Limitations of the Commercially Available Gene Expression Test in Predicting Cutaneous Squamous Cell Carcinoma Metastasis and Clinical Outcomes

BACKGROUND

Clinical genetic tests are integral to healthcare decision-making. However, the unclear regulatory framework, especially regarding products that evade stringent FDA oversight, may compromise test validity and transparency.

OBJECTIVE

To critically evaluate the DecisionDx® cutaneous squamous cell carcinoma test by Castle Biosciences for its dataset biases, gene panel selection, and reported accuracy metrics, providing insight into broader challenges in the clinical genetic testing landscape.

METHODS

Independent analyses of the DecisionDx®-SCC 40-GEP test data from Castle Biosciences were conducted. These included comparisons to clinical genetic testing standards, analysis of prevalence metrics against national cSCC rates, gene ontology of 34 genes for cSCC associations, and evaluation of accuracy metrics.

RESULTS

The DecisionDx®-SCC met 11 of 44 CDC's ACCE criteria for clinical genetic testing. Its dataset showed a metastasis prevalence higher than the national average. Out of 34 genes, 15 had known associations with cSCC. Inconsistencies in accuracy metrics presentation were noted, particularly in moderate and high-risk stratifications.

CONCLUSION

Analysis of DecisionDx®-SCC indicates potential biases and ambiguities, exacerbated by differences between FDA and CLIA standards. This highlights the need for systematic validation and a unified regulatory approach, stressing the necessity for precise and dependable genetic testing in patient care.

Frequency-dependent ecological interactions increase the prevalence, and shape the distribution, of pre-existing drug resistance

The evolution of resistance remains one of the primary challenges for modern medicine from infectious diseases to cancers. Many of these resistance-conferring mutations often carry a substantial fitness cost in the absence of treatment. As a result, we would expect these mutants to undergo purifying selection and be rapidly driven to extinction. Nevertheless, pre-existing resistance is frequently observed from drug-resistant malaria to targeted cancer therapies in non-small cell lung cancer (NSCLC) and melanoma. Solutions to this apparent paradox have taken several forms from spatial rescue to simple mutation supply arguments. Recently, in an evolved resistant NSCLC cell line, we found that frequency-dependent ecological interactions between ancestor and resistant mutant ameliorate the cost of resistance in the absence of treatment. Here, we hypothesize that frequency-dependent ecological interactions in general play a major role in the prevalence of pre-existing resistance. We combine numerical simulations with robust analytical approximations to provide a rigorous mathematical framework for studying the effects of frequency-dependent ecological interactions on the evolutionary dynamics of pre-existing resistance. First, we find that ecological interactions significantly expand the parameter regime under which we expect to observe pre-existing resistance. Next, even when positive ecological interactions between mutants and ancestors are rare, these resistant clones provide the primary mode of evolved resistance because even weak positive interaction leads to significantly longer extinction times. We then find that even in the case where mutation supply alone is sufficient to predict pre-existing resistance, frequency-dependent ecological forces still contribute a strong evolutionary pressure that selects for increasingly positive ecological effects (negative frequency-dependent selection). Finally, we genetically engineer several of the most common clinically observed resistance mechanisms to targeted therapies in NSCLC, a treatment notorious for pre-existing resistance. We find that each engineered mutant displays a positive ecological interaction with their ancestor. As a whole, these results suggest that frequency-dependent ecological effects can play a crucial role in shaping the evolutionary dynamics of pre-existing resistance.

Most cancers carry a substantial deleterious load due to Hill-Robertson interference

Cancer genomes exhibit surprisingly weak signatures of negative selection^1,2. This may be because selective pressures are relaxed or because genome-wide linkage prevents deleterious mutations from being removed (Hill-Robertson interference)³. By stratifying tumors by their genome-wide mutational burden, we observe negative selection (dN/dS ~ 0.56) in low mutational burden tumors, while remaining cancers exhibit dN/dS ratios ~1. This suggests that most tumors do not remove deleterious passengers. To buffer against deleterious passengers, tumors upregulate heat shock pathways as their mutational burden increases. Finally, evolutionary modeling finds that Hill-Robertson interference alone can reproduce patterns of attenuated selection and estimates the total fitness cost of passengers to be 46% per cell on average. Collectively, our findings suggest that the lack of observed negative selection in most tumors is not due to relaxed selective pressures, but rather the inability of selection to remove deleterious mutations in the presence of genome-wide linkage.

Quantitative In Vivo Analyses Reveal a Complex Pharmacogenomic Landscape in Lung Adenocarcinoma

The lack of knowledge about the relationship between tumor genotypes and therapeutic responses remains one of the most critical gaps in enabling the effective use of cancer therapies. Here, we couple a multiplexed and quantitative experimental platform with robust statistical methods to enable pharmacogenomic mapping of lung cancer treatment responses in vivo. The complex map of genotype-specific treatment responses uncovered that over 20% of possible interactions show significant resistance or sensitivity. Known and novel interactions were identified, and one of these interactions, the resistance of KEAP1-mutant lung tumors to platinum therapy, was validated using a large patient response data set. These results highlight the broad impact of tumor suppressor genotype on treatment responses and define a strategy to identify the determinants of precision therapies. SIGNIFICANCE: An experimental and analytical framework to generate in vivo pharmacogenomic maps that relate tumor genotypes to therapeutic responses reveals a surprisingly complex map of genotype-specific resistance and sensitivity.

A Functional Taxonomy of Tumor Suppression in Oncogenic KRAS-Driven Lung Cancer

Cancer genotyping has identified a large number of putative tumor suppressor genes. Carcinogenesis is a multistep process, but the importance and specific roles of many of these genes during tumor initiation, growth, and progression remain unknown. Here we use a multiplexed mouse model of oncogenic KRAS-driven lung cancer to quantify the impact of 48 known and putative tumor suppressor genes on diverse aspects of carcinogenesis at an unprecedented scale and resolution. We uncover many previously understudied functional tumor suppressors that constrain cancer in vivo. Inactivation of some genes substantially increased growth, whereas the inactivation of others increases tumor initiation and/or the emergence of exceptionally large tumors. These functional in vivo analyses revealed an unexpectedly complex landscape of tumor suppression that has implications for understanding cancer evolution, interpreting clinical cancer genome sequencing data, and directing approaches to limit tumor initiation and progression. SIGNIFICANCE: Our high-throughput and high-resolution analysis of tumor suppression uncovered novel genetic determinants of oncogenic KRAS-driven lung cancer initiation, overall growth, and exceptional growth. This taxonomy is consistent with changing constraints during the life history of cancer and highlights the value of quantitative in vivo genetic analyses in autochthonous cancer models.This article is highlighted in the In This Issue feature, p. 1601.

Abstract IA26: Multiplexed functional cancer genomics

Extensive cancer genotyping has identified a large number of putative tumor-suppressor genes. However, genome sequencing data alone are insufficient to uncover the importance and role of these genes during carcinogenesis. Moreover, whether tumor suppressors restrict the rate of tumor growth, decrease the probability of incipient tumorigenesis, or limit the emergence of particularly fast-growing clones remains unknown for even the most well-studied tumor suppressors. I will discuss the development and use of multiplexed autochthonous mouse models of human lung adenocarcinoma that integrate tumor barcoding, CRISPR/Cas9-based genome editing, and high-throughput barcode sequencing (Tuba-seq). Using these methods, the impact of large panels of putative tumor suppressor genes can be quantified in parallel. We find that many previously understudied genes have strong effects on multiple aspects of tumor growth in vivo. While some functional tumor-suppressor genes have particularly strong effects only during some but not other phases of tumorigenesis, many genes impacted multiple facets of cancer development. Tuba-seq-based studies across oncogenic contexts and with combinatorial tumor suppressor gene inactivation further highlight the context dependency of gene function during carcinogenesis. These cause-and-effect analyses uncover an unexpectedly complex taxonomy of tumor suppression. We have further employed these multiplexed and quantitative in vivo models to assess the relationship between tumor genotype and therapeutic responses at scale. We coupled Tuba-seq with robust statistical methods to enable the generation of a pharmacogenomic map of lung cancer treatment responses in vivo. We uncover a surprisingly informative map of genotype-specific therapeutic responses, which highlights the importance of tumor-suppressor genotype in driving precision treatment approaches. These blueprints of the multifaceted nature of carcinogenesis have implications for understanding cancer evolution, interpreting clinical cancer genome sequencing data, and directing approaches for precision cancer treatment.

Citation Format: Hongchen Cai, Chuan Li, Su Kit Chew, Maryam Yousefi, Giorgia Foggetti, Wen-Yang Lin, Zoë N. Rogers, Ian P. Winters, Christopher D. McFarland, Katherina Politi, Charlie Swanton, Dmitri A. Petrov, Monte M. Winslow. Multiplexed functional cancer genomics [abstract]. In: Proceedings of the AACR Special Conference on the Evolving Landscape of Cancer Modeling; 2020 Mar 2-5; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2020;80(11 Suppl):Abstract nr IA26.

Abstract IA03: Functional lung cancer genomics through in vivo genome editing

Cancer genome sequencing has been instrumental in identifying the genomic alterations that occur in human tumors. However, the functional importance of the vast majority of these alterations, both alone and in combination, remains unknown. I will describe methods that integrate tumor barcoding with CRISPR/Cas9-mediated genome editing and ultradeep barcode sequencing to interrogate multiple tumor genotypes simultaneously in autochthonous mouse models of human cancer. We initially used this method to quantify the effects of eleven of the most frequently inactivated genes in human lung adenocarcinoma on tumor growth in vivo. We also investigated the in vivo fitness advantage conferred by inactivation of each of these eleven genes in combination with inactivation of p53 or Lkb1. This map of tumor-suppressive effects for >30 common lung adenocarcinoma genotypes revealed a complex landscape of context dependence and variability in effect strength. I will also describe a platform that integrates multiplexed Cas9-mediated homology-directed repair (HDR) with DNA barcoding and high-throughput sequencing to simultaneously investigate multiple oncogenic alterations in parallel. Using this approach, we introduced a library of nonsynonymous mutations into endogenous Kras in adult somatic cells to initiate tumors. High-throughput sequencing of barcoded Kras HDR alleles from bulk tumor-bearing lung uncovered surprising diversity in Kras variant oncogenicity. These in vivo approaches may redefine our ability to understand how diverse genomic alterations impact tumor initiation, growth, malignant transformation, and therapy responses.

Citation Format: Ian P. Winters, Zoë N. Rogers, Christopher D. McFarland, Pranav V. Lalgudi, Shin-Heng Chiou, Mark A. Kay, Dmitri Petrov, Monte M. Winslow. Functional lung cancer genomics through in vivo genome editing [abstract]. In: Proceedings of the Fifth AACR-IASLC International Joint Conference: Lung Cancer Translational Science from the Bench to the Clinic; Jan 8-11, 2018; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2018;24(17_Suppl):Abstract nr IA03.

Mapping the in vivo fitness landscape of lung adenocarcinoma tumor suppression in mice

The functional impact of most genomic alterations found in cancer, alone or in combination, remains largely unknown. Here we integrate tumor barcoding, CRISPR/Cas9-mediated genome editing and ultra-deep barcode sequencing to interrogate pairwise combinations of tumor suppressor alterations in autochthonous mouse models of human lung adenocarcinoma. We map the tumor suppressive effects of 31 common lung adenocarcinoma genotypes and identify a landscape of context dependence and differential effect strengths.

Multiplexed in vivo homology-directed repair and tumor barcoding enables parallel quantification of Kras variant oncogenicity

Large-scale genomic analyses of human cancers have cataloged somatic point mutations thought to initiate tumor development and sustain cancer growth. However, determining the functional significance of specific alterations remains a major bottleneck in our understanding of the genetic determinants of cancer. Here, we present a platform that integrates multiplexed AAV/Cas9-mediated homology-directed repair (HDR) with DNA barcoding and high-throughput sequencing to simultaneously investigate multiple genomic alterations in de novo cancers in mice. Using this approach, we introduce a barcoded library of non-synonymous mutations into hotspot codons 12 and 13 of Kras in adult somatic cells to initiate tumors in the lung, pancreas, and muscle. High-throughput sequencing of barcoded Kras HDR alleles from bulk lung and pancreas reveals surprising diversity in Kras variant oncogenicity. Rapid, cost-effective, and quantitative approaches to simultaneously investigate the function of precise genomic alterations in vivo will help uncover novel biological and clinically actionable insights into carcinogenesis.

The Damaging Effect of Passenger Mutations on Cancer Progression

Genomic instability and high mutation rates cause cancer to acquire numerous mutations and chromosomal alterations during its somatic evolution; most are termed passengers because they do not confer cancer phenotypes. Evolutionary simulations and cancer genomic studies suggest that mildly deleterious passengers accumulate and can collectively slow cancer progression. Clinical data also suggest an association between passenger load and response to therapeutics, yet no causal link between the effects of passengers and cancer progression has been established. To assess this, we introduced increasing passenger loads into human cell lines and immunocompromised mouse models. We found that passengers dramatically reduced proliferative fitness (∼3% per Mb), slowed tumor growth, and reduced metastatic progression. We developed new genomic measures of damaging passenger load that can accurately predict the fitness costs of passengers in cell lines and in human breast cancers. We conclude that genomic instability and an elevated load of DNA alterations in cancer is a double-edged sword: it accelerates the accumulation of adaptive drivers, but incurs a harmful passenger load that can outweigh driver benefit. The effects of passenger alterations on cancer fitness were unrelated to enhanced immunity, as our tests were performed either in cell culture or in immunocompromised animals. Our findings refute traditional paradigms of passengers as neutral events, suggesting that passenger load reduces the fitness of cancer cells and slows or prevents progression of both primary and metastatic disease. The antitumor effects of chemotherapies can in part be due to the induction of genomic instability and increased passenger load. Cancer Res; 77(18); 4763-72. ©2017 AACR.

A quantitative and multiplexed approach to uncover the fitness landscape of tumor suppression in vivo

Cancer growth is a multi-stage, stochastic evolutionary process. While cancer genome sequencing has been instrumental in identifying the genomic alterations that occur in human tumors, the consequences of these alterations on tumor growth remains largely unexplored. Conventional genetically engineered mouse models enable the study of tumor growth in vivo, but they are neither readily scalable nor sufficiently quantitative to unravel the magnitude and mode of action of many tumor suppressor genes. Here, we present a method that integrates tumor barcoding with ultra-deep barcode sequencing (Tuba-seq) to interrogate tumor suppressor function in mouse models of human cancer. Tuba-seq uncovers genotype-dependent distributions of tumor sizes with great precision. By combining Tuba-seq with multiplexed CRISPR/Cas9-mediated genome editing, we quantified the effects of eleven tumor-suppressor pathways that are frequently altered in human lung adenocarcinoma. With unprecedented resolution, parallelization, and precision Tuba-seq enables broad quantification of tumor suppressor gene function.

An in vivo multiplexed small-molecule screening platform

Phenotype-based small-molecule screening is a powerful method to identify molecules that regulate cellular functions. However, such screens are generally performed in vitro under conditions that do not necessarily model complex physiological conditions or disease states. Here, we use molecular cell barcoding to enable direct in vivo phenotypic screening of small-molecule libraries. The multiplexed nature of this approach allows rapid in vivo analysis of hundreds to thousands of compounds. Using this platform, we screened >700 covalent inhibitors directed toward hydrolases for their effect on pancreatic cancer metastatic seeding. We identified multiple hits and confirmed the relevant target of one compound as the lipase ABHD6. Pharmacological and genetic studies confirmed the role of this enzyme as a regulator of metastatic fitness. Our results highlight the applicability of this multiplexed screening platform for investigating complex processes in vivo.

A modified ziggurat algorithm for generating exponentially- and normally-distributed pseudorandom numbers

The Ziggurat Algorithm is a very fast rejection sampling method for generating PseudoRandom Numbers (PRNs) from statistical distributions. In the algorithm, rectangular sampling domains are layered on top of each other (resembling a ziggurat) to encapsulate the desired probability density function. Random values within these layers are sampled and then returned if they lie beneath the graph of the probability density function. Here, we present an implementation where ziggurat layers reside completely beneath the probability density function, thereby eliminating the need for any rejection test within the ziggurat layers. In the new algorithm, small overhanging segments of probability density remain to the right of each ziggurat layer, which can be efficiently sampled with triangularly-shaped sampling domains. Median runtimes of the new algorithm for exponential and normal variates is reduced to 58% and 53% respectively (collective range: 41-93%). An accessible C library, along with extensions into Python and MATLAB/Octave are provided.

Abstract 24: A genetic model of metastatic evolution: Driver and passenger mutations affect metastatic fitness

Background: The metastatic process requires that cells accomplish two feats: 1) migrate from the primary tumor into the blood stream and arrest in foreign stroma, and 2) grow from a colony of only a few cells into a metastasis of macroscopic size. This second step is highly inefficient for reasons not well understood. Our previous research in tumorigenesis found that many mildly deleterious passenger mutations accumulate in cancer and can influence neoplastic progression, yet the effect of these mutations in metastatic evolution has not been studied.

Methods: We developed a model of somatic evolution of neoplastic progressions and metastatic growth. Using a modified Gillespie algorithm, individual cells in our model stochastically divide and die, occasionally acquiring advantageous driver mutations or mildly deleterious (for cancer) passenger mutations. The rate of cell division and death depend on the collective effect of accumulated passenger and driver mutations. Evolutionary pressures lead to clonal expansion and occasionally carcinogenesis. In our in silico experiments, aliquots of cells were taken from successful primary tumors and “injected” into a new micro-environment with new stromal interactions, where they were assayed for metastatic success. This allowed us to track the entire evolutionary history of metastatic cells and genomic determinants of their success.

Results: Accumulation of passengers during tumorigenesis as well as metastatic progression frequently prevents initial colonies and micrometastases from developing into metastatic tumors. The population bottleneck experienced during colony formation accelerates the accumulation of deleterious passenger mutations in cancerous populations. We find that in the model, metastatic efficiency depends heavily on primary tumor size, age, and genetic diversity in a manner consistent with clinical observations. A favorable stromal environment was also critical for metastatic success. Surprisingly, we found that as the total number of mutations in neoplastic cells increased, metastatic efficiency, on average, decreased as many of these additional mutations were deleterious passengers.

Conclusions: Accumulation of deleterious passenger mutations helps explain metastatic inefficiency after extravasation, as well as many other features of metastasis. Our model makes several novel predictions testable in vivo_most notably that increased mutational load may prevent or impede metastasis. Cancer treatments exploiting the deleterious effects of passenger mutations may prevent metastatic disease.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 24. doi:10.1158/1538-7445.AM2011-24

The effect of adsorbed fibrinogen, fibronectin, von Willebrand factor and vitronectin on the procoagulant state of adherent platelets

Procoagulant (activated) platelets provide a site for assembly of the prothrombinase complex which can rapidly convert prothrombin into thrombin (a potent inducer of clot formation). Previously, we reported that adhesion of platelets to surfaces preadsorbed with blood plasma caused them to become procoagulant. In the present study we investigated the effect of adsorbed adhesion proteins (fibrinogen (Fg), fibronectin (Fn), von Willebrand factor (vWF) and vitronectin (Vn)) on the procoagulant activity of adherent platelets. Adsorbed Fn, vWF and Fg promoted platelet adhesion in the following order: Fn < vWF = Fg. However, these proteins promoted platelet activation (thrombin generation per adherent platelet) in the following order: Fg < Fn < vWF. Adsorption with a series of dilutions of normal plasma, serum, and plasmas deficient in or depleted of von Willebrand factor (de-vWF), fibronectin (de-Fn), vitronectin (de-Vn), or both vitronectin and fibronectin (de-VnFn) resulted in varied platelet adhesion, but little difference in platelet activation. However, preadsorption with dilute de-vWF plasma induced lower procoagulant activity than normal plasma. Preadsorption with normal plasma resulted in higher levels of platelet activation than preadsorption with Fg, suggesting that adsorption of plasma proteins other than Fg caused the high levels of activation observed for plasma preadsorbed surfaces.

Protein adsorption and cell attachment to patterned surfaces

To better understand the events involved in the generation of defined tissue architectures on biomaterials, we have examined the mechanism of attachment of human bone-derived cells (HBDC) to surfaces with patterned surface chemistry in vitro. Photolithography was used to generate alternating domains of N-(2-aminoethyl)-3-aminopropyl-trimethoxysilane (EDS) and dimethyldichlorosilane (DMS). At 90 min after seeding, HBDC were localized preferentially to the EDS regions of the pattern. Using sera specifically depleted of adhesive glycoproteins, this spatial organization was found to be mediated by adsorption of vitronectin (Vn) from serum onto the EDS domains. In contrast, fibronectin (Fn) was unable to adsorb in the face of competition from other serum components. These results were confirmed by immunostaining, which also revealed that both Vn and Fn were able to adsorb to EDS and DMS regions when coated from pure solution, i.e., in the absence of competition. In this situation, each protein was able to mediate cell adhesion across a range of surface densities. Cell spreading was constrained on the EDS domains, as indicated by cell morphology and the lack of integrin receptor clustering and focal adhesion formation. This spatial constraint may have implications for the subsequent expression of differentiated function.

Surfaces designed to control the projected area and shape of individual cells

Materials with spatially resolved surface chemistry were designed to isolate individual mammalian cells to determine the influence of projected area on specific cell functions (e.g., proliferation, cytoskeletal organization). Surfaces were fabricated using a photolithographic process resulting in islands of cell binding N-(2-aminoethyl)-3-aminopropyl-trimethoxysilane (EDS) separated by a nonadhesive interpenetrating polymer network [poly (acrylamide-co-ethylene glycol); P (AAm-co-EG)]. The surfaces contained over 3800 adhesive islands/cm2, allowing for isolation of single cells with projected areas ranging from 100 microns 2 to 10,000 microns 2. These surfaces provide a useful tool for researching how cell morphology and mechanical forces affect cell function.

Attachment of cultured human bone cells to novel polymers

The initial attachment of human bone-derived cells (HBDC) to several polymer systems has been studied in vitro. A novel polymer system based on poly(ethyl methacrylate) polymer and tetrahydrofurfuryl methacrylate monomer (PEMA/THFMA) was compared with a variant in which THFMA was replaced by 2-hydroxyethyl methacrylate (PEMA/HEMA). Tissue culture polystyrene (TCPS) and polystyrene (PS) were used as reference materials. The ability of the substrates to adsorb the attachment glycoproteins fibronectin (Fn) and vitronectin (Vn) from serum and the subsequent effect on radiolabeled HBDC attachment were examined. Initial cell attachment from the medium containing 10% (v/v) serum was highest on TCPS; on PEMA/THFMA and PEMA/HEMA substrates it was about 25% of this level, and on PS it was only 10% of that on TCPS. Attachment of HBDC to all substrates was dependent on the presence of Vn, which, unlike Fn, was able to adsorb in the face of competition from other serum components. Both Vn and Fn were able to support cell attachment when precoated onto all substrates. In comparison to TCPS, PEMA/THFMA did not show enhanced adsorption of either Fn or Vn from serum, and this was reflected in the level of cell attachment. Interestingly, the potency of preadsorbed Fn for cell attachment was much higher on this substrate than on any other: 45 ng/cm2 Fn when adsorbed to PEMA/THFMA gave a level of cell attachment 1.6-fold higher than the same density of Fn on PS or TCPS. The maximum Fn surface density achieved on HEMA/PEMA was 16 ng/cm2. Cells on PEMA/THFMA showed typical clustering of the alpha5 beta1 Fn receptor, but this was not evident in cells attached to PEMA/HEMA even when precoated with Fn. This study indicates that the initial attachment of HBDC to all substrates was Vn dependent. It also indicates that on PEMA/THFMA the favorable presentation of subsequently adsorbed Fn may assist matrix assembly.

Albumin-binding surfaces: synthesis and characterization

The nature of the proteinaceous film deposited on a biomaterial surface following implantation is a key determinant of the subsequent biological response. To achieve selectivity in the formation of this film, monoclonal antibodies have been coupled to a range of solid substrates using avidin-biotin technology. Antibody clones varied in their antigen-binding activity following insertion of biotin groups into lysine residues. Biotinylated antibodies coupled to solid substrates via an immobilized avidin bridge retained their biological activity. During immobilization of avidin a significant proportion of the protein molecules were passively adsorbed rather than covalently attached to the surface. This loosely bound material could be removed by stringent elution procedures which resulted in a surface density of 5.4 pmol avidin cm(-2). Although these conditions would be harsh enough to denature monoclonal antibodies, they did not destroy the biotin-binding activity of the residual surface-coupled avidin, enabling the subsequent immobilization of biotinylated antibodies. The two-step immobilization technique allowed the use of gentle protein modification procedures, reduced the risk of surface-induced denaturation and removed loosely bound material from the surface. The versatility of the technique encourages its application to a wide range of immobilization systems where retention of biological activity is a key requirement.

Albumin-binding surfaces: in vitro activity

Immobilized monoclonal antibodies (Mabs) have been used to attract specific molecules to a solid surface from complex mixtures such as blood, plasma or serum, thereby directing the response to the modified substrate, a key goal in rational biomaterial design. The nature of the Mab dictated the nature of the response: anti-albumin antibodies were used to prevent cell and platelet adhesion in vitro, whilst anti-fibronectin Mabs promoted attachment. Patterned surfaces could be formed, bearing Mabs that generated adhesive and non-adhesive regions. Fibrinogen adsorption from plasma showed a Vroman peak on unmodified control polymer, which was reduced by 64% in the presence of surface-bound anti-albumin Mab. Immobilization of a control Mab reduced fibrinogen adsorption only slightly, implying an albumin-mediated effect. In static tests, platelet adhesion from human platelet rich plasma was significantly reduced by the immobilization of anti-HSA Mab when compared to the untreated FEP surface (p < 0.0001). This effect was also seen with citrated blood flowing through Mab-treated polyurethane tubing at a shear rate of 132 s(-1) (p=0.034). Since platelets and proteins (as blood, plasma or serum) were introduced to the surface simultaneously, the generation of a defined protein film must have been sufficiently rapid as to shape the platelet or cell response.

Polymer surface chemistry and bone cell migration

Implant devices for orthopaedic applications may be improved if the surface of the biomaterial provides for osteointegration. To understand the effect of hydrophilicity on colonisation by human bone derived (HBD) cells, we compared untreated polystyrene (PS) and a sulfuric acid-treated PS surface for mechanisms of cell migration. The chemical composition of the acid-treated PS surface was analysed by monochromatic X-ray photoelectron spectroscopy and found to contain various oxidatively produced groups and a minor amount of sulfonate groups. It was found that migration of HBD cells on both PS and acid-treated PS surface was dependent on the presence of vitronectin (Vn) and was higher on the hydrophilic acid-treated surface. Minimal migration of HBD cells occurred on either surface in the absence of Vn, even when fibronectin was present in the culture medium. Using radiolabelled protein, it was shown that Vn adsorption onto the acid-treated surface was two to three fold greater than that on the hydrophobic PS. When HBD cells were seeded onto a patterned surface in a medium containing Vn, the cells preferentially colonised the hydrophilic region and few, if any, cells traversed the haptotactic boundary from the hydrophilic to the hydrophobic side. Thus the enhanced HBD cell migration seen on the acid-treated PS compared with the untreated PS surface and the haptotactic boundary phenomenon, relate to Vn adsorption.

The role of vitronectin in the attachment and spatial distribution of bone-derived cells on materials with patterned surface chemistry

In recent years a central objective of tissue engineering has been understanding the interaction of cells with biomaterial surfaces. In this study we examined the protein adsorption events necessary to control the attachment and the subsequent spatial distribution of bone-derived cells exposed to chemically modified surfaces. Silane chemistry and photolithography techniques were used to create substrates with alternating regions of an aminosilane, N-(2-aminoethyl)-3-aminopropyl-trimethoxysilane (EDS), along side an alkylsilane, dimethyldichlorosilane (DMS), on quartz surfaces. Sera depleted of fibronectin (Fn), vitronectin (Vn), or both were used to determine if these proteins were necessary for the initial attachment and spatial distribution of bone-derived cells exposed to modified surfaces in vitro. The kinetics and mechanisms of the spatial distribution of cells were examined using light microscopy and digital image acquisition and subsequently were analyzed. Compared to complete serum, the use of serum depleted of fibronectin with vitronectin included had minimal effect on the cell attachment, spreading, and spatial distribution on the EDS regions of the surface. However, the use of serum depleted of vitronectin with or without fibronectin included resulted in greatly reduced cell attachment and spreading. Thus the presence of vitronectin was required for the attachment, spreading, and spatial distribution of bone-derived cells exposed to EDS/DMS-patterned surfaces.

The role of vitronectin in the attachment and spatial distribution of bone-derived cells on materials with patterned surface chemistry

In recent years a central objective of tissue engineering has been understanding the interaction of cells with biomaterial surfaces. In this study we examined the protein adsorption events necessary to control the attachment and the subsequent spatial distribution of bone-derived cells exposed to chemically modified surfaces. Silane chemistry and photolithography techniques were used to create substrates with alternating regions of an aminosilane, N-(2-aminoethyl)-3-aminopropyl-trimethoxysilane (EDS), along side an alkylsilane, dimethyldichlorosilane (DMS), on quartz surfaces. Sera depleted of fibronectin (Fn), vitronectin (Vn), or both were used to determine if these proteins were necessary for the initial attachment and spatial distribution of bone-derived cells exposed to modified surfaces in vitro. The kinetics and mechanisms of the spatial distribution of cells were examined using light microscopy and digital image acquisition and subsequently were analyzed. Compared to complete serum, the use of serum depleted of fibronectin with vitronectin included had minimal effect on the cell attachment, spreading, and spatial distribution on the EDS regions of the surface. However, the use of serum depleted of vitronectin with or without fibronectin included resulted in greatly reduced cell attachment and spreading. Thus the presence of vitronectin was required for the attachment, spreading, and spatial distribution of bone-derived cells exposed to EDS/DMS-patterned surfaces.

Role of the heparin binding domain of fibronectin in attachment and spreading of human bone-derived cells

Human bone-derived cells are known to attach and spread on surfaces which have been precoated with fibronectin, but the contributions made by specific domains of the molecule have not yet been defined. Here we refer to the osteoblast-like cells as human bone cells. We have determined the relevance of separate regions of fibronectin, particularly the heparin-binding region, for the initial attachment and spreading of these cells. Human bone cells attached to fragments from each of the cell- and heparin-binding regions of fibronectin, but failed to attach to a fragment from the gelatin-binding region. Bovine corneal epithelial cells, which were included as an example of an alternative primary cell strain, attached to the cell-binding fragment but showed no specific short-term attachment to the heparin or gelatin-binding fragments. Monoclonal antibody MAb17, which binds to the cell binding region of fibronectin, partially inhibited the attachment of both human bone cells and corneal epithelial cells to intact fibronectin when present at 50 micrograms/ml and reduced human bone cell attachment to the cell-binding region fragment of fibronectin. Monoclonal antibody, MAb 32, which binds to the heparin-binding region of fibronectin, failed to inhibit attachment of the human bone cells to fibronectin but reduced the attachment of these cells to the heparin-binding region fragment. Heparin and chondroitin sulphate were able to inhibit human bone cell attachment to the heparin-binding fragment of fibronectin but had no effect on their attachment to intact fibronectin or the cell-binding region of fibronectin. Immunofluorescent staining and confocal microscopy showed extensive spreading and actin filament formation when human bone cells were cultured on intact fibronectin. Cells cultured on the heparin-binding fragment showed only minimal spreading coinciding with less extensive actin filament organisation. On the cell-binding fragment of fibronectin more spreading was seen than on the heparin-binding fragment but it was not as extensive as on intact fibronectin. Taken together, these data suggest that human bone cells, unlike bovine corneal epithelial cells, have an attachment mechanism for the heparin-binding region of fibronectin. Attachment to this region is probably mediated by cell surface proteoglycans. However, interaction with the cell-binding domain is required for effective cell spreading of human bone cells on fibronectin during the first 90 minutes after seeding into culture.

Regulation of growth plate cartilage degradation in vitro: effects of calcification and a low molecular weight angiogenic factor (ESAF)

Endothelial cell stimulating angiogenic factor (ESAF) is an activator of matrix metalloproteinases, including latent collagenase, and is released by chondrocytes during calcification. ESAF, added to cultured growth plate chondrocytes, elicited a time-dependent, 2.4-fold increase in matrix lysis (compared with 30% for IL-1). Matrix breakdown was suppressed by addition of tissue inhibitor of metalloproteinase (TIMP). Although calcification has been previously reported to stimulate ESAF production, no corresponding increase in cartilage lysis was seen in the present study. However, the level of ESAF that cultures produce during calcification is many times less than that added to the cultures in this series of experiments. We conclude that ESAF can produce dramatic increases in cartilage breakdown (apparently by activation of latent enzymes), but only at levels in excess of those stimulated by calcification. This indicates that ESAF may operate in concert with other initiators, perhaps from the invading endothelial cells.

Production of endothelial cell stimulating angiogenesis factor (ESAF) by chondrocytes during in vitro cartilage calcification

The aim of this study was to identify the stimulus for production of the latent collagenase and angiogenic activator ESAF by growth plate chondrocytes. Stimulation correlated most closely with matrix calcification. Alkaline phosphatase was necessary for calcification (and so stimulation of ESAF production) but we could find no evidence for a direct link with ESAF production. ESAF production was also stimulated by addition of preformed mineral to non-calcified cultures but was inhibited by dexamethasone. Protein synthesis was necessary for the stimulation of ESAF production by calcification, though ESAF is not itself a protein. Based on these findings we suggest that chondrocytes, at a suitable stage of maturation in the growth plate, are stimulated to produce ESAF by the proximity of crystals in the matrix. Stimulation, which may consist of the induction of an enzyme or transport protein, leads to the release of this potent activator of collagenolysis as part of the angiogenic cascade.

Epiphyseal growth plate cartilage and chondrocytes in mineralising cultures produce a low molecular mass angiogenic procollagenase activator

We have investigated one aspect of the biochemical control of angiogenesis in the mammalian growth plate. This is the first report of an angiogenic compound from normal growing skeletal tissue. The active agent was indistinguishable from ESAF (endothelial cell-stimulating angiogenic factor), previously identified from tumour, retina and synovial fluid. 1. Using a procollagenase activation assay, ESAF was detected in low molecular weight extracts of growth plate cartilage from two species (foetal calf and young rabbit). ESAF content declined with increasing age of the foetal calf growth plate cartilage. 2. Rabbit growth plate chondrocytes in high density culture also produced ESAF 24-48 h after calcification of the culture. Inhibition of alkaline phosphatase in these cultures reduced the stimulation of ESAF production. These findings suggest that ESAF production occurs as a stage in chondrocyte differentiation which is linked with the onset of matrix calcification. A model is proposed for the control of angiogenesis in the growth plate.