Effect of separase depletion on ionizing radiation-induced cell cycle checkpoints and survival in human lung cancer cell lines

Abstract.   Objectives : This study is to evaluate the effect of separase depletion on cell cycle progression of irradiated and non-irradiated cells through the G₂/M phases and consecutive cell survival. Materials and methods : Separase was depleted with siRNA in two human non-small cell lung carcinoma (NSCLC) cell lines. Cell cycle progression, mitotic fraction, DNA repair, apoptotic and clonogenic cell death were determined. Results : By depletion of endogenous separase with siRNA in NSCLCs, we showed that separase affects progression through the G₂ phase. In non-irradiated exponentially growing cells, separase depletion led to an increased G₂ accumulation from 17.2% to 29.1% in H460 and from 15.7% to 30.9% in A549 cells and a decrease in mitotic cells. Depletion of separase significantly ( P <  0.01) increased the fraction of radiation-induced G₂ arrested cells 30–56 h after irradiation and led to decrease in the mitotic fraction. This was associated with increased double-strand break repair as measured by γ-H2AX foci kinetics in H460 cells and to a lesser extent in A549 cells. In addition, a decrease in the expression of mitotic linked cell death after irradiation was found. Conclusions : These results indicate that separase has additional targets involved in regulation of G₂ to M progression after DNA damage. Prolonged G₂ phase arrest in the absence of separase has consequences on repair of damaged DNA and cell death.     

Influence of denervation on the regeneration of Pleurodele limbs

Abstract. A cytophotometric study of Feulgen-stained mesenchymal cell nuclei from regeneration blastemas of both innervated and denervated limbs over the 1st 7 days following the midbud stage showed a diminution of the percentage of cells in the S + G₂ phases and a corresponding augmentation of the percentage of cells in the G₀+ G₁ phases. This change, which was temporally correlated with the redifferentiation of the innervated blastemas, was greater in denervated blastemas, even though they do not redifferentiate. From these results, it is concluded that the denervation of midbud blastemas brings about either an extension of the G₁ phase or an exiting from the cell cycle to G₁ (G_0–1), or both phenomena.     

Long duration of mitosis and consequences for the cell cycle concept, as seen in the isthmal cells of the mouse pyloric antrum. II. Duration of mitotic phases and cycle stages, and their relation to one another

Abstract. The kinetics of isthmal cells in mouse antrum were examined in three ways: (a) the duration of cell cycle and DNA-synthesizing (S) stage was measured by the 'fraction of labelled mitoses' method; (b) the duration of interphase and mitotic phases was determined from how frequently they occurred; and (c) mice were killed at various intervals after an intravenous injection of ³H-thymidine to time the acquisition of label by the various phases of mitosis.
The duration of the isthmal cell cycle was found to be 13.8 hr and that of the DNA-synthesizing (S) stage, 5.8 h. Estimates for the duration of the G₁ and G₂ stages were 6.8 and 1.0 hr, respectively.
From the frequency of mitotic phases, defined as indicated in the preceding article (El-Alfy & Leblond, 1987) and corrected for the probability of their occurence, it was estimated that prophase lasted 4.8 hr; metaphase, 0.2 hr; anaphase, 0.06 hr and telophase, 3.3 hr, while the interphase lasted 5.4 hr. In accordance with this, the duration of the whole mitotic process was 8.4 hr.
Ten minutes after an intravenous injection of ³H-thymidine, 38% of labelled isthmal cells were in interphase and 62% in early or mid prophase, while cells in late prophase and other mitotic phases were unlabelled. After 60 min, label was in late prophase, after 120 min, in mid telophase and after 180 min, in late telophase.
We conclude that there is overlap between some mitotic phases and cycle stages. Thus, while nuclei are at interphase during the early third of S, they are in prophase during the late two-thirds as well as during G₂. Also, nuclei are in telophase during the early half of G₁ but at interphase during the late half. Differences in nuclear diameter show that subdivision of both S and G₁ into early and late periods is practical.     

Monocytic differentiation of HL60 cells induced by potent analogs of vitamin D3 precedes the G1/G0 phase cell cycle block

Abstract. Differentiation of mammalian cells is accompanied by reduced rates of proliferation and an exit from the cell cycle. Human leukemic cells HL60 present a widely used model of neoplastic cell differentiation, and acquire the monocytic phenotype when exposed to analogs of vitamin D₃ (VD₃). The maturation process is accompanied by two blocks in the cell cycle: an arrest in the G₁/G₀ phase, and a recently described G₂+ M block. In this study we have analyzed the traverse of the cell cycle phases of the well-differentiating HL60-G cells exposed to one of ten analogs of VD₃, and compared the cell cycle effects of each compound with its potency as a differentiation-inducing agent. We found that in general there was a good correlation between the effects of these compounds on the cell cycle and on differentiation, but the best cell cycle predictor of differentiation potency was the extent of accumulation of the cells in the G₂ compartment. All analogs induced a marked decrease in the mitotic index, and polynucleation of HL60 cells was produced, especially by compounds which were effective as inducers of differentiation. Time course studies showed that induction of differentiation was accompanied by a transient increase of the proportion of cells in the G₂+ M compartment, but preceded the G₁ to S, and the G₂ compartment blocks. These studies indicate that complex changes in the cell cycle traverse accompany, but do not precede, the acquisition of the monocytic phenotype by HL60 cells.     

Laser scanning cytometry (LCS) allows detailed analysis of the cell cycle in PI stained human fibroblasts (TIG-7)

We have demonstrated a method for the in situ determination of the cell cycle phases of TIG-7 fibroblasts using a laser scanning cytometer (LSC) which has not only a function equivalent to flow cytometry (FCM) but also has a capability unique in itself. LSC allows a more detailed analysis of the cell cycle in cells stained with propidium iodide (PI) than FCM. With LSC it is possible to discriminate between mitotic cells and G₂ cells, between post-mitotic cells and G₁ cells, and between quiescent cells and cycling cells in a PI fluorescence peak (chromatin condensation) vs. fluorescence value (DNA content) cytogram for cells stained with PI. These were amply confirmed by experiments using colcemid and adriamycin. We were able to identify at least six cell subpopulations for PI stained cells using LSC; namely G₁, S, G₂, M, postmitotic and quiescent cell populations. LSC analysis facilitates the monitoring of effects of drugs on the cell cycle.     

EFFECTS OF PHYTOCHROME AND BLUE-NEAR ULTRAVIOLET LIGHT-ABSORBING PIGMENT ON DURATION OF COMPONENT PHASES OF THE CELL CYCLE IN ADIANTUM GAMETOPHYTES

Single-celled protonemata of the fern Adiantum capillus-veneris, kept under continuous red light, grew with a very low rate of cell division, and the cell cycle was arrested in the early G₁ phase. Cell division was induced by transferring the protonemata to the dark after various light treatments, and the duration of component phases in the cell cycle was determined by a continuous-labelling technique with 3H-thymidine. Blue light irradiation greatly reduced the duration of the G₁ phase but did not affect that of other phases. The greater the fluence of blue light, the shorter was the duration of G₁ phase was observed. In contrast, a brief exposure of red-light-grown protonemata to far-red light given immediately before the dark incubation showed no effect on the duration of G₁ S and M phases but significantly extended that of the G₂ phase. The effect of far-red light on the G₂ phase was reversed by red light, and the effects of red and far-red light were repeatedly reversible. The progression in the M phase was shown by means of a time-lapse video system to be not at all influenced by any pre-irradiation described above.     

Expression of G1 and G2 cyclins measured in individual cells by multiparameter flow cytometry: a new tool in the analysis of the cell cycle

Abstract. Multivariate analysis of the expression of cyclin proteins and DNA content has opened new possibilities for the study of the cell cycle. By virtue of their cell cycle phase specificity, the expression of cyclins may serve, in addition to DNA content, as another marker of a cell's position in the cycle, and provide information about the proliferative potential of cell populations. Several applications of the methodology based on bivariate analysis of DNA content v . expression of B, E and D type cyclins are reviewed: 1 expression of cyclins by individual cells during their progression through the cycle can be studied, using exponentially growing cells without the necessity of cell synchronization or other perturbations of the cycle; 2 cells having the same DNA content but residing in different phases of the cycle (e.g. G₂ diploid v. G₁ tetraploid) can be distinguished; 3 cell transition from G₀ to G₁ and progression through G₁ (e.g. mitogen stimulated lymphocytes) can be assayed; 4 the population of proliferating cells can be distinguished from noncycling cells based on dual cell labelling with a G₁ and G₂ cyclin antibody; 5 cyclin restriction points can serve as additional cell cycle landmarks to map the point of action of antitumour drugs; 6 unscheduled expression of cyclins (e.g. the presence of cyclin B1 during G₁ and S) can be detected in several tumour transformed cell lines, possibly indicating disregulation of the machmery of cell cycle progression. The last finding 6 is of special importance, because such disregulation may be of prognostic consequence in human tumours.     

Cytokinetic studies on the switch from glucose to uridine metabolism, and vice versa, of Ehrlich ascites tumour cells in vitro

Abstract. Glucose is normally required as the energy source and for the proliferation of neoplastic cells. For Ehrlich ascites tumour cells, kept under glucose-free culture conditions, this requirement was alleviated by uridine, indicating that the supply of ribose is obligatory for sustaining growth capacity.
In a 96-hr culture experiment with mouse-derived cells, the increase in cell number from cultures supplemented with 5 mM uridine was 50–70%, whilst lactate production was 5% that of controls. An increase in the number of multinucleate cells was observed from cell-smears; DNA histograms indicated the presence of cells with a DNA content higher than 4c and an increased portion of cells in G₂ phase. For precise determination of changes in cell cycle distribution on transfer of cells from glucose-supplemented to glucose-free conditions, the progression of phase-accumulated cells (by centrifugal elutriation) was monitored by DNA distribution analysis; G₂ cells continued the cycle at a rate comparable to controls but were delayed, in the following cycle, predominantly in S and G₂ phases. This was also observed with G₁ cells from a G₁-accumulated fraction in the first cycle.
The addition of glucose to cells kept for some hours in glucose-free, uridine-supplemented medium resulted in an immediate increase in mitotic index (amplification by the colcemid method).
The results are interpreted and support our concept that the delivery of compounds, necessary for normal growth, i.e. hexoses for glycoproteins and glycolipids, are limited as a consequence of the 'metabolic channelling' of pentose from uridine in Ehrlich ascites tumour cells. Therefore, the constantly lowered growth-rate in uridine-supplemented cells observed with long-term culture experiments could reflect an adaption of growth-cycle to these limitations.     

INCORPORATION OF TRITIATED THYMIDINE INTO THE SKIN AND HAIR FOLLICLES

The incorporation of tritiated thymidine into the skin has been studied after the initiation of new hair growth by plucking. Techniques are described whereby the incorporation into hair follicles and into the basal layer of the epidermis can be studied independently. Fluctuations are observed in the levels of incorporation in both cell populations through the hair growth cycle. These fluctuations show great regularity. the early fluctuations have been attributed to synchronous progression of cells through the cell division cycle. the significance of the fluctuations is discussed with particular reference to (I) possible persistence of cell cycle synchrony, (II) regulatory feedback mechanisms controlling DNA synthesis, perhaps through specific mitotic inhibitors, (III) the existence of G_o, G₁ and G₂ cell sub-populations.     

Laser scanning cytometer (LSC) analysis of fraction of labelled mitoses (FLM)

Abstract. In this report we describe the successful application of a novel microscope-based multiparameter laser scanning cytometer (LSC) to measure duration of different phases of cell cycle in HL-60 human leukaemic cell lines by the fraction of labelled mitoses (FLM) method. Exponentially growing cells were harvested after various time intervals following pulse-labelling with 5'-bromo-2'-deoxyuridine (BrdUrd), cytocentrifuged, fixed in ethanol, and then exposed to UV light to induce DNA strand breaks at the sites of incorporated BrdUrd. The 3'OH termini of the photolytically generated DNA strand breaks were labelled with BrdUTP in the reaction catalysed by exogenous terminal deoxynucleotidyl transferase (TdT), followed by FITC-labelled BrdUrd antibodies. DNA was counterstained with propidium iodide (PI). Due to differences in chromatin structure between the interphase and mitotic cells, the LSC identified the latter by virtue of their higher red (PI) fluorescence intensity values among all pixels over the measured cell. To confirm that the cells selected were indeed cells in mitosis, predominantly in metaphase, the recorded X-Y coordinates of selected cells were used to re-position the cell for their visual examination. From the time lapse analysis of percentage BrdUrd-labelled cells progressing through mitosis it was possible to calculate the duration of individual phases of the cell cycle. The duration of S (T_s) and G₂+ M (T_G2+M) was 8 and 3 h, respectively, and the minimal duration of G₂ (T_G2) was 2 h. The cell cycle time (T_c) estimated for the cohort of the most rapidly progressing cells was 13 h. The ability to automatically and rapidly discriminate mitotic cells combined with the possibility of their subsequent identification by image analysis makes LSC the instrument of choice for the FLM analysis.     

A dual block to cell cycle progression in HL60 cells exposed to analogues of vitamin D,

Abstract. The physiologically active form of vitamin D₃, 1,25-dihydroxy-vitamin D₃, (1,25(OH)₂, D₃), induces differentiation of several types of myeloid leukaemia cells. The acquisition of monocyte-like phenotype is accompanied by slower progression through the cell cycle, and G₁, block has been reported to be the basis of this effect. It is shown here that human promyelocytic leukaemia HL60 cells treated with analogues of vitamin D₃, which are potent inducers of monocytic differentiation, have an additional cell cycle block. Exposure to 10-⁷m 1,25(OH)₂, D₃, or 1,25-(OH)₂,-16-ene-D₃ resulted in monocytic differentiation and the expected G₁, block evident at approximately 48 h in a rapidly differentiating variant of HL60 cells (HL60-G), and at 96 h in the more slowly differentiating HL60-240 cells. In addition, a G₂,+M block was noted at approximately 72 h in HL60-G and HL60-240 cells. Exposure to vitamin D₃, analogues also markedly increased the number of dikaryons, suggesting that cytokinesis was impaired more than karyokinesis. Treatment with a third analogue 25-hydroxy-16,23-diene-D₃, produced little differentiation and had minimal effects on the cell cycle parameters. These findings indicate that vitamin D₃, analogues regulate cell proliferation by control of the transition of G₁, and G₂,+M phases, reminiscent of the cdc2/CDK2 type of cell cycle control.     

Synthesis of protein and mRNA is necessary for transition of suspension-cultured Catharanthus roseus cells from the G1 to the S phase of the cell cycle

The effects of inhibition of the synthesis of protein, mRNA or rRNA on the progression of the cell cycle have been analyzed in cultures of Catharanthus roseus in which cells were induced to divide in synchrony by the double phosphate starvation method. The partial inhibition of protein synthesis at the G₁ phase by anisoniycio or cycloheximide caused the arrest of cells in the G₁ phase or delayed the entry of cells into the S phase. When protein synthesis was partially inhibited at the S phase, cell division occurred to about the same extent as in the control. When asynchronously dividing cells were treated with cycloheximide, cells accumulated in the G₁ phase, as shown by flow-cytometric analysis. The partial inhibition of mRNA synthesis by α-amanitin at the G₁ phase caused the arrest of cells in the G₁ phase, although partial inhibition of mRNA synthesis at the S phase had little effect on cell division. In the case of inhibition of synthesis of rRNA by actinomycin D at the G₁ phase, initiation of DNA synthesis was observed, but no subsequent DNA synthesis or the division of cells occurred. However, the addition of actinomycin D during the S phase had no effect on cell division. These results suggest that specific protein(s), required for the progression of the cell cycle, are synthesized in the G₁ phase, and that the mRNA(s) that encode these proteins are also synthesized at the G₁ phase.     

Method For Probing Cells In Radiation-Induced G2Arrest: Demonstration of Potentially Lethal Damage Repair

Abstract. Chinese hamster ovary cells were arrested in the G₂ phase of the cell cycle by X-irradiation. When subsequently treated with 5 mM caffeine the arrested population progressed into mitosis as a synchronous cohort where it was harvested by mitotic cell selection. This procedure provides a means to isolate cell populations treated in G₂, for the investigation of G₂ arrest. Comparisons were made of the number of cells retrieved from G₂ arrest with the number suffering arrest, as determined by flow cytometry and by matrix algebraic simulations of irradiated cell progression. the retrieved population was not significantly less than expected for doses up to 3.5 Gy, indicating that the retrieval process does not favour the isolation of any population subset below this dose. Cell populations retrieved from arrest at varying intervals (0-3 h) after irradiation (0-3.5 Gy) showed an increase in survival with increase in interval, consistent with repair of potentially lethal damage. the repair curves (surviving fraction us time) were each described by a single exponential. G₂ cells that were brought to mitosis without a period of arrest exhibited the same radiation response as cells irradiated in mitosis.     

A Changing Pattern of the Cell Cycle during the First Two Cleavage Divisions of the Mouse

The cell cycles of the early cleavage stages of the mouse were analyzed by examining Feulgen-stained ova. The period from ovulation to the completion of second cleavage division was investigated. The ova donors were C57BL/6 × DBA/2 female mice, which were hormonally superovulated. To estimate the durations of DNA synthesis and mitotic phases of the cleavage divisions, the ova were pooled into culture medium, and as a function of time, aliquots were removed from the batch of pooled ova. The ova specimens were Feulgen-stained and classified as the ova nuclei in G₁, S, G₂ or mitosis by use of a cytophotometric technique and then the durations of these phases were determined by probit analysis.
The pronuclear stage had a generation time of 11 hr, with a G₁ phase of 6 hr and a short S phase of 1.7 hr. In contrast the two-cell stage had a generation time of 18 hr, with a G₁ phase of 2 hr and an S phase of 3 hr. The duration of cleavage division also changed; the first cleavage division spanned 3 hr while the second spanned 1 hr.     

cAMP IN THE CELL CYCLE OF THE DINOFLAGELLATE CRYPTHECODINIUM COHNII (DINOPHYTA)

The second messenger cAMP is a key regulator of growth in many cells. Previous studies showed that cAMP could reverse the growth inhibition of indoleamines in the dinoflagellate Crypthecodinium cohnii Biecheler. In the present study, we measured the level of intracellular cAMP during the cell cycle of C. cohnii . cAMP peaked during the G₁ phase and decreased to a minimum during S phase. Similarly, cAMP-dependent protein kinase activities peaked at both G₁ and G₂+M phases of the cell cycle, decreasing to a minimum at S phase. Addition of N6, O₂'-dibutyryl (Bt₂)-cAMP directly stimulated the growth of C. cohnii . Flow cytometric analysis of synchronized C. cohnii cells suggested that 1 mM cAMP shortened the cell cycle, probably at the exit from mitosis. The size of Bt₂-cAMP treated cells at G₁ was also larger than the control cells. The present study demonstrated a regulatory role of cAMP in the cell cycle progression in dinoflagellates.     

CELL CYCLE PROPERTIES OF DIFFERENTIATING SPERMATOGONIA IN ADULT SPRAGUE-DAWLEY RATS

The duration of the mitotic cycle and of its components was analysed for each of the six successive generations of differentiating spermatogonia (A₁, A₂, A₃, A4, intermediate and B), using radioautographed whole mounts of seminiferous tubules from testes of adult Sprague-Dawley rats. Cell cycles were determined from two successive waves of per cent labeled metaphases obtained during the period of 81 hr after a single dose of ³H-thymidine. Except for the A₁ spermatogonia, all spermatogonial types (A₂ to B) had similar cell cycle durations of 41-42.5 hr and comparable pre-DNA synthesis phases (G₁) of 11-13 hr. Although the combined duration of DNA synthesis (S) and the post-synthesis phase (G₂) remained identical for all the cell types including A₁, there was a progressive lengthening of the S period at the expense of G₂ during the process of spermatogonial maturation. This change was most marked during the transition from A₁ to A₃ spermatogonia when the S period increased from 14 hr to 21 hr, and the G₂ phase shortened from 13 hr to 7.5 hr. This feature seems to be unique to germ cells and may be associated with an increasing amount of heterochromatin in the nucleus. Excluding the development of type A₁ cells, the entire process of spermatogonial maturation lasted for 208 hr. Combined data on cell cycle times indicated that every 313 hr or 13 days, a new sequence of spermatogonial differentiation was initiated by the A₁ cells. This was equivalent to the duration of one 'cycle' of the seminiferous epithelium as measured by other techniques.     

New methods for calculating kinetic properties of cells in vitro using pulse labelling with bromodeoxyuridine

Abstract. The transit times of Chinese hamster ovary cells through the phases of their cell cycle were measured using dual parameter flow cytometry to measure DNA content and the presence of monoclonal antibodies to bromodeoxyuridine. Up to four separate populations can be accurately measured: unlabelled cells in G₂+ M; labelled cells that have not yet divided; labelled cells that have already divided; and the unlabelled cells that were originally in G₁ plus the cells that were originally in G₂+ M and have since divided. The fractions of cells in these populations can be easily followed in time and the usual kinetic properties can be estimated from these fractions, or combinations thereof, including the times through G₁, S, G₂+ M and the cycle time. We present equations for analysing this type of data and comment on which equations are most appropriate for measuring specific kinetic properties of the cells.     

Cell cycle progression in human cells following re-oxygenation after extreme hypoxia: consequences concerning initiation of DNA synthesis

Abstract. The initiation of DNA synthesis and further cell cycle progression in cells during and following exposure to extremely hypoxic conditions in either G₁ or G₂+M has been studied in human NHIK 3025 cells. Populations of cells, synchronized by mitotic selection, were rendered extremely hypoxic (< 4 p.p.m. O₂) for up to 24n h. Cell cycle progression was studied from flow cytometric DNA recordings. No accumulation of DNA was found to take place during extreme hypoxia. Cells initially in G₁ at the onset of treatment did not enter S during up to 24 h exposure to extreme hypoxia, but started DNA synthesis in a highly synchronous manner within 1.5 to 2.25 h after reoxygenation. The duration of S phase was only slightly affected (increased by ≅10%) by the hypoxic treatment. This suggests that the DNA synthesizing machinery either remains intact during hypoxia or is rapidly restored after reoxygenation. Cells initially in G₂ at the onset of hypoxia were able to complete mitosis, but further cell cycle progression was blocked in the subsequent G^ Following reoxygenation, these cells progressed into S phase, but the initiation of DNA synthesis was delayed for a period corresponding to at least the duration of normal G₁ and did not appear in a synchronous manner. In fact, cell cycle variability was found to be increased rather than decreased as a result of exposure to hypoxia starting in G₂. We interpret these findings as an indication that important steps in the preparation for initiation of DNA synthesis take place before mitosis. Furthermore, the change in cell cycle duration induced by hypoxia commencing in G₁ is of a nature other than that induced by hypoxia commencing in other parts of the cell cycle.     

CELL CYCLE COMPARTMENT ANALYSIS OF CHINESE HAMSTER CELLS IN STATIONARY PHASE CULTURES

The distribution of Chinese hamster cells with respect to the compartments of the cell generation cycle was studied in cultures in the stationary phase of growth in two different media. A measure of the state of depletion of the nutrient medium was formulated by defining a quantity termed the nutritive capacity of the medium. This quantity was used to verify that the cessation of cell proliferation is due to nutrient deficiencies and not to density dependent growth inhibition. Cell cultures in stationary phase were diluted into fresh medium and as growth resumed, mitotic index, cumulative mitotic index, label index and viability were measured as a function of time. The distribution of cells with respect to compartments of the cell generation cycle in stationary phase populations was reconstructed from these data. Stationary phase populations of Chinese hamster cells that retained the capacity for renewed growth when diluted into fresh medium were found to be arrested in the G₁ and G₂ portions of the cycle; the relative proportion of these cells in G₁ increased with time in the stationary phase, but the sequence differs in the two media. In early stationary phase, in the less rich medium, more cells are in G₂ than in G₁. Also in this medium a fraction of the population was observed to be synthesizing DNA during stationary phase, but this fraction was not stimulated to renewed growth by dilution into fresh medium.     

Histones and the first cell cycle in maize germination

The timing of the onset of cell division during seed germination in maize and the role of histones for this process have been studied. Embryonic axes of maize seeds ( Zea mays L. hybrid H-30) were incubated in a sterile nutrient medium for different periods of time. For some experiments putrescine was also added. Mesocotyl, root tip and scutellar node were dissected at specific periods after incubation and the mitotic indices were determined in these tissues. Embryonic axes were incubated in the same medium either with [¹⁴C]-lysine or [³²P]-phosphate. The incorporation of either ¹⁴C or ³²P into histones was followed, both in postribosomal supernatant and in nuclei. It was found that during germination, there is specific timing for meristematic cells entering into cell division. Among the tissues tested, the mesocotyl meristem was the first to initiate this process. De novo synthesis of histones was detected as early as after 6 h of imbibition and the rate increased up to 12 h. Putrescine stimulated cell division and phosphorylation of the histones. The implications of these findings are discussed.