Mouse Strain Variability in the Expression of the ...

From www.bloodjournal.org by guest on May 10, 2017. For personal use only.

Mouse Strain Variability in the Expression of the Hematopoietic Stem Cell Antigen Ly-6A/E by Bone Marrow Cells By Gerald J. Spangrude and Diane M. Brooks The cell surface molecule Ly-GA/E provides a convenient marker for primitive stem cells in the hematopoietic tissues of both fetal and adult mice. However, previous studies have shown that Ly-GA/E expression by lymphocytes is variable depending on the haplotype of the Ly-6 locus. Therefore, strain-specific variation in Ly-GA/E expression by bone marrow (BM) cells was investigated. The results show that Ly-6" mice have, on average, 50% of the number of BM cells expressing Ly-GA/E relative to that for Ly-sb mice. Furthermore, among the 5% of BM cells that do not express antigens characteristic of mature T, B, myeloid, or erythroid lineages, which include the primitive hematopoietic stem cell compartment, Ly-6" mice have, on average, more than fivefold fewer Ly-GA/E+ cells relative to that for Ly-sbmice. Isolation of Ly-GA/E- and Ly-GA/E+ cells from mice of both haplotypes showed that, whereas 99%of the marrow repopulating activity (MRA) of C57BL/Ka (Ly-sb) mice could be recovered in the Ly-GA/E+ fraction, only about 25% of the MRA of BALB/c (Ly-6") was recoverable in the same population. On a per-cell basis, the Ly-GA/E+ cells that were isolated from BALB/c mice were essentially

equivalent in MRA t o those isolated from C57BL/Ka mice. Thus, whereas a large percentage of the hematopoietic stem cells of Ly-6' mice do not express the Ly-GA/E molecule, the antigen may be used t o isolate a subset of stem cells from these mice. These results show that hematopoietic stem cell phenotype can vary between mouse strains and imply that caution should be exercised in the identification of human stem cell antigens such as CD34, because a similar variability may occur between individual humans. To further explore the influence of Ly-6 haplotype on Ly6A/E expression by specific cell subsets, lymph-node lymphocytes from a panel of mouse strains were analyzed by multiparameter flow cytometry for correlated expression of Ly-GA/E. CD4, and CD8. All Ly-6' strains examined had less than 20% Ly-GA/E+ cells, and those cells were predominantly CD8+ T lymphocytes. In contrast, the Ly-sb strains had greater than 30% Ly-GA/E+ cells, and those cells included CD4+, CD8+, and B lymphocytes. This is a US government work. There are no restrictions on its use.

T

between the two parental phenotypes and the IFN-y induc~ ~ parent.I6 tion being more similar to the L Y -haplotype Previous studies have shown that virtually all hematopoietic stem cells of C57BL mice ( L Y - ~express ~) Ly-6A/E.",'* However, in light of the influence of Ly-6 haplotype on basal levels of Ly-6AfE expression by lymphocytes, it seemed possible that the cells comprising the hematopoietic stem cell compartment of Ly-6" haplotype mice may not express Ly-6A/E. The present experiments were conducted to determine the answer to this question.

HE MOUSE stem cell antigen-] (Sca-1) was identified as a marker for hematopoietic stem cells based on functional analysis of Sca- I cells isolated by fluorescenceactivated cell sorting (FACS).Is2Subsequent investigation of the molecular characteristics of the Sca- 1 molecule showed it to be a member of the Ly-6 antigen family, a group of molecules expressed primarily by lymphocytes but present on other cells as welL3v4Based on a variety of biochemical analyses, including immunoprecipitation and cross-blocking studies with defined anti-Ly-6 monoclonal antibodies (MoAbs), the Sca-1 specificity was shown to be identical to LY-~A/E.~ The murine Ly-6 locus controls the expression of a family of phosphatidylinositol-anchored membrane proteins.6 The various members of the family differ in tissue distribution but are highly related at the nucleic acid and protein levels.'-'' Although 6 members of the Ly-6 family were defined by serology and tissue distribution (Ly-6A, B, C, D, E, and ThB), some of these specificities may be carried by single proteins or are allelic variants. For example, MoAbs specific for the Ly-6D.2 determinant react with cells transfected by a cDNA clone of the Ly-6A.2 molecule.'' Furthermore, the Ly-6E. I and Ly-6A.2 antigens were originally identified as distinct specificities because of dramatic differences in tissue distribution but, on cloning, were found to differ by only two amino acids.* Because the Ly-6E. 1 and Ly-6A.2 specificities are expressed by the Ly-6" and L Y - haplo~ ~ types, respectively, it has been proposed that they represent allelic variants ofthe same molecule (now termed Ly-6AfE) that are under distinct genetic ~ o n t r o l . ' ~The . ' ~ Ly-6A/E proteins are both induced by interferon-y (IFN-y),I4 but basal levels of expression differ in that Ly-6E. 1 is normally expressed by 10% to 15% of peripheral lymphoid cells, whereas Ly-6A.2 is expressed by 50% to 70%.'3Js(Ly-6" X L Y - ~F,~ hybrid ) mice are distinct from either parent, with the basal level of Ly-6AfE expression being intermediate +

Blood, Vol82, No 1 1 (December l ) , 1993:pp 3327-3332

MATERIALS AND METHODS

Animals. C57BL/Ka, C57BL/Ka-Thy-l.l, C57BL/6-a-17, AKR/J, and SJL/J mice were bred and maintained in the animal facility at the National Institute of Allergy and Infectious Diseases (Rocky Mountain Laboratories, Hamilton, MT). A/J, BALB/c, C3H/J, CBA/J, and C57BL/6J mice were purchased from Jackson Labs (Bar Harbor, ME). All animals were maintained on acidified (pH 2.5) drinking water and autoclaved chow (Purina Mills Inc, St Louis, MO) ad libitum.

From the Laboratory of Persistent Viral Diseases, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Hamilton, MT. Submitted May 17, 1993; accepted August 6, 1993. The contents of this publication do not necessarily reflect the views or policies of the Department of Health and Human Services, nor does the mention of trade names, commercial products, or organizations imply endorsement by the US government. Address reprint requests to Gerald J. Spangrude, PhD, Rocky Mountain Laboratories, 903 S 4th St, Hamilton, M T 59840. The publication costs of this article were defayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C.section I734 solely to indicate this fact. This is a USgovernment work. Thereare no restrictions on its use. 0006-4971/93/821I-0019$0.00/0 3327


3328

Cell preparation. All manipulations of cells were performed using Hanks' Balanced Salt Solution containing 3% fetal calf serum and IO mmol/L HEPES buffer, pH 7.2 (HBSS). Bone marrow (BM) cells were prepared from young adult mice (4 to 8 weeks old) by crushing femora and tibia in HBSS with a mortar and pestle. Lowdensity cells were enriched by equilibrium centrifugation on a cushion of metrizamide (Nycomed analytical grade; Accurate Chemical and Scientific Corp, Westbury, NY) at a density of 1.085 g/mL. Lymph node cells were prepared from pooled inguinal, axillary, brachial, and cervical lymph nodes by gentle teasing with forceps followed by repeated pipetting and unit gravity sedimentation to remove connective tissue and to give a single cell suspension. Immunomagnetic depletions. Low-density BM cells at a cell density of 5 X IO7 cells/mL were reacted with a saturating solution of MoAb specific for antigens expressed by lymphoid, myeloid, and erythroid lineages for 20 minutes on ice, followed by a wash, as previously described.'* Twice-washed immunomagnetic particles (sheep antirat Ig specificity; Dynal Inc, Great Neck, NY) were added to the washed cells at a bead-to-cell ratio of 1:1, and the mixture was incubated for 30 minutes at 4°C with constant mixing. The particles with attached cells were then removed from the solution by application of a magnetic field, and the remaining cells were collected and concentrated by centrifugation, and a second round of bead depletion was performed. Cell staining and jlow cytometry. For analysis of Ly-6A/E expression on normal and lineage-depleted BM cells, the two populations were reacted with the lineage MoAbs followed by a fluorescein-conjugated antirat Ig reagent (Caltag Laboratories, South San Francisco, CA). After a wash, rat Ig binding sites were blocked with I O pg rat Ig (Pel-Freez, Rogers, AR) for 5 minutes, and the cells were then stained with phycoerythrin-conjugated anti-Ly-6A/E (clone E13 161-7,19conjugated to phycoerythrin using reagents obtained from Molecular Probes, Inc, Eugene, OR). Lymph node lymphocytes were reacted with phycoerythrin-CD4 (clone GKI .5), fluorescein-CD8 (clone 53-6.7), and biotin-Ly-6A/E followed by streptavidin-Red6 13 (Immunoselect; Life Technologies, Inc, Grand Island, NY). For separation of Ly-6A/E+ and Ly-6A/E- BM populations, low-density BM cells were reacted with biotin-Ly-6A/ E followed by streptavidin-fluorescein(Life Technologies, Inc). All cell populations were resuspended in HBSS containing 10 pg/mL propidium iodide and filtered through nylon mesh (pore size, 88 pm; Small Parts, Inc, Miami Lakes, FL) before analysis and/or separation by flow cytometry. A FACStar instrument, modified for 5-parameter operation (Becton Dickinson Immunocytometry Systems, San Jose, CA), was used for analysis and sorting. Ly-6A/E+ and Ly-6A/E- BM populations were reanalyzed after sorting, and the Ly-6A/E+ population was usually sorted a second time to achieve a high level of purity (>98%) in both populations. Marrow repopulation assay. BM recipients were exposed to 1 1 Gy of radiation from a 137Cs source (Mark I gamma irradiator; J.L. Sheperd and Associates, Glendale, CA) at a dose rate of I .4 Gy/ min, with the total dose being delivered in two equal fractions separated by a 3-hour rest. To assess the marrow repopulating activity (MRA) of Ly-6A/E+ and Ly-6A/E- BM populations, 1 to 2 X IO4 Ly-6A/E+ cells and 1 to 2 x IO6 Ly-6A/E- cells were transplanted by retroorbital injection under methoxyflurane anesthesia into lethally irradiated recipient mice. Thirteen days later, BM suspensions were prepared from these animals and transferred into secondary groups of irradiated animals. Usually, each BM suspension (pooled from 2 to 4 primary recipients) was transferred into 2 or 3 groups of 4 mice, with each group receiving a fivefold dilution ofthe previous group's marrow dose. Secondary recipient animals were maintained on aqueous antibiotics (Neomycin Sulfate, 2 mg/mL; AgriLabs, Omaha Vaccine CO, Omaha, NE). Thirteen days later,

SPANGRUDE AND BROOKS

the secondary groups of animals were killed, and their spleens were removed and fixed in Telleyesniczky's solution (70% ethanol-acetic acid-formalin, 20: 1 :1 by volume) before macroscopic surface colony count. The resulting colony counts were averaged, and the data were calculated to reflect the number of colony-forming unitsspleen (CFU-S) per femur of the primary recipient animal per lo6 BM cells injected. RESULTS

Variability in the expression of Ly-M/E by BM cells. To investigate the expression of the Ly-6A/E molecule by hematopoietic stem cells from various mouse strains, BM cells were isolated from a variety of mouse strains of both the Ly-6" and L Y -haplotypes. ~ ~ These cells were stained with an MoAb specific for Ly-6A/E, either before or after immunomagnetic depletion of cells expressing lineage markers. An example of Ly-6A/E staining of lineage-depleted BM from one strain of each haplotype is ~) Lyshown in Fig 1. Whereas C57BL mice ( L Y - ~express 6A/E on 4.6% of lineage-negative cells, BALB/c mice (Ly6") express the antigen on only 0.3% of these cells. Table I summarizes results from a variety of mouse strains of each haplotype. As a group, the Ly-6" haplotype animals showed expression of the Ly-6A/E molecule by 2. I % of BM cells, whereas the L Y -haplotype ~ ~ animals expressed the marker on an average of 4.7% of cells (P< .002). Because much of the expression of Ly-6A/E by BM cells can be accounted for by mature T and B lymphocytes,2 depletion of lineage-positive cells results in a lower frequency of cells expressing the antigen (Table I). However, this effect is more pronounced for Ly-6" haplotype mice, which express the antigen on less than 0.5% of lineage-negative BM, in contrast to L Y - ~ ~ haplotype mice in which an average of 3.790 oflineage-negative BM cells express the antigen (P< ,001). Variability in the expression of the antigen is evident among the various mouse strains, because C57BL strains consistently had more positive cells than the other L Y - haplotype ~ ~ strains, whereas BALB/c mice had fewer positive cells compared with other Ly-6" haplotype strains. Distribution of MRA between Ly-6A/F and Ly-6A/EBM cells. To assess the relative contributions of the Ly6A/E+ and Ly-6A/E- subsets of cells to the total MRA of normal BM, cells from BALB/c or C57BL/Ka mice were separated into the two subsets by flow cytometry, without prior lineage depletion. The Ly-6A/E+ subsets were usually sorted twice to minimize contamination by Ly-6A/E- cells, whereas the Ly-6A/E- subset was usually obtained in high purity (>98%) after one sort. These cells were transplanted into lethally irradiated syngeneic animals, and the BM of transplant recipients was harvested 13 days later for CFU-S determination. As shown in Table 2, the MRA/106 cells of C57BL/Ka mice ( L Y - ~was ~ ) concentrated in the Ly-6A/E+ cells relative to the Ly-6A/E- cells, with a ratio of 1233:l (4,070:3.3). In contrast, the MRA/106 cells of BALB/c mice (Ly-6") was split between the two populations, with about 16X more activity among Ly-6A/E+ cells compared with that for Ly-6A/E- cells (2,570: 156). Interestingly, the Ly6A/E+ cells isolated from BALB/c BM were comparable (within a factor of 2 ) with C57BL/Ka cells with respect to


3329

Ly-GA/E EXPRESSION BY MOUSE STEM CELLS 1wO

1oM)

BALBk ( L Y - ~ ~ )

1W

100

10

10

1

1

0.1

0.1

1

C57BL/Ka-Thy-l.1 (Ly-Gb)

0

Forward Scatter Fig 1. Expression of Ly-GA/E by lineage-negative BM cells. Low-density BM cells isolatedfrom BALB/c (left) or C57BL/Ka-Thy-l.l (right) mice were immunomagnetically depleted of cells expressing lymphoid, myeloid, or erythroid antigens, as described in Materials and Methods. The remaining cells were reacted with biotinylated anti-Ly-GA/E followed by fluorescein-streptavidin and propidium iodide. The plots depict 8,000 viable (propidium iodide-negative) cells plotted with respect to size (forward scatter) and expression of Ly-GA/E. The boxed areas represent cells with fluorescence above a background level defined by staining with the fluorescein-streptavidin and propidium iodide reagents only.

MRA/106cells. A significant amount of MRA was observed among Ly-6AfE- BM cells isolated from BALBfc mice, whereas very little MRA was recovered in the same population of C57BLfKa cells, in agreement with previous observations.'' When the total amount of MRA in BM of the two strains was calculated based on the relative frequencies of the two subsets of cells, 99% of C57BLfKa MRA was localized in the Ly-6A/E+population, whereas 75.6% of BALBfc MRA was localized in the Ly-6AfE- population (Table 2). Expression of Ly-6A/E by subsets of T lymphocytesfrom various mouse strains. To further investigate cell-specific regulation of Ly-6A/E expression,lymph node lymphocytes isolated from Ly-6" and L Y -haplotype ~ ~ mouse strains were evaluated for LydAfE expression in the context of the func-

Table 1. Percentage of BM Cells Expressing Ly-GA/E in Various Mouse Strains

tional T-cell antigens CD4 and CD8, which are generally associated with helper and cytotoxic effector functions, respectively. Typical results for two mouse strains, shown in Fig 2 , indicate that a rather small fraction of CBAfJ (Ly-6") lymphocytes express the Ly-6A/E antigen, and those cells that do express the antigen predominantly have the cytotoxic (CD8+) phenotype. In contrast, more than half of C57BL/Ka lymphocytes express the antigen, and these cells include more CD4+ cells relative to CD8+ cells. Similar observations were reported by Codias et This pattern was consistent for the mouse strains tested here (Table 3) in that Ly-6" haplotype strains expressed the Ly-6A/E antigen on relatively few cells (8%to 18%),and those cells were predominantly CD8+ (CD4:CD8 ratio