Yang Liu

← Primer

HSC · Healing Source Code

细胞会把酶分给邻居——一个 1968 年的意外,如何让造血干细胞变成体内的酶工厂 Cells lend their enzymes to the neighbors—how a 1968 accident turned hematopoietic stem cells into an enzyme factory inside the body

2026-07-25 · 溶酶体贮积症, cross-correction, 造血干细胞, 基因治疗, MLD

1968 年,两种病人的皮肤成纤维细胞被养进同一个培养皿。一种来自 Hurler 综合征患者,一种来自 Hunter 综合征患者——两种病都让细胞里堆满降解不掉的黏多糖,单独培养时都病得很清楚。混在一起之后,两边都好了。遗传缺陷写在细胞自己的基因组里,凭什么隔壁细胞能把它纠正过来?这个意外后来撑起了一整类疾病的治疗逻辑:溶酶体里的酶,是可以借给邻居用的。

培养基里那个”校正因子”

Fratantoni、Hall 与 Neufeld 把这个观察发表在 1968 年的 Science 上,并点出关键:纠正效应不靠细胞本身,而由释放到培养基里的物质介导。次年,同一组人在 PNAS 上把话说得更实——Hurler 细胞的代谢异常,可以被其他基因型成纤维细胞的分泌物纠正,反过来 Hunter 细胞也一样。这个未知物被朴素地称作”校正因子”。多年后 Neufeld 在 Annu Rev Biochem 的回顾里把来龙去脉说得很直白:那个关键实验一开始是个意外——不同基因型的细胞被混在了一起,结果它们的生化缺陷被纠正了。

1972 年,Bach 等人在 PNAS 上把谜底揭开:从正常人尿液中纯化出的”Hurler 校正因子”带有 α-L-iduronidase 活性,而缺乏该因子的细胞系检测不到这个酶;缺陷细胞被纠正的过程,伴随着它们从培养基中高效摄取该酶。所谓校正因子,就是那个缺失的酶本身。细胞会把一部分新合成的溶酶体酶分泌出去,别的细胞再把它捡回来送进自己的溶酶体——这个后来被称作 cross-correction 的现象,是溶酶体贮积症与其他遗传病最不一样的地方:它的缺陷产物可扩散、可共享。

让它从巧合变成工程学的,是 1970 年代 Sly 实验室搞清楚的一件事:溶酶体酶身上带着”寄送地址”,细胞表面有专门读它的受体。1977 年,Kaplan、Achord 与 Sly 在 PNAS 上给出两条证据:人 β-glucuronidase 被成纤维细胞快速摄取的那些形式,靠的是甘露糖型糖链上或其附近的一个磷酸基团;同年他们在 J Clin Invest 上提示,这种”磷酸己糖识别”很可能是溶酶体糖苷酶被摄取的普遍规律。1979 年,Natowicz 等人给出直接证据:人脾脏纯化的 β-glucuronidase 各形式间摄取速率相差 18 倍,而最酸性、也就是摄取最快的那些形式,每分子携带 4.4 个 mannose 6-phosphate。于是有了一个很实用的推论:要治这类病,不必修好每一个细胞,只需要在体内安置一群持续分泌酶的细胞。

把工厂装进骨髓

造血干细胞几乎是为这件事量身定做的:能被移植、能长期植入、能不断产出遍布全身的髓系后代,而这些后代天然分泌溶酶体酶。1981 年,Hobbs 等人在 Lancet 上报告了最早的尝试之一——一名一岁的 Hurler 男孩接受母亲的骨髓移植,37 天内白细胞的 α-L-iduronidase 活性升到携带者水平,肝脾肿大消退,角膜混浊转清,发育的倒退看上去被止住了。

四十年积累下来的随访也把天花板照了出来。2015 年,Aldenhoven 等人在 Blood 上汇总 217 例成功植入的 Hurler 患者,中位随访年龄 9.2 岁:多数人仍留有可观的残余病负荷;而移植后能否达到正常水平的酶活性,是多数器官系统长期预后的高度显著预测因子。异基因移植能给的酶,最多是一个健康供者的量;这个量,对骨骼和大脑往往不够。

为什么酶能进脑,而它并不”自然”

静脉输注重组酶在外周确实奏效:2001 年 Kakkis 等人在 N Engl J Med 报告的 MPS I 试验里,10 名患者每周输注重组 α-L-iduronidase 共 52 周,肝脾肿大全部缩小,尿糖胺聚糖平均下降 63%;但该试验报告的获益集中在外周,并未报告中枢神经系统的改善。而要让酶在脑内持续存在,后来走通的路是让产酶的细胞本身进到脑里。

移植后确实会发生这件事,但它并不”自然”。2010 年,Ginhoux 等人在 Science 上用谱系追踪证明,成年小鼠脑内的 microglia 来自胚胎第 8 天之前就出现的原始髓系前体,出生后的造血祖细胞对成年脑内 microglia 的稳态维持并无显著贡献——血液和大脑在常态下是两套彼此独立的髓系系统。2012 年,Capotondo 等人在 PNAS 上于小鼠里拆开了移植后的过程:确有一小部分造血祖细胞短暂涌入脑内,这一波与是否做过预处理、脑内有无病变都无关;但只有当预处理足以清除脑内原有的髓系前体时,供者细胞才真正完成对 microglia 的更替,靠的还是早期进入者在脑内就地增殖。

在小鼠里,这也给了 conditioning 另一重角色:清髓不只是给骨髓腾位置,还决定了脑内原有的髓系前体会不会让位。

把产量调上去

如果异基因移植的问题是”酶不够”,那基因治疗最直接的优势就不是纠正病因,而是超生理表达。2004 年,Biffi 等人在 J Clin Invest 上用慢病毒把 ARSA 基因转入小鼠造血干细胞,移植后转基因细胞广泛重建了 CNS 的 microglia 与外周神经的 endoneurial macrophage,并阻止了异染性脑白质营养不良模型的病理与行为异常;关键在于,其疗效显著优于野生型造血干细胞移植,作者把这一差别指向酶的过表达。2010 年,Visigalli 等人在 Blood 上于 MPS I 小鼠里把话说得更死:疗效严格依赖造血系统中达到超正常的酶活性,正是这个过量,才把酶送到了脑和骨骼。产量之外还有产地——2013 年,Sergijenko 等人在 Mol Ther 上发现,在 MPS IIIA 小鼠里,髓系特异的 CD11b 启动子在骨髓中给出的 SGSH 活性与泛表达的 PGK 相当(473% vs 576% 正常值),脑内表达却显著更高(11% vs 7%)。

走到病床

人体数据在 2013 年落地。Biffi 等人在 Science 上报告了三名症状前的晚婴型 MLD 患儿,回输慢病毒转导的自体造血干细胞后,各造血谱系与脑脊液中都出现高水平 ARSA;在超过预测发病年龄 7 至 21 个月的时点上,疾病没有出现也没有进展。2022 年,Fumagalli 等人在 Lancet 上给出整合分析:29 名接受 arsa-cel 的早发型 MLD 患儿中 26 名在分析时存活,中位随访 3.16 年;与年龄和亚型匹配的自然史对照相比,粗大运动功能总分的平均差异在晚婴型为 66%、早幼年型为 42%,症状前接受治疗者获益尤其明显(这个窗口值得单独讲,留到 MLD 那篇)。该疗法即 atidarsagene autotemcel,2020 年 12 月 17 日在欧盟以 Libmeldy 获批,2024 年 3 月 18 日在美国以 Lenmeldy 获批。

Hurler 那边也补上了闭环。2021 年,Gentner 等人在 N Engl J Med 报告了 8 名 MPS IH 患儿(治疗时平均年龄 1.9±0.5 岁,中位随访 2.10 年):清髓后回输 IDUA 慢病毒转导的自体造血干祖细胞,一个月内血中 IDUA 活性即达超生理水平并持续维持,原本检测不到的脑脊液 IDUA 活性变得可测,并伴随局部糖胺聚糖清除。1968 年那个”从隔壁借来的酶”,此时由孩子自己骨髓里长出来的细胞造出;在清髓性预处理之后,它出现在了他自己的脑脊液里。

还没解决的部分

每一环都还留着钉子。酶大致沿着浓度梯度分布,离产酶细胞越远,拿到的就越少,骨骼至今是难点。在小鼠里,脑内更替 microglia 需要足够强的 conditioning;而在临床上,conditioning 恰恰是这类疗法毒性的主要来源——arsa-cel 那组的 3 级及以上不良事件多与预处理或基础疾病相关。把预处理做轻做无,与目前赖以更替脑内髓系细胞的路径存在张力(这道两难留到讲 conditioning 时再细说)。免疫也没走远:Fumagalli 那组里有四名患者出现一过性的抗 ARSA 抗体,虽未影响临床结局。更基础的一层限制在于,cross-correction 的前提是酶能被分泌、被 mannose 6-phosphate 受体捡回、并在受体细胞里保持活性,而并非每一种溶酶体蛋白都满足这三条。

回到那个培养皿。让 Hurler 细胞和 Hunter 细胞互相救活的,不是任何一方的基因被修好,而是它们之间存在一条可扩散的货币。半个多世纪以来,这个领域做的事情本质上只有一件:把这条货币的印钞机从培养基挪进人的骨髓,再把印钞量调到足够高、印钞的位置挪到足够近。

参考文献

  1. Fratantoni JC, Hall CW, Neufeld EF. Hurler and Hunter syndromes: mutual correction of the defect in cultured fibroblasts. Science. 1968;162(3853):570-2. DOI
  2. Fratantoni JC, Hall CW, Neufeld EF. The defect in Hurler and Hunter syndromes. II. Deficiency of specific factors involved in mucopolysaccharide degradation. Proc Natl Acad Sci U S A. 1969;64(1):360-6. DOI
  3. Neufeld EF. From serendipity to therapy. Annu Rev Biochem. 2011;80:1-15. DOI
  4. Bach G, Friedman R, Weissmann B, Neufeld EF. The defect in the Hurler and Scheie syndromes: deficiency of α-L-iduronidase. Proc Natl Acad Sci U S A. 1972;69(8):2048-51. DOI
  5. Kaplan A, Achord DT, Sly WS. Phosphohexosyl components of a lysosomal enzyme are recognized by pinocytosis receptors on human fibroblasts. Proc Natl Acad Sci U S A. 1977;74(5):2026-30. DOI
  6. Kaplan A, Fischer D, Achord D, Sly W. Phosphohexosyl recognition is a general characteristic of pinocytosis of lysosomal glycosidases by human fibroblasts. J Clin Invest. 1977;60(5):1088-93. DOI
  7. Natowicz MR, Chi MM, Lowry OH, Sly WS. Enzymatic identification of mannose 6-phosphate on the recognition marker for receptor-mediated pinocytosis of β-glucuronidase by human fibroblasts. Proc Natl Acad Sci U S A. 1979;76(9):4322-6. DOI
  8. Hobbs JR, Hugh-Jones K, Barrett AJ, et al. Reversal of clinical features of Hurler’s disease and biochemical improvement after treatment by bone-marrow transplantation. Lancet. 1981;2(8249):709-12. DOI
  9. Aldenhoven M, Wynn RF, Orchard PJ, et al. Long-term outcome of Hurler syndrome patients after hematopoietic cell transplantation: an international multicenter study. Blood. 2015;125(13):2164-72. DOI
  10. Kakkis ED, Muenzer J, Tiller GE, et al. Enzyme-replacement therapy in mucopolysaccharidosis I. N Engl J Med. 2001;344(3):182-8. DOI
  11. Ginhoux F, Greter M, Leboeuf M, et al. Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science. 2010;330(6005):841-5. DOI
  12. Capotondo A, Milazzo R, Politi LS, et al. Brain conditioning is instrumental for successful microglia reconstitution following hematopoietic stem cell transplantation. Proc Natl Acad Sci U S A. 2012;109(37):15018-23. DOI
  13. Biffi A, De Palma M, Quattrini A, et al. Correction of metachromatic leukodystrophy in the mouse model by transplantation of genetically modified hematopoietic stem cells. J Clin Invest. 2004;113(8):1118-29. DOI
  14. Visigalli I, Delai S, Politi LS, et al. Gene therapy augments the efficacy of hematopoietic cell transplantation and fully corrects mucopolysaccharidosis type I phenotype in the mouse model. Blood. 2010;116(24):5130-9. DOI
  15. Sergijenko A, Langford-Smith A, Liao AY, et al. Myeloid/microglial driven autologous hematopoietic stem cell gene therapy corrects a neuronopathic lysosomal disease. Mol Ther. 2013;21(10):1938-49. DOI
  16. Biffi A, Montini E, Lorioli L, et al. Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy. Science. 2013;341(6148):1233158. DOI
  17. Fumagalli F, Calbi V, Natali Sora MG, et al. Lentiviral haematopoietic stem-cell gene therapy for early-onset metachromatic leukodystrophy: long-term results from a non-randomised, open-label, phase 1/2 trial and expanded access. Lancet. 2022;399(10322):372-83. DOI
  18. Gentner B, Tucci F, Galimberti S, et al. Hematopoietic stem- and progenitor-cell gene therapy for Hurler syndrome. N Engl J Med. 2021;385(21):1929-40. DOI
  19. European Medicines Agency. Libmeldy (atidarsagene autotemcel) — EPAR. Marketing authorisation issued 17/12/2020. https://www.ema.europa.eu/en/medicines/human/EPAR/libmeldy
  20. U.S. Food and Drug Administration. LENMELDY (atidarsagene autotemcel) — approval letter, March 18, 2024. https://www.fda.gov/media/177122/download

In 1968, skin fibroblasts from two kinds of patients were grown together in a single dish. One line came from a patient with Hurler syndrome, the other from a patient with Hunter syndrome—both diseases leave cells stuffed with mucopolysaccharide they cannot degrade, and in separate culture both were plainly sick. Mixed together, both got better. The genetic defect is written into each cell’s own genome; by what right could the cell next door correct it? That accident would go on to support the entire treatment logic of a class of diseases: the enzymes inside the lysosome can be lent to the neighbors.

The corrective factor in the medium

Fratantoni, Hall and Neufeld published the observation in Science in 1968, and named the crucial point: the correction did not depend on the cells themselves but was mediated by substances released into the medium. The following year, the same group made it more concrete in PNAS—the aberrant metabolism of Hurler cells could be corrected by the secretions of fibroblasts of a genotype other than Hurler, and the same held in reverse for Hunter cells. The unknown agent was given the plain name of a corrective factor. Years later, in a retrospective in Annu Rev Biochem, Neufeld laid out how it had happened: the key experiment started as an accident—cells of different genotypes were mixed together, and the result was correction of their biochemical defect.

In 1972, Bach and colleagues solved it in PNAS: the Hurler corrective factor, purified from normal human urine, carried α-L-iduronidase activity, while cell lines lacking the factor had no detectable enzyme; and the correction of defective cells was accompanied by their efficient uptake of that enzyme from the medium. The corrective factor was simply the missing enzyme itself. Cells secrete a fraction of their newly made lysosomal enzymes, and other cells pick them up and route them into their own lysosomes—this phenomenon, later called cross-correction, is what most sets lysosomal storage disorders apart from other genetic diseases: the missing product is diffusible and shareable.

What turned the coincidence into engineering was something the Sly lab worked out in the 1970s: lysosomal enzymes carry a shipping address, and the cell surface has receptors dedicated to reading it. In 1977, Kaplan, Achord and Sly presented two lines of evidence in PNAS that the forms of human β-glucuronidase rapidly taken up by fibroblasts depend on a phosphate on, or in proximity to, a mannose-type carbohydrate; in the same year, in J Clin Invest, they suggested that this phosphohexosyl recognition is likely a general characteristic of how lysosomal glycosidases are taken up. In 1979, Natowicz and colleagues supplied direct evidence: forms of β-glucuronidase purified from human spleen differed over an 18-fold range in uptake rate, and the most acidic forms—that is, the most rapidly internalized ones—carried 4.4 mol of mannose 6-phosphate per mol of enzyme. From which follows a very practical corollary: to treat this class of disease you do not have to fix every cell; you only have to install, inside the body, a population of cells that keeps secreting the enzyme.

Putting the factory into the bone marrow

Hematopoietic stem cells are almost purpose-built for this: they can be transplanted, they engraft long-term, and they keep producing myeloid progeny that spread throughout the body—progeny that secrete lysosomal enzymes naturally. In 1981, Hobbs and colleagues reported in Lancet one of the earliest attempts—a one-year-old boy with Hurler syndrome received a bone marrow transplant from his mother, leukocyte α-L-iduronidase activity reached heterozygote levels within 37 days, hepatosplenomegaly resolved, corneal clouding cleared, and the developmental regression seemed to have been arrested.

Forty years of accumulated follow-up have also exposed the ceiling. In 2015, Aldenhoven and colleagues pooled 217 successfully engrafted Hurler patients in Blood, at a median follow-up age of 9.2 years: the majority still carried a considerable residual disease burden; and whether a normal enzyme level was achieved after transplant was a highly significant predictor of long-term outcome in most organ systems. The most enzyme an allogeneic transplant can supply is the amount a healthy donor makes; that amount is often not enough for bone and brain.

Why enzyme can reach the brain, and why that is not natural

Intravenous recombinant enzyme does work in the periphery: in the MPS I trial Kakkis and colleagues reported in N Engl J Med in 2001, 10 patients received weekly infusions of recombinant α-L-iduronidase for 52 weeks, hepatosplenomegaly shrank in all of them, and urinary glycosaminoglycan fell by an average of 63%; but the benefits that trial reported were concentrated in the periphery, and no improvement in the central nervous system was reported. And for enzyme to persist inside the brain, the route that was eventually made to work was to get the enzyme-producing cells themselves into the brain.

That does indeed happen after transplant, but it is not natural. In 2010, Ginhoux and colleagues showed by fate mapping in Science that microglia in the adult mouse brain derive from primitive myeloid progenitors that appear before embryonic day 8, and that postnatal hematopoietic progenitors do not significantly contribute to microglia homeostasis in the adult brain—under normal conditions, blood and brain are two independent myeloid systems. In 2012, Capotondo and colleagues took the post-transplant process apart in PNAS in mice: a fraction of transplanted hematopoietic progenitors does briefly infiltrate the brain, and this wave is independent of whether a preparative regimen was given and of whether there is disease in the brain; but only when the conditioning regimen was capable of ablating the brain-resident myeloid precursors did donor cells actually achieve turnover of microglia—and that turnover was mediated by local proliferation of the early immigrants.

In mice, this also gives conditioning a second role: myeloablation is not only about making room in the marrow, it also determines whether the brain’s resident myeloid precursors give way.

Turning up the output

If the problem with allogeneic transplant is that there is not enough enzyme, then the most direct advantage of gene therapy is not correcting the cause of disease but supraphysiological expression. In 2004, Biffi and colleagues used a lentiviral vector in J Clin Invest to transfer the ARSA gene into mouse hematopoietic stem cells; after transplant, transgene-expressing cells extensively repopulated CNS microglia and peripheral-nerve endoneurial macrophages, and prevented the pathological and behavioral abnormalities of a metachromatic leukodystrophy model. The key point is that the therapeutic impact was significantly higher than that of wild-type hematopoietic stem cell transplantation, a difference the authors took as indicating a critical role for enzyme overexpression. In 2010, Visigalli and colleagues put it more starkly in Blood in MPS I mice: efficacy was strictly dependent on achieving supranormal enzyme activity in the hematopoietic system, and it was that excess which allowed enzyme delivery to the brain and skeleton. Beyond output there is also the site of production—in 2013, Sergijenko and colleagues found in Mol Ther that in MPS IIIA mice the myeloid-specific CD11b promoter gave bone marrow SGSH activity similar to that of the ubiquitous PGK (473% versus 576% of normal activity), but significantly higher expression in the brain (11% versus 7%).

To the bedside

The human data landed in 2013. Biffi and colleagues reported in Science on three presymptomatic children with late-infantile MLD; after reinfusion of lentivirally transduced autologous hematopoietic stem cells, high ARSA levels appeared across hematopoietic lineages and in the cerebrospinal fluid, and at time points 7 to 21 months beyond the predicted age of disease onset, the disease had neither appeared nor progressed. In 2022, Fumagalli and colleagues gave an integrated analysis in Lancet: of 29 children with early-onset MLD who received arsa-cel, 26 were alive at the time of analysis, with a median follow-up of 3.16 years; compared with age- and subtype-matched natural history controls, the mean difference in total gross motor function score was 66% in the late-infantile group and 42% in the early-juvenile group, with the benefit especially clear in those treated presymptomatically (that window deserves its own piece, and is left to the MLD entry). The therapy is atidarsagene autotemcel, approved in the European Union as Libmeldy on 17 December 2020 and in the United States as Lenmeldy on 18 March 2024.

The Hurler side closed its loop too. In 2021, Gentner and colleagues reported in N Engl J Med on 8 children with MPS IH (mean age at treatment 1.9±0.5 years, median follow-up 2.10 years): after myeloablation and reinfusion of IDUA lentivirally transduced autologous hematopoietic stem and progenitor cells, blood IDUA activity reached supraphysiological levels within a month and was sustained, and cerebrospinal fluid IDUA activity, previously undetectable, became measurable, accompanied by local glycosaminoglycan clearance. The enzyme borrowed from next door in 1968 was now made by cells grown out of the child’s own bone marrow; after myeloablative conditioning, it turned up in his own cerebrospinal fluid.

What is still unsolved

Every link still has a nail sticking out. Enzyme distributes roughly along a concentration gradient—the farther from the producing cell, the less there is—and bone remains a hard problem to this day. In mice, replacing microglia in the brain requires conditioning strong enough to do it; and in the clinic, conditioning is precisely the main source of toxicity in this class of therapy—in the arsa-cel cohort, the grade 3 or higher adverse events were mostly related to conditioning or to background disease. Making the preparative regimen lighter, or removing it altogether, is in tension with the route currently relied on to replace myeloid cells in the brain (that dilemma is left for the piece on conditioning). Immunology has not gone away either: four patients in the Fumagalli cohort developed transient anti-ARSA antibodies, though without effect on clinical outcome. A more fundamental limit is that cross-correction presupposes that the enzyme can be secreted, retrieved by the mannose 6-phosphate receptor, and remain active inside the recipient cell—and not every lysosomal protein satisfies all three conditions.

Back to that dish. What let Hurler cells and Hunter cells rescue each other was not that either side’s gene was repaired, but that there existed between them a diffusible currency. For more than half a century, what this field has done amounts essentially to one thing: moving the printing press for that currency out of the culture medium and into human bone marrow, then turning the print run high enough and moving the press close enough.

References

  1. Fratantoni JC, Hall CW, Neufeld EF. Hurler and Hunter syndromes: mutual correction of the defect in cultured fibroblasts. Science. 1968;162(3853):570-2. DOI
  2. Fratantoni JC, Hall CW, Neufeld EF. The defect in Hurler and Hunter syndromes. II. Deficiency of specific factors involved in mucopolysaccharide degradation. Proc Natl Acad Sci U S A. 1969;64(1):360-6. DOI
  3. Neufeld EF. From serendipity to therapy. Annu Rev Biochem. 2011;80:1-15. DOI
  4. Bach G, Friedman R, Weissmann B, Neufeld EF. The defect in the Hurler and Scheie syndromes: deficiency of α-L-iduronidase. Proc Natl Acad Sci U S A. 1972;69(8):2048-51. DOI
  5. Kaplan A, Achord DT, Sly WS. Phosphohexosyl components of a lysosomal enzyme are recognized by pinocytosis receptors on human fibroblasts. Proc Natl Acad Sci U S A. 1977;74(5):2026-30. DOI
  6. Kaplan A, Fischer D, Achord D, Sly W. Phosphohexosyl recognition is a general characteristic of pinocytosis of lysosomal glycosidases by human fibroblasts. J Clin Invest. 1977;60(5):1088-93. DOI
  7. Natowicz MR, Chi MM, Lowry OH, Sly WS. Enzymatic identification of mannose 6-phosphate on the recognition marker for receptor-mediated pinocytosis of β-glucuronidase by human fibroblasts. Proc Natl Acad Sci U S A. 1979;76(9):4322-6. DOI
  8. Hobbs JR, Hugh-Jones K, Barrett AJ, et al. Reversal of clinical features of Hurler’s disease and biochemical improvement after treatment by bone-marrow transplantation. Lancet. 1981;2(8249):709-12. DOI
  9. Aldenhoven M, Wynn RF, Orchard PJ, et al. Long-term outcome of Hurler syndrome patients after hematopoietic cell transplantation: an international multicenter study. Blood. 2015;125(13):2164-72. DOI
  10. Kakkis ED, Muenzer J, Tiller GE, et al. Enzyme-replacement therapy in mucopolysaccharidosis I. N Engl J Med. 2001;344(3):182-8. DOI
  11. Ginhoux F, Greter M, Leboeuf M, et al. Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science. 2010;330(6005):841-5. DOI
  12. Capotondo A, Milazzo R, Politi LS, et al. Brain conditioning is instrumental for successful microglia reconstitution following hematopoietic stem cell transplantation. Proc Natl Acad Sci U S A. 2012;109(37):15018-23. DOI
  13. Biffi A, De Palma M, Quattrini A, et al. Correction of metachromatic leukodystrophy in the mouse model by transplantation of genetically modified hematopoietic stem cells. J Clin Invest. 2004;113(8):1118-29. DOI
  14. Visigalli I, Delai S, Politi LS, et al. Gene therapy augments the efficacy of hematopoietic cell transplantation and fully corrects mucopolysaccharidosis type I phenotype in the mouse model. Blood. 2010;116(24):5130-9. DOI
  15. Sergijenko A, Langford-Smith A, Liao AY, et al. Myeloid/microglial driven autologous hematopoietic stem cell gene therapy corrects a neuronopathic lysosomal disease. Mol Ther. 2013;21(10):1938-49. DOI
  16. Biffi A, Montini E, Lorioli L, et al. Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy. Science. 2013;341(6148):1233158. DOI
  17. Fumagalli F, Calbi V, Natali Sora MG, et al. Lentiviral haematopoietic stem-cell gene therapy for early-onset metachromatic leukodystrophy: long-term results from a non-randomised, open-label, phase 1/2 trial and expanded access. Lancet. 2022;399(10322):372-83. DOI
  18. Gentner B, Tucci F, Galimberti S, et al. Hematopoietic stem- and progenitor-cell gene therapy for Hurler syndrome. N Engl J Med. 2021;385(21):1929-40. DOI
  19. European Medicines Agency. Libmeldy (atidarsagene autotemcel) — EPAR. Marketing authorisation issued 17/12/2020. https://www.ema.europa.eu/en/medicines/human/EPAR/libmeldy
  20. U.S. Food and Drug Administration. LENMELDY (atidarsagene autotemcel) — approval letter, March 18, 2024. https://www.fda.gov/media/177122/download