Regeneration and Cancer—The Two Sides of a “Stem Cell” Coin: The 2025 Mildred Weisenfeld Award Lecture

Abstract

Stem cells endowed with properties of self-renewal and differentiation have opened new avenues for restoration and regeneration of damaged tissues. The same adult stem cells, when they undergo irreversible mutations, transform into cancer cells and generate a tumor. In this lecture, I present our work on these two aspects of stem cells: one focuses on regeneration of the ocular surface, whereas the other targets cancer stem cells (CSCs) in retinoblastoma (Rb). In regeneration, I present our experience of transplantation of the cultured limbal epithelial cells to restore the ocular surface in severe limbal stem cell deficiency, with 60% to 70% success. The technique proved that transplanted cells survive, network, integrate, contribute to function, and cause no harm to the patient. We also developed a technique wherein both limbal and conjunctival cells could be cultured on a single membrane utilized ring barrier. This approach was effective in reconstructing the conjunctiva and fornices in patients with severe ocular disease. Furthermore, our work led identification of a subpopulation of limbal-derived stromal cells with features of mesenchymal stem cells. We also saw success with culturing human lacrimal gland tissues marked by the presence of epithelial stem cells and mesenchymal stem cells with secretory potential. These cultures we believe could pave the way of lacrimal gland regeneration through cell therapy. We then flip to the “dark side” of stem cells: the isolation and characterization of CSCs in Rb. We worked on both primary tumors as well as Rb y79 cell lines, with a special focus on reversing their chemoresistant properties. Our work demonstrated the possible role of small extracellular vesicles in Rb as a diagnostic and prognostic marker. The reduction of chemoresistance using nano-formulated drugs and targeted delivery of drug-loaded small extracellular vesicles derived from the tumor cells shows promise.

 

I would like to thank Professor Matt Wilson, Board of Trustees (AP), for introducing me so elegantly and Professor Martine Jager, my good friend, for such a lovely and touching presentation about my personal life and milestones. A very good morning, ladies and gentlemen and my dear ARVO family. It is my distinct pleasure and honor to be here in front of this august audience to deliver the prestigious Mildred Weisenfeld Award Lecture, in beautiful Salt Lake City, Utah. I make this presentation on behalf of all my team and collaborators.

Although I didn't have the opportunity to meet Ms Mildred Weisenfeld, I would like to pay my humble respects and regards to her. I am truly grateful to ARVO and to Ms Weisenfeld for setting up the precedent of raising grants to do translation research in ophthalmology.

Before I proceed with my presentation on the two different aspects of adult stem cells which have intrigued me across my career, I would like to pay my respects and gratitude to my two gurus. They introduced me to the field of ophthalmic pathology during my pathology residency at the Nizam's Institute of Medical Sciences (NIMS), Hyderabad, and supervised my thesis entitled “Surgical Pathology of Cornea: 1000 corneal specimens.” The late professor Kamaraju S. Ratnakar from NIMS insisted that I re-learn slit lamp examination, so that I could identify the clinical conditions myself. Dr Gullapalli Nageswara Rao from L. V. Prasad Eye Institute (LVPEI), Hyderabad—who is here to cheer me on today—stressed on highlighting the clinico-pathological correlation of all cases that we studied.

After my pathology residency, I took up surgical pathology at Medwin Hospitals, Hyderabad, and was trained in Oncopathology at the world-renowned Tata Memorial Hospital, Mumbai. After 4 years in surgical pathology, including cytology and hematology, I was invited by Dr Rao to lead the Department of Ophthalmic Pathology at LVPEI, which today is a leading quaternary eye hospital and a World Health Organization (WHO) collaboration center for the prevention of blindness. This decade and a half at the start of my career provided me with the richest experience in ocular pathology and also paved the way for my interests in ophthalmic research. While pursuing this field, another call provided me with an opportunity to work in an academic leadership position at the University of Hyderabad, an Institute of Eminence in India. This equipped me with a very rich experience in academic leadership, academic diversity, teaching, and mentoring. With the cumulative experience of both institutes, I have now returned to LVPEI after my superannuation. I now share my experience of working on two unique aspects of adult stem cells—regeneration and cancer.

The Two Sides of Stem Cells

As a pathology resident, we were taught that all the cells in the body, from head to foot, can be divided into three types of cells. Labile cells (bone marrow, mucosa, and skin cells) regenerate throughout one's lifetime. They tend to have a “stem cell reserve” and regenerate the tissue.¹ Stable adult cells—liver, bone, endothelial cells, and so on—harbor stem cells that come to the rescue as and when required. The permanent tissues in our body, like the central nervous system, the musculoskeletal system, the heart, and glomeruli, do not regenerate in our lifetime.^2,3 So, this knowledge of adult stem cells and the various types of culture techniques available, has helped us progress from 2D to 3D culture systems. This progression has revolutionized stem cell research, and has opened up the potential to regenerate or replace damaged tissues, like those in the skin or limbus.^4–7 On the flip side, these same adult stem cells, when they undergo mutations or irreparable damage, evade the repair and apoptotic pathways and transform into cancer initiating cells that generate a tumor.⁸ In this scenario, the focus shifts to targeting such cancer stem cells (CSCs) to prevent further tumor growth and relapses. In a clinical setting, we do not have the opportunity to identify these CSCs. Therefore, efforts are made to identify cancer initiating cells in the tumor tissue, characterize them, and attempt to target them using various approaches. This talk is about our work involving these two characteristics of the same adult cell (Fig. 1).

Figure 1.

The two aspects of adult stem cells. In regeneration and cancer initiation.

Stem cells are undifferentiated progenitor cells capable of self-renewal and lineage-specific differentiation, which itself is dependent on the developmental hierarchy of stem cells.^9–12 It is now well understood that the stem cells are located in a special environment called the “niche,” which also plays a crucial role in maintaining the special properties of these stem cells; properties that are governed both by intrinsic and extrinsic factors.

Limbal Stem Cells

I will start with my story of exploring the nature of limbal stem cells with Dr Virender S. Sangwan, a corneal surgeon, my friend and colleague at LVPEI. In 2000, I had begun working on Rb Y79 cultures, and one day, I shared my progress with Virender over a cup of tea. He demurred, noting that the need of the hour was to apply the cell culture technique and knowledge of stem cells to help patients with severe ocular surface burns, quoting Graziella Pellegrini's work.⁷ He highlighted that there are many patients suffering from severe ocular surface disease due to chemical burns who are young and otherwise healthy. Fortunately for us, it was already established by the giants in the field that the presence of limbal stem cells can be exploited to make a sheet of cultured epithelium in vitro, for a potential clinical transplant.^6,9 This procedure reduces the amount of donor tissue required and thus avoids any potential threat to the donor site.^10,11

After seeking the institute's review board's approval, obtaining patients’ consent, and financial support from the Indian government's Department of Biotechnology, we developed a unique limbal stem cell culture technique; a xeno-free method using denuded human amniotic membrane (HAM) as a substrate^12,13 as summarized in Figure 2 that became known as the Cultivated Limbal Epithelial Transplant (CLET). The technique avoided the use of feeder cells and used a glass slide in place of the culture insert. Human corneal epithelial cell media, free of xeno products, was used to culture these cells for 2 to 3 weeks. HAM's biocompatible nature and its prior use in clinical practice for ocular surface reconstruction was a big advantage. It was prepared in-house by LVPEI's Ramayamma International Eye Bank using the protocol established by Tseng et al.⁹ We standardized the limbal explant culture technique using 2 to 3 mm of limbal tissues obtained from cadaveric corneo-scleral limbal tissues as well as limbal tissue obtained from patients undergoing ocular surface surgeries. It was noted that the success of establishing cultures with live tissues was 100% as compared to 50% using cadaveric tissues.¹⁴ Subsequently, clinical cases were performed using autologous or allogenic tissues from contralateral healthy eyes or donor limbal tissues.

Figure 2.

Collection of limbal biopsy and explant culture on HAM. Stepwise display of the procedure and culture outcomes. (A) Limbal biopsy (2 × 2 mm²) from a healthy eye; (B) collection of tissue biopsy specimens in a sterile microcentrifuge tube containing HCE medium; (C) peeling of HAM from the nitrocellulose paper; (D) denuded HAM spread and tucked around a glass slide; (E) mincing of limbal tissue on a sterile glass slide; (F) proliferating cells migrating out from the periphery of the explant on days 2 to 3; (G) appearance of the growth zone of a limbal monolayer on HAM on days 4 to 5 (arrow points to an advancing growth zone); (H) a confluent monolayer of limbal epithelial cells on days 10 to 14; (I) hematoxylin and eosin staining of paraffinized sections of limbal epithelial cells grown on HAM (arrow points to the cell-basement membrane junction between cultured cells and HAM at the bottom); (J) patchy growth (arrows point to patches on HAM without cells); (K) explant showing only fibroblastic outgrowth; and (L) a failed culture with only stromal keratocytes. Reprinted from Mariappan I, et al. In vitro culture and expansion of human limbal epithelial cells. Nat Protoc. 2010;5:1470–1479.

The cultured limbal cells were evaluated for the presence of stem cells and differentiated cells, by immunophenotyping (for ABCG2 a key ATP-binding cassette transporter protein; p63, a key nuclear transcription factor; and Cytokeratins [CK] 12/14), label retaining index, BrdU [5-bromo-2′-deoxyuridine] pulse chase, histopathology, and electron microscopy [EM] studies; Supplementary Fig. S1).^15,16 The cells were positive for CK12/14 and ABCG2 with a label retaining index of 4% to 8% at the end of 4 weeks. Histology showed a two to three layered stratified epithelium with intercellular junctions as evident under EM.^12,17

After standardizing the process for the clinical trial, patients clinically diagnosed with severe limbal stem cell deficiency were enrolled for the study. Most of the patients had a history of chemical burns with alkalis at home or at the workplace. The mean age of patients was 10 ± 6.66 years, approximately 67.3% were less than 35 years of age. Patients with Stevens–Johnson syndrome (SJS) and other systemic diseases were excluded. A few patients with vernal keratoconjunctivitis (VKC) who developed severe limbal stem cell deficiency (LSCD) as a complication of severe burnt-out disease were included in the study.¹⁸ For unilateral LSCD, autologous limbal tissue from the contralateral eye was used, and for bilateral LSCD, allogenic limbal tissue was used, along with immunosuppressive therapy. Rare cases of bilateral LSCD with a foci of preserved limbal tissue also benefited from autologous limbal culture transplantation.^19,20 Those who developed a stable ocular surface after transplantation, but had persistent visual disability, were taken up for penetrating keratoplasty.

The surgeries were performed by Dr Virender Sangwan, and later supported by Dr Sayan Basu²¹ (Fig. 3). Ocular surface restoration was seen in 70% to 80% of patients with this procedure alone, without any complications related to surgery or cell transplantation^22,23 (Table). Those who continued to have corneal opacity and visual disability underwent penetrating keratoplasty²⁴ (see Table). The availability of this cornea tissue was a most valuable specimen, as it was now possible to document that the transplant had survived and integrated into the cornea with the formation of a basement membrane.²⁵

Figure 3.

Clinical photographs of eyes before and after autologous cultivated limbal epithelial transplantation. Eyes of four different patients with total limbal stem cell deficiency and variable amounts of corneal stromal scarring (A–D); same eyes 1 year after limbal transplantation (E–H). Right eye of a 26-year-old female patient with a history of alkali injury showing a stable corneal surface, minimal stromal scarring, and a best-corrected visual acuity (BCVA) of 20/30 E. Right eye of a 19-year-old male patient with a history of acid injury showing a stable corneal surface with residual stromal scarring with a BCVA of 20/100 F. Right eye of a 45-year-old male patient with a history of alkali injury showing recurrence of conjunctivalization inferiorly (failure) between 4 and 7 o'clock at the limbus, with a BCVA of 20/40 G. Right eye of a 37-year-old female patient with a history of acid injury showing a stable corneal surface and a clear corneal graft with a BCVA of 20/20 H. Reprinted from Basu S, et al. Clinical outcomes of xeno-free allogeneic cultivated limbal epithelial transplantation for bilateral limbal stem cell deficiency. The British Journal of Ophthalmology. 2012;96:1504–1509.

Table.

Summary of the Publications on Clinical Applications

Sl. No.	Type of Transplantation	No. of Cases	Clinical Outcome	Reference
1.	Autologous cultivated limbal epithelial transplantation	125	Stable ocular surface was noted in all patients within 2 wk of transplantation; 15 patients underwent subsequent penetrating keratoplasty for visual rehabilitation.	25
2.	Autologous cultivated limbal epithelial transplantation for bilateral LSCD	1	Single case with a visual outcome of 20/40 at 1 y follow up	19
3.	Autologous cultivated limbal epithelial transplantation	86	Stable surface noted in 73% of patients, with successful outcome of PK in 11 of 18 cases	22
4.	Autologous cultivated limbal epithelial transplantation	200 with 10-y FU	Avascular, stable corneal surface was seen in 71% cases	23
5.	Allogenic cultivated limbal epithelial transplantation for bilateral LSCD	28 eyes	Stable ocular surface in 715 and successful PK in 80% of cases	26
6.	CLET in children	107 eyes	54% success after one or two CLET procedures	27
7.	CLET in VKC	2	Both cases showed amelioration in symptoms and improvement in vision	20
8.	CLET with PK	47	2 state CLET and PK—80% success	28
9.	CLET with PK	15/125	Of the 15 cases that underwent PK after CLET, 90% success noted	24
10.	Co-culture of limbal and conjunctiva	40 eyes	The cumulative survival probability was 60% at 1 y and 45% at 4	29
11.	Oral mucosa cultures	1	Adapts to limbal phenotype	30

CLET, cultivated limbal epithelial transplant; LSCD, limbal stem cell deficiency; PK, penetrating keratoplasty; VKC, vernal keratoconjunctivitis.

Culmination Into SLET

Although CLET was successful with 70% clinical success, it required a dedicated laboratory with “Good Manufacturing Practice and Good Laboratory Practice” (GMP/GLP) infrastructure and trained personnel. Hence, a very promising and logical outcome was that Dr Sangwan combined the technique of direct transplantation with that of an explant culture technique. The combined version named “simple limbal epithelial transplantation” (SLET) involved the explanting of limbal-derived tissue directly on an amniotic membrane that was placed on the ocular surface. This technique facilitated the expansion and stratification of the epithelial cultures in situ.^26,31,32

Cultivated Oral Mucosa

Patients with bilateral severe injuries lacking an autologous source of limbal tissue were advised an allogenic transplant (cadaveric/live-related) along with immunosuppression. The clinical success of Kinoshita's group prompted us to explore the option of autologous, cultivated oral mucosa epithelium as a substitute for limbal tissue.³³ In our experience using explant culture, cells derived from limbal, conjunctival, and oral tissues share many markers.³⁴ One interesting observation made was the high secretion of sFlt-1 by limbal cells as compared with oral or conjunctival cells. Hence, we speculated that the avascularity of the cornea is possibly a proactive phenomenon, specific to limbal-derived cells. Although the transplanted oral mucosa cells survived and integrated onto the corneal stroma, the transparency and maintenance of avascularity was the limiting factor.

A unique and novel contribution from our group was the reconstruction of the ocular surface in cases with severe ocular surface injuries involving the limbus and the entire conjunctiva, resulting in symblepharon. This was done by designing a “composite or co-culture” of the limbal and conjunctival epithelium in a single amnionic membrane,^25,35 using a fabricated ring barrier made of Perspex. The limbal tissues were explanted within the ring barrier while the conjunctival tissue was explanted outside the barrier, which helped in segregating both cells on the denuded human amniotic membrane (Supplementary Fig. S2). This facilitated cultures of both limbal and conjunctival cells on a single membrane with a cell-free zone where the ring barrier was placed. This technique was used in cases of severe ocular surface disorders with lid deformities and symblepharon and proved to be more than 50% successful.²⁹

Limbal Stromal Cells

The presence of spindle cells at the bottom of a flask prompted us to explore the nature of these cells. Based on our experience with bone marrow *(BM)-derived stromal cells,^36,37 we decided to expand and characterize these spindle cells as per the International Society for Cell & Gene Therapy (ISCT) criteria for mesenchymal stem cells (MSCs).³⁸ We could demonstrate that these spindle cells showed a striking similarity to BM-derived stromal cells in morphology, trilineage differentiation, phenotype, and gene expression, including being HLA-DR negative. The gene expression studies and the secretome of BM-MSCs, limbal-derived MSCs, and limbal-derived epithelial cells suggested that they have a unique signature and that there is complimentary support for limbal epithelial cells and limbal stromal cells³⁹ (Fig. 4; Supplementary Fig. S3). The limbal stromal cells secreted specific growth factors, cytokines, for which there were receptors on the limbal epithelial cells.⁴⁰ One of the criticisms of our initial work was how we could sustain the stemness of our limbal cells in the absence of feeder cells. With cumulative evidence, we postulated that the presence of limbal stromal cells, derived from the limbal stroma, possibly served as a limbal tissue specific feeder cell culture. Subsequently, these stromal cells were used by our group to treat corneal scars to achieve better wound healing.⁴¹

Figure 4.

(A) Mesenchymal phenotype of MSC-BM by laser scanning confocal microscopy. LSCM pictures of MSC-BM show positivity for vimentin (20X), CD90 (20X), and CD29 (20X), and negativity for CD11c (20X), CD11a (20X), CD45 (20X), CD34 (20X), and CD25 (20X). The nuclei are counterstained with propidium iodide (red). (B) Epithelial phenotype of LEC and MC-L by laser scanning confocal microscopy and FACS analysis. LSCM pictures of limbal epithelial cells (LECs) show positivity for K3 (20X) and K14 (20X; a, b), and LSCM pictures of MC-L show negativity for K3 (40X) and K14 (20X; c, d). The nuclei are counterstained with propidium iodide (blue and red). FACS histograms of MC-L confirm the absence of K3 (e) and K14 (f) expression. The blue line represents the test sample, and the purple line represents the isotype control. Reprinted from Polisetty N, Fatima A, Madhira SL, Sangwan VS, Vemuganti GK. Mesenchymal cells from limbal stroma of human eye. Mol Vis. 2008;14:431–442.

Lacrimal Gland Cultures

While our group was working on ocular surface regeneration, the International Atomic Energy Agency (IAEA) invited us to explore the possibility of a stem cell therapy to combat radiation-induced damage. In response, we proposed that the dry eye disease (DED) that develops in eyes after radiation exposure could potentially be treated with cultured lacrimal gland cells. With funding and support from the IAEA, we first documented in a retrospective cohort the development of aqueous deficit DED in more than 50% of patients who had received radiation therapy (RT) to the ocular adnexal region. We went on to establish the first human lacrimal gland cultures in 2D and 3D spheres and characterized the cultured cells using different scaffolds, such as collagen, matrigel, and HAM. The cultured cells from human lacrimal gland tissues demonstrated the presence of a heterogeneous population comprising stem-like cells and differentiated epithelial, mesenchymal, and myofibroblastic cells. This promising finding was demonstrating the secretory functions of these cultured cells by the release of key proteins, such as secretory IgA, lysozyme, and lactoferrin. The identification of sphere-like structures prompted us to name them as lacrispheres, similar to the term coined for mamospheres and sailospheres.⁴² These lacrispheres could secrete quantifiable levels of tear proteins (lysozyme, lactoferrin, and scIgA) into the conditioned media (Fig. 5). The addition of carbocol could further enhance the secretions. Whereas ABCG2 and C-kit were identified in the cultured cells, further investigation is warranted for identification of specific markers, if any, in LG cultured cells.

Figure 5.

(A) Dry eye in patient population. a Graphical representations of dry eye grading in patient population. (b) Kaplan Meier survival analysis curve for age of patient and dry eye development. The analysis showed that it takes 2.9 years for 50% in age group A to develop DES; 1.4 years for those in age group B and 0.5 years for those in age group C to develop ADDE. (B) Comparison of normal (a–d) and post-radiated (e–h) human lacrimal gland. The normal lacrimal gland shows intact tissue architecture. The secretory cells stain positive (arrowhead) for lysozyme, myoepithelial cells for S-100 protein, and the putative stem cells for CD117. The post-radiated tissue shows loss in lobular architecture with widespread periductal and intralobular fibrosis and lymphocytic inundation. Immunophenotyping shows mild to moderate focal positivity for lysozyme, S-100, and CD117, which is reduced both in intensity and number when compared to the control. Scale bar = 100 µm. Reprinted from Tiwari S, et al. Aqueous deficient dry eye syndrome post orbital radiotherapy: a 10-year retrospective study. Translational Vision Science & Technology. 2017;6:19. (C) Lacrispheres from human lacrimal gland. The isolated human lacrimal gland cells were grown as spheres. These lacrispheres increase in size from 60.38 µm on day 7 to approximately 200 µm by days 14 to 16 in vitro. Reprinted from Tiwari S, et al. Establishing human lacrimal gland cultures with secretory function. PLoS One. 2012;7:e29458.

After a few years, along with our collaborators at LVPEI, the presence of mesenchymal stem cells was established in lacrimal gland cultures following the ISCT criteria for stemness. The data revealed that these spindle-shaped cells exhibited similar phenotypic expression, colony-forming ability, and trilineage differentiation like BM-MSCs (Supplementary Fig. S4). Moreover, the gene expression also did not show any significant difference, except for increased IL1-β in LG-MSCs. The LG-MSCs did not express any lacrimal epithelial markers unlike LG tissue ruling out its epithelial origin.⁴³

Summary of the Stem Cell Applications for Regeneration

In summary, our experience of application of adult ocular stem cells was variable. The most successful application was that of limbal cultures to treat patients with severe ocular surface disease due to chemical burns. The long-term outcome shows 70% success using autologous cultures and 60% success with allogenic transplantation for bilateral chemical injuries, which were supported by the use of immunosuppressive drugs. In our series, more than 70% patients were younger than 35 years of age. In a pediatric population, the success of co-cultures using limbal and conjunctival stem cells was of great significance. CLET's key limitation is that it requires a dedicated cGMP laboratory and trained personnel. To overcome these limitations, in 2012, Dr Sangwan and a team from LVPEI combined the two techniques of direct limbal transplantation and explant culture method to establish a cost-effective, single-step clinical procedure which they termed as SLET.²⁶

Cancer Stem Cells

My other emerging interest in stem cell research is the role of CSCs in tumors. It is now evident that all tumors are initiated from one cell—the cancer-initiating cell or CSC—that survives a mutation and divides consistently to make a clone of tumor cells.^44,45 Even though all tumors start as a clonal population, they end up being heterogenous due to various compounding factors.^46,47 The interest in CSCs has huge clinical implications due to their chemo-resistant properties which result in a relapse after treatment.⁴⁸ Most protocols for radiotherapy or chemotherapy target the rapidly dividing cells, but what remains is an under-the-bunker enemy, the CSC, which resides in the niche in its quiescent phase. These quiescent cells resurface, proliferate, and contribute to the relapse of the tumor, metastasizing at distant sites ultimately leading to the death of the patient.^49,50 Our interest in CSCs was kindled after observing the presence of viable tumor cells in many enucleated specimens of retinoblastoma (Rb), even after chemotherapy. Thus began our journey to identify and characterize CSCs in Rb tumor samples.

Cancer Stem Cells in Primary Rb Tumor Cells

To evaluate CSCs in Rb, the tumor cells were isolated from freshly harvested tumor tissues of enucleated specimens. After evaluating multiple markers using flow cytometry, we hypothesized that the cells which were small and FSC^low/SSC^low were the CSC population.⁵¹ After sorting these cells using MACS, we noted that this subset of tumor cells (CD 44 + and CD 133 low) expressed CXCr4, Oct 4 unlike non-CSC which had low or no expression of these primitive genes⁵¹ (Fig. 6).

Figure 6.

(A) Differential expression of stem and differentiated cell markers in Rb cells. (B) The P1 selected cells in FSC versus SSC plot. (C) P2 gate showed the FSC^lo/SSC^lo subpopulation (green), and P3 gate showed the subpopulation FSC^hi/SSC^lo (blue). The population in red was eliminated from analysis. (B) The RT-PCR analysis for human Syntaxin1A, PROX1, CD133, NSE, and GAPDH. Reprinted from Balla MM, Vemuganti GK, Kannabiran C, Honavar SG, Murthy R. Phenotypic characterization of retinoblastoma for the presence of putative cancer stem-like cell markers by flow cytometry. Investigative Ophthalmology & Visual Science. 2009;50:1506–1514.

Because all the assays of stem cells like clonal assays, cytotoxicity, etc. could not be done in primary tumors, we moved on to evaluate CSC in a well-established cell line, RbY79, for different stem cell properties.⁵² One hurdle was that this cell line lacked the expression of CD 44, possibly due to long passaging, which meant we had to use CD 133 negative cells. This subset of CD 133 negative cells showed the properties of CSC: they had high clone forming efficiency, higher migration properties, stem cell gene expression signatures, and were resistant to cytotoxic drugs like cyclophosphamide and ethambutol, which are routinely used in Rb treatment. To demonstrate their CSC function, we also attempted to derive drug-resistant clones in these cells by serial exposure to carboplatin. The cells showed overexpression of ABCG2 genes with a high IC50 value of drugs in cytotoxicity assays.⁵²

Another property of CSC is their tumor-initiating property. We explored this property in a very cost-effective xenograft model using Chick embryo-chorio allantoic membrane (CE-CAM). This old animal model has received attention in recent times for its low cost and ease of handling. CAM, a vascular organ, is very suitable for exploring tumor forming efficiency, angiogenesis, and for drug efficacy studies. We used this model to evaluate the tumor forming efficiency in a CSC and non-CSC subset of the Rb cell line. The eggs were incubated and a window was made on day 3 to facilitate transplantation and observation (Fig. 7). One day 7, the labeled tumor cells were transplanted on the CAM after gentle abrasion. On day 14, after sacrificing CE, the CAM and the embryo were dissected and studied separately. The CAM was evaluated for tumor size and fluorescent intensity using confocal microscopy. The embryo was evaluated using in vivo imaging for evidence of metastatic foci. The study showed that CSC formed larger tumor nodules as compared to the non-CSC lines, and had led to enhanced metastatic lesions in the brain, liver, and bones. Similarly, tumor formation and spontaneous potential was enhanced significantly in the drug-resistant clones derived from this cell line. Histological examination of the CAM and the metastatic site showed evidence of tumor cells with high NC ratio and high ki67.⁵³ There was overexpression of VEGF and PCNA in the CSC population.

Figure 7.

(A) Assessment of tumor nodules formation on CAM and quantitative measurement of tumor nodules Tumor nodules of (a) CM-Dil CD133⁻ transplanted cells, (b) CM-Dil CD133⁺ transplanted cells, (c) eGFP Y79 CD133⁻ transplanted cells, and (d) eGFP Y79 CD133⁺ transplanted cells. (e) Quantitative measurement of tumor volume of eGFP Y79 transplanted cells (CD133⁻ vs. CD133⁺, **P = 0.0014). (B) Evidence of Rb spontaneous metastasis in different parts of chick embryo analyzed using in vivo IVIS spectral imaging. Representative image of in vivo fluorescence observed in chick embryo transplanted with (a) CM-Dil Y79 CD133⁻ cells, (b) CM-Dil Y79 CD133⁺ cells, (c) Quantification of fluorescence intensity with CM Dil labelled cells (d) eGFP Y79 CD133⁻ cells, and (e) eGFP Y79 CD133⁺ cells. (f) Quantification of fluorescence intensity with GFP labelled cells within the chick embryo revealed that CD133⁻ cells had increased spontaneous metastasis when compared to CD133⁺ cells (**P = 0.0077, 0.0017) Reprinted from Nair RM, et al. A short-term chick embryo in vivo xenograft model to study retinoblastoma cancer stem cells. Indian Journal of Ophthalmology. 2022;70:1703–1711.

We then attempted to reverse this chemo-resistance of the CSCs using nanoformulations of the standard chemotherapeutic agents The efficacy and cytotoxicity of clinical-grade carboplatin (CPT) and etoposide (ETP) loaded lactoferrin nanoparticles (Lf-Nps) was evaluated on total, CD133-enriched (non-CSC), and CD133-depleted (CSC) populations of Rb Y79 cells. It was found that compared to the standard chemotherapeutic drugs, – CPT and ET, lactoferrin loaded drugs overcame chemoresistance in the CSC population by more than 50%, which is a promising start for translational research⁵⁴ (Fig. 8).

Figure 8.

(A) Physicochemical characterization of CPT- or ETP-loaded Lf-Nps. HPLC analysis of standard drugs and Lf-NPs (a) CPT and (b) ETP (c) pH-dependent drug release of Lf-CPT and Lf-ETP. (B) Quantitative measurement of drug uptake and release in Rb Y79 CSCs and non-CSCs. (a) Standard CPT and Lf-CPT drug uptake Rb Y79 CSCs and non-CSCs. (b) CPT available in conditioned media after drug treatment of CSCs and non-CSCs at different time points. (c) Standard ETP and Lf-ETP drug uptake Rb Y79 CSCs and non-CSCs and (d) ETP available in conditioned media after drug treatment of CSCs and non-CSCs at different time points (*P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001). Reprinted from Narayana RVL, et al. Carboplatin- and etoposide-loaded lactoferrin protein nanoparticles for targeting cancer stem cells in retinoblastoma in vitro. Investigative Ophthalmology & Visual Science. 2021;62:13.

One unexplored aspect of Rb was the nature of its secretome, especially the role of extracellular vesicles (EVs). The cargo of the small EVs (sEVs)/exosomes secreted by diverse cell types and present in all biological fluids are now being explored for their role in early diagnosis, prognosis as well as potential treatment options. They are also emerging as an ideal drug delivery vehicle as they have a long-circulating half-life, intrinsic homing capability, optimal biocompatibility, minimal or no inherent toxicity, and penetrate different biological barriers. To understand the role of sEVs in Rb, we standardized the optimal method of isolating and characterizing EVs using multiple methods. We found that the sucrose ultracentrifuge method demonstrated mono-dispersive vesicles, and were immunoreactive for CD63, TSG101, CD9, and HSP7.^55,56 We could document in a pilot study that the serum EVs could serve as potential biomarkers in diagnosing extraocular Rb and also serve as markers for resistance to therapy.^57,58

With emerging evidence on the role of sEVs as a drug-delivery carrier system, we evaluated the efficacy of sEVs loaded with CPT (sEVs-CPT) to reverse chemoresistance. Our study provided evidence that sEVs-CPT were highly effective in enhancing cytotoxicity in a drug-resistant clone of Y79 cells, and appear to hold promise as a novel complimentary drug delivery system⁵⁵ (Fig. 9).

Figure 9.

(A) Characterization of sEVs–CPT. (a) Morphology under FE-SEM images were captured at 200 nm. (b) Morphology under a TEM and images were captured at 100 nm. (c) Average size distribution and peak diameter of sEVs-CPT by NTA. (d) The 3D dimension of sEVs-CPT by NTA. (e) Zeta potential of sEVs-CPT at 4°C measured by (f) Polydisperse index of sEVs and sEVs-CPT measured by DLS. (B) CPT release from sEVs in PBS with pH 5.0 and pH 7.2, respectively. (C) Confocal microscopy analysis of sEVs uptake by Y79 cells at different time points (4 hours, 6 hours, 12 hours, and 24 hours). Cell nucleus stained with DAPI (blue), were incubated with 30 µg of Dil-labeled sEVs (red). Bar = 10 µm. (D) Quantitative measurement of cellular uptake (CPT and sEVS) and release from Y79 cells. (a) CPT and sEV-CPT drug uptake Y79 at different time points (1 hour, 2 hours, 4 hours, 8 hours, 16 hours, 24 hours, and 48 hours; n = 3). (b) CPT and sEV-CPT drug uptake Y79 at 4 hours. (c) CPT available in media after sEVs- CPT and CPT treatment on Y79. (d) Percentage drug retention of sEVs-CPT and CPT treated Y79 cells. Values represent mean ±SD (****P < 0.0001, n = 3). All experiments were performed in triplicate. Reprinted from Attem J, Narayana RVL, Manukonda R, Kaliki S, Vemuganti GK. Small extracellular vesicles loaded with carboplatin effectively enhance the cytotoxicity of drug-resistant cells from Y79 cells in-vitro. Biomed Pharmacother. 2024;173:116403.

To understand the interaction of tumor cells at a metastatic site, we created an in vitro experiment where we could mimic the interplay of BM stromal cells and tumor cells through their EVs. The effect of BM-derived EVs on tumor cells and tumor EVs on BM stromal cells were evaluated, specifically for viability, proliferation rate, apoptosis, pro/anti-angiogenic factors, and cytokine profile.⁵⁹ It was noted that sEVs from both sources were internalized by the cells. There was no difference in cell viability of the BM cells after being exposed to tumor EVs. On exposure to tumor cell-derived sEVs, however, the BM stromal cells show proliferation, and migrate with a change in the secretome profile including pro-angiogenic signaling. On the other hand, the sEVs from BM stromal cells when exposed to tumor cells appear to enhance proliferation, migration of cells, with reduced apoptosis and enhanced stem cell properties. So, based on this in vitro evidence of interactions between sEVs of tumor cells and BM stromal cells on each other, we speculated that at the metastatic niche these interactions favor a pro-metastatic milieu that helps the tumor cells proliferate and migrate ultimately supporting the progression to metastasis.

To summarize, we documented the presence of CSCs in primary Rb tumor cells, based on the size and phenotype, with characteristic features including chemoresistance. Chemoresistance was enhanced by serial exposure to cytotoxic drugs and reversed by the use of nano-formulated drugs and drugs loaded on to sEVs. Another novel contribution of our work was the development of a low-cost chick embryo model which could document tumor initiating properties of tumor cells, both from cell lines and primary tumors. It also provided evidence for specific tumor-initiating properties and metastatic properties of the CSC subset of tumor cells, like increased tumor size and evidence of spontaneous metastasis. We believe that this is a good model to evaluate drug efficacy and the angiogenic properties of tumors. Our preliminary evidence points toward a highly communicative environment established between tumor cells and niche cells through their secreted cargo of exosomes.

To conclude, I have worked on these two powerful aspects of adult stem cells over 2.5 decades. We were able to successfully translate the regenerating properties of adult stem cells in the limbus to restore structure and function in severe ocular surface disease due to chemical burns. On the contrary, our work on CSCs in Rb has provided basic and preclinical evidence that targeting of these cells using novel formulations that could potentially reverse their chemo-resistant properties thus opening up new avenues of therapy.