Trihalomethanes in drinking water and human colorectal cancer.

The relation of trihalomethanes (THM) to colorectal cancer was evaluated. A total of 395 colorectal cancer deaths among white women teachers in New York State was compared to an equal number of deaths of teachers from noncancerous causes. Cumulative chloroform (CHCl3) exposure was estimated by the application of a statistical model to operational records from the individual water treatment facilities that served the home and work addresses of each study subject during the 20 years prior to death. The odds of exposure to a surface source containing little or no THM was no greater for cases than for controls. The odds ratio = 1.07; the 90% confidence interval = 0.79-1.43; and the P = .68. The distribution of CHCl3 exposure was not significantly different between cases and controls (rated by Wilcoxon signed rank statistic = -0.52; P = .60). No effect of cumulative CHCl3 exposure on outcome was seen in a logistic analysis controlling for average source type, population density, marital status, age, and year of death (likelihood ratio test statistic = 0.047; P = .83).

A rapid, semi-automated procedure for the evaluation of leukocyte locomotion in the micropore filter assay.

A commerically available bacterial colony counter has been modified to allow rapid, accurate, semi-automated evaluation of cell numbers in the micropore filter assay for chemotaxis. The method is valuable for objective, rapid evaluation of cell counts at various levels through the filter, as well as counts on the distal surface of the filter. Coupled with a programmable calculator, this instrument had made feasible a new method of assessing random migration by the regression line analysis, which discriminates between migration rate and mass migration of cells. This combination of equipment may thus serve as a considerable time saving accessory to laboratories engaged in cell locomotion research, but also will allow more rigorous assessment of differences among specific populations of cells.

Mutagenicity of 1,3-butadiene at the Hprt locus of T-lymphocytes following inhalation exposures of female mice and rats.

The species specific response to 1,3-butadiene (BD), an important industrial chemical, was investigated by determining the influence of exposure duration and exposure concentration on the mutagenicity of BD in mice and rats and by defining the spectra of mutations in the Hprt gene T-cell mutants from control and BD-exposed mice. Female B6C3F1 mice and F344 rats (4-5 weeks old) were exposed by inhalation to 0, 20, 62.5, or 625 ppm of BD for up to 4 weeks (6 h/day, 5 days/week). Groups of control and exposed animals (n=4-12/group) were necropsied at multiple time points after exposure and the T-cell cloning assay was used to measure Hprt mutant frequencies in lymphocytes isolated from spleen. Mutant clones collected from control and BD-exposed mice were propagated and analyzed by RT-PCR to produce Hprt cDNA for sequencing. In animals necropsied 4 weeks after 2 or 4 weeks of BD exposure (0 or 625 ppm), the rate of accumulation of mutations was greater in mice than in rats. Supra-linear dose-response curves were observed in BD-exposed mice, indicating a higher efficiency of mutant induction at lower concentrations of BD. The mutagenic potency estimates (represented by the differences in the areas under the mutant T-cell 'manifestation' curves of treated vs. control animals) in mice were 11 and 61 following 4 weeks of exposures to 62.5 and 625 ppm of BD, respectively, while mutant frequencies (Mfs) in rats were significantly increased only at 625 ppm BD (mutagenic potency of 7). Molecular analysis of Hprt cDNA from expanded T-cell clones from control and BD-exposed mice demonstrated an increased frequency of mutants in exposed animals that likely contain large deletions in the Hprt gene (P=0.016). These data indicate that both exposure duration and exposure concentration are important in determining the magnitude of mutagenic response to BD, and that mutagenic and carcinogenic properties of BD in mice may be related more to the ability of its metabolites to cause chromosomal deletions than to produce point mutations.

Mutagenicity of the racemic mixtures of butadiene monoepoxide and butadiene diepoxide at the Hprt locus of T-lymphocytes following inhalation exposures of female mice and rats.

The purpose of this study was to determine if Hprt mutant frequency (Mf) data from rodents exposed directly to individual epoxy metabolites of 1,3-butadiene (BD) can be used to identify the relative significance of each intermediate in the mutagenicity of BD in mice vs. rats. To this end, the relative contributions of the racemic mixtures of BD monoepoxide (BDO) and BD diepoxide (BDO(2)) to BD-induced mutagenicity was investigated by exposing mice and rats to selected concentrations of BDO and BDO(2) (i.e., 2.5 and 4.0 ppm, respectively) and comparing the mutagenic potency of each intermediate to that of BD (at 62.5 ppm) when comparable blood levels of metabolites are achieved (in the mouse). Female B6C3F1 mice and F344 rats (4-5 weeks old) were exposed to rac-BDO (0, 2.5, or 25 ppm) or (+/-)-BDO(2) (0, 2, 4 ppm) by inhalation for 4 weeks (6 h/day, 5 days/week), and then groups of control and exposed animals (n=3-12/group) were necropsied at multiple time points post-exposure for measuring Hprt Mfs in splenic lymphocytes (via the T-cell cloning assay) and estimating mutagenic potencies (represented by the difference in the areas under the mutant T-cell 'manifestation' curves of treated vs. control animals). The resulting Mf data, along with the extant metabolism data, suggest that at lower BD exposures (

Relationships between exposure, cell loss and proliferation, and manifestation of Hprt mutant T cells following treatment of preweanling, weanling, and adult male mice with N-ethyl-N-nitrosourea.

Experiments were performed to characterize the age-related patterns of appearance and frequency of hypoxanthine-guanine phosphoribosyl transferase (Hprt) mutant T lymphocytes in thymus and spleen following exposure of preweanling (12-day-old), weanling (22-day-old), and young adult (8-week-old) male B6C3F1 mice to ethylnitrosourea (ENU). Mice were given single i.p. injections of 0 or 40 mg ENU/kg and then groups of animals were necropsied from 2 h to 116 days after treatment to examine the relationships between exposure, cell loss and proliferation, and the frequency of Hprt mutant T cells in thymus and spleen. Hprt mutant frequency (Mf) data for thymus of ENU-exposed (0, 11.7, 35, 58, or 72 mg/kg, or five weekly doses of 1.7 mg/kg i.p.) male C57BL/6 mice (12- or 62-week-old), obtained during an earlier study of spleen cells [I.M. Jones, K. Burkhart-Schultz, C.L. Strout, T.L. Crippen, Factors that affect the frequency of thioguanine-resistant lymphocytes in mice following exposure to ethylnitrosourea, Environ. Mutagen, 9 (1987) 317-329.], were compared to results in B6C3F1 mice. Isolated T cells were cultured in the presence of mitogen, growth factor, and 6-thioguanine to detect Hprt mutants. The time required to achieve maximum Mfs in thymus was uniformly found at 2 weeks after ENU treatment, while the times needed to reach peak values in spleen were proportional to animal age at treatment. These data indicate that age-related differences in the appearance of Hprt mutant cells in spleen are largely defined by the physiologically based, age-dependent trafficking of mutant cells from or through the thymus. Three modes of handling the resulting Hprt Mf data were evaluated: (i) comparing the Mfs at a single time point, (ii) comparing the maximum Mfs observed, and (iii) comparing the change in Mfs over time (or the mutant T cell 'manifestation' curves in treated vs. control mice) in each age group post-exposure. Measuring the Mfs in spleen at multiple time points after cessation of exposure and integrating the frequency of mutants as a function of time appeared to be the superior method for comparing mutagenic responses in different age groups. Some of the underlying assumptions of this approach, as well as its strengths and weaknesses, are discussed.

Patterns of blood-vessel invasion by mammary tumor cells.

Rat mammary tumor TMT-081 was employed as a model for blood vessel invasion because its mode of metastasis resembles that of human tumors. The invasive mechanism was studied with two methods of serial transplantation: transfer of enzymatically dispersed solid tumors, and transfer of buffy coat containing circulating tumor cells. The latter method produced greater invasion of blood vessels, including larger veins and occasionally arteries, perhaps by obviating damage to tumor cells during enzyme treatment. The course of migration was traced by three-dimensional examination in the high voltage electron microscope, as well as the light microscope. Two broad patterns were found for the course of invasion of small and large vessels respectively.

Maximum likelihood estimation of subsequence conservation.

A statistical method is presented for comparing protein sequences by partitioning the polymers and estimating each subsegment's degree of conservation. Conservation is measured as a function of the number of transitions occurring in the underlying time homogeneous Markov process assumed to govern amino acid mutations. The Markovian assumption also permits estimation of the ancestral sequence. Partitioning and estimation are carried out via maximum likelihood. The method is contrasted with the commonly utilized percent homology measure. A moving likelihood ratio plot to aid in identifying regions of high conservation is suggested as an analogue to moving hydrophobicity plots. An application is presented which identifies highly conserved regions in thymidylate synthase from L. casei and E. coli.

An expectation maximization (EM) algorithm for the identification and characterization of common sites in unaligned biopolymer sequences.

Statistical methodology for the identification and characterization of protein binding sites in a set of unaligned DNA fragments is presented. Each sequence must contain at least one common site. No alignment of the sites is required. Instead, the uncertainty in the location of the sites is handled by employing the missing information principle to develop an "expectation maximization" (EM) algorithm. This approach allows for the simultaneous identification of the sites and characterization of the binding motifs. The reliability of the algorithm increases with the number of fragments, but the computations increase only linearly. The method is illustrated with an example, using known cyclic adenosine monophosphate receptor protein (CRP) binding sites. The final motif is utilized in a search for undiscovered CRP binding sites.

Evidence that the interferon-induced Daudi cell human lupus inclusions are de novo synthesized complexes of ribonucleoprotein and membrane.

Quantitative electron microscope autoradiography has been used to define the macromolecular composition of the interferon-induced human lupus-type inclusions (LI) in the human B lymphoblastoid cell line, Daudi. LI were first apparent in Daudi cell cultures 12 h after the addition of 100 units/ml of the purified recombinant human leukocyte interferon, IFLrA. Radiolabels were added at this time and allowed to incorporate over the following 12 h during which an estimated greater than 99% of the LI material present at 24 h was formed. The LI-incorporated radiolabels were present only during this discrete 12-h period after the interferon activation of LI cell pathways in order to detect LIs de novo synthesized macromolecular components. The estimate relative specific activities of the LI-incorporated radiolabels were: choline at 4.042, mannose at 2.631, uridine at 0.664, glucosamine at 0.578, and amino acids at 0.477. With thymidine the estimated LI specific activity was 0.000. LI isolated from whole cells retained the tubular elements and the interwoven membrane network. These results provide direct evidence that the interferon-induced Daudi cell LI are de novo synthesized complexes of ribonucleoprotein and membrane.

Observational study of erythrocyte protoporphyrin screening test for detecting low lead exposure in children: impact of lowering the blood lead action threshold.

We examined a retrospective sample of 1800 children on whom both erythrocyte protoporphyrin (EP) and blood lead (BPb) measurements were taken. The primary objective was to ascertain whether EP is a cost-effective screening test for low but increased BPb concentrations and to establish the optimal thresholds. The data did not provide evidence of an EP threshold at low BPb concentrations; however, the data did show a significant age effect. A subset of 500 children for whom both EP and hematocrit data were available showed no correlation between those variables. Age-specific operating characteristic curves, total error, and cost analyses are presented. The latter sets bounds on the relative cost of EP testing, above which only BPb determination should be performed. The implications of these findings are discussed in light of impending changes in U.S. federal guidelines for preventing lead poisoning in young children.

Effect of the ratio of free to total prostate-specific antigen on interassay variability in proficiency test samples.

BACKGROUND: Up to sevenfold differences were observed between total prostate-specific antigen (PSA) methods for New York State Proficiency Test samples prepared with seminal fluid PSA in human female serum. Because the PSA was mainly in its free form under these conditions, we wanted to determine whether a defined mixture of free and complexed PSA would reduce the interassay differences. METHODS: We prepared a series of five solutions of 60 g/L bovine serum albumin with 10 microgram/L total PSA consisting of varied proportions of free, noncomplexible PSA, and alpha(1)-antichymotrypsin (ACT)-complexed PSA from 0% to 100%. Two hundred seventy laboratories measured the total PSA in these samples, and 16 laboratories also analyzed the samples for free PSA. The results were used to calculate free/total PSA ratios. RESULTS: Interassay CVs for total PSA measurements were approximately 7% at 10-15% free PSA but became gradually larger as the free/total PSA ratio increased. Measured free-PSA concentrations were similar within each sample (mean CV, 12%), and the results were relatively independent of the proportion of free PSA in the samples. Twofold discrepancies between actual and expected ratios were observed with some methods at 100% free PSA and to a lesser degree at 30% free PSA. At 100% free PSA, the relatively higher total-PSA values measured by nonequimolar methods yielded low free/total PSA ratios of 50-60%. In contrast, the lower total PSA values obtained by equimolar methods yielded ratios close to the expected 100%. CONCLUSIONS: Preparing proficiency test samples with a 10:90 mixture of free, noncomplexible PSA:PSA-ACT is a viable alternative to the use of seminal fluid PSA. Furthermore, the method used to measure total PSA may have a substantial impact on the calculated proportion of free PSA and hence may have clinical relevance.

Multivariate discrimination for phenylketonuria (PKU) and non-PKU hyperphenylalaninemia after analysis of newborns' dried blood-spot specimens for six amino acids by ion-exchange chromatography.

Ion-exchange HPLC was developed for testing dried blood-spot specimens from newborns. The method is suitable for quantitative confirmatory testing of abnormal specimens detected in the New York State Newborn Screening Program. Positive specimens were initially identified among all New York State newborns with semiquantitative bacterial inhibition assays (BIA) for aminoacidopathies, including phenylketonuria (PKU) and non-PKU hyperphenylalaninemia (HP), maple syrup urine disease, and homocystinuria. A selection of 1346 specimens from routine BIA screening, including 131 newborns with PKU or persistent HP, were tested by HPLC. Of 179 BIA results that were falsely positive, 98 (55%) were also falsely positive by HPLC in which the Phe/Tyr ratio was the discriminator and the threshold was set to attain 100% sensitivity. Investigation of three multivariate discriminatory methods revealed that linear discriminant analysis excluded all but 35 (20%) of the BIA false-positives.

Screening children exposed to lead: an assessment of the capillary blood lead fingerstick test.

We describe results of a 3-year study in which 499 paired venous and capillary blood specimens, collected by fingerstick on the same day, were analyzed for lead (BPb) and erythrocyte protoporphyrin (EP). False-positive rates (FPRs) and the proportion of false positives were calculated at four BPb thresholds. At the 100 microg/L threshold, the FPR for all data was 13%, but the proportion of false positives was only 5%. The log ratios of capillary-to-venous BPb data indicate that, with the exception of eight outliers, two subpopulations exist that follow a log-normal distribution. These two subpopulations, the "core" (n = 303) and "shifted" (n = 188) groups, on average generated a positive bias at 100 microg/L BPb of 8.6% and 30.3%, respectively. The log ratios of capillary-to-venous EP data followed a normal distribution, indicating that capillary EP is not statistically different from venous EP.

Estimating serum polychlorinated biphenyl levels in highly exposed workers: an empirical model.

A regression model estimating high-homolog polychlorinated biphenyl (PCB) serum concentration on the basis of job exposure categorizations was developed. The model assumes first-order kinetics with a half-life determined empirically and uses variables that incorporate both intensity and duration of exposure over a 30-yr period. In order to compare the efficiency of these regression-based exposure estimates relative to often-used epidemiological parameters, models with dichotomized, ordinal, and continuous exposure surrogates were also investigated. Among the alternative exposure categorizations the most straightforward measure, ever versus never direct, was a particularly poor predictor of serum PCB level (r2 = .01). Nearly all of the candidate exposure measures we tried predicted serum levels poorly. The best of these after the fact was with total months employed in direct-exposure jobs (r2 = .43). None of the logical deductive models approached the predictability of the empirical model developed here (r2 = .69).

Regression line analysis of neutrophil chemotaxis.

Polymorphonuclear leukocyte chemotaxis and chemokinesis, in the micropore filter assay, have been evaluated using Regression Line Analysis. The method discriminates between migration rate and the number of migrating cells by statistical comparison of the slope of the regression line and the y-intercept, respectively. The derivation of this new method of assessment of cell migration is reported. This report also includes critical evaluation of the chemokinesis dose/response curve and the directional component of chemotaxis. The effect of inhibitors of chemotaxis were evaluated using the new method. Colchicine was found to effect the directional component of chemotaxis, as indicated by decreased rate of locomotion in chemotaxis experiments, and had no effect on random migration and chemokinesis. Cytochalasin B had a general inhibitory effect upon cell locomotion by depressing the rate of random migration and chemotaxis. 1-Tosylamide-2-phenylethyl chloromethylketone (TPCK) had the apparent effect of uncoupling the receptor and effector mechanisms of chemotaxis or altering the binding of the chemotactic ligand, since random migration was unaffected but the rate of chemotaxis and chemokinesis were depressed. Ionophore A23187 had a generalized inhibitory effect on the cell rate under all three conditions; however, at low concentrations it induced an increase in the number of migrating cells. These studies indicate that the method of regression line analysis is effective in differentiating in vitro the mechanisms of chemotaxis inhibition. The method may be used effectively in the assessment of defective neutrophil chemotaxis in disease states and may prove a useful adjunct to the study of cell locomotory mechanisms.

Intercellular transfer of a soluble viral superantigen.

Mouse mammary tumor virus (MMTV) superantigens (vSAgs) can undergo intercellular transfer in vivo and in vitro such that a vSAg can be presented to T cells by major histocompatibility complex (MHC) class II proteins on antigen-presenting cells (APCs) that do not express the superantigen. This process may allow T-cell activation to occur prior to viral infection. Consistent with these findings, vSAg produced by Chinese hamster ovary (CHO) cells was readily transferred to class II IE and IA (H-2(k) and H-2(d)) proteins on a B-cell lymphoma or mouse splenocytes. Fixed class II-expressing acceptor cells were used to demonstrate that the vSAg, but not the class II proteins, underwent intercellular transfer, indicating that vSAg binding to class II MHC could occur directly at the cell surface. Intercellular transfer also occurred efficiently to splenocytes from endogenous retrovirus-free mice, indicating that other proviral proteins were not involved. Presentation of vSAg7 produced by a class II-negative, furin protease-deficient CHO variant (FD11) was unsuccessful, indicating that proteolytic processing was a requisite event and that proteolytic activity could not be provided by an endoprotease on the acceptor APC. Furthermore, vSAg presentation was effected using cell-free supernatant from class II-negative, vSAg-positive cells, indicating that a soluble molecule, most likely produced by proteolytic processing, was sufficient to stimulate T cells. Because the membrane-proximal endoproteolytic cleavage site in the vSAg (residues 68 to 71) was not necessary for intercellular transfer, the data support the notion that the carboxy-terminal endoproteolytic cleavage product is an active vSAg moiety.

Habituation of activity in an open field: A survey of inbred strains and F1 hybrids.

To determine if there is genetic variability in habituation of activity in an open field, we examined a number of inbred strains and F1 hybrids. Using 5-min exposures to a dark open field, we measured changes in exploratory behavior over 3 consecutive days in 129S3/SvImJ, A/J, BALB/ cByJ, C3H/HeJ, C57BL/6J, CBA/J, DBA/2J, FVB/NJ, (B6 x 129)F1/J, and (B6 x C3H) F1/J male and female mice. Strain differences in open-field activity and in habituation were evident. Some of the strain differences were further modified by sex. The strains and F1's could be separated into groups that increased, decreased, or did not modify their activities across testing sessions. In a second study, the effects of altering the floor surface on habituation were examined in male 129S3/SvImJ, C57BL/6J, DBA/2J, and (B6 x 129)F1/J mice. When the floor was altered after 3 consecutive days of habituation, increased activity levels were evident. There were strain differences in the responsiveness to the changes in the floor. These results confirm a genetic role in intersession habituation to an open field.

Population-based studies reveal differences in the allelic frequencies of two functionally significant human interleukin-4 receptor polymorphisms in several ethnic groups.

PURPOSE: The presence of functionally significant human interleukin-4 receptor sequence variants, Gln551Arg and Ile50Val, was examined in four anonymous New York State populations defined by ethnic origin. These variants were studied because they are associated with atopy or atopic asthma whose prevalence varies in different populations. METHODS: PCR/RFLP (Ile50Val) and PCR/allele-specific oligonucleotide hybridization (Gln551Arg) assays were developed to detect both polymorphisms in 855 newborn screening specimens. RESULTS: Arg551 was most frequently found in Blacks (allele frequency of 68%). However, the Ile50 allele was most common in Whites (allele frequency, 87%). Significantly more Blacks had chromosomes bearing both of the "enhanced signaling" variants (Ile50/Arg551). CONCLUSIONS: Enhanced IL-4R signaling is associated with increased IgE production (atopy). Therefore, our data suggest that the African American population may be at increased risk for diseases, including asthma, which are associated with atopy. These data also emphasize the importance of determining the frequencies of single nucleotide polymorphisms in different populations before drawing conclusions from allele association studies, since the background allele frequencies may be disparate between different populations.

Comparison of the mutagenic potency of 1,3-butadiene at the hprt locus of T-lymphocytes following inhalation exposure of female B6C3F1 mice and F344 rats.

1,3-Butadiene (BD) is an indirect alkylating agent that has greater cancer potency in the mouse than in the rat. The purpose of the present study was to compare the mutagenic potency of BD at the hprt locus of T-lymphocytes of exposed mice and rats and to determine whether mutations induced in this marker gene can be used as a quantitative indicator for species differences in susceptibility to cancer. To this end, experiments were conducted to define the effects of exposure duration and the time elapsed after exposures on the frequency of hprt mutations (Mf) in T-cells from female B6C3F1 mice and F344 rats of similar age (4-5 weeks) when exposed to BD by inhalation. The accumulation of hprt mutations in T-cells from thymus was assessed in animals necropsied 2 weeks after exposure to 0 or 1250 ppm BD for 1 or 2 weeks, while the time course for the appearance of hprt mutant T-cells (i.e., the phenotypic expression and cell migration) in thymus and spleen was evaluated in animals necropsied at weekly/biweekly intervals up to 10 weeks after exposure for 2 weeks. At necropsy, T-cells were isolated from thymus and spleen and cultured in the presence of IL-2, concanavalin A, and 6-thioguanine (Walker and Skopek, Mutat. Res., 288, 151-162, 1993). BD exposures of 1 and 2 weeks led to mutagenic effects in mouse thymus, with the average Mfs being 3- and 5-fold greater than background values, respectively. In rat thymus, there was only a 1.7-fold increase in Mfs after 2 weeks of BD exposure. In the mutant expression experiment, hprt Mfs in thymus and spleen of both species increased for several weeks post-exposure and then declined. Hprt Mfs in thymus reached maximum levels at 2 weeks post-exposure in mice (Mfs = 11.3 +/- 2.4 x 10(-6)) and at 3 weeks post-exposure in rats (4.9 +/- 1.2 x 10(-6)), while hprt Mfs in spleen reached peak levels at 5 weeks post-exposure in mice (19.7 +/- 1.9 x 10(-6)) and 4 weeks post-exposure in rats (10.1 +/- 1.8 x 10(-6)). Background Mfs for mouse and rat thymus and spleen ranged from 1.6 +/- 0.3 x 10(-6) to 3.0 +/- 1.1 x 10(-6). Statistical analyses of the hprt Mf data for spleen demonstrated that, under these exposure conditions, the mutagenic potency of BD (represented by the difference in the areas under the phenotypic expression curves of treated versus control animals) was 5-fold greater in mice than in rats. The magnitude of the species differences in mutagenic potency, observed after 2 weeks of BD exposure, resembles the species differences in metabolism more closely than the species differences in cancer potency.

CYP1B1 expression in human lung.

Cytochrome P450 1B1 is a recently recognized phase I bioactivating enzyme with high affinity for both inhaled tobacco carcinogens and 17beta-estradiol. We evaluated the human lung expression of this multifunctional member of the P450 superfamily across 16 individuals. Expression of CYP1B1 was evaluated by qualitative reverse transcription-polymerase chain reaction and Western immunoblots performed on human tumor and nontumor lung tissue. Expression at both mRNA and protein levels was then correlated with smoking history, plasma biomarkers of tobacco exposure (nicotine and cotinine), gender, and tumor histology. CYP1B1 mRNA and protein were detected in 94 and 100% of individuals, respectively. Multivariate analysis confirmed that there were more subjects displaying CYP1B1 mRNA expression in tumor than nontumor tissue (p = 0.0003). Correlation of CYP1B1 protein with plasma cotinine levels was statistically marginal (p = 0.027). Self-reported smoking history, gender, and tumor histology did not correlate with gene expression in the multivariate model. After multivariate modeling for confounding factors, the expression patterns of 5 of 16 individuals appeared to differ from the group as a whole for mRNA and/or protein. We conclude that CYP1B1 is commonly expressed in human lung and hypothesize that it may be an important phase I enzyme with respect to human lung carcinogen metabolism, warranting an understanding of regulatory control and coding region polymorphisms.