New approaches to genetic manipulation of mice: tissue-specific expression of ACE

The renin-angiotensin system (RAS) plays a central role in body physiology, controlling blood pressure and blood electrolyte composition. ACE.1 (null) mice are null for all expression of angiotensin-converting enzyme (ACE). These mice have low blood pressure, the inability to concentrate urine, and a maldevelopment of the kidney. In contrast, ACE.2 (tissue null) mice produce one-third normal plasma ACE but no tissue ACE. They also have low blood pressure and cannot concentrate urine, but they have normal indices of renal function. These mice, while very informative, show that the null approach to creating knockout mice has intrinsic limitations given the many different physiological systems that no longer operate in an animal without a functioning RAS. To investigate the fine control of body physiology by the RAS, we developed a novel promoter swapping approach to generate a more selective tissue knockout of ACE expression. We used this to create ACE.3 (liver ACE) mice that selectively express ACE in the liver but lack all ACE within the vasculature. Evaluation of these mice shows that endothelial expression of ACE is not required for blood pressure control or normal renal function. Targeted homologous recombination has the power to create new strains of mice expressing the RAS in selected subsets of tissues. Not only will these new genetic models be useful for studying blood pressure regulation but also they show great promise for the investigation of the function of the RAS in complicated disease models.

Endothelium dysfunction in LDL receptor knockout mice: a role for H2O2

1. In this study, the role of endogenous H(2)O(2) as an endothelium-dependent relaxant factor was characterised in aortas from C57BL/6J and LDL receptor-deficient mice (LDLR(-/-)). 2. Aortic rings from LDLR(-/-) mice showed impaired endothelium-dependent relaxation to acetylcholine (ACh; 0.001-100 micro M) and to the Ca(2+) ionophore A23187 (0.001-3 micro M) compared with aortic rings from control mice. Endothelium-independent relaxation produced by the NO donor, 3-morpholino-sydnonimine (SIN-1) was not different between strains. 3. Pretreatment of vessels with L-NNA (100 micro M) or L-NNA (100 micro M) plus L-NAME (300 micro M) plus haemoglobin (10 micro M) markedly decreased, but did not abolish the relaxation to ACh in control mice. In the aortas from LDLR(-/-) mice treated with L-NNA (100 micro M), ACh induced a contractile effect. Catalase (800 and 2400 U ml(-1)) shifted to the right the endothelium-dependent relaxation to ACh in aortas from control but not from LDLR(-/-) mice. Aminotriazole (50 mM), which inhibits catalase, abolished its effect on control mice. Treatment of vessels with L-NNA and catalase abolished vasorelaxation induced by ACh. Indomethacin (10 micro M) did not modify the concentration-response curve to ACh. Superoxide dismutase (300 U ml(-1)) did not change ACh-induced relaxation in both strains. 4. Exogenous H(2)O(2) produced a concentration-dependent relaxation in endothelium-denuded aortic rings, which was not different between strains. 5. It is concluded that H(2)O(2) greatly contributes to relaxation to ACh in aorta from control mice. Endothelial-dependent relaxation to ACh is impaired in LDLR(-/-) mice. Reduced biosynthesis or increased inactivation of H(2)O(2) is the possible mechanism responsible for endothelial dysfunction in aortas of atherosclerosis-susceptible LDLR(-/-) mice.

Family 1 G protein-coupled receptor function in the CNS. Insights from gene knockout mice

Family 1 G protein-coupled receptors (GPCRs) are activated by a large number of ligands including photons, odorants, neurotransmitters and hormones and are involved in a wide variety of central and peripheral functions. Due to their wide distribution in the central nervous system (CNS), family 1 GPCRs play a major role in the regulation of neuronal activity and behaviour. In general, the lack of selective ligands for each member of the GPCR subfamilies has made it difficult to assign specific central functions to each receptor subtype. Advances in gene targeting techniques have allowed the inactivation of receptor genes in the mouse through homologous recombination leading to the generation of mouse 'knockout' models lacking one or more GPCRs. In this review, we have listed the family 1 GPCR knockout models produced in the past decade and we have summarized the findings obtained from studies on these mice with respect to CNS function.

Knockout mice as model systems for studying nm23/NDP kinase gene functions. Application to the nm23-M1 gene

Mice carrying a homozygous germ-line mutation in the nm23-M1 gene that eliminates its protein expression and drives expression of beta-galactosidase by nm23-M1 promoter have been generated. nm23-M1 gene inactivation is not teratogenic and the pups can grow to adult age without apparent health problems. However, they undergo a growth retardation and knocked out females cannot feed their pups. Both effects are background dependent. Beta-galactosidase mapping of nm23-M1 promoter activation during embryogenesis shows that the nm23-M1 gene is principally expressed in epithelial layer of tissues which require inductive epithelial-mesenchymal interactions for their formation. In conclusion, invalidated mice could be interesting models to analyze the role of nm23-M1 on signal transduction pathway regulation, or cancer induction and proliferation.

Formation of multiple hearts in mice following deletion of beta-catenin in the embryonic endoderm

Using Cre/loxP, we conditionally inactivated the beta-catenin gene in cells of structures that exhibit important embryonic organizer functions: the visceral endoderm, the node, the notochord, and the definitive endoderm. Mesoderm formation was not affected in the mutant embryos, but the node was missing, patterning of the head and trunk was affected, and no notochord or somites were formed. Surprisingly, deletion of beta-catenin in the definitive endoderm led to the formation of multiple hearts all along the anterior-posterior (A/P) axis of the embryo. Ectopic hearts developed in parallel with the normal heart in regions of ectopic Bmp2 expression. We provide evidence that ablation of beta-catenin in embryonic endoderm changes cell fate from endoderm to precardiac mesoderm, consistent with the existence of bipotential mesendodermal progenitors in mouse embryos.

An 'oligarchy' rules neural development

Recent reports show that Olig genes, which encode the basic helix-loop-helix Olig transcription factors, are essential for development of oligodendrocytes. Surprisingly, Olig function is also required for formation of somatic motor neurons. These findings alter our views of how the oligodendrocyte lineage is generated and raise further questions about the underlying developmental relationships between neurons and glia.

Maternal loading with very low-density lipoproteins stimulates fetal surfactant synthesis

We examined whether administration of very low-density lipoproteins (VLDL) to pregnant rats increases surfactant phosphatidylcholine (PtdCho) content in fetal pre-type II alveolar epithelial cells. VLDL-triglycerides are hydrolyzed to fatty acids by lipoprotein lipase (LPL), an enzyme activated by heparin. Fatty acids released by LPL can incorporate into the PtdCho molecule or activate the key biosynthetic enzyme cytidylyltransferase (CCT). Dams were given BSA, heparin, VLDL, or VLDL with heparin intravenously. Radiolabeled VLDL given to the pregnant rat crossed the placenta and was distributed systemically in the fetus and incorporated into disaturated PtdCho (DSPtdCho) in pre-type II cells. Maternal administration of VLDL with heparin increased DSPtdCho content in cells by 45% compared with control (P < 0.05). VLDL produced a dose-dependent, saturable, and selective increase in CCT activity. VLDL did not significantly alter immunoreactive CCT content but increased palmitic, stearic, and oleic acids in pre-type II cells. Furthermore, hypertriglyceridemic apolipoprotein E knockout mice contained significantly greater levels of DSPtdCho content in alveolar lavage and CCT activity compared with either LDL receptor knockout mice or wild-type controls that have normal serum triglycerides. Thus the nutritional or genetic modulation of serum VLDL-triglycerides provides specific fatty acids that stimulate PtdCho synthesis and CCT activity thereby increasing surfactant content.

EphB forward signaling controls directional branch extension and arborization required for dorsal-ventral retinotopic mapping

We report that EphB receptors direct unique axonal behaviors required for mapping the dorsal-ventral (D-V) retinal axis along the lateral-medial (L-M) axis of the superior colliculus (SC). EphBs are expressed in a D-V gradient, ephrin-B1 in a L-M gradient in SC, and ephrin-B3 at its midline. EphBs and ephrin-Bs are expressed in countergradients in retina and SC. Developmental analyses reveal that retinal axons lack D-V ordering along the L-M axis, but directionally extend branches along it to establish ordered arbors. Directed branch extension is disrupted in EphB2; EphB3-deficient mice resulting in lateral ectopic arbors. Mice with kinase-inactive EphB2 have similar D-V mapping defects indicating that forward signaling dominates over reverse signaling. Our data suggest that branches of EphB expressing axons are attracted medially by ephrin-B1, and provide molecular mechanisms for D-V mapping in visual centers.

REV-ving up the clock

Circadian rhythms are generated by a transcription/translation feedback loop consisting of two limbs, one positive and one negative. The nuclear orphan receptor, REV-ERBalpha, is identified as a molecular link coupling these two limbs.

Quantitative autoradiographic mapping of opioid receptors in the brain of delta-opioid receptor gene knockout mice

Using quantitative receptor autoradiography we have determined if deletion of the delta-opioid receptor gene (Oprd1) results in compensatory changes in the expression of other opioid receptors. Gene targeting was used to delete exon 1 of the mouse delta-opioid receptor gene and autoradiography was carried out on brains from wild-type, heterozygous and homozygous knockout mice. Delta-opioid receptors were labeled with [(3)H]deltorphin I (7 nM), mu- with [(3)H]DAMGO (4 nM), and kappa- with [(3)H]CI-977 (2.5 nM) or [(3)H]bremazocine (2 nM in the presence of DPDPE and DAMGO) and non-specific binding determined with naloxone. [(3)H]Deltorphin I binding was reduced by approximately 50% in heterozygous animals. In homozygous animals specific binding could only be detected after long-term film exposure (12 weeks). Regions exhibiting this residual [(3)H]deltorphin I binding correlated significantly with those demonstrating high levels of the mu-receptor and were abolished in the presence of the mu-agonist DAMGO. Autoradiographic mapping showed significant overall reductions in [(3)H]DAMGO and [(3)H]CI-977 binding throughout the brain following loss of both copies of the Oprd1 gene. In contrast, overall levels of [(3)H]bremazocine binding were higher in brains from -/- than +/+ mice. Our findings suggest that residual [(3)H]deltorphin I binding in the brain of delta-receptor gene knockout mice is the result of cross-reactivity with mu-sites and that there are no delta-receptor subtypes derived from a different gene. Changes in mu- and kappa-receptor labeling suggest compensatory changes in these subtypes in response to the absence of the delta-receptor. The differences in [(3)H]CI-977 and [(3)H]bremazocine binding indicate these ligands show differential recognition of the kappa-receptor.

Cellular prion protein: implications in seizures and epilepsy

1. Cellular prion (PrPc) is a plasma membrane protein involved with copper uptake, protection against oxidative stress, cell adhesion, differentiation, signaling, and survival in the central nervous system. 2. Deletion of PrPc gene (Pmp) in mice enhances sensitivity to seizures in vivo and neuronal excitability in vitro which can be related to: (i) disrupted Ca(+2)-activated K+ currents, with loss of IHAP conductance in hippocampus; (ii) abnormal GABA-A inhibition in the hippocampus; (iii) mossy fiber reorganization in the hippocampus; (iv) changes in ectonucleotidases in both hippocampus and neocortex; and (v) higher levels of neocortical and subcortical oxidative stress. Moreover, postnatal Prnp knockout mice showed a significant reduction of after hyperpolarization potentials in hippocampal CA1 cells. 3. Taken together, these findings suggest that loss of PrPc function contributes to the hyperexcitable and synchronized activities underlying epileptic seizures generated in neocortex and hippocampus. Hence, the role of PrPc on human symptomatic, cryptogenic or idiopathic epileptic syndromes deserves further investigation.

Cortical and thalamic axon pathfinding defects in Tbr1, Gbx2, and Pax6 mutant mice: evidence that cortical and thalamic axons interact and guide each other

During development, cortical areas establish precise reciprocal projections with corresponding thalamic nuclei. Pioneer axons from the cortex and thalamus first meet in the intermediate zone of the subcortical telencephalon (subpallium). Their close interactions in the subpallium suggest that they may use each other for guidance. To test this hypothesis, the development of corticothalamic and thalamocortical connections was studied in mice with mutations of transcription factor genes expressed specifically in the cortex (Tbr1), the dorsal thalamus (Gbx2), or both (Pax6). In Tbr1 mutants, cortical pioneer axons entered the subpallium at the appropriate time, but most stopped growing without entering the diencephalon. Surprisingly, thalamic axons (which do not express Tbr1) deviated into the external capsule and amygdala regions, without entering the cortex. Conversely, in most Gbx2 mutants, thalamic axons were reduced in number and grew no farther than the subpallium. Cortical axons (which do not express Gbx2) grew into the subpallium but did not enter the diencephalon. In one Gbx2- /- case, sparse thalamocortical and corticothalamic projections both developed, but in no case did one projection reach its target and not the other. In Pax6 mutants, neither corticothalamic nor thalamocortical axons reached their targets. These results suggest that thalamocortical and corticothalamic projections may not form independently. After reaching the subpallium, each projection may require a molecularly intact reciprocal projection for further guidance. This type of mechanism ensures that thalamocortical and corticothalamic axons project reciprocally. However, the exact nature of the interaction between cortical and thalamic pioneer axons remains to be elucidated.

Increased glycosphingolipid levels in serum and aortae of apolipoprotein E gene knockout mice

The apolipoprotein E gene knockout (apoE-/-) mouse develops atherosclerosis that shares many features of human atherosclerosis. Increased levels of glycosphingolipid (GSL) have been reported in human atherosclerotic lesions; however, GSL levels have not been studied in the apoE-/- mouse. Here we used HPLC methods to analyze serum and aortic GSL levels in apoE-/- and C57BL/6J control mice. The concentrations of glucosyl ceramide (GlcCer), lactosyl ceramide (LacCer), GalNAcbeta1-4Galbeta1-4Glc-Cer (GA2), and ceramide trihexoside (CTH) were increased by approximately 7-fold in the apoE-/- mouse serum compared with controls. The major serum ganglioside, N-glycolyl GalNAcbeta1-4[NeuNAcalpha2-3]Galbeta1-4Glc-Cer (N-glycolyl GM2), was increased in concentration by approximately 3-fold. A redistribution of GSLs from HDL to VLDL populations was also observed in the apoE-/- mice. These changes were accompanied by an increase in the levels of GSLs in the aortic sinus and arch of the apoE-/- mice. The spectrum of gangliosides present in the aortic tissues was more complex than that found in the lipoproteins, with the latter represented almost entirely by N-glycolyl GM2 and the former comprised of NeuNAcalpha2-3Galbeta1-4Glc-Cer (GM3), GM2, N-glycolyl GM2, GM1, GD3, and GD1a. In conclusion, neutral GSL and ganglioside levels were increased in the serum and aortae of apoE-/- mice compared with controls, and this was associated with a preferential redistribution of GSL to the proatherogenic lipoprotein populations. The apoE-/- mouse therefore represents a useful model to study the potential role of GSL metabolism in atherogenesis.

Increased sensitivity of homozygous Sod2 mutant mice to oxygen toxicity

Induction or overexpression of pulmonary manganese superoxide dismutase (MnSOD) has been shown to protect against oxygen (O2) toxicity. Genetic inactivation of MnSOD (Sod2) results in multiple organ failure and early neonatal death. However, lungs or O2-tolerance of Sod2 knockout mice have not been investigated. We evaluated survival, lung histopathology, and other pulmonary antioxidants (glutathione cycle) of homozygous (-/-) and heterozygous (+/-) Sod2 mutant mice compared with wild-type controls (Sod2+/+) following 48 h exposure to either room air or to O2. The ability of antioxidant N-acetylcysteine to compensate for the loss of MnSOD was explored. Mortality of Sod2-/- mice increased from 0% in room air to 18 and 83% in 50 and 80% O2, respectively. N-acetylcysteine did not alter mortality of Sod2-/- mice. Histopathological analysis revealed abnormalities in saccules of Sod2-/- mice exposed either to room air or to 50% O2 suggestive of delayed postnatal lung development. In 50% O2, activities of glutamate-cysteine ligase (GCL) (previously known as gamma-glutamylcysteine synthetase, gamma-GCS) and glutathione peroxidase increased in Sod2-/- (35 and 70%, respectively) and Sod2+/- (12 and 70%, respectively) mice, but glutathione levels remained unaltered. We conclude that MnSOD is required for normal O2 tolerance and that in the absence of MnSOD there is a compensatory increase in pulmonary glutathione-dependent antioxidant defense in hyperoxia.

Hormonal and genetic influences on arousal--sexual and otherwise

Genetic influences on lordosis, a mammalian social behavior, are amenable for study because of the relative simplicity of both stimuli and response. The neural circuit for lordosis involves a supraspinal loop, which is controlled by an estrogen- and progesterone-dependent signal from the medial hypothalamus and results in heightened sexual motivation. In turn, this involves elevated states of arousal, defined by increased sensory alertness, motor activity and emotional reactivity. Mice in which the gene encoding the alpha form of the estrogen receptor (ERalpha) has been knocked out show that ERalpha is crucial for lordosis behavior. Comparing ERalpha-, ERbeta- and double knockouts reveals that different patterns of sexual behaviors in mice require different patterns of ER activity. Understanding how hormonal and genetic effects on deep motivational and arousal processes contribute to their effects on specific sexual and aggressive behaviors pose significant challenges for mouse functional genomics.

The CYP2D6 humanized mouse: effect of the human CYP2D6 transgene and HNF4alpha on the disposition of debrisoquine in the mouse

CYP2D6 is a highly polymorphic human gene responsible for a large variability in the disposition of more than 100 drugs to which humans may be exposed. Animal models are inadequate for preclinical pharmacological evaluation of CYP2D6 substrates because of marked species differences in CYP2D isoforms. To overcome this issue, a transgenic mouse line expressing the human CYP2D6 gene was generated. The complete wild-type CYP2D6 gene, including its regulatory sequence, was microinjected into a fertilized FVB/N mouse egg, and the resultant offspring were genotyped by both polymerase chain reaction and Southern blotting. CYP2D6-specific protein expression was detected in the liver, intestine, and kidney from only the CYP2D6 humanized mice. Pharmacokinetic analysis revealed that debrisoquine (DEB) clearance was markedly higher (94.1 +/- 22.3 l/h/kg), and its half-life significantly reduced (6.9 +/- 1.6 h), in CYP2D6 humanized mice compared with wild-type animals (15.2 +/- 0.9 l/h/kg and 16.5 +/- 4.5 h, respectively). Mutations in hepatic nuclear factor 4alpha (HNF4alpha), a hepatic transcription factor known to regulate in vitro expression of the CYP2D6 gene, could affect the disposition of CYP2D6 drug substrates. To determine whether the HNF4alpha gene modulates in vivo pharmacokinetics of CYP2D6 substrates, a mouse line carrying both the CYP2D6 gene and the HNF4alpha conditional mutation was generated and phenotyped using DEB. After deletion of HNF4alpha, DEB 4-hydroxylase activity in CYP2D6 humanized mice decreased more than 50%. The data presented in this study show that only CYP2D6 humanized mice but not wild-type mice display significant DEB 4-hydroxylase activity and that HNF4alpha regulates CYP2D6 activity in vivo. The CYP2D6 humanized mice represent an attractive model for future preclinical studies on the pharmacology, toxicology, and physiology of CYP2D6-mediated metabolism.

Genetic perspectives on the serotonin transporter

The serotonin transporter (5-HTT) is most well known as the site of action of the serotonin reuptake inhibitors, which were initially developed as antidepressants, but now are the most widely used agents in the treatment of many additional neuropsychiatric and related disorders. The discovery that the gene that expresses the 5-HTT possesses a functional promoter-region polymorphism, which is associated with temperament and personality traits such as anxiety and negative emotionality as well as some behaviors, led to many studies examining this polymorphism in individuals with different neuropsychiatric disorders. The subsequent development of mice with a targeted disruption of the 5-HTT in our laboratory has provided an experimental model to examine the many consequences of diminished (in +/-, heterozygote mice) or absent (in -/-, homozygote knockout mice) function of the 5-HTT. The 5-HTT-deficient mice were also crossed with other knockout mice, allowing the study of multiple neurobiologic dysfunctions. As multiple genes are probably involved in the expression of complex behaviors such as anxiety, as well as neuropsychiatric disorders, these more genetically complex mice may more closely model disorders with complex etiologies. Thus, the combination of these comparative human and mouse studies may extend the opportunities to examine genetic alterations from a novel "bottom-up" approach [gene knockout or partial gene knockout in a combinational gene x gene x (yet unknown) gene approach], which is complementary to the traditional "top-down" genetic approach based upon studies of individuals with diagnosed neuropsychiatric disorders and their family members.

Development of the visual pathway is disrupted in mice with a targeted disruption of the calcium channel beta(3)-subunit gene

Refinement of the retinal pathways to the superior colliculus (SC) and dorsal lateral geniculate nucleus (dLGN) is mediated by nitric oxide (NO). Long-term depression (LTD) can also be induced in SC and LGN during the time at which these pathways are refined, and this LTD is partially dependent on NO and L-type Ca(2+) channel function. In an effort to determine whether NO-mediated pathway refinement is also mediated by Ca(2+) channel function, we have examined the refinement of the retinocollicular and retinogeniculate pathways in mice which lack the gene for the Ca(2+) channel beta(3) subunit (CCKO) and which have significantly reduced L-type Ca(2+) currents. Injections of the anterograde tracer cholera toxin subunit B/HRP were made into one eye of these knockout animals and in wild-type mice ages postnatal day (P) 13, P19, and P26. After 48 hours, mice were perfused and sections processed by using tetramethylbenzidine histochemistry. Labeling distribution in some animals was analyzed quantitatively. Obvious differences in the distribution of the ipsilateral retinocollicular pathway were observed at P15, with the pathway being more exuberant in CCKO mice. This difference was statistically significant. More subtle differences were seen at P21 and P28. Obvious differences were also seen in the contralateral retinogeniculate pathway which in CCKO mice filled most of the domain normally occupied by ipsilateral eye fibers. This difference was also statistically significant. We conclude that reduction in L-type Ca(2+) currents has an effect on axonal refinement similar to that which occurs in NO knockout mice, which supports the possibility that L-type Ca(2+) channel-dependent LTD mediates NO-dependent axonal refinement.

trkA modulation of developing somatosensory neurons in oro-facial tissues: tooth pulp fibers are absent in trkA knockout mice

To investigate the nerve growth factor requirement of developing oro-facial somatosensory afferents, we have studied the survival of sensory fibers subserving nociception, mechanoreception or proprioception in receptor tyrosine kinase (trkA) knockout mice using immunohistochemistry. trkA receptor null mutant mice lack nerve fibers in tooth pulp, including sympathetic fibers, and showed only sparse innervation of the periodontal ligament. Ruffini endings were formed definitively in the periodontal ligament of the trkA knockout mice, although calcitonin gene-related peptide- and substance P-immunoreactive fibers were reduced in number or had disappeared completely. trkA gene deletion had also no obvious effect on the formation of Meissner corpuscles in the palate. In the vibrissal follicle, however, some mechanoreceptive afferents were sensitive for trkA gene deletion, confirming a previous report [Fundin et al. (1997) Dev. Biol. 190, 94-116]. Moreover, calretinin-positive fibers innervating longitudinal lanceolate endings were completely lost in trkA knockout mice, as were the calretinin-containing parent cells in the trigeminal ganglion.These results indicate that trkA is indispensable for developing nociceptive neurons innervating oral tissues, but not for developing mechanoreceptive neurons innervating oral tissues (Ruffini endings and Meissner corpuscles), and that calretinin-containing, trkA dependent neurons in the trigeminal ganglion normally participate in mechanoreception through longitudinal lanceolate endings of the vibrissal follicle.

Long-term depression: a cellular basis for learning?

Long-term depression (LTD) comprises a persistent activity-dependent reduction in synaptic efficacy which typically occurs following repeated low frequency afferent stimulation. Hippocampal LTD has been a subject of particular interest due to the established role of the hippocampus in certain forms of information storage and retrieval. Recently, it was reported that LTD in the CA1 region may be associated with novelty acquisition in rats. CA1 LTD expression may also be increased in stressful conditions. This suggests a more complex role for this form of plasticity than the oft-cited hypothesis that it simply serves to prevent synapse saturation, by means, for example, of enabling reversal of long-term potentiation (LTP). One possibility is that LTD may be directly involved in the creation of a memory trace. Alternatively, LTD may prime a synapse in readiness for the expression of LTP, thereby contributing indirectly to information storage. There is increasing evidence that LTD is not mechanistically the reverse of LTP. Although some common processes exist, molecular, biochemical, electrophysiological and pharmacological studies all point to several quite distinct induction and maintenance mechanisms for this form of synaptic plasticity. Taken together these findings suggest that hippocampal LTD must be considered in a new light. This review focuses on the interpretation of novel and established information with regard to LTD in the hippocampal CA1 region in terms of its possible role as a cellular basis for learning and memory.

Arterial wall chondroitin sulfate proteoglycans: diverse molecules with distinct roles in lipoprotein retention and atherogenesis

Chondroitin sulfate proteoglycans (CSPGs) of the arterial wall are generally considered to be atherogenic because of their ability to trap cholesterol-rich lipoproteins in vitro. Nevertheless, CSPGs are a diverse group of molecules with a long evolutionary history and distinct biologic functions. The three principal CSPGs in the arterial wall are versican, which is part of the hyalectan gene family; and decorin and biglycan, which are members of a separate gene family, the small leucine-rich proteoglycans. Importantly, there is now substantial evidence that the different molecular species of CSPGs participate unequally in lipoprotein retention, and that they exert unequal regulatory effects that are related to atherogenesis. Recently available murine models with genetic manipulations that affect CSPGs now allow causal studies of the roles of these molecules to be conducted in vivo, with occasionally surprising results. Moreover, tools are being developed to examine human genetic variations that are relevant to CSPGs, which may provide additional important insights into the human disease. The era in which proteoglycans are regarded as a nondescript backdrop, playing purely nonspecific structural roles, is over. Studies in manipulated animals and in human populations will continue to reveal precise, dynamic roles for these fascinating and ancient molecules.