Hormonal patterns in normal and hyperleptinemic mares in response to three common feeding-housing regimens.

We previously reported that a rise in plasma leptin concentrations followed the rise in insulin and glucose in meal-fed horses, whereas horses maintained on pasture had little fluctuations in hormonal patterns. We have also described a hyperleptinemic-hyperinsulinemic condition that occurs in about 30% of our light horse mares of high body condition maintained on pasture. The present experiment was designed to 1) study the effect of 3 common feeding-housing regimens on leptin and other metabolic hormones in mares and 2) determine whether the hyperleptinemic condition interacted with these regimens. Six light horse mares with high body condition (average score = 7) were assigned to 2 simultaneous 3 x 3 Latin squares, 1 with normal mares (leptin = 0.1 to 6 ng/mL) and 1 with mares displaying hyperleptinemia (>10 ng/mL). Three feeding-housing regimens were compared: ad libitum pasture, ad libitum native grass hay in an outdoor paddock, and single morning feedings of a pelleted concentrate and hay at 0700 in a barn. Five days of acclimation to the feeding regimens were followed by a 36-h period of hourly blood collection to characterize the hormonal characteristics. Leptin concentrations were elevated (P < 0.001) in mares predetermined to be hyperleptinemic compared with normal mares, regardless of the feeding regimen. Leptin was greatest (P < 0.01) in mares on pasture and least in mares fed hay. Variations over time (P < 0.01) were present for all hormones and metabolites studied. Glucose and insulin concentrations were greatest (P < 0.01) in mares on pasture, with meal-fed mares exhibiting an immediate rise in plasma concentrations of both after feeding. Mares on hay had low and constant concentrations of glucose, insulin, and leptin, with no apparent fluctuations. Cortisol, prolactin, and IGF-I did not differ with leptin status, whereas GH differed due to feeding-housing regimen (P < 0.02); there was also an interaction of leptin status and feeding-housing regimen for GH concentrations (P = 0.094). It was concluded that 1) estimates of hormonal secretion in horses based on frequent sampling, depending upon the hormone in question, can be profoundly affected by the feeding-housing regimens, and 2) the hyperleptinemic condition persists under differing conditions of feeding-housing.

Effect of dexamethasone, feeding time, and insulin infusion on leptin concentrations in stallions.

Three experiments tested the hypotheses that daily cortisol rhythm, feeding time, and/or insulin infusion affect(s) leptin secretion in stallions. Ten mature stallions received ad libitum hay and water and were fed a grain concentrate once daily at 0700. In Exp. 1, stallions received either a single injection of dexamethasone (125 microg/kg BW i.m.; n = 5) or vehicle (controls; n = 5) at 0700 on d -1. Starting 24 h later, blood samples were collected every 2 h for 36 h via jugular venipuncture. Cortisol in control stallions varied (P < 0.01) with time, with a morning peak and evening nadir; dexamethasone suppressed (P < 0.01) cortisol concentrations. Leptin and insulin were greater (P < 0.01) in the treated stallions, as was the insulin response to feeding (P < 0.01). Leptin in control stallions varied (P < 0.01) in a diurnal pattern, peaking approximately 10 h after onset of eating. This pattern of leptin secretion was similar, although of greater magnitude (P < 0.01), in treated stallions. In Exp. 2, five stallions were fed the concentrate portion of their diet daily at 0700 and five were switched to feeding at 1900. After 14 d on these regimens, blood samples were collected every 4 h for 48 h and then twice daily for 5 d. Cortisol varied diurnally (P = 0.02) and was not altered (P = 0.21) by feeding time. Insulin and leptin increased (P < 0.01) after feeding, and the peaks in insulin and leptin were shifted 12 h by feeding at 1900. In Exp. 3, six stallions were used in two 3 x 3 Latin square experiments. Treatments were 1) normal daily meal at 0700; 2) no feed for 24 h; and 3) no feed and a bolus injection of insulin (0.4 mIU/kg BW i.v.) followed by infusion of insulin (1.2 mIU.kg BW(-1).min(-1)) for 180 min, which was gradually decreased to 0 by 240 min; sufficient glucose was infused to maintain euglycemia. Plasma insulin increased (P < 0.01) in stallions when they were meal-fed (to approximately 150 microIU/mL) or infused with insulin and glucose (to approximately 75 microIU/mL), but insulin remained low (10 microIU/mL or less) when they were not fed. The increases in insulin were paralleled by gradual increases (P < 0.01) in leptin concentrations 3 to 4 h later in stallions fed or infused with insulin and glucose. When stallions were not fed, leptin concentrations remained low. These results demonstrate that feeding time, and more specifically the insulin increase associated with a meal, not cortisol rhythm, drives the postprandial increase in plasma leptin concentrations in horses.

Quantitative analysis of cancer invasion in vitro: comparison of two new assays and of tumour sublines with different metastatic capacity.

Various murine tumour sublines which differed considerably in their in vivo metastatic capacity were tested in vitro for their ability to invade normal tissue. For this purpose we developed two quantitative tests, a Boyden chamber endothelial cell invasion assay and a brain tissue microsphere invasion assay. The invasion of [75Se]methionine-prelabelled tumour cells into the normal tissues was followed by measuring the percentage of tumour-associated label in the brain microspheres or the endothelial monolayers after 12-48 h of co-cultivation. Clear and comparable differences existed in both assays between the amount of radiolabel found in the normal tissues after a co-cultivation with the different tumour lines. In three of the four tumour lines invasiveness correlated with metastatic capacity. The fourth line, a plastic adherent variant, was highly invasive but low metastatic. The ability of tumour cells to invade normal tissue, therefore, while necessary for the generation of metastases, is not in itself sufficient. Since both assays are independent of time-consuming histological sectioning and staining and allow a quantitative determination of invasive capacity of tumour cells grown as single cell suspensions they appear well suited for experimental manipulation and for screening of anti-invasive drugs.

Identification of asparagine-linked oligosaccharides involved in tumor cell adhesion to laminin and type IV collagen.

MDW4, a wheat germ agglutinin-resistant nonmetastatic mutant of the highly metastatic murine tumor cell line called MDAY-D2 has previously been shown to attach to fibronectin and type IV collagen, whereas MDAY-D2 and phenotypic revertants of MDW4 attached poorly to these substrates. The increased adhesiveness of the mutant cells appeared to be closely related to a lesion in cell surface carbohydrate structures. In an effort to identify the carbohydrates involved in cell attachment, glycopeptides isolated from mutant and wild-type cells as well as from purified glycoproteins were tested for their ability to inhibit the attachment of MDW4 cells to plastic surfaces coated with fibronectin, laminin, or type IV collagen. The addition of mannose-terminating glycopeptide to the adhesion assay inhibited MDW4 cell attachment to type IV collagen. In contrast, a sialylated poly N-acetyllactosamine-containing glycopeptide, isolated from wheat germ agglutinin-sensitive MDAY-D2 cells but absent in MDW4 cells, inhibited MDW4 attachment to laminin. None of the glycopeptides used in this study inhibited attachment of MDW4 cells to fibronectin-coated plastic. Peptide N-glycosidase treatment of the cells to remove surface asparagine-linked oligosaccharides inhibited MDW4 adhesion to type IV collagen, but not to laminin, and the same treatment of the wheat germ agglutinin-sensitive cells enhanced attachment to laminin. Tumor cell attachment to, and detachment from, the sublaminal matrix protein laminin and type IV collagen are thought to be important events in the metastatic process. Our results indicate that tumor cell attachment to these proteins may be partially modulated by the expression of specific oligosaccharide structures associated with the cell surface.

Monoclonal antibodies which prevent experimental lung metastases. Interference with the adhesion of tumour cells to laminin.

Cellular adhesion is important during metastasis, as metastatic cells must escape from the primary site into lymph and blood systems, there to adhere specifically to sites in distant organs. We have recently selected monoclonal antibodies which prevent adherence of B16 mouse melanoma cells to tissue culture dishes, and also markedly reduce experimental lung metastasis in mice when injected before or with the tumor cells. Here, we investigated which step in the metastatic process may be affected by the antibodies. The possible inhibitory effect of antibody on tumour cell adherence to vascular endothelial monolayers and to purified components of the underlying extracellular matrix - fibronectin, laminin and collagen type IV - was studied using in vitro assays. We found that the antibodies significantly blocked attachment to laminin, suggesting that specific basement membrane components play an important role in attracting or otherwise modifying the behaviour of metastatic tumour cells.

Phenotypic and genotypic differences between high- and low-metastatic related tumor lines and the problem of tumor progression and variant generation.

We have described a number of phenotypic and genotypic differences between defined tumor sublines from the Eb/ESb model system that differ greatly in metastatic capacity. Some of these differences are summarized in Figure 5. It can be seen that the differences are both structural and functional. They include differences in the glycosylation of membrane glycoproteins as well as differences in total membrane protein composition. The functional differences refer particularly to enzymatic activities, cell motility, and invasive capacity. In addition, a change in membrane fluidity and membrane dynamics was seen when we investigated membrane vesicles shed spontaneously from these tumor lines. ESb type cells were found to shed about three times as much membrane material as the low-metastatic, Eb cells. Since such vesicles carried degradative enzyme activity, they could play a role in the invasion process. The vesicles also carried TATA and thus could also play a role in preventing immunological attack against these tumor cells. Another immunological escape mechanism was found to be based on the ability of the high-metastatic cells to generate immunoresistant variants that no longer expressed the respective TATA. A cytogenetic comparison of such immunoresistant variants with the TATA+ line revealed several chromosomal changes. The contribution of chromosomal rearrangements to the ability of malignant cells to generate variants was discussed.

Quantitative determination of disseminated tumor cells by [3H]thymidine incorporation in vitro and by agar colony formation.

A new microradioassay for the detection and quantification of disseminated tumor cells in blood and organ samples of tumor-bearing animals has been worked out in a murine total system for tumor metastasis. In contrast to previous prelabeling or in situ labeling procedures, the basis of this assay is a postlabeling in vitro of tumor cell-containing material with [3H]thymidine. Percoll gradient fractionation and autoradiography revealed that most of the isotope was incorporated into tumor cells. Titration curves with tumor cells from tissue culture were run in parallel and allowed to calculate from the radioactivity of a sample that actual number of proliferating tumor cells. The postlabeling assay correlated fairly well with an agar colony test which measure the clonogenic or stem cell potential of a tumor cell population. About 70% of syngeneic animals which had been inoculated s.c. with ESb tumor cells showed increased [3H]thymidine uptake in their blood, particularly at certain time intervals (11 and 21 days). None of these animals lived for more than 2 days longer. The advantages of the new microradioassay and its possible prognostic significance will be discussed.

Enhancement of lymphocyte-mediated cytotoxicity after tumor resection in patients with colorectal cancer.

Lymphocyte numbers and function in 37 patients with colorectal cancer were compared with those in 23 healthy controls of the same age range. No significant difference in lymphocyte count, numbers of cells forming erythrocyte rosettes or bearing surface immunoglobulin, cell-mediated cytotoxicity, or mitotic response to phytohemagglutinin (PHA) was found. Six to 12 months after tumor resection, patients showed a rise in spontaneous, antibody-induced, and PHA-induced cytotoxicities to levels significantly higher than those found in age-matched controls. This rise occurred irrespective of whether patients had received adjuvant immunotherapy with Corynebacterium parvum.

Effect of intravenous corynebacterium parvum on peripheral-blood effector cells of cancer patients.

The i.v. administration of Corynebacterium parvum (CP) to patients who had recently undergone resection of colorectal tumours was found to have the following effects: 1. Polymorphonuclear leucocyte counts were raised 24 h after CP administration, while both lymphocyte and monocyte counts fell during this period. Polymorph and lymphocyte counts had returned to pre-infusion levels at one week, but monocyte counts were significantly increased at this time. 2. The lymphocyte mitotic response to PHA was reduced during the 24 h after CP infusion. 3 The spontaneous, antibody-induced, and PHA-induced lymphocyte-mediated cytotoxicity against a nucleated target cell fell significantly 3 h after CP infusion, but these functions recovered by 7 days. 4. A rise in serum lysozyme was found 3 and 24 h after CP administration. However, these increased levels were not maintained beyond 24 h.

Amino acid sequences around the sites of phosphorylation in the pig heart pyruvate dehydrogenase complex.

1. When pig heart pyruvate dehydrogenase complex was phosphorylated to completion with [gamma-32P]ATP by its intrinsic kinase, three phosphorylation sites were observed. The amino acid sequences around these sites were: sequence 1, Tyr-Gly-Met-Gly-Thr-Ser(P)-Val-Glu-Arg; and sequence 2, Tyr-His-Gly-His-Ser(P)-Met-Ser-Asp-Pro-Gly-Val-Ser(P)-Tyr-Arg. 2. When phosphorylated to inactivation by repetitive additions of limiting quantities of [gamma-32P]ATP, phosphate was incorporated mainly (more than 90%) into Ser-5 of sequence 2. Phosphorylation of this site thus results in activation of pyruvate dehydrogenase. 3. If Ser-5 is phosphorylated with ATP and the enzyme then incubated with [gamma-32P]ATP, phosphorylation of the remaining sites occurred. Ser-12 of sequence 2 is phosphorylated about twice as rapidly as Ser-6 of sequence 1. 4. Incubation of pyruvate dehydrogenase with excess [gamma-32P]ATP with termination of phosphorylation at about 50% complete inactivation showed that Ser-5 of sequence 2 was phosphorylated most rapidly, but also that Ser-12 of sequence 2 was significantly (15% of total) phosphorylated. Ser-6 sequence 1 contained about 1% total P. 5. These results suggest that addition of limiting amounts of ATP produces primarily phosphorylation of Ser-5 of sequence 2 (inactivating site). This also occurs during incubation with excess ATP before complete inactivation occurs, but a greater occupancy of other sites also occurs during this treatment.

A pen smear technique for assays of rosette-forming lymphocytes.

A simple technique is described for the preparation of permanent stained smears of human E-rosette lymphocytes. Smears of the cells suspended in foetal calf serum were drawn as thin strips on slides with a pen nib. Such smears, after Romanowsky staining, and incorporating latex-particles phagocytosis as a marker for non-lymphoid cells, show excellent rosette preservation and permit easy distinction and counting of rosetting lymphocytes, non-rosetting lymphocytes, monocytes and granulocytes. The percentages of rosette-forming lymphocytes estimated on such smears correspond closely to those obtained from counts made on viable cell suspensions.

Relationships between different functional properties of human monocytes.

Various methods for measuring monocyte numbers and function in human blood are compared. The assays used were: phagocytosis of antibody-sensitized sheep red cells; staining for diffuse non-specific esterase activity; capacity to adhere to polystyrene and ability to lyse human A1 red cells sensitized with allogeneic anti-A1 antibody. The following conclusions are drawn from the results: (1) Previous observations showing that sensitized A1 red cells are lysed by phagocytic cells and not K cells are confirmed. (2) Granulocytes lyse sensitized A1 red cells more rapidly than monocytes and this assay is only useful for assessing monocyte function if granulocytes are first removed from preparations. (3) Phagocytosis is important in the lysis of sensitized A1 red cells by monocytes. (4) Esterase positive cells correlate significantly with the number of cells phagocytosing sensitized sheep red blood cells, r = 0.83 (P less than 0.001) and with the number of cells adhering to polystyrene, r = 0.53 (P less than 0.05). (5) The lysis of sensitized A1 red cells correlated significantly with esterase positive cell numbers, r = 0.57 (P less than 0.001) and phagocytosis of sensitized sheep red cells, r = 0.50 (P less than 0.01 but not with the numbers of adherent cells, r = 0.34 (P greater than 0.05).

The effects of prednisolone in leucocyte function in man. A double blind controlled study.

The effect of prednisolone on various immunological parameters was studied in patients with ulcerative colitis in complete remission. The study was designed as a double blind trial in which patients received either prednisolone or a dummy preparation and the following observations were made:(1) The mean lymphocyte count fell from 1738 cells per mm3 to 501 cells/mm3 4 hr after prednisolone was given but by 24 hr was significantly elevated to 2399 cells/mm3; thereafter it returned to normal levels. (2) Surface marker assays of lymphocytes forming spontaneous sheep cell (E), Fc (EA), and C3 (EAC) rosettes; and cells bearing surface immunoglobulin fluctuated in approximately the same pattern as the total lymphocyte count. (3) The mitotic response to a sub-maximal stimulating dose of phytohaemagglutinin (PHA) was significantly depressed 4 hr after steroid administration but returned to normal by 24 hr. (4) Spontaneous and PHA-induced lymphocyte mediated cytotoxicity fell significantly by four hours and remained depressed to the end of steroid administration. The PHA-induced cytotoxicity was still significantly depressed 7 days after steroid administration was stopped. (5) K-cell cytotoxicity did not follow the general pattern and was only slightly reduced at four hours being lowest after 24 hr and still depressed 7 days after cessation of steroid administration. (6) The number of plasma cells in the rectal lamina propria showed no significant change after one week of steroid administration. (7) No significant changes occurred in any of the above assays, in the control group. (8) Polymorphonuclear leucocyte counts rose sharply by 4 hr in the patients receiving prednisolone. There was also a smaller but significant rise in the control group. They remained elevated for 7 days in the group receiving prednisolone, and subsequently fell to normal levels. The control group had returned to initial levels by 24 hr.

Analysis of treatment in childhood leukaemia. III. Independence of lymphopenia induced by irradiation and by chemotherapy.

Lymphopenia induced by treatment for acute lymphoblastic leukaemia is analysed and discussed in relation to the type and incidence of infection occurring in those patients during complete remission. Blood lymphocytes can be placed into three largely independent groups: (1) those lymphocytes susceptible to long-term depletion following irradiation; (2) those lost from the blood during and for a short period after maintenance chemotherapy with methotrexate and 6-mercaptopurine; and (3) the remainder which are not depleted by irradiation or maintenance chemotherapy. The number of cells in each compartment varies from child to child and probably with age but on average is about 1.2 x 10(9)/1. for group 1, 0.7 x 10(9)/1. for group 2 and 0.4 x 10(9)/1. for group 3. Conventional lymphocyte typing crosses these barriers in that: Group 1 consists mainly of E-rosetting cells and cells which show a mitotic response to phytohaemagglutinin; Group 2 also contains E-rosetting cells but contains a major proportion of blood lymphocytes with surface immunoglobulin and essentially all antibody dependent cytotoxic lymphocyte (K-cell) activity; Group 3 comprises E-rosetting cells and a few immunoglobulin-staining cells.

Immunosuppression in the treatment of inflammatory bowel disease. II. The effects of azathioprine on lymphoid cell populations in a double blind trial in ulcerative colitis.

Blood lymphocytes and rectal plasma cells have been studied in patients with ulcerative colitis taking part in a double-blind trial of treatment with azathioprine. Treatment for 1 year resulted in a modest fall in blood lymphocyte count, with little change in neutrophils or platelets. There was no major change in the proportions of circulating T and B lymphocytes, suggesting that the number of such cells per millilitre of blood fell in proportion to the change in lymphocyte count. The number of plasma cells in the rectal lamina propria was reduced to a mean less than half that of the control patient group. Blood K-cell cytotoxic activity fell at least 25-fold after 1 year's treatment. PHA-induced cytotoxicity was also reduced, but less consistently. Reduced K-cell activity is interpreted as reflecting depletion of effector cells from the circulation. The fall in lymphocyte count, K-cell activity and gut plasma cells was slow, indicating continuous inhibition of lymphopoiesis or differentiation throughout the trial period. Thus, azathioprine has some immunosuppressive effects which develop only after prolonged treatment. The clinical results of the trial did not show a major beneficial effect of azathioprine in the treatment of ulcerative colitis, nor were there clear correlations between the results of lymphocyte assays and clinical response in individual patients.

Two populations of lymphocytes involved in phytohaemagglutinin-induced cytotoxicity of a dividing target cell.

A quantitative study of the surface characteristics of cells cytotoxic against Chang cells permits the following conclusions: [1] In the presence of phytohaemagglutinin two types of cell are involved; one has Fc receptors and the other does not. Approximately two thirds of the cytotoxic activity is lost if the Fc-receptor-bearing cells are removed. [2] Both spontaneous and antibody-induced cytotoxicity is almost completely abolished by removal of Fc-receptor-bearing cells. [3] None of the cytotoxic activity, spontaneous or induced, appears to be mediated by cells that are plastic-adherent or bear surface immunoglobulin. These results are compared with similar studies by other authors.

Lymphocyte reactivity in pregnant women and newborn infants.

The mitotic response to phytohaemagglutinin (PHA) was determined in lymphocytes of mothers and their newborn infants obtained at delivery and seven days later by measuring the rate of 125 I-idoxuridine uptake into DNA in lymphocytes cultured in their own plasma and after washing and resuspension in fetal bovine serum. There was no difference in the unstimulated counts of maternal lymphocytes taken at delivery, whether unwashed or washed, compared with those from nonpregnant controls. With PHA stimulation the mitotic response of the maternal lymphocytes cultured in their own plasma was reduced compared with that of the control lymphocytes but washed maternal cells showed a similar response to the controls. These findings suggest that the reduced lymphocyte mitotic response to PHA in pregnancy is due to a plasma inhibitory factor This inhibition was not evident in maternal blood taken seven days after delivery. DNA synthesis in unstimulated cultures from newborn infants at birth and seven days after birth was greater than that in adult control cultures. With PHA stimulation the mitotic response of cord-blood lymphocytes cultured in their own plasma paralleled that of control lymphocytes but washed newborn cells showed a greater response. Thus plasma suppression similar to that observed in the mother seems also to affect infants at birth. This inhibition was not demonstrable in blood taken from infants of 7 days.