded as in the case of lymphocytosis or certain clonally expanded tumors. To determine if the lymphocytosis observed in these patientsrepresented a clonal, biclonal, polyclonal, or random expansion of cells, the immunoglobulin gene rearrangement pattern was examined. High molecular weight genomic DNA was isolated, digested with the restriction endonuclease Bam HI, and hybridized to the Cu probe, which encodes the synthesis of IgM. Hybridization of the Cu. probe to the digested DNA taken from the two patients at a time when they were smoking heavily (1 to 2 packs/d) revealed identical rearrangements of one Cp. gene (16.0 kb) with the other Cp. gene retained in germline (nonrearranged) configuration (19.0 kb) (Figure 1, lane 1). The small decrease in size of the rearranged band compared to the germline band was unusual compared to that seen with other B cells in which rearrangement of Cp. gene has occurred and suggested that rearrangement was the result of a small deletion (1 to 2 kb) within the J H gene segments. In addition to the samples taken when the patients were smoking heavily, one of the patients was monitored at a time after she had reduced smoking to a minimum (1
pack/wk) for 6 mo. Figure 1, lane 2 indicates that during periods of relative abstention, the characteristic rearranged band (close doublet) was barely detectable. When she resumed heavy smoking (1.5 packs/d), the rearranged band returned (Figure 1, lane 3). This suggests that a direct correlation exits between smoking and the appearance of cells that carry this characteristically rearranged Cj1.. Because immunoglobulin heavy chain gene rearrangements of the type (close doublet) described here have been described in leukemic cells of both T and B cell origin (4, 7-9), we determined whether the rearranged band occurred in a subpopulation of T or B cells. Peripheral blood lymphocytes were first separated into two populations: E-rosetting T cells (E+) and the E-nonrosetting (E-) (predominantly Bcontaining) cells. DNA was extracted and digested with restriction endonucleases. When the separated populations were examined with Bam HI and the Cp. probe, a rearranged allele was observed in the E- population (Figure 2A, lane 2). In the E+ population, the Cu. genes were found to be only in gerrnline configuration (Figure 2A, lane 1). These observations suggest that the B lymphocytes (E- population), and not the T cells, harbor the rearranged band. Hybridization of DNA to both CK and C)I. DNA gene probes showed germline configuration for kappa and lambda genes in all cell populations (data not shown). Because no clonally rearranged kappa or lambda bands were detected, kappa and lambda genes were probably in germline configuration in the population of cells that contain the characteristic rearranged Cp. band. To determine whether the nongermline p. band represented a legitimate rearrangement occurring upstream (5) to Cu, an aberrant rearrangement occurring downstream (3) to Cu, or a DNA sequence abnormality (polymorphism), DNA was digested with Hind III, which cleaves between JH and Cp. at a point 5' to the p. switch region. The resulting blots were hybridized to the Cp. probe and a J H probe. A single p. germline band (11.5 kb) was seen in both the E+ and Epopulation (Figure 2B, lanes 1 and 2, respectively). In addition, when a similar blot was hybridized to the J H gene probe (Figure 2C), two rearranged alleles and a gerrnline allele were observed in the E- population (lane 2) and not in the E+ cells (lane 1). Detection of the rearranged band with a second enzyme suggests that the band observed with Bam HI was not due to restriction enzyme polymorphism. The results using Hind III showed that no alteration in the size of the Cp. band was observed 3' to the switch region. This indicates that the rearrangement occurred 5' to the p. switch region and was, therefore, most likely a legitimate rearrangement.
Discussion A unique syndrome of persistent lymphocytosis with small numbers of binucleate cells has been described in women who are smokers and share the HLA-DR7 (and in six of seven patients, Bw44) antigen in common (1, 2). Immunofluorescence studies of surface and cytoplasmic immunoglobulin staining suggested that the lymphocytosis was polyclonal in nature. Results presented here suggest that an unusual rearrangement of the Cu gene is present in B-lineage cells during the time when the patient is smoking heavily. The small de-
Chan, Benedict, Carstairs et al.: Lymphocytosis and Cigarette Smoking
A
551
Figure 2. Hybridization of digested DNA from one patient with a 1.3 kb Eco RI CIl DNA gene fragment and a 6.0 kb Hind III-Bam HI J H DNA gene fragment. Lanes 1: P population (T cells). Lanes 2: E- population (B cells). (A) Bam HI-digested DNA probed with CIl. The germline band is 19.0 kb. (B) Hind III-digested DNA probed with CIl. The germline band is 11.5 kb. (C) Hind III-digested DNA probed with J H • The germline band is 11.5 kb. Rearranged bands are indicated by asterisks.
c
B
2
2
2
11.5kb-
19k b-
11.5kb-
*
crease in size of the rearranged allele, in conjunction with germline configuration of the J1- switch region, suggests that the rearrangement involved a small deletion within or close to the JH gene segments. We have previously reported a similar rearrangement pattern in T cell acute lymphoblastic leukemia (7) and acute myelogenous leukemia (4). The same rearrangement pattern has also been observed by Mizutani and associates (10) in a number of T cell leukemias and in a few immature B cell leukemias. Detailed analysis of these B cell leukemias showed rearrangement involving the DH segment (DQS2) nearest to the JHgene segments. Our observations are compatible with such a DHJ H recombination. DH-J Hrecombinations are rarely observed because they are transient and the subsequent step, rearrangement of VHto DHJ H, usually occurs soon after, during maturation of the B cell subpopulation. Examination of immunoglobulin gene rearrangement patterns in more than 30 independently derived B cell lines from normal individuals (smokers and nonsmokers) has not revealed this unique pattern of rearrangement. The expansion of cells containing this immature DHJ H rearrangement appears to be linked to cigarette smoking. In one patient, abnormal lymphocyte numbers and morphology were observed when she smoked. However, when she abstained, blood counts and lymphocyte morphology reverted to normal. This phenomenon correlated well with the DNA analysis. A more prominent band representing the rearranged DHJ Hgenes was observed in the DNA sample taken at the time she had resumed smoking than was observed in the DNA sample taken after she had stopped for several
months. It was interesting that when the population of cells was reexamined after the patient had resumed smoking, the rearranged band was less distinct (Figure 1, lane 3) than observed earlier (not shown) and less distinct when compared with the rearranged band from the other patient who was continually a heavy smoker. Because the band was less distinct after reexpansion, it is possible that some individual evolution occurred with the DHJ H rearranged population, resulting in small changes in the size of the rearranged bands. This evolution could have taken the form of additional (upstream) D Hto JHrearrangements. The intensity of the DHJ H rearranged band makes it unlikely that the small percentage (l to 2 %) of binucleate cells could account entirely for the rearranged band. It is possible that the close doublet is delineated by two populations of cells. The binucleate cells may represent either precursor cells that develop into the expanded B cell population or a small subset of the cells that are undergoing expansion. Alternatively, heavy smoking by HLA-DR7+ females could result in expansion of two separate cell populations: one defined by increased size and binucleate morphology and the other defined by a rearranged DHJ Hband which is characteristic of certain immature B cells. There are at least two possibilities that could explain the appearance of these cells in susceptible individuals. First, smoking may elict expansion of a subpopulation of cells that have both alleles rearranged to DHJ H and, unlike normal cell populations, are maintained in this state during the period of smoking. Because such a population has not completed the first series of rearrangements (VHto DHJ H), the
552
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 2 1990
recombinational apparatus may remain continually active and render the cells prone to secondary, potentially abnormal rearrangements. A second, less likely, possibility is that the DHJH rearrangement occurs on one allele of every peripheral B cell in these patients. The other allele may be rearranged normally. This type of cell may potentially be more stable than the first population. The changes that occur in the B cells of these HLA-DR7+ female patients seem to be, at least in part, driven by smoking, and this raises the possibility that smoking may elicit or stabilize certain subpopulations. Such an effect could be precursor to further abnormalities. Acknowledgments: This work was supported by grants from the National Institutes of Health, National Cancer Institute (Canada), the Medical Research Council (Canada), the Addiction Research Foundation (Toronto), and the Ontario Cancer Treatment Research Fund. Dr. Gelfand is a scholar of the Raymond and Beverly Sackler Foundation.
References 1. Gordon, D. S., B. M. Jones, S. W. Browning et al. 1982. Persistent polyclonal lymphocytosis of B lymphocytes. N. Engl. J. Med. 307:232-236.
2. Carstairs, K. c., W. H. Francombe, J. G. Scott, and E. W. Gelfand. 1985. Persistent polyclonal lymphocytosis of B lymphocytes, induced by cigarette smoking? Lancet i:1094. 3. Cheung, R. K., S. Grinstein, and E. W. Gelfand. 1982. Volume regulation by human lymphocytes. Identification of differences between the two major lymphocyte subpopulations. J. Clin. Invest. 70:632-638. 4. Ha, K., M. Minden, N. Hozumi, and E. W. Gelfand. 1984. Immunoglobulin gene rearrangement in acute myelogenous leukemia. Cancer Res. 44: 4658--4660. 5. Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA restriction fragments to high specific activity. Anal. Biochem. 132: 6-13. 6. Ravetch, J. V., U. Siebenlist, S. Korsmeyer, T. Waldmann, and P. Leder. 1981. Structure of the human immunoglobulin J1. locus: characterization of embryonic and rearranged J and D genes. Cell 27:583-591. 7. Ha, K., M. Minden, N. Hozumi, and E. W. Gelfand. 1984. Immunoglobulin J1. chain gene rearrangement in a patient with T cell acute lymphoblastic leukemia. J. Clin. Invest. 73:1232-1236. 8. Korsmeyer, S. J., A. Arnold, A. Bakhshi et al. 1983. Immunoglobulin gene rearrangement and cell surface antigen expression in acute lymphocytic leukemia of T cell and B cell precursor origin. J. Clin. Invest. 71:301-313. 9. Ford, A. M., H. V. Melgaard, M. F. Greaves, andH. J. Gould. 1983. Immunoglobulin gene organization in haemopoietic stem cell leukemia. EMBO J. 2:997-1001. 10. Mizutani, S., A. M. Ford, L. M. Wiedemann et al. 1986. Rearrangement of immunoglobulin heavy chain genes in human T leukemic cells shows preferential utilization of the D segment (DQS2) nearest to the J region. EMBO J. 5:3467-3473.
Persistent polyclonal lymphocytosis has been described in a group of female patients who all have the HLA-DR7 antigen in common and who are all heavy cigarette smokers. Immunoglobulin heavy chain gene rearrangement was analyzed by hybridization with specific immunoglobulin heavy chain genes to restriction enzyme-digested genotnic DNA samples. The results in two of these patients showed that the lymphocytosis was associated with an expanded subpopulation of B-lineage cells represented by the presence of an unusual immunoglobulin gene rearrangement pattern. Expansion of this subpopulation of B cells appeared to be linked to cigarette smoking since the intensity of the cell population harboring the rearranged gene was much stronger in patients who were smoking heavily compared with the same patients who were temporarily not smoking.
Seven patients have now been described with persistent polyclonal lymphocytosis of B lymphocytes (1, 2). All patients were female, expressed the HLA-DR7 antigen, and were heavy cigarette smokers. The lymphocytosis was characterized by raised serum IgM levels and by the presence of atypical lymphocytes, including small numbers (1 to 2 %) of unusual binucleate forms. The lymphocytosis appeared to be polyclonal in nature, based on surface and cytoplasmic immunoglobulin staining of lymphocytes which revealed an equal distribution of kappa and lambda positive cells. In the present report, we extend our previous findings in two of the patients and demonstrate that the lymphocytosis is characterized by the presence of an unusual DHJ H rearranged immunoglobulin band. The intensity of this rearranged band was stronger when the patient was smoking heavily and was only just distinguishable after several months of reduced smoking.
Materials and Methods Peripheral blood mononuclear cells and T and non-T subpopulations were obtained following Ficoll-Hypaque gra-
Key Words: smoking, B lymphocytes, immunoglobulin, gene rearrangement (Received in original form June 29, 1989 and in revised form January 11, 1990)
Addresscorrespondence to: Dr. Marcia A. Chan, Department of Pediatrics, National Jewish Center for Immunology and Respiratory Medicine, 1400 Jackson Street, Denver, CO 80206. Am. J. Respir. Cell Mol. BioI. Vol. 2. pp. 549-552, 1990
dient centrifugation as previously described (3). High molecular weight DNA was obtained by standard methods, digested with specific restriction endonucleases, separated by 0.6% agarose gel electrophoresis, and blotted to nylon membranes (4). Filters were hybridized to DNA probes, as described (4), that had been radiolabeled with 32P-dCTP by the random primer technique (5). Hybridizing bands were detected by autoradiography. Fragment sizes were determined by reference to bacteriophage A DNA digested with Hind III. The DNA probes described were a 1.3 kilobase (kb) Eco RI CJl DNA gene fragment (6) and a 6.0 kb Hind III-Bam HI J HDNA gene fragment (6).
Results As B lymphocytes differentiate into different antibody-secreting cells, a series of changes take place at the DNA level. One such change occurs at the immunoglobulin heavy chain gene locus. This locus comprises a variable (V) region gene element, a diversity (D) gene element, a joining (J) gene element, and constant region genes encoding the various isotypes. As the B cell matures, rearrangements of the DNA encoding these genes occur. These rearrangements are accompanied by deletion of intervening DNA, resulting in the creation of a new and unique segment of DNA. This new region contains a V gene, a D gene, and a J gene, each selected at random from a large pool. Because the newly created DNA segment is unique, and because each cell rearranges the segments differently, the newly created DNA segment is specific for and characteristic of a single cell. The only way
550
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 2 1990
1
3
2
* 19kb-
19kb
*
Figure 1. Hybridization of Bam HI-digested DNA with a 1.3 kilobase (kb) Eco RI Cj1. DNA gene fragment. Lane 1: DNA pattern of two patients at a time when they were smoking heavily. Lane 2: DNA pattern of one patient taken at a time when she severely curtailed smoking. Lane 3: DNA pattern of the same patient in lane 2 taken when she resumed smoking. The germline band is 19.0 kb. The rearranged band is indicated by asterisks.
a single one of these rearranged genetic loci can be identified in a mixed population of peripheral lymphocytes is if the specific cell is expanded as in the case of lymphocytosis or certain clonally expanded tumors. To determine if the lymphocytosis observed in these patientsrepresented a clonal, biclonal, polyclonal, or random expansion of cells, the immunoglobulin gene rearrangement pattern was examined. High molecular weight genomic DNA was isolated, digested with the restriction endonuclease Bam HI, and hybridized to the Cu probe, which encodes the synthesis of IgM. Hybridization of the Cu. probe to the digested DNA taken from the two patients at a time when they were smoking heavily (1 to 2 packs/d) revealed identical rearrangements of one Cp. gene (16.0 kb) with the other Cp. gene retained in germline (nonrearranged) configuration (19.0 kb) (Figure 1, lane 1). The small decrease in size of the rearranged band compared to the germline band was unusual compared to that seen with other B cells in which rearrangement of Cp. gene has occurred and suggested that rearrangement was the result of a small deletion (1 to 2 kb) within the J H gene segments. In addition to the samples taken when the patients were smoking heavily, one of the patients was monitored at a time after she had reduced smoking to a minimum (1
pack/wk) for 6 mo. Figure 1, lane 2 indicates that during periods of relative abstention, the characteristic rearranged band (close doublet) was barely detectable. When she resumed heavy smoking (1.5 packs/d), the rearranged band returned (Figure 1, lane 3). This suggests that a direct correlation exits between smoking and the appearance of cells that carry this characteristically rearranged Cj1.. Because immunoglobulin heavy chain gene rearrangements of the type (close doublet) described here have been described in leukemic cells of both T and B cell origin (4, 7-9), we determined whether the rearranged band occurred in a subpopulation of T or B cells. Peripheral blood lymphocytes were first separated into two populations: E-rosetting T cells (E+) and the E-nonrosetting (E-) (predominantly Bcontaining) cells. DNA was extracted and digested with restriction endonucleases. When the separated populations were examined with Bam HI and the Cp. probe, a rearranged allele was observed in the E- population (Figure 2A, lane 2). In the E+ population, the Cu. genes were found to be only in gerrnline configuration (Figure 2A, lane 1). These observations suggest that the B lymphocytes (E- population), and not the T cells, harbor the rearranged band. Hybridization of DNA to both CK and C)I. DNA gene probes showed germline configuration for kappa and lambda genes in all cell populations (data not shown). Because no clonally rearranged kappa or lambda bands were detected, kappa and lambda genes were probably in germline configuration in the population of cells that contain the characteristic rearranged Cp. band. To determine whether the nongermline p. band represented a legitimate rearrangement occurring upstream (5) to Cu, an aberrant rearrangement occurring downstream (3) to Cu, or a DNA sequence abnormality (polymorphism), DNA was digested with Hind III, which cleaves between JH and Cp. at a point 5' to the p. switch region. The resulting blots were hybridized to the Cp. probe and a J H probe. A single p. germline band (11.5 kb) was seen in both the E+ and Epopulation (Figure 2B, lanes 1 and 2, respectively). In addition, when a similar blot was hybridized to the J H gene probe (Figure 2C), two rearranged alleles and a gerrnline allele were observed in the E- population (lane 2) and not in the E+ cells (lane 1). Detection of the rearranged band with a second enzyme suggests that the band observed with Bam HI was not due to restriction enzyme polymorphism. The results using Hind III showed that no alteration in the size of the Cp. band was observed 3' to the switch region. This indicates that the rearrangement occurred 5' to the p. switch region and was, therefore, most likely a legitimate rearrangement.
Discussion A unique syndrome of persistent lymphocytosis with small numbers of binucleate cells has been described in women who are smokers and share the HLA-DR7 (and in six of seven patients, Bw44) antigen in common (1, 2). Immunofluorescence studies of surface and cytoplasmic immunoglobulin staining suggested that the lymphocytosis was polyclonal in nature. Results presented here suggest that an unusual rearrangement of the Cu gene is present in B-lineage cells during the time when the patient is smoking heavily. The small de-
Chan, Benedict, Carstairs et al.: Lymphocytosis and Cigarette Smoking
A
551
Figure 2. Hybridization of digested DNA from one patient with a 1.3 kb Eco RI CIl DNA gene fragment and a 6.0 kb Hind III-Bam HI J H DNA gene fragment. Lanes 1: P population (T cells). Lanes 2: E- population (B cells). (A) Bam HI-digested DNA probed with CIl. The germline band is 19.0 kb. (B) Hind III-digested DNA probed with CIl. The germline band is 11.5 kb. (C) Hind III-digested DNA probed with J H • The germline band is 11.5 kb. Rearranged bands are indicated by asterisks.
c
B
2
2
2
11.5kb-
19k b-
11.5kb-
*
crease in size of the rearranged allele, in conjunction with germline configuration of the J1- switch region, suggests that the rearrangement involved a small deletion within or close to the JH gene segments. We have previously reported a similar rearrangement pattern in T cell acute lymphoblastic leukemia (7) and acute myelogenous leukemia (4). The same rearrangement pattern has also been observed by Mizutani and associates (10) in a number of T cell leukemias and in a few immature B cell leukemias. Detailed analysis of these B cell leukemias showed rearrangement involving the DH segment (DQS2) nearest to the JHgene segments. Our observations are compatible with such a DHJ H recombination. DH-J Hrecombinations are rarely observed because they are transient and the subsequent step, rearrangement of VHto DHJ H, usually occurs soon after, during maturation of the B cell subpopulation. Examination of immunoglobulin gene rearrangement patterns in more than 30 independently derived B cell lines from normal individuals (smokers and nonsmokers) has not revealed this unique pattern of rearrangement. The expansion of cells containing this immature DHJ H rearrangement appears to be linked to cigarette smoking. In one patient, abnormal lymphocyte numbers and morphology were observed when she smoked. However, when she abstained, blood counts and lymphocyte morphology reverted to normal. This phenomenon correlated well with the DNA analysis. A more prominent band representing the rearranged DHJ Hgenes was observed in the DNA sample taken at the time she had resumed smoking than was observed in the DNA sample taken after she had stopped for several
months. It was interesting that when the population of cells was reexamined after the patient had resumed smoking, the rearranged band was less distinct (Figure 1, lane 3) than observed earlier (not shown) and less distinct when compared with the rearranged band from the other patient who was continually a heavy smoker. Because the band was less distinct after reexpansion, it is possible that some individual evolution occurred with the DHJ H rearranged population, resulting in small changes in the size of the rearranged bands. This evolution could have taken the form of additional (upstream) D Hto JHrearrangements. The intensity of the DHJ H rearranged band makes it unlikely that the small percentage (l to 2 %) of binucleate cells could account entirely for the rearranged band. It is possible that the close doublet is delineated by two populations of cells. The binucleate cells may represent either precursor cells that develop into the expanded B cell population or a small subset of the cells that are undergoing expansion. Alternatively, heavy smoking by HLA-DR7+ females could result in expansion of two separate cell populations: one defined by increased size and binucleate morphology and the other defined by a rearranged DHJ Hband which is characteristic of certain immature B cells. There are at least two possibilities that could explain the appearance of these cells in susceptible individuals. First, smoking may elict expansion of a subpopulation of cells that have both alleles rearranged to DHJ H and, unlike normal cell populations, are maintained in this state during the period of smoking. Because such a population has not completed the first series of rearrangements (VHto DHJ H), the
552
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 2 1990
recombinational apparatus may remain continually active and render the cells prone to secondary, potentially abnormal rearrangements. A second, less likely, possibility is that the DHJH rearrangement occurs on one allele of every peripheral B cell in these patients. The other allele may be rearranged normally. This type of cell may potentially be more stable than the first population. The changes that occur in the B cells of these HLA-DR7+ female patients seem to be, at least in part, driven by smoking, and this raises the possibility that smoking may elicit or stabilize certain subpopulations. Such an effect could be precursor to further abnormalities. Acknowledgments: This work was supported by grants from the National Institutes of Health, National Cancer Institute (Canada), the Medical Research Council (Canada), the Addiction Research Foundation (Toronto), and the Ontario Cancer Treatment Research Fund. Dr. Gelfand is a scholar of the Raymond and Beverly Sackler Foundation.
References 1. Gordon, D. S., B. M. Jones, S. W. Browning et al. 1982. Persistent polyclonal lymphocytosis of B lymphocytes. N. Engl. J. Med. 307:232-236.
2. Carstairs, K. c., W. H. Francombe, J. G. Scott, and E. W. Gelfand. 1985. Persistent polyclonal lymphocytosis of B lymphocytes, induced by cigarette smoking? Lancet i:1094. 3. Cheung, R. K., S. Grinstein, and E. W. Gelfand. 1982. Volume regulation by human lymphocytes. Identification of differences between the two major lymphocyte subpopulations. J. Clin. Invest. 70:632-638. 4. Ha, K., M. Minden, N. Hozumi, and E. W. Gelfand. 1984. Immunoglobulin gene rearrangement in acute myelogenous leukemia. Cancer Res. 44: 4658--4660. 5. Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA restriction fragments to high specific activity. Anal. Biochem. 132: 6-13. 6. Ravetch, J. V., U. Siebenlist, S. Korsmeyer, T. Waldmann, and P. Leder. 1981. Structure of the human immunoglobulin J1. locus: characterization of embryonic and rearranged J and D genes. Cell 27:583-591. 7. Ha, K., M. Minden, N. Hozumi, and E. W. Gelfand. 1984. Immunoglobulin J1. chain gene rearrangement in a patient with T cell acute lymphoblastic leukemia. J. Clin. Invest. 73:1232-1236. 8. Korsmeyer, S. J., A. Arnold, A. Bakhshi et al. 1983. Immunoglobulin gene rearrangement and cell surface antigen expression in acute lymphocytic leukemia of T cell and B cell precursor origin. J. Clin. Invest. 71:301-313. 9. Ford, A. M., H. V. Melgaard, M. F. Greaves, andH. J. Gould. 1983. Immunoglobulin gene organization in haemopoietic stem cell leukemia. EMBO J. 2:997-1001. 10. Mizutani, S., A. M. Ford, L. M. Wiedemann et al. 1986. Rearrangement of immunoglobulin heavy chain genes in human T leukemic cells shows preferential utilization of the D segment (DQS2) nearest to the J region. EMBO J. 5:3467-3473.
