More than black and white: differences in predictors of obesity among Native Hawaiian/Pacific Islanders and European Americans

Although Native Hawaiians and Pacific Islanders exhibit the highest rates of obesity and associated chronic diseases of any racial/ethnic group, they remain vastly underrepresented in health research. In a cross-sectional survey of college students (N = 402) we examined BMI and health outcomes in an ethno-racially diverse rural sample of Native Hawaiian/Pacific Islanders (25.1%), Asian Americans (39.8%), and European Americans (35.1%). Measures assessed BMI, health status, health behaviors, frequency of exercise, and symptoms of psychiatric disorders (i.e., depression, anxiety, posttraumatic stress, and substance abuse and dependence). Regression analyses revealed that an overall model of five predictors (gender, race, regular exercise, difficulty sleeping, and anxiety) was significantly associated with obesity (P < 0.001) and correctly classified 84.2% of cases. A 30.7% of Native Hawaiians/Pacific Islanders were obese as compared with 9.2% of European Americans and 10.6% of Asian Americans. These findings suggest that Native Hawaiian/ Pacific Islanders are at high risk for obesity and associated medical comorbidities, but that regular physical activity may ameliorate this risk. Further, these results support the consideration of Native Hawaiians/Pacific Islanders as a distinct racial/ethnic subgroup separate from other Asian populations.     

X‐chromosome lineages and the settlement of the Americas

Most genetic studies on the origins of Native Americans have examined data from mtDNA and Y‐chromosome DNA. To complement these studies and to broaden our understanding of the origin of Native American populations, we present an analysis of 1,873 X‐chromosomes representing Native American (n = 438) and other continental populations (n = 1,435). We genotyped 36 polymorphic sites, forming an informative haplotype within an 8‐kb DNA segment spanning exon 44 of the dystrophin gene. The data reveal continuity from a common Eurasian ancestry between Europeans, Siberians, and Native Americans. However, the loss of two haplotypes frequent in Eurasia (18.8 and 7%) and the rise in frequency of a third haplotype rare elsewhere, indicate a major population bottleneck in the peopling of the Americas. Although genetic drift appears to have played a greater role in the genetic differentiation of Native Americans than in the latitudinally distributed Eurasians, we also observe a signal of a differentiated ancestry of southern and northern populations that cannot be simply explained by the serial southward dilution of genetic diversity. It is possible that the distribution of X‐chromosome lineages reflects the genetic structure of the population of Beringia, itself issued from founder effects and a source of subsequent southern colonization(s). Am J Phys Anthropol, 2009. © 2009 Wiley‐Liss, Inc.     

Nutritive composition of green and ripe pods of honey mesquite (Prosopis glandulosa,fabaceae)

Pods from Prosopis glandulosa (honey mesquite) were harvested at seven different growth stages. Protein content (N × 6.25) of the whole pod decreased with maturation with small differences occurring after stage 3. Seed contained about 82% of the total pod protein. Carbohydrate accumulated primarily in the pericarp, and ripe pods (stage 7) contained about 80% total carbohydrate. Fat content of the pod increased slightly with maturation while total ash and fiber (ADF) decreased. Small differences existed between stages 4 and 7 in the essential amino acid pattern of whole pods. The total sulfur-containing amino-acids were the most limiting in both growth stages; the pericarp protein had a relatively better amino-acid pattern than did the seed protein. Both green and ripe honey mesquite pods provide similar amounts of micro- and macro-minerals.     

Distribution of mtDNA haplogroup X among Native North Americans.

Mitochondrial DNA (mtDNA) samples of 70 Native Americans, most of whom had been found not to belong to any of the four common Native American haplogroups (A, B, C, and D), were analyzed for the presence of Dde I site losses at np 1715 and np 10394. These two mutations are characteristic of haplogroup X which might be of European origin. The first hypervariable segment (HVSI) of the non-coding control region (CR) of mtDNA of a representative selection of samples exhibiting these mutations was sequenced to confirm their assignment to haplogroup X. Thirty-two of the samples exhibited the restriction site losses characteristic of haplogroup X and, when sequenced, a representative selection (n = 11) of these exhibited the CR mutations commonly associated with haplogroup X, C --> T transitions at np 16278 and 16223, in addition to as many as three other HVSI mutations. The wide distribution of this haplogroup throughout North America, and its prehistoric presence there, are consistent with its being a fifth founding haplogroup exhibited by about 3% of modern Native Americans. Its markedly nonrandom distribution with high frequency in certain regions, as for the other four major mtDNA haplogroups, should facilitate establishing ancestor/descendant relationships between modern and prehistoric groups of Native Americans. The low frequency of haplogroups other than A, B, C, D, and X among the samples studied suggests a paucity of both recent non-Native American maternal admixture in alleged fullblood Native Americans and mutations at the restriction sites that characterize the five haplogroups as well as the absence of additional (undiscovered) founding haplogroups.     

Development of a sorghum chain in the Italian Campania Region: From the field to the celiac patient's table

Abstract

Although sorghum has been used for centuries as a food-crop in Africa and India, researchers in the United States and Europe have only relatively recently become interested in the potential of this unique cereal. Much of this interest focuses on the potential use of sorghum in food product development for individuals with allergies to foods containing wheat-based flours. Because it lacks gluten, sorghum is considered safe for people diagnosed with celiac disease, a condition marked by intolerance to gluten. Recent studies have shown that certain sorghum varieties, tan-plant sorghums, can be used to produce high-quality food and beverage products including cookies, waffles, flour, bread, noodles and beer. The intention is to promote the use and marketing of these sorghums in the Italian Campania Region for sorghum flours suitable as food for celiac patients.     

Deoxynivalenol in Mehlproben des Jahres 1999 aus dem Einzelhandel

To study the levels of deoxynivalenol (DON) in retail cereal products, wheat and rye samples were purchased in 1999 from supermarkets and “organic food” shops in Munich, Germany. DON was analysed by an enzyme immunoassay (EIA), 78 of these samples were additionally analysed by HPLC. The following contamination rates (%) and mean DON levels were found: wheat flour type 405 (n=42): 71%, 200 µg/kg; flour type 550 (n=9): 33%, 410 µg/kg; flour type 1050 (n=11): 91%, 370 µg/kg; bread-baking wheat premixes (n=14): 79%, 210 µg/kg; whole grain flour (n=20): 65%, 300 µg/kg; whole grain wheat (n=8): 75%, 280 µg/kg, wheat bran (n=20): 85%, 830 µg/kg; rye flour and grits (n=7): 29%, 120 µg/kg. HPLC confirmed the results obtained by EIA. Further analysis of 16 wheat flour (405) samples in May 2000 showed a similar frequency (69%) and mean DON level (270 µg/kg) as for samples from 1999. It is concluded that with DON levels in wheat for human consumption ranging from 200–400 µg/kg, the intake of DON has to be taken seriously in the light of the temporary tolerable daily intake of 1 µg DON per kg body weight as proposed within the European Union.     

Birthweight‐specific infant mortality risks for native Americans and whites,United States, 1960 and 1984

Abstract

We used NCHS natality and linked‐birth/death certificate tapes to compare birthweight‐specific neonatal and postneonatal mortality risks for Native Americans and whites in 1960 and in 1984. The birthweight distributions for the two groups were similar both years. Native American neonatal mortality risk dropped from 20.2 in 1960 to 5.2 in 1984, and the relative risk for Native Americans with respect to whites fell from 1.31 in 1960 to a nonsignificant difference in 1984. Postneonatal mortality risks for Native Americans fell from 27.5 in 1960 to 6.2 in 1984, with a drop in the relative risk from 5.2 to 2.1. Although the relative improvements for Native Americans were highest in postneonatal survival, Native Americans still had over twice the level of white postneonatal mortality. Birthweight was positively associated with survival for both groups, but the odds of Native American neonatal death were affected less by low and very low birthweights. For both groups, improvements in neonatal mortality were highest at the lower birthweights, while the gains in postneonatal survival benefitted normal and high birthweight infants most.     

Toward Comprehensive Obesity Prevention Programs in Native American Communities

Obesity is a particularly important challenge to the health status of Native Americans. This challenge is manifest in the increasing rates of non-insulin-dependent diabetes mellitus among Native Americans. Most studies of Native American infants, preschool children, schoolchildren, and adults have confirmed a high prevalence of over weight Historical studies suggest that for many Native American communities the high rates of obesity are a relatively recent phenomenon. The specific reasons for the increase in obesity among Native Americans have not been determined, although it has been hypothesized that Native Americans have a genetic predisposition to over weight in a “westernized” environment of abundant food and decreased energy expenditure. Few detailed studies of diet or of physical activity levels of contemporary Native Americans have been published. Community-based interventions to modify diet and activity levels to prevent obesity in Native American communities are needed. Preliminary evidence from two formative school-based programs in the Southwest suggest that Native American communities are receptive to school-based interventions, and that such programs may be able to slow the rate of excess weight gain and to improve fitness in schoolchildren. Because of the cultural diversity among Native Americans, future studies should focus on collecting community- and region-specific data, and should emphasize the need for obesity prevention through culturally appropriate community-and school-based behavioral interventions.     

The Relationship between Native American Ancestry,Body Mass Index and Diabetes Risk among Mexican-Americans

Higher body mass index (BMI) is a well-established risk factor for type 2 diabetes, and rates of obesity and type 2 diabetes are substantially higher among Mexican-Americans relative to non-Hispanic European Americans. Mexican-Americans are genetically diverse, with a highly variable distribution of Native American, European, and African ancestries. Here, we evaluate the role of Native American ancestry on BMI and diabetes risk in a well-defined Mexican-American population. Participants were randomly selected among individuals residing in the Houston area who are enrolled in the Mexican-American Cohort study. Using a custom Illumina GoldenGate Panel, we genotyped DNA from 4,662 cohort participants for 87 Ancestry-Informative Markers. On average, the participants were of 50.2% Native American ancestry, 42.7% European ancestry and 7.1% African ancestry. Using multivariate linear regression, we found BMI and Native American ancestry were inversely correlated; individuals with <20% Native American ancestry were 2.5 times more likely to be severely obese compared to those with >80% Native American ancestry. Furthermore, we demonstrated an interaction between BMI and Native American ancestry in diabetes risk among women; Native American ancestry was a strong risk factor for diabetes only among overweight and obese women (OR = 1.190 for each 10% increase in Native American ancestry). This study offers new insight into the complex relationship between obesity, genetic ancestry, and their respective effects on diabetes risk. Findings from this study may improve the diabetes risk prediction among Mexican-American individuals thereby facilitating targeted prevention strategies.     

Aboriginal overkill

Prior to European influence, predation by Native Americans was the major factor limiting the numbers and distribution of ungulates in the Intermountain West. This hypothesis is based on analyses of (1) the efficiency of Native American predation, including cooperative hunting, use of dogs, food storage, use of nonungulate foods, and hunting methods; (2) optimal-foraging studies; (3) tribal territory boundary zones as prey reservoirs; (4) species ratios, and sex and age of aboriginal ungulate kills; (5) impact of European diseases on aboriginal populations; and (6) synergism between aboriginal and carnivore predation. Native Americans had no effective conservation practices, and the manner in which they harvested ungulates was the exact opposite of any predicted conservation strategy. Native Americans acted in ways that maximized their individual fitness regardless of the impact on the environment. For humans, conservation is seldom an evolutionarily stable strategy. By limiting ungulate numbers and purposefully modifying the vegetation with fire, Native Americans structured entire plant and animal communities. Because ecosystems with native peoples are entirely different than those lacking aboriginal populations, a “hands-off” or “natural regulation” approach by today’s land managers will not duplicate the ecological conditions under which those ecosystems developed. The modern concept of wilderness as areas without human influence is a myth. North America was not a “wilderness” waiting to be discovered, instead it was home to tens of millions of aboriginal peoples before European-introduced diseases decimated their numbers.     

Did Native Americans influence the northward migration of plants during the Holocene?

Abstract Long‐distance plant dispersal explains the rapid northward migration of plant species during the Holocene but the mechanisms by which it occurred are poorly understood. Given that Native Americans spread numerous cultigens over thousands of kilometres during the late Holocene, I examined historical literature for evidence of non‐cultigen dispersal or cultivation in North America's eastern woodlands. Cultivation references are included because a strong relationship between dispersal and indigenous flora husbandry is assumed. Sixty‐seven texts describing Native American lifestyle, cultural activities, and land management reported some form of plant use. Most accounts, however, focus on cultigen production or the use of indigenous flora for medicine or food without mention of dispersal. Twenty‐four of the texts described the trade, transport, or cultivation of plants indigenous to eastern North American woodlands. Most accounts focus on the informal production of food plants, especially trees and shrubs. Confounding these reports was clear evidence of observer bias, limited botanical knowledge, acculturation, and secrecy by Native American informants. Because of these shortcomings, the likelihood of widespread long‐distance plant dispersal by Native Americans could not be determined using historical literature. This activity was either not widespread or was not observed by, or revealed to, Europeans. To adequately test the Native American plant dispersal hypothesis, direct evidence from other sources (e.g. archaeobotancial data) will be required.     

Consumption of Calcium-Fortified Cereal Bars to Improve Dietary Calcium Intake of Healthy Women: Randomized Controlled Feasibility Study

Calcium is an important structural component of the skeletal system. Although an adequate intake of calcium helps to maintain bone health and reduce the risk of osteoporosis, many women do not meet recommended daily intakes of calcium. Previous interventions studies designed to increase dietary intake of women have utilized primarily dairy sources of calcium or supplements. However, lactose intolerance, milk protein allergies, or food preferences may lead many women to exclude important dairy sources of dietary calcium. Therefore, we undertook a 9 week randomized crossover design trial to examine the potential benefit of including a non-dairy source of calcium in the diet of women. Following a 3 week run-in baseline period, 35 healthy women > 18 years were randomized by crossover design into either Group I or Group II. Group I added 2 calcium-fortified cereal bars daily (total of 400 mg calcium/day) (intervention) to their usual diet and Group II continued their usual diet (control). At the end of 3 weeks, diets were switched for another 3 weeks. Intakes of calcium and energy were estimated from 3-day diet and supplemental diaries. Wilcoxon signed-rank tests were used for within group comparisons and Mann Whitney U tests were used for between group comparisons of calcium and energy intake. Dietary calcium was significantly higher during intervention (1071 mg/d) when participants consumed 2 calcium-fortified cereal bars daily than during the baseline (720 mg/d, P <0.0001) or control diets (775 mg/d, P = 0.0001) periods. Furthermore, the addition of 2 calcium-fortified cereal bars daily for the 3 week intervention did not significantly increase total energy intake or result in weight gain. In conclusion, consumption of calcium-fortified cereal bars significantly increased calcium intake of women. Further research examining the potential ability of fortified cereal bars to help maintain and improve bone health of women is warranted.

Trial Registration

ClinicalTrials.gov NCT01508689     

A Behavioral Mechanism for Incorporating an Unpalatable Food in the Diet of a Generalist Herbivore (Orthoptera: Acrididae)

We investigated mechanisms that could lead to incorporation of unpalatable foods into the diet of a generalist grasshopper, Schistocerca americana: nutritional stress, habituation, learning, and attraction to novelty. The model system involved mesquite, a palatable but inferior food, and mulberry, an unpalatable but adequate food. Nutritional stress, due to prolonged intake of the inferior food, mesquite, did not increase the acceptability of mulberry. Habituation to the deterrent compounds in mulberry and associative learning of the nutritional benefits of mulberry also did not occur. However, mulberry became more acceptable after a day of restriction to a single food type other than mulberry, and even deterrent and nutritionally worthless alternatives such as filter paper became acceptable after a day on any one food type. A tendency to feed on novel food types may be a proximate mechanism for the incorporation of relatively unpalatable, but nutritionally valuable foods into the diet. Novelty and the apparent need for diversity of foods are discussed in the context of exploratory foraging behavior by generalist herbivores.     

Mitochondrial DNA analysis reveals substantial Native American ancestry in Puerto Rico

To estimate the maternal contribution of Native Americans to the human gene pool of Puerto Ricans--a population of mixed African, European, and Amerindian ancestry--the mtDNAs of two sample sets were screened for restriction fragment length polymorphisms (RFLPs) defining the four major Native American haplogroups. The sample set collected from people who claimed to have a maternal ancestor with Native American physiognomic traits had a statistically significant higher frequency of Native American mtDNAs (69.6%) than did the unbiased sample set (52.6%). This higher frequency suggests that, despite the fact that the native Taíno culture has been extinct for centuries, the Taíno contribution to the current population is considerable and some of the Taíno physiognomic traits are still present. Native American haplogroup frequency analysis shows a highly structured distribution, suggesting that the contribution of Native Americans foreign to Puerto Rico is minimal. Haplogroups A and C cover 56.0% and 35.6% of the Native American mtDNAs, respectively. No haplogroup D mtDNAs were found. Most of the linguistic, biological, and cultural evidence suggests that the Ceramic culture of the Taínos originated in or close to the Yanomama territory in the Amazon. However, the absence of haplogroup A in the Yanomami suggests that the Yanomami are not the only Taíno ancestors.     

Development of a Quantitative Food Frequency Questionnaire for Use among the Yup'ik People of Western Alaska

Alaska Native populations are experiencing a nutrition transition and a resulting decrease in diet quality. The present study aimed to develop a quantitative food frequency questionnaire to assess the diet of the Yup''ik people of Western Alaska. A cross-sectional survey was conducted using 24-hour recalls and the information collected served as a basis for developing a quantitative food frequency questionnaire. A total of 177 males and females, aged 13-88, in six western Alaska communities, completed up to three 24-hour recalls as part of the Alaska Native Dietary and Subsistence Food Assessment Project. The frequency of the foods reported in the 24-hour recalls was tabulated and used to create a draft quantitative food frequency questionnaire, which was pilot tested and finalized with input from community members. Store-bought foods high in fat and sugar were reported more frequently than traditional foods. Seven of the top 26 foods most frequently reported were traditional foods. A 150-item quantitative food frequency questionnaire was developed that included 14 breads and crackers; 3 cereals; 11 dairy products; 69 meats, poultry and fish; 13 fruit; 22 vegetables; 9 desserts and snacks; and 9 beverages. The quantitative food frequency questionnaire contains 39 traditional food items. This quantitative food frequency questionnaire can be used to assess the unique diet of the Alaska Native people of Western Alaska. This tool will allow for monitoring of dietary changes over time as well as the identification of foods and nutrients that could be promoted in a nutrition intervention program intended to reduce chronic disease.     

Woody plant seed dispersal and gap formation in a North American subtropical savanna woodland: the role of domestic herbivores

The relationship between domestic cattle and vegetation change in a savanna woodland was evaluated with respect to dung deposition and the dispersal and establishment of mesquite (Prosopis glandulosa var. glandulosa, Mimosaceae), a cosmopolitan woody invader of grasslands in the southwestern USA. Dung deposited in autumn disintegrated rapidly, leaving patches of bare ground ranging from 50 to 900 cm². Herbaceous cover on gaps created by dung deposition recovered to levels comparable to neighboring vegetation by the end of the following growing season. Vegetation colonizing gaps consisted primarily of grasses not found in the surrounding vegetation. Dung deposition increased species diversity and spatial heterogeneity of the herbaceous vegetation and contributed to the development of a fine-grain mosaic of small patches of varying successional age-states.The role of cattle in facilitating the ingress and establishment of mesquite has broader implications with regard to the conversion of grasslands to woodlands. On the site with cattle, mesquite seedlings were found in 75% of dung pats surveyed in September (mean =4.2 seedlings per pat; maximum =50). Although seedling survival in dung (79%) was only 16% greater than that of mesquite emerging from seeds experimentally sown away from dung, no seedlings were found on areas without cattle. Mean (± SE) density of mesquite seedlings ranged from 12±2 to 15±2 m^-2 on the site with cattle. Seed densities away from parent plants averaged 10.7 m^-2 and 0.0 m^-2 on areas with and without cattle, respectively. Seed densities beneath adult plants were comparable between sites.The high density of seedlings on areas with cattle, in contrast to absence of seedlings on the area without cattle, suggests rates of invasion of grasslands by mesquite would have increased substantially in North America following the settlement and introduction of domestic ungulates. Prior to the introduction of livestock, poor seed dissemination and germination may have limited its Holocene spread.     

Ethnically mismatched cord blood transplants in African Americans: the Saint Louis Cord Blood Bank experience

BACKGROUND: For ethnic minority patients where a suitably matched BM or peripheral blood donor is frequently unavailable, cord blood offers an opportunity for hematopoietic stem cell transplantation. Focused recruitment of ethnic minorities for cord blood donation has been proposed as the preferred strategy to improve access for minority recipients to cord blood for transplantation. The aim of this study was to evaluate cord blood characteristics for Caucasian and African American donors and the success of ethnically mismatched UC blood transplantation in African American recipients. METHODS: Retrospective data analysis was performed comparing the characteristics of 556 cord blood units from African American and Caucasian donors. The outcomes of 18 African American ethnically mismatched transplant recipients were compared with a paired sample of 18 ethnically matched Caucasian recipients. RESULTS: The fraction of collected units meeting acceptability criteria from African Americans was significantly lower compared with Caucasians (P = <0.0001). Additionally, African Americans had a significantly lower post-processing total nucleated cell count (TNC) compared with Caucasians (P=0.007) but there were no other significant differences in conventionally measured product characteristics. In the transplant analysis, there was no difference in overall survival at 1 year (P=0.85) or time to neutrophil engraftment (P=0.92) between the two patient populations. DISCUSSION: At comparable levels of TNC dose and HLA matching, the use of ethnically mismatched UC blood units as a source for allogeneic unrelated transplant can result in successful transplant outcomes for African American patients.     

Self-reported ethnicity, genetic structure and the impact of population stratification in a multiethnic study

It is well-known that population substructure may lead to confounding in case–control association studies. Here, we examined genetic structure in a large racially and ethnically diverse sample consisting of five ethnic groups of the Multiethnic Cohort study (African Americans, Japanese Americans, Latinos, European Americans and Native Hawaiians) using 2,509 SNPs distributed across the genome. Principal component analysis on 6,213 study participants, 18 Native Americans and 11 HapMap III populations revealed four important principal components (PCs): the first two separated Asians, Europeans and Africans, and the third and fourth corresponded to Native American and Native Hawaiian (Polynesian) ancestry, respectively. Individual ethnic composition derived from self-reported parental information matched well to genetic ancestry for Japanese and European Americans. STRUCTURE-estimated individual ancestral proportions for African Americans and Latinos are consistent with previous reports. We quantified the East Asian (mean 27%), European (mean 27%) and Polynesian (mean 46%) ancestral proportions for the first time, to our knowledge, for Native Hawaiians. Simulations based on realistic settings of case–control studies nested in the Multiethnic Cohort found that the effect of population stratification was modest and readily corrected by adjusting for race/ethnicity or by adjusting for top PCs derived from all SNPs or from ancestry informative markers; the power of these approaches was similar when averaged across causal variants simulated based on allele frequencies of the 2,509 genotyped markers. The bias may be large in case-only analysis of gene by gene interactions but it can be corrected by top PCs derived from all SNPs.     

Development of a genetics education workshop curriculum for Native American college and university students.

The long-term goal of Genetic Education for Native Americans (GENA), a project funded by the National Human Genome Research Institute (NHGRI), is to provide a balance of scientific and cultural information about genetics and genetic research to Native Americans and thereby to improve informed decision making. The project provides culturally sensitive education about genetic research to Native American medical students and college and university students. Curriculum development included focus groups, extensive review of available curricula, and collection of information about career opportunities in genetics. Special attention was focused on genetic research to identify key concepts, instructional methods, and issues that are potentially troublesome or sensitive for Native Americans. Content on genetic research and careers in genetics was adapted from a wide variety of sources for use in the curriculum. The resulting GENA curriculum is based on 24 objectives arranged into modules customized for selected science-related conference participants. The curriculum was pretested with Native American students, medical and general university, health care professionals, and basic scientists. Implementation of the curriculum is ongoing. This article describes the development and pretesting of the genetics curriculum for the project with the expectation that the curriculum will be useful for genetics educators working in diverse settings.     

DEVELOPMENT OF AN ACCEEPTABILITY CONSTRAINT FOR A LINEAR PROGRAMMING MODEL IN FOOD FORMULATION

Control of a product's market acceptability can be a difficulty when using linear programming models in food formulation. The development of an acceptability constraint was demonstrated for a linear programming model used for the formulation of fresh turkey bratwurst, a coarse ground type sausage. Development was in two stages. First, an experimental design and in-house panel determined quantitative relationships between the product's textural attributes and turkey meat ingredients. Second, the product toughness relationship was utilized to develop three formulations with different levels of toughness. These formulations were market tested using the acceptor set size as the measure of market acceptability. A relationship between product toughness and acceptor set size was determined, into which was substituted the toughness f (ingredients) relationship. This yielded acceptor set size as a f (ingredients) that was added to the least cost linear programming model in the form of an acceptability constraint.