Search

Intra-process memory isolation can improve security by enforcing least-privilege at a finer granularity than traditional operating system controls without the context-switch overhead associated with inter-process communication. Because the process has traditionally been a fundamental security boundary, assigning different levels of trust to components within a process is a fundamental change in secure systems design. However, so far there has been little research on the challenges of securely implementing intra-process isolation on top of existing operating system abstractions. We find that frequently-used assumptions in secure system design do not precisely hold under realistic conditions, and that these discrepancies lead to exploitable vulnerabilities. We evaluate two recently-proposed memory isolation systems and show that both are vulnerable to the same generic attacks that break their security model. We then extend a subset of these attacks by applying them to a fully-precise model of control-flow integrity, demonstrating a data-only attack that bypasses both static and dynamic control-flow integrity enforcement by overwriting executable code in-memory even under typical w^x assumptions. From these two results, we propose a set of kernel modifications called Xlock that systemically addresses weaknesses in memory permissions enforcement on Linux, bringing them into line with w^x assumptions. Finally, we present modifications to intra-process isolation systems that preserve efficient userspace component transitions while drastically reducing risk of accidental kernel mismanagement by modeling intra-process components as separate processes from the kernel's perspective. Taken together, these mitigations represent a more robust architecture for efficient and secure intra-process isolation.

Absence of functional phenylalanine hydroxylase results in phenylketonuria (PKU). Viable treatments remain few, expensive and secondary conditions such as osteopenia occur in most PKU patients. Objective 1: Given the recently described roles of gut microbes to aid host digestion, an orally administered genetically engineered probiotic as the delivery vehicle for enzyme replacement therapy was created. The engineered probiotic, pHENOMMenal, produced phenylalanine ammonia lyase with significant production of trans-cinnamate (phenylalanine cleavage product) in vitro and resulted in a reduction of 515 μM in blood phenylalanine when fed to PKU animals for 14 days (from 2307µM ± 264µM to 1792µM ± 261µM, n = 6, P < 0.05). The control probiotic produced no change in blood phenylalanine. Thus, pHENOMMenal treatment in PKU mice demonstrated engineered microbes could compensate for a metabolic deficiency of the host. Objective 2: Evaluate the PAHenu2 mouse model of PKU for a genetic discrepancy causing ocular enlargement and delayed development observed only after the PAHenu2 mutation was crossed to the C57BL/6J mouse. When compared to healthy littermates, ELISA indicated a consistent but insignificant decrease in plasma IGF-1 and an increase in ocular IGF-1 in PKU animals. SNP screening demonstrated a differential inheritance of IGF-1 alleles in healthy and PKU animals based on PAH allele inheritance. Ocular and developmental phenotypes in the PAHenu2 colony match those described in previous IGF-1 studies. Understanding the IGF-1 inheritance discrepancy will enable better osteopenia research using PAHenu2 mice and allow breeding of a healthier mouse colony for continued research. Collectively the results from this work describe a new therapeutic approach for treatment of PKU as well as a better understanding of the PAHenu2 mouse model to study this disease.

Maternal phenylketonuria [MPKU] is a syndrome of cardiovascular malformations [CVMs], microcephaly, intellectual impairment, and small size for gestational age caused by in-utero exposure to elevated serum phenylalanine [Phe] due to PKU in the mother. It is becoming a public health concern as more women with PKU reach child bearing age. Although a mouse model of PKU, BTBRPahenu2, has been available for 20 years, it has not been well utilized for studying MPKU. We used this model to delineate critical parameters in Phe cardiovascular teratogenicity and to study the effects of genetic background. Dosing and timing experiments were performed with the BTBRPahenu2 mouse. A dose response curve was constructed, with CVM rates at maternal serum Phe levels 900μM [high] of 11.86%, 16.67%, 30.86%, and 46.67% respectively. A variety of CVMs were noted on the BTBR background, including double outlet right ventricle, aortic arch abnormalities, and ventricular septal defects. Phe exposure during specific gestational time points identified a teratogenic window from embryonic day 8.5-13.5, with higher rates of conotruncal and valve defects occurring with exposure during the early time period [E8.5-11.5] and persistent truncus arteriosus and aortic arch branching abnormalities occurring with mid time period [E12.5-14.5].iiiCompared to the BTBRPahenu2 strain, the C3H.BTBR.Pahenu2 and B6.BTBR. Pahenu2 offspring had comparable rates of CVMs in general but a higher rate of left ventricular outflow tract malformations. We have delineated key parameters of Phe cardiovascular teratogenicity, demonstrated the utility of this MPKU model on different mouse strains, and shown how genetic background profoundly affects the phenotype.We conducted initial mechanistic studies investigating the effect of high Phe on the pharyngeal and cardiac region. Defects affecting the septation and remodeling of the outflow tract [OFT] and aortic arch arteries [AAA] implicate the neural crest cells as a potential cell population important in the manifestation of CVMs in MPKU. Utilizing neural tube explants from chick, we investigated the impact of high Phe on the migration and development of cardiac neural crest cells [cNCCs]. High Phe exposure resulted in a significant increase in the area of neural crest cell migration. Immunohistochemistry [IHC] analysis of caspase-3 [a marker of apoptosis] and phosphoHistone H3 [a marker of proliferation] revealed decreases in both apoptosis and proliferation. Gene expression analysis of a panel of genes important in heart development revealed significant global decrease in many genes at E10.5 [n=3 litters in triplicate] and E11.5 [n=2 litters in triplicate], with up regulation of few genes including Gata6.These studies are limited by the number of genes we selected for expression analysis and the number of litters assayed. Though these preliminary results are inconclusive, these data warrant further investigation of other genes important in heart development.

Phenylketonuria (PKU) is a genetic disorder caused by the deficiency of the enzyme, phenylalanine hydroxylase. Phenylalanine accumulation in brain causes cognitive impairment and behavioral problems, which can be prevented by dietary restriction of phenylalanine. People with PKU have to maintain a highly restrictive low protein diet throughout their entire life. PKU is typically diagnosed at birth through newborn screening. In order for a patient to reach their optimal IQ and health status, they need to maintain their blood phenylalanine (phe) level within the recommended range of 60-360 µmol/L. The purpose of this study is to investigate attitudes of adults diagnosed with phenylketonuria (PKU) and how they relate to their actions of management therapy recommendations. Adults with PKU completed an online survey through Survey Monkey. After receiving consent via email, the survey link was sent to them so they completed it at their convenience. The study asked questions such as, how often they tested their blood phe level and compared that to how often their clinic recommends getting tested. Then subjects were asked on a scale of one to five (with five meaning they strongly agree) if they agreed that getting their blood phe level tested as often as their clinic recommends is important. From the data we gathered information about different management practices, the perceived importance of certain dietary management practices, and whether certain factors, such as relationships between patient and clinic influence dietary management. There was no set universal method for counting phe intake. While there are three main methods, the two that were used the most were by counting exactly using milligrams (mg) of phe and the other was an estimation of counting in grams of protein. It was also found that the relationship with the clinic may have an impact on dietary management. Patients who had a good relationship with their clinic were more likely to follow their daily protein restriction, test their blood phe level more often, and realize the importance of testing their blood phe level as often as their clinic recommends. Advisors: Kaye Stanek-Krogstrand and Wanda Koszewski