MetFinder: A Tool for Automated Quantitation of Metastatic Burden in Histological Sections From Preclinical Models

As efforts to study the mechanisms of melanoma metastasis and novel therapeutic approaches multiply, researchers need accurate, high-throughput methods to evaluate the effects on tumor burden resulting from specific interventions. We show that automated quantification of tumor content from whole slide images is a compelling solution to assess in vivo experiments. In order to increase the outflow of data collection from preclinical studies, we assembled a large dataset with annotations and trained a deep neural network for the quantitative analysis of melanoma tumor content on histopathological sections of murine models. After assessing its performance in segmenting these images, the tool obtained consistent results with an orthogonal method (bioluminescence) of measuring metastasis in an experimental setting. This AI-based algorithm, made freely available to academic laboratories through a web-interface called MetFinder, promises to become an asset for melanoma researchers and pathologists interested in accurate, quantitative assessment of metastasis burden.

Abstract LB052: Melanoma-secreted amyloid beta suppresses neuroinflammation and promotes brain metastasis

Brain metastasis is a significant cause of morbidity and mortality in multiple cancer types and represents an unmet clinical need. The mechanisms that mediate metastatic cancer growth in the brain parenchyma are largely unknown. Melanoma, which has the highest rate of brain metastasis among common cancer types, is an ideal model to study how cancer cells adapt to the brain parenchyma. Our unbiased proteomics analysis of melanoma short-term cultures revealed that proteins implicated in neurodegenerative pathologies are differentially expressed in melanoma cells explanted from brain metastases compared to those derived from extracranial metastases. We showed that melanoma cells require amyloid beta (Aβ) for growth and survival in the brain parenchyma. Melanoma-secreted Aβ activates surrounding astrocytes to a prometastatic, anti-inflammatory phenotype and prevents phagocytosis of melanoma by microglia. Finally, we demonstrate that pharmacological inhibition of Aβ decreases brain metastatic burden. Our results reveal a novel mechanistic connection between brain metastasis and Alzheimer’s disease - two previously unrelated pathologies, establish Aβ as a promising therapeutic target for brain metastasis, and demonstrate suppression of neuroinflammation as a critical feature of metastatic adaptation to the brain parenchyma.

Citation Format: Kevin Kleffman, Grace Levinson, Indigo V. Rose, Lili Blumenberg, Sorin A. Shadaloey, Avantika Dhabaria, Eitan Wong, Francisco Galán-Echevarría, Alcida Karz, Diana Argibay, Richard Von-Itter, Alfredo Floristán, Gillian Baptiste, Nicole Eskow, James Tranos, Jenny Chen, Eleazar C. Vega Saenz de Miera, Melissa Call, Robert Rogers, George Jour, Youssef Zaim Wadghiri, Iman Osman, Yue Ming Li, Paul Mathews, Ronald Demattos, Beatrix Ueberheide, Kelly Ruggles, Shane A. Liddelow, Robert J. Schneider, Eva Hernando. Melanoma-secreted amyloid beta suppresses neuroinflammation and promotes brain metastasis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr LB052.

Melanoma-secreted Amyloid Beta Suppresses Neuroinflammation and Promotes Brain Metastasis

Brain metastasis is a significant cause of morbidity and mortality in multiple cancer types and represents an unmet clinical need. The mechanisms that mediate metastatic cancer growth in the brain parenchyma are largely unknown. Melanoma, which has the highest rate of brain metastasis among common cancer types, is an ideal model to study how cancer cells adapt to the brain parenchyma. Our unbiased proteomics analysis of melanoma short-term cultures revealed that proteins implicated in neurodegenerative pathologies are differentially expressed in melanoma cells explanted from brain metastases compared to those derived from extracranial metastases. We showed that melanoma cells require amyloid beta (AB) for growth and survival in the brain parenchyma. Melanoma-secreted AB activates surrounding astrocytes to a pro-metastatic, anti-inflammatory phenotype and prevents phagocytosis of melanoma by microglia. Finally, we demonstrate that pharmacological inhibition of AB decreases brain metastatic burden.

Evolutionarily informed machine learning enhances the power of predictive gene-to-phenotype relationships

Inferring phenotypic outcomes from genomic features is both a promise and challenge for systems biology. Using gene expression data to predict phenotypic outcomes, and functionally validating the genes with predictive powers are two challenges we address in this study. We applied an evolutionarily informed machine learning approach to predict phenotypes based on transcriptome responses shared both within and across species. Specifically, we exploited the phenotypic diversity in nitrogen use efficiency and evolutionarily conserved transcriptome responses to nitrogen treatments across Arabidopsis accessions and maize varieties. We demonstrate that using evolutionarily conserved nitrogen responsive genes is a biologically principled approach to reduce the feature dimensionality in machine learning that ultimately improved the predictive power of our gene-to-trait models. Further, we functionally validated seven candidate transcription factors with predictive power for NUE outcomes in Arabidopsis and one in maize. Moreover, application of our evolutionarily informed pipeline to other species including rice and mice models underscores its potential to uncover genes affecting any physiological or clinical traits of interest across biology, agriculture, or medicine.

Preimplantation genetic diagnosis for spinal muscular atrophy type I

OBJECTIVE

Couples with children who have spinal muscular atrophy type I (SMA) face a 25% risk of having affected offspring with spontaneous conception. Preimplantation genetic testing (PGT) is possible for the deletions in the survival motor neuron (SMN) gene that have been identified in 98% of SMA type I cases. PGT would provide new reproductive options for families at risk for SMA.

METHODS

Three couples with previously affected children confirmed by DNA testing each underwent in vitro fertilization (IVF) and PGT of the resulting embryos. One or two blastomeres were biopsied from each embryo and analyzed for deletions in exons 7 and 8 of the SMN gene.

RESULTS

Nine embryos were predicted to be unaffected, three to be affected, and one embryo could not be interpreted. One of three patients receiving transfer of unaffected embryos became pregnant with twins.

CONCLUSIONS

Preimplantation genetic testing provides a means for couples at risk for spinal muscular atrophy type I to reduce their chance of initiating an affected pregnancy.

Efficiency of MicroSort flow cytometry for producing sperm populations enriched in X- or Y-chromosome haplotypes: a blind trial assessed by double and triple colour fluorescent in-situ hybridization

Using fluorescent in-situ hybridization (FISH) we have evaluated, on a blind basis, the efficiency of flow cytometry to separate human X- and Y-chromosome bearing spermatozoa. Our data demonstrate that human spermatozoa can be sorted to a purity of 80-90% for X spermatozoa and of 60-70% for Y spermatozoa. Our results using triple FISH fully agree with the sorting treatment used in each case and corroborate the efficiency of the flow sorting technique for sperm sex selection. In these limited samples (200-500 sperm/donor), the frequencies of disomic or diploid spermatozoa were not increased when comparing the sorted samples with unselected samples or with our control series.

Preimplantation genetic diagnosis in Marfan syndrome

The in vitro fertilization technology coupled with the ability to amplify DNA from a single cell has been used for the preimplantation genetic diagnosis of Marfan syndrome. An intragenic FBN1 gene marker has been used to track the inheritance of this disorder in a family. Marker genotyping was established following two rounds of amplification. Whenever possible, two blastomeres were separately assayed per embryo. The transfer of five embryos resulted in a singleton pregnancy and the birth of a full-term male infant.

Preimplantation genetic testing for Marfan syndrome

Marfan syndrome (MFS) is an autosomal dominant disease that affects the skeletal, ocular and cardiovascular systems. Defects in the gene that codes for fibrillin (FBN-1) are responsible for MFS. Here we report the world's first use of preimplantation genetic testing (PGT) to achieve a clinical pregnancy and live birth of a baby free of a Marfan mutation. One or two blastomeres from each embryo were tested for a CA repeat within the FBN-1 gene. The prospective mother is homozygous for the CA repeat (2/2) and has two normal copies of the FBN-1 gene, while the prospective father is heterozygous for the CA repeat (1/2), and is affected with the Marfan syndrome. In the father's family, allele 2 segregates with the mutated FBN-1 gene. For PGT, any embryo diagnosed as heterozygous for the CA repeat (1/2) would be presumed to have inherited normal FBN-1 genes from the father and the mother and be unaffected. One in-vitro fertilization (IVF) cycle yielded 12 embryos for preimplantation testing; six of the embryos were heterozygous for the CA repeat (1/2) and presumed to be free of the Marfan mutation. Five of the six embryos were subsequently transferred into the uterus. The fetus was tested by chorionic villus sampling and found to be free of the Marfan mutation by the same linkage analysis, had a normal fetal echocardiogram, and was normal at birth.

Molecular fragile X screening in normal populations

In December, 1993, we initiated a pilot project in which DNA fragile X (fraX) testing was offered during routine prenatal or genetic counseling to all pregnant women seen at the Genetics & IVF Institute, most of whom were referred for the indication of advanced maternal age. A brochure on fragile X syndrome was sent to each patient prior to her appointment and was reviewed by a counselor or physician during the counseling session. As of June 1995, 3,345 patients were offered testing; 474 women with no identified family history of mental retardation or learning disability and 214 women with a positive family history accepted the test on a self-pay basis. The second population screened was 271 potential donors in our anonymous egg donor program. DNA from blood was tested by Southern blot using EcoRI/EagI and StB12.3. If an expansion was detected, CGG repeat number was determined by PCR-based analysis. Among the 474 patients with unremarkable family histories, three fraX carriers were identified (repeat sizes = 60+), whereas none were found in the 214 patients with a positive family history. Among the potential egg donors, two high borderline patients were identified (repeat sizes = between 50 and 59). Our ongoing study indicates that screening of pregnant or preconceptual populations for fraX carrier status using DNA testing is accepted by many patients and is an important addition to current medical practice.

Recent advances in reproductive genetic technologies

New possibilities for the diagnosis and treatment of reproductive and genetic disorders are becoming available as a result of a series of recent technical advances. Intracytoplasmic sperm injection (ICSI) allows treatment of numerous infertile men whose sperm cannot penetrate the egg to initiate fertilization. Molecular genetic testing provides clients of reproductive age with additional information that permits prevention of genetic diseases such as fragile X syndrome, the leading cause of inherited mental retardation. Preimplantation genetic testing (PGT) offers couples who carry genetic disorders the prospect of having children with a greatly decreased risk of initiating a pregnancy involving an affected individual. Flow-cytometric sperm separation offers a new, effective approach for prevention of X-linked genetic disorders. Two major causes of recurrent pregnancy loss (RPL) involve recurrent trisomies and immunological disorders. Of the latter, 70% of studied populations of patients can attain live births with simple treatment protocols. Maternal serum assays involving multiple markers reduce both false positives and false negatives in detection of trisomies. Despite these advances in research, many safe and effective methods of diagnosis and treatment remain under-utilized in the clinical arena.

DNA-based X-enriched sperm separation as an adjunct to preimplantation genetic testing for the prevention of X-linked disease

We report the world's first clinical pregnancy resulting from DNA-based enrichment for X-bearing human spermatozoa, for prevention of X-linked hydrocephalus. Sperm separation was followed by embryo biopsy and nested multiplex polymerase chain reaction (PCR) for gender determination. Enriched populations of X-bearing spermatozoa ranging from 80 to 89% pure as determined by fluorescence in-situ hybridization (FISH) resulted in in-vitro fertilization (IVF) rates indistinguishable from normal IVF procedures (65%). In two separate biopsy procedures, 7/9 and 15/16 of the resulting embryos were determined to be female by multiplex PCR. Embryo transfer resulted in a karyotypically normal female fetus. This technique should be widely applicable to gender selection for the prevention of genetic disorders.

Improved sizing of fragile X CCG repeats by nested polymerase chain reaction

We have developed an improved method for polymerase chain reaction (PCR)-based sizing of the CCG repeat region at the fragile X locus, FMR-1. This method is designed to optimize denaturation and replication of long repeats with high G + C content, which are otherwise refractory to amplification. The method utilizes nested PCR primers to increase sensitivity and specificity. Alkaline denaturation of the genomic template DNA, combined with addition of glycerol and deaza-dGTP, facilitates strand separation. Labeled PCR products are sized on denaturing polyacrylamide gels. For alleles in the normal-to-premutation size range, strong reproducible signals are routinely obtained from small amounts of rapidly prepared DNA. This allows precise determination of the CCG repeat number, providing data related to the expansion potential of the repetitive segment. Detection of large premutations and some full mutations is also enhanced by the improved procedure.

Prenatal diagnosis in known fragile X carriers

Prenatal diagnosis for fragile X syndrome was performed in 34 pregnancies of 33 known carriers, on 22 chorionic villus samples (CVS), and 15 amniocentesis samples. Fetal and maternal DNA were analyzed by the EagI/EcoRI Southern blot of Rousseau et al. [1991: N Engl J Med 325:1673-1681], with detection of full mutations ensured by a second loading with brief electrophoresis. As a supplemental assay for full mutations, cytogenetic induction was performed in 20 cases. Positive cytogenetic results were helpful in confirming full mutations in CVS cases where the fetal DNA was intermediate in appearance, between a large premutation and a small full mutation. Of 8 mothers with full mutations, the fetal results were 5 full, 2 normal, and 1 premutation (whose mother was a full/pre compound heterozygote). Of 26 mothers with premutations, the fetal results were 5 full, 13 normal, 7 premutation, and 1 uninterpretable (maternal contamination). Maternal premutations were sized in kb by Southern blot and in CGG repeat number by PCR; the predicted correlation between maternal length and penetrance was seen. Follow-up studies include 3 full mutations and 2 premutations confirmed by DNA analysis at birth. Maternal contamination of CVS samples was encountered in 3 of 22 cases, illustrating the value of EagI in detecting maternal (lyonized) chromosomes.

Reliable gender screening for human preimplantation embryos, using multiple DNA target-sequences

Dependable methods were developed for preimplantation sexing of human IVF embryos, for use in clinical settings where prospective parents are at high risk for transmission of X-linked diseases. Using single cultured cells and blastomeres from human embryos as model systems, a multiplex protocol was developed for rapid analysis via nested polymerase chain reaction (PCR). Reliability was enhanced by co-amplification of conserved amelogenin gene segments from both X and Y chromosomes, as well as Y-linked DYZ1 repetitive elements. Each cell was manually isolated and individually washed to avoid potential contaminants. Multiplex amplification allowed recognition of spurious amplification failures specific to particular amelogenin single-copy targets. The X-linked internal control and multiple Y-linked markers allowed recognition and exclusion of most aberrant samples, thus averting potential misdiagnosis. The optimized single-cell protocol reduced experimental sexing errors to < 2% (1/60), but also revealed potential pitfalls of single-cell analysis. With human triploid embryos, separate sampling of individual blastomeres provided concordant female or male signals. Slight modification adapted the procedure for diagnosis of biopsy material from blastocyst stage embryos, allowing separate analysis of multiple tubes containing multiple cells for improved reliability.

Predominant use of a T-cell receptor V beta gene family in simian immunodeficiency virus Gag-specific cytotoxic T lymphocytes in a rhesus monkey

To explore the structural basis for AIDS virus recognition by CD8+ lymphocytes, we sought to determine whether there is a diverse or restricted usage of T-cell receptors (TCR) by simian immunodeficiency virus of macaques (SIVmac) Gag-specific cytotoxic T lymphocytes (CTL) in the rhesus monkey. Six Gag-specific CTL clones were independently generated from an SIVmac-infected rhesus monkey. All six CTL clones recognized a single SIVmac Gag peptide in association with a single major histocompatibility complex class I gene product, Mamu-A*01. TCR alpha-chain sequences from these six CTL clones employed four different V alpha families and five different J alpha gene segments. In contrast, five of the six CTL clones expressed V beta genes that were members of the same family, a human V beta 23 homolog. Furthermore, only one J beta gene was expressed by four of the six CTL clones. These results indicate that TCR of SIVmac Gag-specific CTL from a rhesus monkey can exhibit a restricted usage of V beta gene families and J beta genes.

Sequence and diversity of rhesus monkey T-cell receptor beta chain genes

We have sequenced 23 rearranged T-cell receptor beta chain (Tcrb) cDNA clones derived from peripheral blood lymphocytes (PBL) of a rhesus monkey. All of the clones have a variable-diversity-joining-constant (V-D-J-C) rearrangement similar to that of humans. Two rhesus constant (C) region genes were found, each closely resembling human Cb 1 and 2. All of the rhesus J region sequences align well with ten of the 13 reported human J regions. 17 of the 23 rhesus V region sequences could be assigned to families homologous with eight different human families (Vb 1, 2, 6, 7, 8, 9, 13, and 14). The remaining six V region sequences are more distantly related to human Vb 1 and 13. Thus, the organization and sequences of studied rhesus Tcrb chains resemble human homologs. An evolutionary tree analysis revealed paralogous relationships between specific members of the rhesus and human V region families. Analysis of synonymous and nonsynonymous nucleotide sequence differences indicated that the evolution of the presumed major histocompatibility complex (MHC)-contact regions of the Tcrb chains is less constrained than that of the framework regions.

Organization and nucleotide sequence of the rat T cell receptor beta-chain complex

We have characterized four overlapping genomic clones containing the DA rat TCR C beta complex, which span a total of 23 kb and bear two closely related complexes of gene segments. The D beta 1-J beta 1-C beta 1 and the D beta 2-J beta 2-C beta 2 complexes each contain a single diversity segment, six joining segments and four exons that encode the C region. All gene segments appear to be functional except J beta 2.5, which has a 5-bp frame-shifting deletion. This organizational pattern is identical to that of the mouse, and the homologous rat and mouse coding regions share about 92% nucleotide sequence identity. Our sequence comparisons indicate that a localized gene correction event has homogenized the sequences of the first exons of C beta 1 and C beta 2 in the evolutionary time since rats and mice became separate species. We have identified three repetitive elements, each flanked by short direct repeats, present in the region "brain-specific" identifier (ID) sequences, another is a truncated member of the LINE I class of repetitive elements, and the third is a member of the Alu type 2 family. The insertion of at least two, and probably all, of these elements has occurred since the time of rat/mouse divergence. We have identified a substantial number of "cryptic" rearrangement signals (heptamer/nonamer) in the C beta locus, which match the consensus sequence as well or better than authentic signals, and may represent sites of nonfunctional rearrangements.

Organization and nucleotide sequence of the rat T cell receptor beta-chain complex

We have characterized four overlapping genomic clones containing the DA rat TCR C beta complex, which span a total of 23 kb and bear two closely related complexes of gene segments. The D beta 1-J beta 1-C beta 1 and the D beta 2-J beta 2-C beta 2 complexes each contain a single diversity segment, six joining segments and four exons that encode the C region. All gene segments appear to be functional except J beta 2.5, which has a 5-bp frame-shifting deletion. This organizational pattern is identical to that of the mouse, and the homologous rat and mouse coding regions share about 92% nucleotide sequence identity. Our sequence comparisons indicate that a localized gene correction event has homogenized the sequences of the first exons of C beta 1 and C beta 2 in the evolutionary time since rats and mice became separate species. We have identified three repetitive elements, each flanked by short direct repeats, present in the region "brain-specific" identifier (ID) sequences, another is a truncated member of the LINE I class of repetitive elements, and the third is a member of the Alu type 2 family. The insertion of at least two, and probably all, of these elements has occurred since the time of rat/mouse divergence. We have identified a substantial number of "cryptic" rearrangement signals (heptamer/nonamer) in the C beta locus, which match the consensus sequence as well or better than authentic signals, and may represent sites of nonfunctional rearrangements.

1F7, a novel cell surface molecule, involved in helper function of CD4 cells

We have developed a monoclonal antibody, anti-1F7, that inhibits soluble Ag-driven T cell proliferation as well as PWM-driven IgG synthesis. Anti-1F7 antibody reacts with approximately 57% of unfractionated T cells, 62% of CD4+ cells, and 54% of CD8+ cells. Although the 1F7 Ag is widely distributed among lymphoid cells, this Ag on CD4+ cells is preferentially expressed on the CDw29(4B4+) helper population. Moreover, anti-1F7 antibody further subdivides the CD4+CDw29+ cell subset into CDw29+1F7+ and CDw29+1F7- populations. The CD4+CDw29+1F7+ population of cells maximally proliferates to recall Ag such as tetanus toxoid, whereas helper function for PWM-driven IgG synthesis by B cells belongs to both the CD4+CDw29+1F7+ and CD4+CDw29+1F7- population of cells. The most prominent structure defined by this antibody is a 110-kDa molecule that is different from the 135-kDa, 160-kDa, and 185-kDa glycoproteins identified by anti-CDw29 antibody and the 180-kDa glycoprotein identified by UCHL-1 antibody. It is, however, related to the molecule recognized by anti-Ta1, an activation Ag on T cells. Furthermore, although the Ta1 molecule is recognized by anti-1F7 mAb, the 1F7 family of structures also includes molecules distinct from Ta1.

1F7, a novel cell surface molecule, involved in helper function of CD4 cells

We have developed a monoclonal antibody, anti-1F7, that inhibits soluble Ag-driven T cell proliferation as well as PWM-driven IgG synthesis. Anti-1F7 antibody reacts with approximately 57% of unfractionated T cells, 62% of CD4+ cells, and 54% of CD8+ cells. Although the 1F7 Ag is widely distributed among lymphoid cells, this Ag on CD4+ cells is preferentially expressed on the CDw29(4B4+) helper population. Moreover, anti-1F7 antibody further subdivides the CD4+CDw29+ cell subset into CDw29+1F7+ and CDw29+1F7- populations. The CD4+CDw29+1F7+ population of cells maximally proliferates to recall Ag such as tetanus toxoid, whereas helper function for PWM-driven IgG synthesis by B cells belongs to both the CD4+CDw29+1F7+ and CD4+CDw29+1F7- population of cells. The most prominent structure defined by this antibody is a 110-kDa molecule that is different from the 135-kDa, 160-kDa, and 185-kDa glycoproteins identified by anti-CDw29 antibody and the 180-kDa glycoprotein identified by UCHL-1 antibody. It is, however, related to the molecule recognized by anti-Ta1, an activation Ag on T cells. Furthermore, although the Ta1 molecule is recognized by anti-1F7 mAb, the 1F7 family of structures also includes molecules distinct from Ta1.