ARF inhibits the growth and malignant progression of non-small-cell lung carcinoma.

Non-small-cell lung carcinoma (NSCLC) is among the deadliest of human cancers. The CDKN2A locus, which houses the INK4a and ARF tumor suppressor genes, is frequently altered in NSCLC. However, the specific role of ARF in pulmonary tumorigenesis remains unclear. KRAS and other oncogenes induce the expression of ARF, thus stabilizing p53 activity and arresting cell proliferation. To address the role of ARF in Kras-driven NSCLC, we compared the susceptibility of NIH/Ola strain wild-type and Arf-knockout mice to urethane-induced lung carcinogenesis. Lung tumor size, malignancy and associated morbidity were significantly increased in Arf(-/-) compared with Arf(+/+) animals at 25 weeks after induction. Pulmonary tumors from Arf-knockout mice exhibited increased cell proliferation and DNA damage compared with wild-type mice. A subgroup of tumors in Arf(-/-) animals presented as dedifferentiated and metastatic, with many characteristics of pulmonary sarcomatoid carcinoma, a neoplasm previously undocumented in mouse models. Our finding of a role for ARF in NSCLC is consistent with the observation that benign adenomas from Arf(+/+) mice robustly expressed ARF, while ARF expression was markedly reduced in malignant adenocarcinomas. ARF expression also frequently colocalized with the expression of p21(CIP1), a transcriptional target of p53, arguing that ARF induces the p53 checkpoint to arrest cell proliferation in vivo. Taken together, these findings demonstrate that induction of ARF is an early response in lung tumorigenesis that mounts a strong barrier against tumor growth and malignant progression.

Greater sensitivity of pigtailed macaques (Macaca nemestrina) than baboons to total body irradiation.

Two juvenile pigtailed macaques (animals 1 and 2) received total body irradiation (TBI) followed by autologous stem cell transplantation, by a procedure known to be well tolerated by baboons. In this procedure, the TBI consisted of treatment on two consecutive days with 255cGy on one side, followed after 1-2 min by a similar dose on the other side. The two pigtailed macaques showed rapid haematopoietic engraftment, but succumbed either to systemic cytomegalovirus (CMV) infection and necrotising colitis or to haemorrhagic cystitis and tubulointerstitial nephritis. For four further pigtailed macaques (animals 3-6) the radiation procedure was changed to four equal doses of 255cGy, given 6-12 h apart. Animals 4-6 all showed engraftment and survived for long periods (>218 days), with no, or only minor treatable, complications. Animal 3 failed to show engraftment and succumbed to radiation-induced vascular lesions and severe multiorgan haemorrhages. The results suggest that pigtailed macaques have a lower tolerance threshold than baboons, rhesus macaques or human beings to TBI, the adverse effects of TBI being indistinguishable from those seen in human patients. The results also suggest that a hyperfractionated radiation procedure can prevent radiation-induced morbidity and mortality in pigtailed macaques.

T-cell receptor vbeta deletion and valpha polymorphism are responsible for the resistance of SWR mouse to arthritis induction.

Collagen type II-induced arthritis (CIA) develops in susceptible mouse strains after intradermal injections of type II collagen (CII) in complete Freund's adjuvant (CFA). Susceptibility to CIA in mice is linked to genes of the major histocompatibility complex (MHC). Although the SWR mouse has a susceptible MHC haplotype (H2q), it is resistant to CIA. SWR exhibits at least two known immunological defects: (1) it contains a germline deletion of about 50% of T-cell receptor (TCR) Vbeta-chain gene segments, and (2) SWR is deficient in complement component C5. It has been shown that T cells that express TCRValpha11.1 and TCRVbeta8.2 play a substantial role in the pathogenesis of arthritis in the DBA/1 mouse (H2q). We generated SWR transgenic (tg) mice to determine whether the expression of pathogenic Valpha11.1 and/or Vbeta8.2 transgenes would confer arthritis susceptibility. Arthritis was induced in the SWR TCRalphabeta tg mice, but not in SWR TCRbeta tg mice. To address the role of Valpha11.1 in arthritis susceptibility, we examined the allelic polymorphisms of the Tcra-V11-gene subfamily members between the arthritis susceptible DBA/1 mouse and the arthritis-resistant SWR mouse strain. The amino acid sequences of the Valpha11.1 alleles differ at two positions (codons 18 and 68). Accordingly, these two amino acid changes are sufficient to allow the production of pathogenic T cells in SWR mice. This is the first demonstration of the association of a particular Tcra-V allele and arthritis susceptibility in mice.

Effects of immunomodulation with interferon-gamma on hepatic ischemia-reperfusion injury.

The development of an inflammatory response after injury depends on the participation of a variety of cell populations and endogenous mediators. Interferon-gamma (IFN-gamma) is a potent cellular immunomodulating cytokine that contributes to acute and chronic inflammation. In this study, the effects of immunomodulation on ischemia-reperfusion injury were examined using increasing doses of recombinant, rabbit-specific IFN-gamma in an in situ model of hepatic ischemia-reperfusion. Pretreatment with low dose IFN-gamma augmented injury as measured by histology, aminotransferase concentrations, and myeloperoxidase activity. By contrast, high dose IFN-gamma pretreatment, equivalent to IFN-gamma supplements used in clinical trials, resulted in a lack of neutrophil infiltration and minimal progression of late phase, neutrophil-mediated reperfusion injury. These results suggest that immunomodulating mediators such as IFN-gamma may play a regulating role in the evolution of ischemia-reperfusion, contributing to the development and resolution of acute hepatic injury.

Expression of a type II collagen-specific TCR transgene accelerates the onset of arthritis in mice.

Animal models of autoimmune diseases have been instrumental in advancing our understanding of autoimmunity in humans. Collagen-induced arthritis in mice is an autoimmune disease model of rheumatoid arthritis, which is MHC class II restricted and CD4 T cell dependent. To better understand the fundamental role of T cells in arthritis, we have generated a transgenic mouse carrying the rearranged Valpha11.1 and Vbeta8.2 TCR chain genes isolated from a type II collagen (CII)-specific T cell hybridoma. Cell surface analysis indicated that Vbeta8.2 chain was expressed on the surface of nearly all peripheral T cells. Analysis of T cell subsets in transgenic mice revealed a profound skewing in peripheral T cells towards the CD4 population. Although peripheral T cells were not tolerant to CII and responded to CII stimulation in vitro, transgenic mice did not develop spontaneous arthritis. However, a rapid onset of arthritis with severe clinical signs was detected in transgenic mice after immunization with CII in complete Freund's adjuvant. Histological analysis of inflamed joints showed a great resemblance to arthritic joints in man. This unique transgenic mouse model provides valuable insights into the mechanism of arthritis and into potential specific immune interventions.

Aerosol delivery of lipid:DNA complexes to lungs of rhesus monkeys.

PURPOSE: The potential use of aerosol delivery for non-viral gene therapy was tested by nebulization of lipid:DNA complexes to the lungs of rhesus monkeys. METHODS: Four female rhesus monkeys were dosed with lipid:DNA formulations via aerosol inhalation, where the DNA coded for the human Cystic Fibrosis Transmembrane Conductance Regulator (hCFTR) protein. Delivery of DNA was determined in lung samples by polymerase chain reaction (PCR) by qualitative and quantitative methods. Transgene specific messenger RNA was measured by reverse transcriptase PCR (RT-PCR) and protein expression and localization were evaluated by immunohistochemistry (IHC). RESULTS: Approximately four mg of DNA, complexed with cationic lipid 1.2-dimyristoyl-sn-glycero-3-ethylphosphatidylcholine (EDMPC) and cholesterol were delivered to the lungs of animals by airjet nebulizer. Three days after dosing, tissue samples from the lung were collected and shown to have vector specific DNA, RNA and the presence of CFFR protein. Specifically, the hCFTR protein was distributed widely, although non-uniformly, throughout airway epithelium being located on the apical surface of epithelial cells. Importantly, no adverse clinical effects were observed and the lungs showed no histological abnormalities or signs of acute inflammation. CONCLUSIONS: This study shows that lipid:DNA formulations based on EDMPC and cholesterol can be administered to primates by nebulization resulting in measurable expression of the hCFTR protein. The absence of inflammation is also encouraging and such systems may have utility for delivery of genes to the lungs for the treatment of a variety of pulmonary diseases including cystic fibrosis.

Evidence of evolutionary up-regulation of the single active X chromosome in mammals based on Clc4 expression levels in Mus spretus and Mus musculus.

Previous studies have shown that the chloride channel gene Clc4 is X-linked and subject to X inactivation in Mus spretus, but that the same gene is autosomal in laboratory strains of mice. This exception to the conservation of linkage of the X chromosome in one of two interfertile mouse species was exploited to compare expression of Clc4 from the X chromosome to that from the autosome. Clc4 was found to be highly expressed in brain tissues of both mouse species. Quantitative analyses of species-specific expression of Clc4 in brain tissues from mice resulting from M. spretus x laboratory strain crosses, demonstrate that each autosomal locus has half the level of Clc4 expression as compared with the single active X-linked locus. In contrast expression of another chloride channel gene, Clc3, which is autosomal in both mouse species is equal between alleles in F1 animals. There is no evidence of imprinting of the Clc4 autosomal locus. These results are consistent with Ohno's hypothesis of an evolutionary requirement for a higher expression of genes on the single active X chromosome to maintain balance with autosomal gene expression [Ohno, S. (1967) Sex Chromosomes and Sex-Linked Genes (Springer, Berlin)].

Triggers of autoimmune disease in a murine TCR-transgenic model for multiple sclerosis.

The combination of genetic and environmental factors that contribute to human autoimmune responses has made potential triggers of these diseases difficult to identify. We examined how experimental allergic encephalomyelitis (EAE), an animal model of multiple sclerosis, is triggered using TCR-transgenic mice specific for myelin basic protein (MBP). In these TCR-transgenic mice, EAE can be actively induced and also occurs spontaneously. The incidence of spontaneous EAE in this model is largely confined to adolescence and early adulthood and is more prevalent among males than females, indicating that hormonal influences may contribute to triggering central nervous system autoimmune disease. Disease induction studies show that not all stimuli that activate MBP-specific T cells in vivo also induce EAE. Immunization with MBP peptide stimulates the transgenic T cells to produce Th1 cytokines; however, the activated T cells do not accumulate in the central nervous system and induce EAE unless pertussis toxin is also administered. EAE can be induced by intrathecal injection of either stimulated or nonstimulated transgenic T cells into nontransgenic or transgenic recipients. Therefore, gaining access to the central nervous system appears to be the critical step in this model for the induction of EAE, regardless of the activation state of the T cells.

Dexamethasone treatment does not ameliorate subglottic ischemic injury in rabbits.

BACKGROUND: Following tracheal intubation, a small proportion of patients develop laryngeal inflammation or tissue necrosis severe enough to result in clinical symptoms. Although corticosteroids are frequently advocated to prevent such injury, human studies have been inconclusive because of the low incidence of the problem. This study developed a rabbit model of endotracheal tube-induced laryngeal injury to test the hypothesis that a corticosteroid, dexamethasone, could ameliorate the inflammation and necrosis. METHODS: Subglottic injury was induced in 21 anesthetized rabbits by inflating the cuff of an endotracheal tube to 100 mm Hg with the cuff just below the vocal cords. Every 30 min for 2 h, the cuff was deflated, the tube turned 90 degrees, and the cuff then reinflated. After 2 h, the rabbits' tracheas were extubated. Rabbits were divided into two groups: the treatment group received dexamethasone (1 mg/kg) i.v. 1 h prior to extubation with the dose repeated 6 h following extubation; the untreated group received a saline solution placebo. Four additional rabbits were anesthetized for the same period but did not have a tracheal tube inserted. All rabbits were killed 24 h later and the larynxes were harvested. Sections through the larynx at the level of the cricoid cartilage were randomized and submitted blindly to a veterinary pathologist. Larynxes were scored and ranked according to the severity of mucosal inflammation and necrosis, and submucosal hemorrhage, edema, inflammation, and necrosis. Specimens were also evaluated for focal vs diffuse disease. RESULTS: Injured rabbits demonstrated focal to diffuse mucosal and submucosal inflammation and necrosis. Inflammatory exudates were present in sections from most of the injured rabbits and large sections of the larynxes were denuded of epithelium. There were no differences in injury scores between the treated and untreated rabbits. The four uninjured control rabbits had normal larynxes. CONCLUSIONS: Two hours of endotracheal tube cuff inflation to 100 mm Hg causes an inflammatory laryngeal injury. The histologic features of the injury are unaltered by treatment with 2 mg/kg dexamethasone.

SWR: an inbred strain suitable for generating transgenic mice.

Production of fertilized oocytes and generation of transgenic mice is generally more efficient using F2 hybrid embryos than embryos from inbred mice. Most F2 hybrids are of the C57BL/6 background because of its genetic and embryologic features. However, our goal of developing a transgenic mouse model for rheumatoid arthritis necessitated using a susceptible mouse strain such as DBA/1. We prepared alpha and beta T-cell receptor (TCR) chain gene constructs and microinjected them into embryos from DBA/1, SWR, (DBA/1 x SWR)F1, and (SWR x DBA/1)F1 strains. We found SWR female mice to be prolific ovulators in response to exogenous hormones, with oocyte numbers comparable to those produced by (C57BL/6 x C3H)F1 female mice. Embryos from the (SWR x DBA/1)F1 or SWR strain were large and had prominent pronuclei, whereas (DBA/1 x SWR)F1 embryos were smaller and had less visible pronuclei, similar to those of DBA/1 embryos. Therefore, the pronuclear size and visibility are features of the SWR female mice and are independent of the genotype of the fertilizing spermatozoa. Resistance to lysis after co-injection of alpha beta TCR constructs and the efficiency of generating DNA-positive founders were comparable in SWR, (SWR x DBA/1)F1, and (C57BL6 x C3H)F2 embryos. Thus, the SWR mouse is another inbred strain, in addition to the FVB inbred strain, found to be highly suitable for propagation of transgenes. Furthermore, the SWR mouse is well defined genetically, and SWR females have a high ovulation rate, comparable to that of F1 hybrid mice.

Factors influencing the efficiency of cationic liposome-mediated intravenous gene delivery.

We have characterized the relationships between the design of cationic liposomes as a gene transfer vehicle, their resulting biodistribution and processing in animals, and the level and sites of gene expression they produce. By redesigning conventional cationic liposomes, incorporating cholesterol (chol) as the neutral lipid and preparing them as multilamellar vesicles (MLV), we increased the efficiency of cationic liposome:DNA complex (CLDC)-mediated gene delivery. Expression of the luciferase gene increased up to 1,740-fold and of the human granulocyte-colony stimulating factor (hG-CSF) gene up to 569-fold due to prolonged circulation time of injected CLDC, and increased uptake and retention in tissues. The level of gene expression per microgram of DNA taken up per tissue was 1,000-fold higher in lung than in liver, indicating that in addition to issues of delivery and retention of injected DNA, tissue-specific host factors also play a central role in determining the efficiency of expression. Vascular endothelial cells, monocytes, and macrophages are the cell types most commonly transfected by intravenous injection of CLDC.

Gene therapy for donor hearts: ex vivo liposome-mediated transfection.

OBJECTIVE: Liposomes may be an appropriate transfection vehicle for transplanted hearts, avoiding the use of viruses in immunosuppressed hosts and allowing transfection of nondividing cells. To study whether liposome-mediated transfection could be accomplished during transplantation, we used a liposome-reporter gene system in a rabbit model of allograft cardiac transplantation. METHODS: After aortic crossclamping, Stauffland donor hearts were injected with 10 ml Stanford cardioplegic solution; then a 1.3 to 2.0 mg/kg dose of chloramphenicol acetyl transferase in 1:1 deoxyribonucleic acid-liposome complexes was injected proximal to the aortic crossclamp for coronary artery perfusion. The hearts were transplanted into New Zealand White rabbit recipients in the heterotopic cervical position (n = 11 transplants). Recipients were sacrificed at 24 hours. Myocardial specimens (right and left ventricles) and vascular specimens (epicardial coronary artery, aortic root, and coronary sinus) from both the transplanted and native hearts were analyzed for chloramphenicol acetyl transferase protein by means of the enzymatic liquid scintillation assay (counts per minute per milligram of total protein). RESULTS: In the recipient, myocardial chloramphenicol acetyl transferase activity was significantly greater in treated donor hearts (mean 4.6 x 10(5) cpm/mg +/- 1.1 x 10(5) [standard error]) than in native hearts (mean 4.1 x 10(2) cpm/mg +/- 72 [standard error], p < 0.01, Mann-Whitney U test). In treated donor hearts, right and left ventricular specimens, as well as apical and basal myocardial specimens, were transfected equally. Chloramphenicol acetyl transferase activity in vascular specimens also indicated transfection (mean 5.4 x 10(5) cpm/mg +/- 2.5 x 10(5) [standard error]). Chloramphenicol acetyl transferase activity in the coronary sinus was comparable with that in the coronary arteries, which suggests that liposomes can transverse the coronary capillary beds. CONCLUSIONS: These findings demonstrate that ex vivo transfection of donor hearts with a liposome-reporter gene system results in significant in vivo expression of the transfected gene product after cardiac transplantation. Genetic alteration of myocardium and cardiac vasculature has potential clinical applications in the prevention of posttransplantation rejection, ischemia-reperfusion injury, and both transplant and nontransplant coronary artery disease.

Osteoporosis induced in mice by overproduction of interleukin 4.

Osteoporosis is a common disease in which loss of bone mass results in skeletal fragility. The development of therapies for this disorder has been hampered by the lack of a convenient animal model. Here we describe a disorder in bone homeostasis in transgenic mice that inappropriately express the cytokine interleukin 4 (IL-4) under the direction of the lymphocyte-specific proximal promoter for the lck gene. Bone disease in lck-IL-4 mice appeared to result from markedly decreased bone formation by osteoblasts, features strikingly similar to those observed in cases of severe low-turnover human involutional osteoporosis. By 2 months of age, female and male lck-IL-4 mice invariably developed severe osteoporosis of both cortical and trabecular bone. Osteoporosis was observed in two independently derived founder animals, indicating that this phenotype was directly mediated by the IL-4 transgene.

Human relaxin decreases collagen accumulation in vivo in two rodent models of fibrosis.

The reproductive hormone, relaxin, inhibits collagen synthesis in vitro by normal human dermal fibroblast. In the present study, recombinant human relaxin is shown to modulate collagen accumulation and organization by mesenchymal cells in vivo in two rodent models of fibrosis: 1) fibrotic infiltration of polyvinyl alcohol sponge implants in rats, and 2) capsule formation around implanted osmotic pumps in mice. In the sponge, relaxin inhibits collagen accumulation, a measured by hydroxyproline content, in a dose-responsive manner by up to 25-29% in animals receiving 30 ng/ml relaxin, a finding supported by a decrease in collagen-specific trichrome staining in sections of sponges from relaxin-treated animals. In mice, the capsules surrounding relaxin-containing pumps are thinner and less dense than are capsules from control pumps. Ultrastructurally, control capsules are composed of densely packed parallel arrays of collagen fibrils, whereas fibrils more frequently are not packed in parallel arrays and are less abundant in treated capsules.

Immune function in offspring of nonhuman primates (Macaca nemestrina) exposed weekly to 1.8 g/kg ethanol during pregnancy: preliminary observations.

A preliminary investigation of immune host response was conducted in a group of fetal alcohol-exposed nonhuman primates (Macaca nemestrina) who were part of a broader ongoing study of ethanol teratogenicity. The mothers of the offspring received weekly oral doses of ethanol (1.8 g/kg) for the first 3 or 6 or the entire 24 weeks of gestation. A control group received sucrose solution weekly throughout pregnancy. Four of the 18 ethanol-exposed animals (22%) died or were euthanized after infectious disease or failure to thrive during the first year of life; none of the seven control animals died. This imbalance in survival prompted the present review of immune function in the remaining offspring. Parameters assessed included: (1) white blood cell count (WBC), (2) peripheral blood leucocyte subsets (CD4+, CD8+, CD20+, and CD11c+), (3) T-cell proliferation after activation with phytohemagglutinin (PHA), staphylococcus enterotoxin B (SEB), and tetanus toxoid (TT), (4) phagocytic activity of monocytes, and (5) serum immunoglobulin levels and serum antibody titers after TT vaccination. Mean T-cell proliferation to TT was significantly decreased (p = 0.01) in all ethanol-exposed animals relative to controls, with near-significant decreases (p = 0.06) in response to SEB in the ethanol-exposed animals. Lymphocyte proliferation in response to PHA was not altered. Ethanol-exposed animals had significantly lower TT titers than controls after initial vaccination and booster. WBC, leukocyte subsets, serum immunoglobulins, and monocyte phagocytic activity were not significantly different from control values. These preliminary observations suggest that T-cell proliferation and antigen-specific memory responses may be altered in offspring exposed to weekly doses of ethanol in utero and warrant further evaluation for confirmation.

Neutrophils contribute to hepatic ischemia-reperfusion injury by a CD18-independent mechanism.

Hepatic ischemia-reperfusion injury is reported to be modulated by neutrophils (PMNs). The adhesion and emigration of PMNs that precede tissue inflammation and necrosis in other organs are mediated, in part, by the leukocyte adhesion complex CD11/CD18. In this study, the role of PMN adhesion via CD11/CD18 in isolated liver ischemia-reperfusion injury was examined in rabbits using a blocking monoclonal antibody (mAb 60.3) specific for the CD18 receptor. Vinblastine-induced neutropenia provided significant protection, confirming participation of neutrophils in the pathogenesis of hepatic injury. Inhibition of PMN adherence with mAb 60.3 did not ameliorate injury, as measured by aminotransferase concentrations or a histologic scoring system for injury severity. Histologic sections were scored for pattern and extent of injury as well as neutrophil association with injury. These results suggest a CD18-independent mechanism of neutrophil adhesion in the evolution of isolated hepatic ischemia-reperfusion injury.

Transgenic animals in the evaluation of compound efficacy and toxicity: will they be as useful as they are novel?

1. Construction of transgenic mice is predicated upon inserting foreign DNA into native host DNA and having this expressed in the germline. This may be accomplished by nuclear injection, retroviral vectors or use of embryonic stem (ES) cells. 2. Expression of novel structural genes may be reasonably directed by the judicious use of an accompanying promoter/enhancer sequence. Insertion of foreign genes may be designed to result in phenotypic expression of a novel trait or ablation of a native gene or gene product. 3. Resulting transgenic mice offer significant utility as models of human diseases and a unique opportunity for investigating immune and metabolic pathways as well as for exploring mechanisms of development, mutagenesis and teratogenesis. 4. Use of transgenic animals in drug development has considerable potential although realization of this potential will take time. Constructing transgenics is only the first step in a complex series of events culminating in understanding the consequences of imposing novel genetic material on an intact, highly integrated living system. Practical use of transgenic animals will depend upon substantial effort being spent in investigating and validating the phenotypic consequences of gene transfer.

Effects of transport on constituents of bronchoalveolar lavage fluid from horses.

To determine whether road transport affected pulmonary phagocyte activity, 7 healthy Thoroughbred horses were shipped 1,160 kilometers over 36 hours. Fluid collected by bronchoalveolar lavage (BAL) 12 hours, and 7 and 14 days after transport was analyzed. Results were compared to those from the same horses pre-transport, and 7 non-transported control horses that had BAL performed at the same times as the transported horses. Of cells recovered with BAL the percentage of viable pulmonary alveolar macrophages (PAMs) declined from 90.0 +/- 0.9% pre-transport to 80.0 +/- 3.7% by 2 weeks post transport. Although the ability of PAMs to inhibit the growth of Staphylococcus epidermidis had decreased by 2 weeks post-transport (19.2 +/- 3.7% vs. 8.8 +/- 2.3% inhibition) this could not be attributed to transport as a similar effect occurred in the control group. In contrast, the ability of PAMs to phagocytose sheep erythrocytes labelled with rabbit anti-erythrocyte antibodies increased from 74.0 +/- 8.1% to 92.3 +/- 1.5% by 12 hours post-transport. As all variables were unchanged or only mildly altered following transport, we conclude that this form of transport did not alter the PAM functions we assessed.

Reduction in intensity of Pneumocystis carinii pneumonia in mice by aerosol administration of gamma interferon.

Pneumocystis carinii pneumonia in patients with AIDS may result from impaired local release of gamma interferon from lung lymphocytes and subsequent failure of macrophage activation. We tested this hypothesis with mice rendered immunodeficient by selective depletion of CD4+ lymphocytes and inoculated intratracheally with P. carinii. After aerosol administration of recombinant murine gamma interferon, the intensity of P. carinii infection in these mice was reduced in comparison with that in mice not treated with gamma interferon. Histologic scoring of lung tissue did not reveal additional pulmonary inflammation attributable to gamma interferon. Aerosol administration of gamma interferon may reduce the intensity of P. carinii infection by augmentation of host defense, despite persistent CD4+ lymphocyte depletion.