Molecularly cloned SHIV-1157ipd3N4: a highly replication- competent, mucosally transmissible R5 simian-human immunodeficiency virus encoding HIV clade C Env.

Human immunodeficiency virus type 1 (HIV-1) clade C causes >50% of all HIV infections worldwide, and an estimated 90% of all transmissions occur mucosally with R5 strains. A pathogenic R5 simian-human immunodeficiency virus (SHIV) encoding HIV clade C env is highly desirable to evaluate candidate AIDS vaccines in nonhuman primates. To this end, we generated SHIV-1157i, a molecular clone from a Zambian infant isolate that carries HIV clade C env. SHIV-1157i was adapted by serial passage in five monkeys, three of which developed peripheral CD4(+) T-cell depletion. After the first inoculated monkey developed AIDS at week 137 postinoculation, transfer of its infected blood to a naïve animal induced memory T-cell depletion and thrombocytopenia within 3 months in the recipient. In parallel, genomic DNA from the blood donor was amplified to generate the late proviral clone SHIV-1157ipd3. To increase the replicative capacity of SHIV-1157ipd3, an extra NF-kappaB binding site was engineered into its 3' long terminal repeat, giving rise to SHIV-1157ipd3N4. This virus was exclusively R5 tropic and replicated more potently in rhesus peripheral blood mononuclear cells than SHIV-1157ipd3 in the presence of tumor necrosis factor alpha. Rhesus macaques of Indian and Chinese origin were next inoculated intrarectally with SHIV-1157ipd3N4; this virus replicated vigorously in both sets of monkeys. We conclude that SHIV-1157ipd3N4 is a highly replication-competent, mucosally transmissible R5 SHIV that represents a valuable tool to test candidate AIDS vaccines targeting HIV-1 clade C Env.

Observed trends for CF3-containing compounds in background air at Cape Meares, Oregon, Point Barrow, Alaska, and Palmer Station, Antarctica.

The concentrations of CF(3)-containing compounds in archived air samples collected at Cape Meares, Oregon, from 1978 to 1997, at Point Barrow, Alaska, from 1995 to 1998, and at Palmer Station, Antarctica, from 1991 to 1997, were determined by high resolution gas chromatography and high resolution mass spectrometry. The CF(3)-containing compounds measured by this method and discussed here are: the perfluorinated compound, C(3)F(8) (FC 218); four perhalogenated compounds, CF(3)Cl (CFC 13), CF(3)CF(2)Cl (CFC 115), CF(3)CFCl(2) (CFC 114a), and CF(3)Br (Halon 1301); and three hydrofluorocompounds, CF(3)H (HFC 23), CF(3)CH(3) (HFC 143a), and CF(3)CH(2)F (HFC 134a). For four of these compounds, very few measurements have been previously reported. The atmospheric concentrations of all of the CF(3)-containing compounds continuously increased in time over the sample collection periods. From these data, the annual rates of emission into the atmosphere have been estimated. The emission rates fall into one of three distinct categories. The annual emission rates of C(3)F(8), CF(3)H, CF(3)CH(3), and CF(3)CH(2)F have continuously increased over the last two decades. That of CF(3)CFCl(2) has decreased continuously. Emission rates for CF(3)Cl, CF(3)CF(2)Cl, and CF(3)Br reached maximum levels in the late 1980s, and have been decreasing in the 1990s. The emission rates of C(3)F(8), CF(3)CH(3) and CF(3)CH(2)F were nearly zero 20 years ago but have increased rapidly during the last decade.

Characterization of non-methane hydrocarbons emitted from various cookstoves used in China.

Emission contributions from cookstoves to indoor, regional, and global air pollution largely depend on stove and fuel types. This paper presents a database on emission factors of speciated non-methane hydrocarbons (NMHCs) for 16 fuel/stove combinations burning 2 types of crop residue, wood, 4 types of coal, kerosene, and 3 types of gaseous fuels. The emission factors are presented both on a fuel mass basis (compound mass per fuel mass) and on a cooking task basis (compound mass per unit energy delivered to the pot). These fuel/stove combinations cover a large spectrum of the cookstoves used in both urban and rural households in China. Up to 54 hydrocarbons were identified, some of which are reactive precursors of photochemical smog. Based on published maximum incremental reactivity (MIR) values for NMHCs, we estimated stove-specific and fuel-specific ozone forming potentials (OFPs). The results indicate that raw coal powder, wood, and crop residues have higher OFP values than the other types of fuels tested. Strikingly, burning the coal briquette and honeycomb coal briquette produced OFP values more than 2 orders of magnitude lower than burning unprocessed (raw) coal, even in the same vented metal stove, for every 1 MJ delivered to the pot.

Postnatal pre- and postexposure passive immunization strategies: protection of neonatal macaques against oral simian-human immunodeficiency virus challenge.

Simian-human immunodeficiency viruses (SHIV) allow the evaluation of antiviral strategies that target the envelope glycoproteins of the human immunodeficiency virus 1 (HIV-1) in macaques. We previously protected neonates from oral challenge with cell-free SHIV-vpu+ by passive immunization with synergistic human neutralizing monoclonal antibodies (mAbs) (Baba et al., Nat Med 6:200-206, 2000). mAbs were administered prenatally to pregnant dams and postnatally to the neonates. Here, we used solely postnatal or postexposure mAb treatment, thus significantly reducing the amount of mAbs necessary. All neonatal monkeys were also protected with these abbreviated mAb regimens. Our results are directly relevant for humans because we used mAbs that target HIV-1 envelope glycoproteins. Thus, the large-scale use of passive immunization with neutralizing mAbs may be feasible in human neonates. The mAbs, being natural human proteins, can be expected to have low toxicity. Passive immunization has promise to prevent intrapartum as well as milk-borne virus transmission from HIV-1-infected women to their infants.

Atmospheric nitrous oxide: patterns of global change during recent decades and centuries.

Data from weekly global measurements of nitrous oxide from 1981 to the end of 1996 are presented. The results show that there is more N2O in the northern hemisphere by about 0.7 +/- 0.04 ppbv, and the Arctic to Antarctic difference is about 1.2 +/- 0.1 ppbv. Concentrations at locations influenced by continental air are higher than at marine sites, showing the existence of large land-based emissions. For the period studied, N2O increased at an average rate of about 0.6 ppbv/year (approximately 0.2%/year) although there were periods when the rates were substantially different. Using ice core data, a record of N2O can be put together that goes back about 1000 years. It shows pre-industrial levels of about 287 +/- 1 ppbv and that concentrations have now risen by about 27 ppbv or 9.4% over the last century. The ice core data show that N2O started increasing only during the 20th century. The data presented here represent a comprehensive view of the present global distribution of N20 and its historical and recent trends.

Heterologous neutralizing antibody induction in a simian-human immunodeficiency virus primate model: lack of original antigenic sin.

According to the principle of original antigenic sin, neutralizing antibodies (NAbs) initially directed against a single virus strain compromise the immune system's ability to subsequently mount adequate responses against antigenically divergent virus strains. In this study, rhesus macaques, after vaccination and breakthrough infection with homologous simian-human immunodeficiency virus (SHIV), developed strong SHIV-IIIB strain-directed NAb responses that were mostly V3 loop specific. After superinfection with heterologous SHIV89.6P, all macaques developed high-titer SHIV89.6P-specific NAbs without significant boosting of SHIV-IIIB-specific NAbs. These results indicate that prior B cell responses against a single immunodeficiency virus strain do not preclude the later development of NAbs against a divergent strain of the same virus.

Protection of neonatal macaques against experimental SHIV infection by human neutralizing monoclonal antibodies.

Neonatal macaques were completely protected against oral challenge with SHIV-vpu+, a simian-human immunodeficiency virus that encodes the envelope gene of a laboratory-adapted HIV strain, by pre- and post-natal treatment with a triple combination of human neutralizing monoclonal antibodies (mAbs). The mAbs were directed either against the CD4 binding site, a glycosylation-dependent gp120 epitope, or against a linear epitope on gp41. This triple combination was highly synergistic in vitro and neutralized primary HIV completely. Subsequently, oral challenge was performed with pathogenic SHIV89.6P, an animal-passaged variant of a chimeric virus that encodes the envelope gene of the primary, dual-tropic HIV89.6. Only post-natal treatment with a similar triple mAb combination was used. One out of 4 mAb-treated infants was completely protected from infection. In the other 3 treated animals, there was a tendency towards lower peak viral RNA loads compared with untreated controls. Two out of 4 mAb-treated infants maintained normal CD4+ T-cell numbers, in contrast to all controls that had steep declines at 2 weeks post-challenge. We conclude that the triple mAb combination significantly protected the neonates, even against mucosal challenge with pathogenic SHIV89.6P. Passively administered synergistic human mAbs may play a role in preventing mother-infant transmission of HIV, both against intrapartum transmission as well as against infection through breast milk. As passive immunization is a tool to assess correlates of immune protection, we conclude that the epitopes recognized by the mAbs in our combinations are important for AIDS vaccine development. Future passive immunization studies may reveal other important conserved epitopes.

Passive immunization against oral AIDS virus transmission: an approach to prevent mother-to-infant HIV-1 transmission?

To develop immunoprophylaxis regimens against mother-to-child human immunodeficiency virus type 1 (HIV-1) transmission, we established a simian-human immunodeficiency virus (SHIV) model in neonatal macaques that mimics intrapartum mucosal virus exposure (T.W. Baba, J. Koch, E.S. Mittler et al: AIDS Res Hum Retroviruses 10:351-357, 1994). We protected four neonates from oral SHIV-vpu+ challenge by ante- and postpartum treatment with a synergistic triple combination of immunoglobulin (Ig) G1 human anti-HIV-1 neutralizing monoclonal antibodies (mAbs) (T.W. Baba, V. Liska, R. Hofmann-Lehmann et al: Nature Med 6:200-206, 2000), which recognize the CD4-binding site of Env, a glycosylation-dependent gp120, or a linear gp41 epitope. Two neonates that received only postpartum mAbs were also protected from oral SHIV-vpu+ challenge, indicating that postpartum treatment alone is sufficient. Next, we evaluated a similar mAb combination against SHIV89.6P, which encodes env of primary HIV89.6. One of four mAb-treated neonates was protected from infection and two maintained normal CD4+ T-cell counts. We conclude that the epitopes recognized by the three mAbs are important determinants for achieving protection. Combination immunoprophylaxis with synergistic mAbs seems promising to prevent maternal HIV-1 transmission in humans.

Postnatal passive immunization of neonatal macaques with a triple combination of human monoclonal antibodies against oral simian-human immunodeficiency virus challenge.

To develop prophylaxis against mother-to-child human immunodeficiency virus (HIV) transmission, we established a simian-human immunodeficiency virus (SHIV) infection model in neonatal macaques that mimics intrapartum mucosal virus exposure (T. W. Baba et al., AIDS Res. Hum. Retroviruses 10:351-357, 1994). Using this model, neonates were protected from mucosal SHIV-vpu(+) challenge by pre- and postnatal treatment with a combination of three human neutralizing monoclonal antibodies (MAbs), F105, 2G12, and 2F5 (Baba et al., Nat. Med. 6:200-206, 2000). In the present study, we used this MAb combination only postnatally, thereby significantly reducing the quantity of antibodies necessary and rendering their potential use in humans more practical. We protected two neonates with this regimen against oral SHIV-vpu(+) challenge, while four untreated control animals became persistently infected. Thus, synergistic MAbs protect when used as immunoprophylaxis without the prenatal dose. We then determined in vitro the optimal MAb combination against the more pathogenic SHIV89.6P, a chimeric virus encoding env of the primary HIV89.6. Remarkably, the most potent combination included IgG1b12, which alone does not neutralize SHIV89.6P. We administered the combination of MAbs IgG1b12, 2F5, and 2G12 postnatally to four neonates. One of the four infants remained uninfected after oral challenge with SHIV89.6P, and two infants had no or a delayed CD4(+) T-cell decline. In contrast, all control animals had dramatic drops in their CD4(+) T cells by 2 weeks postexposure. We conclude that our triple MAb combination partially protected against mucosal challenge with the highly pathogenic SHIV89.6P. Thus, combination immunoprophylaxis with passively administered synergistic human MAbs may play a role in the clinical prevention of mother-to-infant transmission of HIV type 1.

1999: a time to re-evaluate AIDS vaccine strategies.

The field of AIDS vaccine development is in flux. Important new findings were reported in 1999 that led to a rethinking of AIDS vaccine strategies. We have been given the challenging task of providing an overview. Rather than attempting to provide a comprehensive summary, we will restrict our discussion to a few major topics, and we ask for understanding if we can only highlight.

Soil-atmosphere exchange of radiatively and chemically active gases.

Exchanges between the soils and the atmosphere may control or significantly affect the global budgets of many environmentally important trace gases, both natural and man-made. Flux measurements, taken in several ecosystems, show that soils are a substantial source of chloroform (8 +/- 4 microg/m(2)/d) and a sink for methyl chloride (-10(-3)(+6) microg/m(2)/d). The known sources and sinks of these gases are insufficient to explain the observed concentrations. Our findings will help to balance the global budget of chloroform but may put the budget of methyl chloride further out of balance. We also found, consistent with previous research, that soils are a substantial source of nitrous oxide and carbon monoxide and take up hydrogen and methane. The uptake of man-made chlorocarbons was observed, but the rates are small. Observed fluxes of non-methane hydrocarbons showed few patterns except that soils may be a source of ethane and butane.

Randomized crossover study comparing the phosphate-binding efficacy of calcium ketoglutarate versus calcium carbonate in patients on chronic hemodialysis.

The objective of the study was to evaluate the phosphate-binding efficacy, side effects, and cost of therapy of calcium ketoglutarate granulate as compared with calcium carbonate tablets in patients on chronic hemodialysis. The study design used was a randomized, crossover open trial, and the main outcome measurements were plasma ionized calcium levels, plasma phosphate levels, plasma intact parathyroid hormone (PTH) levels, requirements for supplemental aluminum-aminoacetate therapy, patient tolerance, and cost of therapy. Nineteen patients on chronic hemodialysis were treated with a dialysate calcium concentration of 1.25 mmol/L and a fixed alfacalcidol dose for at least 2 months. All had previously tolerated therapy with calcium carbonate. Of the 19 patients included, 10 completed both treatment arms. After 12 weeks of therapy, the mean (+/-SEM) plasma ionized calcium level was significantly lower in the ketoglutarate arm compared with the calcium carbonate arm (4.8+/-0.1 mg/dL v 5.2+/-0.1 mg/dL; P = 0.004), whereas the mean plasma phosphate (4.5+/-0.3 mg/dL v 5.1+/-0.1 mg/dL) and PTH levels (266+/-125 pg/mL v 301+/-148 pg/mL) did not differ significantly between the two treatment arms. Supplemental aluminum-aminoacetate was not required during calcium ketoglutarate treatment, while two patients needed this supplement when treated with calcium carbonate. Five of 17 (29%) patients were withdrawn from calcium ketoglutarate therapy within 1 to 2 weeks due to intolerance (anorexia, vomiting, diarrhea, general uneasiness), whereas the remaining 12 patients did not experience any side effects at all. The five patients with calcium ketoglutarate intolerance all had pre-existing gastrointestinal symptoms; four of them had received treatment with cimetidine or omeprazol before inclusion into the study. Calculations based on median doses after 12 weeks showed that the cost of the therapy in Denmark was 10 times higher for calcium ketoglutarate compared with calcium carbonate (US$6.00/d v US$0.65/d). Calcium ketoglutarate may be an effective and safe alternative to treatment with aluminum-containing phosphate binders in patients on hemodialysis who are intolerant of calcium carbonate or acetate because of hypercalcemia. However, care must be exercised when dealing with patients with pre-existing gastrointestinal discomfort. Due to the high cost of the therapy, calcium ketoglutarate should be used only for selected patients.

A method for collection, long-term storage, and bioassay of labile volatile chemosignals.

A procedure for headspace sampling and long-term storage of organic volatiles coupled with gas chromatographic-mass spectrometric (GC-MS) analysis was used to study the volatile chemosignals in a biological secretion prior to bioassay. The approach involved collecting the volatiles in evacuated canisters from an apparatus in which 1 ml of secretion was dispersed for headspace sampling. These canisters, stainless steel, 850 ml, and 100% internally electropolished, have been demonstrated to store volatile compounds, in chemically stable form, for several weeks. The GC-MS analyses provided the quantitation and identification of compounds from C3 through C14 at concentrations as low as 0.10 parts per billion volume. The approach was used to study chemosignals of musth temporal gland secretions (TGS) from a male Asian elephant (Elephas maximus). Fresh TGS material loses its biological activity within 1 hr. TGS material stored at -20°C usually looses its activity within 30 days. The usefulness of this method for long-term storage of the volatile chemosignals was demonstrated by the retention of biologically active TGS headspace compounds, as determined through bioassays, stored in these canisters for one year.

Monocytes are required to prime peripheral blood T cells to undergo apoptosis.

Freshly isolated, human peripheral blood T (PBT) cells are largely resistant to the apoptotic effects of anti-CD3 monoclonal antibody, ionomycin, or phorbol 12-myristate 13-acetate (PMA). We demonstrate here, however, that PBT cells, including both CD4+ and CD8+ cell populations, can be readily induced to undergo apoptosis when cocultured with either autologous or allogeneic monocytes (Mo) in PMA-containing medium. Incubation of PBT cells with Mo at a ratio of 1:1 for 18 hr resulted in maximal levels (80%) of apoptotic cell death. The mechanism whereby Mo enable PBT cells to undergo apoptosis in PMA-containing medium appeared to depend on cell-cell contact or close proximity between Mo and PBT cells rather than solely via soluble mediators. It was demonstrated that Mo acquire the ability to prime PBT cells for apoptosis after treatment with PMA and that treated Mo maintain this ability even after fixation with formaldehyde. It was also found that once PBT cells became primed for apoptosis by incubation with PMA-pretreated Mo, the primed PBT cells were susceptible to apoptosis triggered not only by PMA but also by either ionomycin or by monoclonal antibody crosslinking of T-cell surface molecules such as CD4 and CD3. Interestingly, the degree of apoptosis of CD4+ T cells by crosslinking of CD4 molecules via a combination of gp120, anti-gp120, and goat anti-mouse IgG was significantly greater for T cells primed with PMA-treated Mo than for unprimed T cells. Together, these findings reveal an important role for accessory cells in priming resting PBT cells for apoptosis and suggest a possible Mo-dependent mechanism by which T cells may become primed for apoptosis in human immunodeficiency virus-infected asymptomatic individuals.

Anti-CD38-blocked ricin: an immunotoxin for the treatment of multiple myeloma.

We report the development of a potent anti-CD38 immunotoxin capable of killing human myeloma and lymphoma cell lines. The immunotoxin is composed of an anti-CD38 antibody HB7 conjugated to a chemically modified ricin molecule wherein the binding sites of the B chain have been blocked by covalent attachment of affinity ligands (blocked ricin). Conjugation of blocked ricin to the HB7 antibody has minimal effect on the apparent affinity of the antibody and no effect on the ribosome-inactivating activity of the ricin A-chain moiety. Four to six logs of CD38+ tumor cell line kill was achieved at concentrations of HB7-blocked ricin in the range of 0.1 to 3 nmol/L. Low level of toxicity for normal bone marrow (BM) granulocyte-macrophage colony-forming units (CFU-GM), burst-forming units-erythroid (BFU-E), colony-forming units-granulocyte/erythroid/monocyte/macrophage (CFU-GEMM) cells was observed. Greater than two logs of CD38+ multiple myeloma cells were depleted from a 10-fold excess of normal BM mononuclear cells (BMMCs) after an exposure to HB7-blocked ricin under conditions (0.3 nmol/L) that were not very toxic for the normal BM precursors. HB7-blocked ricin was tested for its ability to inhibit protein synthesis in fresh patients' multiple myeloma cells and in normal BMMCs isolated from two healthy volunteers; tumor cells from four of five patients were 100-fold to 500-fold more sensitive to the inhibitory effect of HB7-blocked ricin than the normal BM cells. HB7 antibody does not activate normal resting peripheral blood lymphocytes, and HB7-blocked ricin is not cytotoxic toward these cells at concentrations of up to 1 nmol/L. The potent killing of antigen-bearing tumor cells coupled with a lack of effects on peripheral blood T cells or on hematopoietic progenitor cells suggests that HB7-blocked ricin may have clinical utility for the in vivo or in vitro purging of human multiple myeloma cells.

Anti-CD6-blocked ricin: an anti-pan T-cell immunotoxin.

We report the development of a potent anti-pan T-cell immunotoxin capable of killing cells in an antigen dependent manner. The immunotoxin is composed of a high affinity anti-CD6 antibody (IgG2a, Kd = 1.3 x 10(-11) M) conjugated to blocked ricin that is a chemically modified ricin molecule wherein the lectin binding sites of the B-chain have been blocked by covalent attachment of affinity ligands. Conjugation of blocked ricin to the antibody has minimal effect on the apparent affinity of the antibody and no effect on the ribosome-inactivating activity of the ricin A-chain moiety. Anti-CD6-blocked ricin is a specific and highly toxic immunoconjugate killing the antigen-positive Molt-4 cell line with an IC37 of 4 x 10(-12) M after a 24 h exposure of cells to the immunotoxin. Nonspecific cytotoxicity of anti-CD6-blocked ricin for the antigen-negative Namalwa cell line was more than 750-fold lower with an IC37 > 3 x 10(-9) M. The cytotoxicity of anti-CD6-blocked ricin is dependent on the length of the incubation of cells with the conjugate ranging from an IC37 of 1.5 x 10(-11) M leaving a surviving fraction of Molt-4 cells of 0.03 after a 2.5 h exposure to an IC37 of 5 x 10(-13) M and leaving a surviving fraction of 3 x 10(-6) after a continuous (3 weeks) exposure. Anti-CD6-blocked ricin is also capable of killing CD6 positive cells in human peripheral blood lymphocyte populations. Systemic toxicity of anti-CD6-blocked ricin in mice is similar to the toxicity of other immunotoxins containing blocked ricin that were found to be tolerated well by patients. An application of this immunoconjugate for the prevention and treatment of graft versus host disease or tissue graft rejection is suggested.

Isolation and characterization of CD6- T cells from peripheral blood.

Antibodies to the CD6 Ag have been described as having pan-T cell reactivity. We have recently demonstrated, however, that after treatment of PBL with an anti-CD6-blocked ricin-conjugated immunotoxin, clonal populations of CD3+, CD6- cells can be identified. Herein we show that through dual parameter staining of freshly isolated E-rosette+ cells, an average of 5 to 6% of either CD3+ or CD5+ cells express little or no CD6 on their surface. After negative selection by antibody-coated paramagnetic bead depletion, expanded CD6- T cells were shown to be CD1a-, CD2+, CD3+, CD5+, CD16-, CD56-, TCR-gamma/delta-, and consisted of both CD4+ and CD8+ cells. Furthermore, staining of digitonin permeabilized cells showed no cytoplasmic expression of the CD6 Ag and CD6 mRNA was not detected by Northern blot analysis. Identical staining patterns were observed for T cell clones isolated through bead depletion or immunotoxin treatment and expanded with either PHA or immobilized anti-CD3 mAb. It was also found that, relative to unfractionated T cells, the surface expression of CD5 was significantly diminished on CD6- T cells. Functionally, freshly isolated CD6- T cells showed substantially reduced alloreactivity in MLR compared with unfractionated E-rosette+ cells, yet both gave similar proliferative responses to either PHA or soluble tetanus toxin Ag. We conclude that there exists a minor subpopulation of mature T cells in peripheral blood that lack CD6. The diminished alloreactivity of these cells may help to explain the low incidence of graft-vs-host disease, despite high levels of engraftment, that has been reported in allogeneic bone marrow transplant patients receiving marrow treated with anti-CD6 (T12) mAb plus C'.

Developmental regulation of a mucinlike glycoprotein selectively expressed on natural killer cells.

Natural killer (NK) cells are CD3:TCR-, CD16+, CD56+ large granular lymphocytes capable of recognizing and eliminating a variety of virus-infected, malignant, and antibody-coated target cells. Two functionally distinct populations of peripheral blood NK cells can be differentiated by their surface expression of an isoform of the neural cell adhesion molecule (CD56). CD56bright NK cells have the attributes of an undifferentiated cell, in that they proliferate in response to exogenous cytokines, but exert poor cytolytic activity. CD56dim NK cells have the attributes of a more differentiated cell, in that they proliferate poorly in response to exogenous cytokines, but are potent cytolytic effector cells. Here we describe the molecular characterization of a NK cell restricted epitope (PEN5) that is selectively expressed on the functionally differentiated CD56dim NK cells. PEN5+ NK cells proliferate poorly in response to interleukin 2 (IL-2), but are potent cytolytic effectors, whereas PEN5- NK cells proliferate in response to IL-2, but are poor cytolytic effectors. Biochemical and immunochemical analyses reveal the PEN5 epitope to be an unusual sulfated poly-N-lactosamine carbohydrate related to keratan sulfate glycosaminoglycans. Immunoprecipitates prepared using a monoclonal antibody reactive with PEN5 include two polydisperse membrane-bound glycoproteins, PEN5 alpha (120-170 kD) and PEN5 beta (210-245 kD). Enzymatic deglycosylation reduces the apparent molecular weight of both PEN5 isoforms by 80-90%, and classifies PEN5 beta as a mucinlike glycoprotein. The surface expression of the PEN5 epitope is downmodulated by stimuli that induce NK cell proliferation, and it is absent from leukemic NK cells of patients with granular lymphocyte proliferative disorder. Taken together, these results indicate that PEN5 is a developmentally regulated poly-N-lactosamine epitope associated with a mucin-type glycoprotein, whose expression is restricted to the population of nonproliferative NK cells fully committed to cytolytic effector function.